diff --git a/.vscode/c_cpp_properties.json b/.vscode/c_cpp_properties.json
index 4aad63c52..2544b3cf8 100644
--- a/.vscode/c_cpp_properties.json
+++ b/.vscode/c_cpp_properties.json
@@ -52,7 +52,6 @@
 				"relegated=",
 				"hidden=",
 				"textstartup=",
-				"initarray=",
 				"returnsnonnull=",
 				"returnspointerwithnoaliases=",
 				"printfesque(x)=",
diff --git a/Makefile b/Makefile
index 0f6122670..1186b7674 100644
--- a/Makefile
+++ b/Makefile
@@ -133,10 +133,10 @@ endif
 
 ifneq ($(findstring aarch64,$(MODE)),)
 ARCH = aarch64
-HOSTS ?= pi pi5 studio freebsdarm
+HOSTS ?= pi studio freebsdarm
 else
 ARCH = x86_64
-HOSTS ?= freebsd rhel7 xnu win10 openbsd netbsd meatball nightmare
+HOSTS ?= freebsd rhel7 xnu openbsd netbsd win10
 endif
 
 ZIPOBJ_FLAGS += -a$(ARCH)
@@ -149,9 +149,9 @@ export MODE
 export SOURCE_DATE_EPOCH
 export TMPDIR
 
-COSMOCC = .cosmocc/3.2
+COSMOCC = .cosmocc/3.3
 TOOLCHAIN = $(COSMOCC)/bin/$(ARCH)-linux-cosmo-
-DOWNLOAD := $(shell build/download-cosmocc.sh $(COSMOCC) 3.2 28b48682595f0f46b45ab381118cdffdabc8fcfa29aa54e301fe6ffe35269f5e)
+DOWNLOAD := $(shell build/download-cosmocc.sh $(COSMOCC) 3.3 d26ec8f4e48f6db004fc6a9677c7ff3b50c3b21e936e9393158aa2ed51b0b549)
 
 AS = $(TOOLCHAIN)as
 CC = $(TOOLCHAIN)gcc
@@ -206,8 +206,7 @@ endif
 .UNVEIL +=					\
 	libc/integral				\
 	libc/stdbool.h				\
-	libc/disclaimer.inc			\
-	rwc:/dev/shm				\
+		rwc:/dev/shm				\
 	rx:.cosmocc				\
 	rx:build/bootstrap			\
 	r:build/portcosmo.h			\
diff --git a/ape/BUILD.mk b/ape/BUILD.mk
index 106504d02..afddd13b6 100644
--- a/ape/BUILD.mk
+++ b/ape/BUILD.mk
@@ -157,7 +157,6 @@ o/$(MODE)/ape/ape-no-modify-self.o:		\
 		ape/ape.S			\
 		ape/ape.h			\
 		ape/macros.internal.h		\
-		ape/notice.inc			\
 		ape/relocations.h		\
 		ape/ape.internal.h		\
 		libc/dce.h			\
@@ -185,7 +184,6 @@ o/$(MODE)/ape/ape-copy-self.o:			\
 		ape/ape.S			\
 		ape/ape.h			\
 		ape/macros.internal.h		\
-		ape/notice.inc			\
 		ape/relocations.h		\
 		ape/ape.internal.h		\
 		libc/dce.h			\
diff --git a/ape/aarch64.lds b/ape/aarch64.lds
index 9866ac748..cec91ae89 100644
--- a/ape/aarch64.lds
+++ b/ape/aarch64.lds
@@ -89,10 +89,12 @@ SECTIONS {
     *(.ubsan.data)
   }
 
-  .comment : {
-    __comment_start = .;
-    KEEP(*(.comment))
+  .notice : {
+    __notices = .;
+    KEEP(*(.notice))
     BYTE(0);
+    BYTE(10);
+    BYTE(10);
   }
 
   .eh_frame_hdr : {
@@ -157,8 +159,11 @@ SECTIONS {
 
   .init_array : {
     __init_array_start = .;
-    KEEP(*(SORT_BY_INIT_PRIORITY(.init_array.*) SORT_BY_INIT_PRIORITY(.ctors.*)))
-    KEEP(*(.init_array EXCLUDE_FILE(*crtbegin.o *crtbegin?.o *crtend.o *crtend?.o ) .ctors))
+    KEEP(*(.preinit_array))
+    KEEP(*(SORT_BY_INIT_PRIORITY(.init_array.*)
+           SORT_BY_INIT_PRIORITY(.ctors.*)))
+    KEEP(*(.init_array))
+    KEEP(*(.ctors))
     __init_array_end = .;
   }
 
diff --git a/ape/ape.S b/ape/ape.S
index 80e788570..eaf0bfaab 100644
--- a/ape/ape.S
+++ b/ape/ape.S
@@ -33,7 +33,6 @@
 │ αcτµαlly pδrταblε εxεcµταblε § program header                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "ape/macros.internal.h"
-#include "ape/notice.inc"
 #include "ape/relocations.h"
 #include "libc/calls/metalfile.internal.h"
 #include "libc/dce.h"
@@ -1772,49 +1771,31 @@ kernel:	movabs	$ape_stack_vaddr,%rsp
 	.type	ape_text_nops,@object
 	.type	__test_end,@object
 
-	.section .commentprologue,"a",@progbits
-	.globl	__comment_start
-	.type	__comment_start,@object
-	.hidden	__comment_start
-__comment_start:/*
-	...
-	decentralized content
-	...
-	*/.previous
-	.section .commentepilogue,"a",@progbits
-	.byte	0
-	.previous
-
 	.section .ape.pad.head,"a",@progbits
 	.type	ape_pad_head,@object
 	.hidden	ape_pad_head
 ape_pad_head:
-	.previous
 
 	.section .ape.pad.text,"a",@progbits
 	.type	ape_pad_text,@object
 	.hidden	ape_pad_text
 ape_pad_text:
-	.previous
 
 	.section .ape.pad.privileged,"a",@progbits
 	.type	ape_pad_privileged,@object
 	.hidden	ape_pad_privileged
 ape_pad_privileged:
-	.previous
 
 	.section .ape.pad.data,"a",@progbits
 	.type	ape_pad_data,@object
 	.hidden	ape_pad_data
 ape_pad_data:
-	.previous
 
 #if SupportsWindows()
 	.section .idata.ro,"a",@progbits
 	.type	ape_idata_ro,@object
 	.hidden	ape_idata_ro
 ape_idata_ro:
-	.previous
 #endif /* SupportsWindows() */
 
 	.section .dataprologue,"aw",@progbits
@@ -1822,32 +1803,45 @@ ape_idata_ro:
 	.globl	__data_start
 	.hidden	__data_start
 __data_start:
-	.previous
 
 	.section .dataepilogue,"aw",@progbits
 	.type	__data_end,@object
 	.globl	__data_end
 	.hidden	__data_end
 __data_end:
-	.previous
 
 	.section .bssprologue,"aw",@nobits
 	.type	__bss_start,@object
 	.globl	__bss_start
 	.hidden	__bss_start
 __bss_start:
-	.previous
 
 	.section .bssepilogue,"aw",@nobits
 	.type	__bss_end,@object
 	.globl	__bss_end
 	.hidden	__bss_end
 __bss_end:
-	.previous
 
 	.section .fstls,"awT",@nobits
 	.align	TLS_ALIGNMENT
-	.previous
+
+	.section .notice,"aR",@progbits
+	.asciz	"\n\n\
+Cosmopolitan\n\
+Copyright 2024 Justine Alexandra Roberts Tunney\n\
+\n\
+Permission to use, copy, modify, and/or distribute this software for\n\
+any purpose with or without fee is hereby granted, provided that the\n\
+above copyright notice and this permission notice appear in all copies.\n\
+\n\
+THE SOFTWARE IS PROVIDED \"AS IS\" AND THE AUTHOR DISCLAIMS ALL\n\
+WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED\n\
+WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE\n\
+AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL\n\
+DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR\n\
+PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER\n\
+TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR\n\
+PERFORMANCE OF THIS SOFTWARE."
 
 .end
 
\ No newline at end of file
diff --git a/ape/ape.lds b/ape/ape.lds
index 1a39e2dc1..d0a27da6b 100644
--- a/ape/ape.lds
+++ b/ape/ape.lds
@@ -330,11 +330,11 @@ SECTIONS {
     *(.ubsan.data)
 
     /* Legal Notices */
-#if !defined(IM_FEELING_NAUGHTY) || defined(EMBED_NOTICES)
-    KEEP(*(.commentprologue))
-    KEEP(*(.comment))
-    KEEP(*(.commentepilogue))
-#endif
+    __notices = .;
+    KEEP(*(.notice))
+    BYTE(0);
+    BYTE(10);
+    BYTE(10);
 
 /*BEGIN: read-only data that's only needed for initialization */
 
@@ -390,6 +390,9 @@ SECTIONS {
 /*BEGIN: NT FORK COPYING */
     KEEP(*(.dataprologue))
     *(.data .data.*)
+    *(.gnu_extab)
+    *(.gcc_except_table .gcc_except_table.*)
+    *(.exception_ranges*)
     *(.PyRuntime) /* for python */
     *(.subrs) /* for emacs */
     KEEP(*(SORT_BY_NAME(.sort.data.*)))
@@ -404,11 +407,11 @@ SECTIONS {
 
     . = ALIGN(. != 0 ? __SIZEOF_POINTER__ : 0);
     __init_array_start = .;
+    KEEP(*(.preinit_array))
     KEEP(*(SORT_BY_INIT_PRIORITY(.init_array.*)
            SORT_BY_INIT_PRIORITY(.ctors.*)))
-    KEEP(*(.ctors))
     KEEP(*(.init_array))
-    KEEP(*(.preinit_array))
+    KEEP(*(.ctors))
     __init_array_end = .;
 
     . = ALIGN(. != 0 ? __SIZEOF_POINTER__ : 0);
@@ -425,8 +428,8 @@ SECTIONS {
     . = ALIGN(. != 0 ? __SIZEOF_POINTER__ : 0);
     KEEP(*(SORT_BY_NAME(.piro.data.sort.*)))
     KEEP(*(.piro.pad.data))
-    KEEP(*(.dataepilogue))
     *(.igot.plt)
+    KEEP(*(.dataepilogue))
 
     . = ALIGN(. != 0 ? CONSTANT(COMMONPAGESIZE) : 0);
 /*END: NT FORK COPYING */
diff --git a/ape/sections.internal.h b/ape/sections.internal.h
index e99503b7e..6bc8cc312 100644
--- a/ape/sections.internal.h
+++ b/ape/sections.internal.h
@@ -2,7 +2,7 @@
 #define COSMOPOLITAN_APE_SECTIONS_INTERNAL_H_
 COSMOPOLITAN_C_START_
 
-extern const char __comment_start[] __attribute__((__weak__));
+extern const char __notices[] __attribute__((__weak__));
 extern unsigned char __executable_start[] __attribute__((__weak__));
 extern unsigned char __privileged_start[] __attribute__((__weak__));
 extern unsigned char _ehead[] __attribute__((__weak__));
diff --git a/build/bootstrap/ape.aarch64 b/build/bootstrap/ape.aarch64
index 65fe2cc16..27fd3d7fb 100755
Binary files a/build/bootstrap/ape.aarch64 and b/build/bootstrap/ape.aarch64 differ
diff --git a/build/bootstrap/ape.elf b/build/bootstrap/ape.elf
index f56c3ef8e..6a87fd728 100755
Binary files a/build/bootstrap/ape.elf and b/build/bootstrap/ape.elf differ
diff --git a/build/bootstrap/ape.macho b/build/bootstrap/ape.macho
index 2b887117c..42b153c6a 100755
Binary files a/build/bootstrap/ape.macho and b/build/bootstrap/ape.macho differ
diff --git a/build/bootstrap/fixupobj.com b/build/bootstrap/fixupobj.com
index fefcebd01..fb2db3588 100755
Binary files a/build/bootstrap/fixupobj.com and b/build/bootstrap/fixupobj.com differ
diff --git a/build/definitions.mk b/build/definitions.mk
index 6950d5667..9c3b3d6e6 100644
--- a/build/definitions.mk
+++ b/build/definitions.mk
@@ -138,6 +138,7 @@ MATHEMATICAL =								\
 DEFAULT_CPPFLAGS +=							\
 	-D_COSMO_SOURCE							\
 	-DMODE='"$(MODE)"'						\
+	-Wno-prio-ctor-dtor						\
 	-Wno-unknown-pragmas						\
 	-nostdinc							\
 	-iquote.							\
@@ -164,6 +165,7 @@ DEFAULT_LDFLAGS =							\
 	-nostdlib							\
 	-znorelro							\
 	--gc-sections							\
+	-z noexecstack							\
 	--build-id=none							\
 	--no-dynamic-linker
 
diff --git a/build/objdump b/build/objdump
index 32a4e218a..3a022d3aa 100755
--- a/build/objdump
+++ b/build/objdump
@@ -6,14 +6,14 @@ if [ -n "$OBJDUMP" ]; then
 fi
 
 find_objdump() {
-  if [ -x .cosmocc/3.2/bin/$1-linux-cosmo-objdump ]; then
-    OBJDUMP=.cosmocc/3.2/bin/$1-linux-cosmo-objdump
-  elif [ -x .cosmocc/3.2/bin/$1-linux-musl-objdump ]; then
-    OBJDUMP=.cosmocc/3.2/bin/$1-linux-musl-objdump
-  elif [ -x "$COSMO/.cosmocc/3.2/bin/$1-linux-cosmo-objdump" ]; then
-    OBJDUMP="$COSMO/.cosmocc/3.2/bin/$1-linux-cosmo-objdump"
-  elif [ -x "$COSMO/.cosmocc/3.2/bin/$1-linux-musl-objdump" ]; then
-    OBJDUMP="$COSMO/.cosmocc/3.2/bin/$1-linux-musl-objdump"
+  if [ -x .cosmocc/3.3/bin/$1-linux-cosmo-objdump ]; then
+    OBJDUMP=.cosmocc/3.3/bin/$1-linux-cosmo-objdump
+  elif [ -x .cosmocc/3.3/bin/$1-linux-musl-objdump ]; then
+    OBJDUMP=.cosmocc/3.3/bin/$1-linux-musl-objdump
+  elif [ -x "$COSMO/.cosmocc/3.3/bin/$1-linux-cosmo-objdump" ]; then
+    OBJDUMP="$COSMO/.cosmocc/3.3/bin/$1-linux-cosmo-objdump"
+  elif [ -x "$COSMO/.cosmocc/3.3/bin/$1-linux-musl-objdump" ]; then
+    OBJDUMP="$COSMO/.cosmocc/3.3/bin/$1-linux-musl-objdump"
   else
     echo "error: toolchain not found (try running 'cosmocc --update' or 'make' in the cosmo monorepo)" >&2
     exit 1
diff --git a/dsp/mpeg/buffer.c b/dsp/mpeg/buffer.c
index 76ed2be36..f0cb0de38 100644
--- a/dsp/mpeg/buffer.c
+++ b/dsp/mpeg/buffer.c
@@ -35,12 +35,7 @@
 #include "libc/stdio/stdio.h"
 #include "libc/str/str.h"
 #include "libc/sysv/consts/madv.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 /* clang-format off */
 // -----------------------------------------------------------------------------
diff --git a/dsp/mpeg/demux.c b/dsp/mpeg/demux.c
index 4ded810e5..66eff844a 100644
--- a/dsp/mpeg/demux.c
+++ b/dsp/mpeg/demux.c
@@ -27,17 +27,12 @@
 │  CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE            │
 │  SOFTWARE.                                                                   │
 ╚─────────────────────────────────────────────────────────────────────────────*/
-#include "dsp/mpeg/buffer.h"
 #include "dsp/mpeg/demux.h"
+#include "dsp/mpeg/buffer.h"
 #include "dsp/mpeg/mpeg.h"
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 /* clang-format off */
 // ----------------------------------------------------------------------------
diff --git a/dsp/mpeg/idct.c b/dsp/mpeg/idct.c
index 87c17ae6a..11312607e 100644
--- a/dsp/mpeg/idct.c
+++ b/dsp/mpeg/idct.c
@@ -28,12 +28,7 @@
 │  SOFTWARE.                                                                   │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "dsp/core/half.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 /**
  * Computes Fixed-Point 8x8 Inverse Discrete Cosine Transform.
diff --git a/dsp/mpeg/mp2.c b/dsp/mpeg/mp2.c
index 88a5a0c10..53fc91a23 100644
--- a/dsp/mpeg/mp2.c
+++ b/dsp/mpeg/mp2.c
@@ -33,12 +33,6 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /* clang-format off */
 // -----------------------------------------------------------------------------
 // plm_audio implementation
diff --git a/dsp/mpeg/mpeg1.c b/dsp/mpeg/mpeg1.c
index f9c29910b..905af23da 100644
--- a/dsp/mpeg/mpeg1.c
+++ b/dsp/mpeg/mpeg1.c
@@ -41,12 +41,7 @@
 #include "libc/str/str.h"
 #include "libc/time/time.h"
 #include "libc/x/x.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 // -----------------------------------------------------------------------------
 // plm_video implementation
@@ -1104,7 +1099,7 @@ plm_video_t *plm_video_create_with_buffer(plm_buffer_t *buffer,
   return self;
 }
 
-static textstartup void plm_video_init(void) {
+__attribute__((__constructor__)) static textstartup void plm_video_init(void) {
   PLM_VIDEO_MACROBLOCK_TYPE[0] = NULL;
   PLM_VIDEO_MACROBLOCK_TYPE[1] = (void *)PLM_VIDEO_MACROBLOCK_TYPE_INTRA;
   PLM_VIDEO_MACROBLOCK_TYPE[2] = (void *)PLM_VIDEO_MACROBLOCK_TYPE_PREDICTIVE;
@@ -1113,5 +1108,3 @@ static textstartup void plm_video_init(void) {
   PLM_VIDEO_DCT_SIZE[1] = (void *)PLM_VIDEO_DCT_SIZE_CHROMINANCE;
   PLM_VIDEO_DCT_SIZE[2] = (void *)PLM_VIDEO_DCT_SIZE_CHROMINANCE;
 }
-
-const void *const plm_video_init_ctor[] initarray = {plm_video_init};
diff --git a/dsp/mpeg/notice.c b/dsp/mpeg/notice.c
new file mode 100644
index 000000000..264a7549b
--- /dev/null
+++ b/dsp/mpeg/notice.c
@@ -0,0 +1,4 @@
+__notice(pl_mpeg_notice, "\
+PL_MPEG (MIT License)\n\
+Copyright(c) 2019 Dominic Szablewski\n\
+https://phoboslab.org");
diff --git a/dsp/mpeg/plm.c b/dsp/mpeg/plm.c
index f4f1dc504..7704643ff 100644
--- a/dsp/mpeg/plm.c
+++ b/dsp/mpeg/plm.c
@@ -32,12 +32,7 @@
 #include "libc/mem/mem.h"
 #include "libc/stdio/stdio.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 /* clang-format off */
 // -----------------------------------------------------------------------------
diff --git a/dsp/mpeg/slowrgb.c b/dsp/mpeg/slowrgb.c
index 16c819c38..7472d82f3 100644
--- a/dsp/mpeg/slowrgb.c
+++ b/dsp/mpeg/slowrgb.c
@@ -29,12 +29,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "dsp/mpeg/mpeg.h"
 #include "libc/macros.internal.h"
-
-asm(".ident\t\"\\n\\n\
-PL_MPEG (MIT License)\\n\
-Copyright(c) 2019 Dominic Szablewski\\n\
-https://phoboslab.org\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("pl_mpeg_notice");
 
 /**
  * @see YCbCr2RGB() in tool/viz/lib/ycbcr2rgb.c
diff --git a/dsp/scale/magikarp.c b/dsp/scale/magikarp.c
index d8f83dd80..ea5aa7d55 100644
--- a/dsp/scale/magikarp.c
+++ b/dsp/scale/magikarp.c
@@ -121,8 +121,7 @@ void *MagikarpY(long dys, long dxs, unsigned char d[restrict dys][dxs],
   return d;
 }
 
-static textstartup void g_magikarp_init() {
+__attribute__((__constructor__)) static textstartup void g_magikarp_init() {
   memcpy(g_magkern, kMagkern[0], sizeof(g_magkern));
   memcpy(g_magikarp, kMagikarp[0], sizeof(g_magikarp));
 }
-const void *const g_magikarp_ctor[] initarray = {g_magikarp_init};
diff --git a/dsp/tty/itoa8.c b/dsp/tty/itoa8.c
index bf968728e..57c903b69 100644
--- a/dsp/tty/itoa8.c
+++ b/dsp/tty/itoa8.c
@@ -21,7 +21,7 @@
 
 struct Itoa8 kItoa8;
 
-static textstartup void itoa8_init(void) {
+__attribute__((__constructor__)) static textstartup void itoa8_init(void) {
   int i;
   uint8_t z;
   uint32_t w;
@@ -41,5 +41,3 @@ static textstartup void itoa8_init(void) {
     kItoa8.data[i] = w;
   }
 }
-
-const void *const itoa8_init_ctor[] initarray = {itoa8_init};
diff --git a/dsp/tty/rgb2ansi.c b/dsp/tty/rgb2ansi.c
index 053aa4ebb..baece9e8a 100644
--- a/dsp/tty/rgb2ansi.c
+++ b/dsp/tty/rgb2ansi.c
@@ -97,7 +97,7 @@ static int uncube(int x) {
   return x < 48 ? 0 : x < 115 ? 1 : (x - 35) / 40;
 }
 
-static textstartup void rgb2ansi_init(void) {
+__attribute__((__constructor__)) static textstartup void rgb2ansi_init(void) {
   uint8_t c;
   uint32_t i;
   memcpy(g_ansi2rgb_, &kCgaPalette, sizeof(kCgaPalette));
@@ -114,5 +114,3 @@ static textstartup void rgb2ansi_init(void) {
     g_ansi2rgb_[i].xt = i;
   }
 }
-
-const void *const rgb2ansi_init_ctor[] initarray = {rgb2ansi_init};
diff --git a/dsp/tty/ttyquant.c b/dsp/tty/ttyquant.c
index a0cee959e..0e9a0e16e 100644
--- a/dsp/tty/ttyquant.c
+++ b/dsp/tty/ttyquant.c
@@ -73,8 +73,6 @@ textstartup void ttyquantsetup(enum TtyQuantizationAlgorithm alg,
   TTYQUANT()->blocks = blocks;
 }
 
-textstartup void ttyquant_init(void) {
+__attribute__((__constructor__)) textstartup void ttyquant_init(void) {
   ttyquantsetup(kTtyQuantXterm256, kTtyQuantRgb, kTtyBlocksUnicode);
 }
-
-const void *const ttyquant_init_ctor[] initarray = {ttyquant_init};
diff --git a/examples/kilo.c b/examples/kilo.c
index 7cc1ddefb..9645634e1 100644
--- a/examples/kilo.c
+++ b/examples/kilo.c
@@ -33,11 +33,10 @@
 │  OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.        │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 
-asm(".ident \"\n\
+__notice(kilo_notice, "\
 Kilo ─ A very simple editor (BSD-2)\n\
 Copyright 2016 Salvatore Sanfilippo\n\
-Contact: antirez@gmail.com\"\n\
-.include \"libc/disclaimer.inc\"");
+Contact: antirez@gmail.com");
 
 /*
  * This software has been modified by Justine Tunney to:
diff --git a/examples/script.c b/examples/script.c
index 3ea419caa..e840aca8a 100644
--- a/examples/script.c
+++ b/examples/script.c
@@ -67,13 +67,10 @@
  * @see https://asciinema.org/
  */
 
-asm(".ident\t\"\\n\\n\
-FreeBSD Script (BSD-3 License)\\n\
-Copyright (c) 2010, 2012 David E. O'Brien\\n\
-Copyright (c) 1980, 1992, 1993\\n\
-\tThe Regents of the University of California.\\n\
-\tAll rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(freebsd_script_notice, "\
+FreeBSD Script (BSD-3 License)\n\
+Copyright (c) 2010, 2012 David E. O'Brien\n\
+Copyright (c) 1980, 1992, 1993 The Regents of the University of California");
 
 #define DEF_BUF 65536
 
diff --git a/examples/whois.c b/examples/whois.c
index bae2bc359..95154787f 100644
--- a/examples/whois.c
+++ b/examples/whois.c
@@ -48,12 +48,9 @@
 #include "third_party/musl/netdb.h"
 // clang-format off
 
-asm(".ident\t\"\\n\\n\
-FreeBSD Whois (BSD-3 License)\\n\
-Copyright (c) 1980, 1993\\n\
-\tThe Regents of the University of California.\\n\
-\tAll rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(freebsd_whois_notice, "\
+FreeBSD Whois (BSD-3 License)\n\
+Copyright (c) 1980, 1993 The Regents of the University of California");
 
 #define	ABUSEHOST	"whois.abuse.net"
 #define	ANICHOST	"whois.arin.net"
diff --git a/libc/calls/BUILD.mk b/libc/calls/BUILD.mk
index 7f1611f9a..a12d7f3e6 100644
--- a/libc/calls/BUILD.mk
+++ b/libc/calls/BUILD.mk
@@ -73,12 +73,10 @@ $(LIBC_CALLS_A_OBJS): private				\
 			-Wframe-larger-than=4096	\
 			-Walloca-larger-than=4096
 
-ifneq ($(ARCH), aarch64)
 # we always want -O3 because:
 #   it makes the code size smaller too
-# we need -mstringop-strategy=loop because:
-#   privileged code might generate memcpy call
 o/$(MODE)/libc/calls/termios2host.o			\
+o/$(MODE)/libc/calls/siginfo2cosmo.o			\
 o/$(MODE)/libc/calls/sigenter-freebsd.o			\
 o/$(MODE)/libc/calls/sigenter-netbsd.o			\
 o/$(MODE)/libc/calls/sigenter-openbsd.o			\
@@ -86,6 +84,19 @@ o/$(MODE)/libc/calls/sigenter-xnu.o			\
 o/$(MODE)/libc/calls/ntcontext2linux.o: private		\
 		COPTS +=				\
 			-O3				\
+			-ffreestanding
+
+ifeq ($(ARCH), x86_64)
+# we need -mstringop-strategy=loop because:
+#   privileged code might generate memcpy call
+o/$(MODE)/libc/calls/termios2host.o			\
+o/$(MODE)/libc/calls/siginfo2cosmo.o			\
+o/$(MODE)/libc/calls/sigenter-freebsd.o			\
+o/$(MODE)/libc/calls/sigenter-netbsd.o			\
+o/$(MODE)/libc/calls/sigenter-openbsd.o			\
+o/$(MODE)/libc/calls/sigenter-xnu.o			\
+o/$(MODE)/libc/calls/ntcontext2linux.o: private		\
+		COPTS +=				\
 			-mstringop-strategy=loop
 endif
 
diff --git a/libc/calls/clock_gettime-nt.c b/libc/calls/clock_gettime-nt.c
index 8ebef10b6..5a6464e42 100644
--- a/libc/calls/clock_gettime-nt.c
+++ b/libc/calls/clock_gettime-nt.c
@@ -87,13 +87,9 @@ textwindows int sys_clock_gettime_nt(int clock, struct timespec *ts) {
   }
 }
 
-static textstartup void winclock_init() {
+__attribute__((__constructor__(40))) static textstartup void winclock_init() {
   if (IsWindows()) {
     QueryPerformanceCounter(&g_winclock.base);
     QueryPerformanceFrequency(&g_winclock.freq);
   }
 }
-
-const void *const winclock_ctor[] initarray = {
-    winclock_init,
-};
diff --git a/libc/calls/fchmodat-linux.c b/libc/calls/fchmodat-linux.c
index 3036be97f..1640eb636 100644
--- a/libc/calls/fchmodat-linux.c
+++ b/libc/calls/fchmodat-linux.c
@@ -32,11 +32,8 @@
 #include "libc/sysv/consts/o.h"
 #include "libc/sysv/consts/s.h"
 #include "libc/sysv/errfuns.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 static void __procfdname(char *buf, unsigned fd) {
diff --git a/libc/calls/finddebugbinary.c b/libc/calls/finddebugbinary.c
index b4bae854e..3ca5bd1c0 100644
--- a/libc/calls/finddebugbinary.c
+++ b/libc/calls/finddebugbinary.c
@@ -27,6 +27,7 @@
 #include "libc/elf/tinyelf.internal.h"
 #include "libc/errno.h"
 #include "libc/intrin/directmap.internal.h"
+#include "libc/intrin/kprintf.h"
 #include "libc/nt/memory.h"
 #include "libc/nt/runtime.h"
 #include "libc/runtime/runtime.h"
diff --git a/libc/calls/getloadavg-nt.c b/libc/calls/getloadavg-nt.c
index 30f940de0..4e8d6d847 100644
--- a/libc/calls/getloadavg-nt.c
+++ b/libc/calls/getloadavg-nt.c
@@ -59,7 +59,7 @@ textwindows int sys_getloadavg_nt(double *a, int n) {
   return rc;
 }
 
-static textstartup void sys_getloadavg_nt_init(void) {
+__attribute__((__constructor__(40))) static textstartup void ntinitload(void) {
   if (IsWindows()) {
     load = 1;
     cpus = __get_cpu_count() / 2;
@@ -67,7 +67,3 @@ static textstartup void sys_getloadavg_nt_init(void) {
     GetSystemTimes(&idle1, &kern1, &user1);
   }
 }
-
-const void *const sys_getloadavg_nt_ctor[] initarray = {
-    sys_getloadavg_nt_init,
-};
diff --git a/libc/calls/getrandom.c b/libc/calls/getrandom.c
index 7c9a7e2f2..cdc7b13e5 100644
--- a/libc/calls/getrandom.c
+++ b/libc/calls/getrandom.c
@@ -190,7 +190,8 @@ ssize_t getrandom(void *p, size_t n, unsigned f) {
   return rc;
 }
 
-__attribute__((__constructor__)) static textstartup void getrandom_init(void) {
+__attribute__((__constructor__(30))) static textstartup void getrandom_init(
+    void) {
   int e, rc;
   if (IsWindows() || IsMetal()) return;
   BLOCK_CANCELATION;
diff --git a/libc/calls/program_invocation_short_name.c b/libc/calls/program_invocation_short_name.c
index 8299de395..396d55f47 100644
--- a/libc/calls/program_invocation_short_name.c
+++ b/libc/calls/program_invocation_short_name.c
@@ -21,7 +21,7 @@
 
 char *program_invocation_short_name;
 
-__attribute__((__constructor__)) static void  //
+__attribute__((__constructor__(10))) static textstartup void
 program_invocation_short_name_init(void) {
   char *p, *r;
   if (!__argc) return;
diff --git a/libc/calls/rdrand_init.c b/libc/calls/rdrand_init.c
index 9751016bf..a3b28ae07 100644
--- a/libc/calls/rdrand_init.c
+++ b/libc/calls/rdrand_init.c
@@ -22,8 +22,8 @@
 #include "libc/nexgen32e/x86info.h"
 #include "libc/stdio/rand.h"
 
-textstartup void rdrand_init(int argc, char **argv, char **envp,
-                             intptr_t *auxv) {
+__attribute__((__constructor__(2))) textstartup void rdrand_init(
+    int argc, char **argv, char **envp, intptr_t *auxv) {
   extern unsigned kMutableCpuids[KCPUIDS_LEN][4] asm("kCpuids");
   /*
    * Clear RDRAND on AMD models before Zen and then some
@@ -39,5 +39,3 @@ textstartup void rdrand_init(int argc, char **argv, char **envp,
     kMutableCpuids[KCPUIDS_7H][KCPUIDS_EBX] &= ~(1u << 18);
   }
 }
-
-const void *const g_rdrand_init[] initarray = {rdrand_init};
diff --git a/libc/calls/sig.c b/libc/calls/sig.c
index 936ed99d4..cab11bec7 100644
--- a/libc/calls/sig.c
+++ b/libc/calls/sig.c
@@ -32,6 +32,7 @@
 #include "libc/intrin/bsf.h"
 #include "libc/intrin/describebacktrace.internal.h"
 #include "libc/intrin/dll.h"
+#include "libc/intrin/kprintf.h"
 #include "libc/intrin/strace.internal.h"
 #include "libc/intrin/weaken.h"
 #include "libc/nt/console.h"
@@ -517,6 +518,15 @@ static int __sig_crash_sig(struct NtExceptionPointers *ep, int *code) {
   }
 }
 
+static char *__sig_stpcpy(char *d, const char *s) {
+  size_t i;
+  for (i = 0;; ++i) {
+    if (!(d[i] = s[i])) {
+      return d + i;
+    }
+  }
+}
+
 static void __sig_unmaskable(struct NtExceptionPointers *ep, int code, int sig,
                              struct CosmoTib *tib) {
 
@@ -540,9 +550,10 @@ static void __sig_unmaskable(struct NtExceptionPointers *ep, int code, int sig,
     intptr_t hStderr;
     char sigbuf[21], s[128], *p;
     hStderr = GetStdHandle(kNtStdErrorHandle);
-    p = stpcpy(s, "Terminating on uncaught ");
-    p = stpcpy(p, strsignal_r(sig, sigbuf));
-    p = stpcpy(p, ". Pass --strace and/or ShowCrashReports() for details.\n");
+    p = __sig_stpcpy(s, "Terminating on uncaught ");
+    p = __sig_stpcpy(p, strsignal_r(sig, sigbuf));
+    p = __sig_stpcpy(
+        p, ". Pass --strace and/or ShowCrashReports() for details.\n");
     WriteFile(hStderr, s, p - s, 0, 0);
 #endif
     __sig_terminate(sig);
@@ -656,12 +667,10 @@ textwindows int __sig_check(void) {
   }
 }
 
-textstartup void __sig_init(void) {
+__attribute__((__constructor__(10))) textstartup void __sig_init(void) {
   if (!IsWindows()) return;
   AddVectoredExceptionHandler(true, (void *)__sig_crash);
   SetConsoleCtrlHandler((void *)__sig_console, true);
 }
 
-const void *const __sig_ctor[] initarray = {__sig_init};
-
 #endif /* __x86_64__ */
diff --git a/libc/calls/unveil.c b/libc/calls/unveil.c
index 5eae25662..d47c84889 100644
--- a/libc/calls/unveil.c
+++ b/libc/calls/unveil.c
@@ -100,7 +100,7 @@ static const struct sock_filter kUnveilBlacklistLatestAbi[] = {
 static int landlock_abi_version;
 static int landlock_abi_errno;
 
-__attribute__((__constructor__)) void init_landlock_version() {
+__attribute__((__constructor__(40))) textstartup void init_landlock_version() {
   int e = errno;
   landlock_abi_version =
       landlock_create_ruleset(0, 0, LANDLOCK_CREATE_RULESET_VERSION);
diff --git a/libc/crt/crt.S b/libc/crt/crt.S
index f55042d09..69c38493e 100644
--- a/libc/crt/crt.S
+++ b/libc/crt/crt.S
@@ -18,7 +18,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/dce.h"
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 .section .start,"ax",@progbits
 
 #if SupportsXnu() && defined(__x86_64__)
diff --git a/libc/disclaimer.inc b/libc/disclaimer.inc
deleted file mode 100644
index e69de29bb..000000000
diff --git a/libc/dlopen/dlopen.c b/libc/dlopen/dlopen.c
index 9dcda52f6..ae5af3e40 100644
--- a/libc/dlopen/dlopen.c
+++ b/libc/dlopen/dlopen.c
@@ -920,17 +920,3 @@ char *cosmo_dlerror(void) {
   STRACE("dlerror() → %#s", res);
   return res;
 }
-
-#ifdef __x86_64__
-static textstartup void dlopen_init() {
-  if (IsLinux() || IsFreebsd()) {
-    // switch from %fs to %gs for tls
-    struct CosmoTib *tib = __get_tls();
-    __morph_tls();
-    __set_tls(tib);
-  }
-}
-const void *const dlopen_ctor[] initarray = {
-    dlopen_init,
-};
-#endif
diff --git a/libc/dlopen/foreign_tramp.S b/libc/dlopen/foreign_tramp.S
index f44c2edbe..dbd036306 100644
--- a/libc/dlopen/foreign_tramp.S
+++ b/libc/dlopen/foreign_tramp.S
@@ -55,7 +55,7 @@ foreign_tramp:
 	mov	%rax,-0xc0(%rbp)
 
 //	switch to foreign tls
-	mov	%fs:0,%rax
+	mov	%gs:0x30,%rax
 	mov	%rax,-0xc8(%rbp)
 	mov	__foreign+8(%rip),%rdi
 	call	__set_tls
diff --git a/libc/integral/c.inc b/libc/integral/c.inc
index b5870b7fa..a1923bd64 100644
--- a/libc/integral/c.inc
+++ b/libc/integral/c.inc
@@ -639,12 +639,6 @@ void abort(void) wontreturn;
 #define _Section(s) __attribute__((__section__(s)))
 #endif
 
-#if defined(__x86_64__) && !defined(__llvm__)
-#define initarray _Section(".init_array,\"a\",@init_array #")
-#else
-#define initarray _Section(".init_array")
-#endif
-
 #ifndef __llvm__
 #pragma GCC diagnostic ignored "-Wformat=0" /* todo: patch gcc */
 #pragma GCC diagnostic ignored "-Wbuiltin-declaration-mismatch"
@@ -758,17 +752,21 @@ void abort(void) wontreturn;
 #endif
 
 #if defined(__x86_64__) && (defined(__GNUC__) || defined(__llvm__)) && \
-    !defined(__chibicc__) && 0 /* TODO: enable with toolchain upgrade */
+    !defined(__chibicc__) && defined(__OPTIMIZE__)
 #define __target_clones(x) __attribute__((__target_clones__(x ",default")))
 #else
 #define __target_clones(x)
 #endif
 
-#if !defined(TINY) && !defined(__AVX__) && 0
+#if !defined(TINY) && !defined(__AVX__)
 #define __vex __target_clones("avx")
 #else
 #define __vex
 #endif
 
+#define __notice(sym, str)                                                   \
+  __attribute__((__section__(".notice"), __aligned__(1))) const char sym[] = \
+      "\n\n" str
+
 #define MACHINE_CODE_ANALYSIS_BEGIN_
 #define MACHINE_CODE_ANALYSIS_END_
diff --git a/libc/integral/normalize.inc b/libc/integral/normalize.inc
index e9cba82d8..1291b101d 100644
--- a/libc/integral/normalize.inc
+++ b/libc/integral/normalize.inc
@@ -3,8 +3,8 @@
 #endif
 
 #define __COSMOPOLITAN_MAJOR__ 3
-#define __COSMOPOLITAN_MINOR__ 2
-#define __COSMOPOLITAN_PATCH__ 4
+#define __COSMOPOLITAN_MINOR__ 3
+#define __COSMOPOLITAN_PATCH__ 0
 #define __COSMOPOLITAN__                                                   \
   (100000000 * __COSMOPOLITAN_MAJOR__ + 1000000 * __COSMOPOLITAN_MINOR__ + \
    __COSMOPOLITAN_PATCH__)
diff --git a/libc/intrin/BUILD.mk b/libc/intrin/BUILD.mk
index 609e05018..2ba6fa39b 100644
--- a/libc/intrin/BUILD.mk
+++ b/libc/intrin/BUILD.mk
@@ -74,11 +74,15 @@ o//libc/intrin/memmove.o: private			\
 			-fno-toplevel-reorder
 
 o//libc/intrin/bzero.o					\
+o//libc/intrin/strlen.o					\
+o//libc/intrin/strchr.o					\
+o//libc/intrin/memchr.o					\
+o//libc/intrin/memrchr.o				\
 o//libc/intrin/memcmp.o					\
 o//libc/intrin/memset.o					\
 o//libc/intrin/memmove.o: private			\
 		CFLAGS +=				\
-			-O2 -finline
+			-O2 -finline -foptimize-sibling-calls
 
 o/$(MODE)/libc/intrin/bzero.o				\
 o/$(MODE)/libc/intrin/memcmp.o				\
@@ -89,7 +93,9 @@ o/$(MODE)/libc/intrin/memmove.o: private		\
 o/$(MODE)/libc/intrin/x86.o: private			\
 		CFLAGS +=				\
 			-ffreestanding			\
-			-fno-jump-tables
+			-fno-jump-tables		\
+			-fpatchable-function-entry=0	\
+			-Os
 
 # these assembly files are safe to build on aarch64
 o/$(MODE)/libc/intrin/aarch64/%.o: libc/intrin/aarch64/%.S
diff --git a/libc/intrin/aarch64/memchr.S b/libc/intrin/aarch64/memchr.S
index 1f423f311..fcfacc661 100644
--- a/libc/intrin/aarch64/memchr.S
+++ b/libc/intrin/aarch64/memchr.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __memchr_aarch64 memchr
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64
diff --git a/libc/intrin/aarch64/memcpy.S b/libc/intrin/aarch64/memcpy.S
index a736ebdb6..fd30eb4e7 100644
--- a/libc/intrin/aarch64/memcpy.S
+++ b/libc/intrin/aarch64/memcpy.S
@@ -26,15 +26,11 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __memcpy_aarch64_simd memcpy
 #define __memmove_aarch64_simd memmove
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD, unaligned accesses.
diff --git a/libc/intrin/aarch64/memrchr.S b/libc/intrin/aarch64/memrchr.S
index d0795c8b1..f24c358f7 100644
--- a/libc/intrin/aarch64/memrchr.S
+++ b/libc/intrin/aarch64/memrchr.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __memrchr_aarch64 memrchr
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD.
diff --git a/libc/intrin/aarch64/memset.S b/libc/intrin/aarch64/memset.S
index 509840c74..f6b96d3a6 100644
--- a/libc/intrin/aarch64/memset.S
+++ b/libc/intrin/aarch64/memset.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __memset_aarch64 memset
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD, unaligned accesses.
diff --git a/libc/intrin/aarch64/stpcpy.S b/libc/intrin/aarch64/stpcpy.S
index f7e39e52e..05fa1b866 100644
--- a/libc/intrin/aarch64/stpcpy.S
+++ b/libc/intrin/aarch64/stpcpy.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __stpcpy_aarch64 stpcpy
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD.
diff --git a/libc/intrin/aarch64/strchr.S b/libc/intrin/aarch64/strchr.S
index ea841ed98..ffceb8eaf 100644
--- a/libc/intrin/aarch64/strchr.S
+++ b/libc/intrin/aarch64/strchr.S
@@ -26,15 +26,11 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strchr_aarch64 strchr
 #define __index_aarch64 index
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64
diff --git a/libc/intrin/aarch64/strchrnul.S b/libc/intrin/aarch64/strchrnul.S
index 9694fca47..c153c8f65 100644
--- a/libc/intrin/aarch64/strchrnul.S
+++ b/libc/intrin/aarch64/strchrnul.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strchrnul_aarch64 strchrnul
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64
diff --git a/libc/intrin/aarch64/strcmp.S b/libc/intrin/aarch64/strcmp.S
index 8d5702f09..dceb9d25c 100644
--- a/libc/intrin/aarch64/strcmp.S
+++ b/libc/intrin/aarch64/strcmp.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strcmp_aarch64 strcmp
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64.
diff --git a/libc/intrin/aarch64/strcpy.S b/libc/intrin/aarch64/strcpy.S
index 3e6599bb8..a92541b40 100644
--- a/libc/intrin/aarch64/strcpy.S
+++ b/libc/intrin/aarch64/strcpy.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strcpy_aarch64 strcpy
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD.
diff --git a/libc/intrin/aarch64/strlen.S b/libc/intrin/aarch64/strlen.S
index a70e7b214..36ecd0e18 100644
--- a/libc/intrin/aarch64/strlen.S
+++ b/libc/intrin/aarch64/strlen.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strlen_aarch64 strlen
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD, unaligned accesses.
diff --git a/libc/intrin/aarch64/strncmp.S b/libc/intrin/aarch64/strncmp.S
index 8741d766b..f91114e51 100644
--- a/libc/intrin/aarch64/strncmp.S
+++ b/libc/intrin/aarch64/strncmp.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strncmp_aarch64 strncmp
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64.
diff --git a/libc/intrin/aarch64/strnlen.S b/libc/intrin/aarch64/strnlen.S
index 2e7c72d31..a556aa61b 100644
--- a/libc/intrin/aarch64/strnlen.S
+++ b/libc/intrin/aarch64/strnlen.S
@@ -26,14 +26,10 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strnlen_aarch64 strnlen
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64, Advanced SIMD.
diff --git a/libc/intrin/aarch64/strrchr.S b/libc/intrin/aarch64/strrchr.S
index f492a4423..73289101f 100644
--- a/libc/intrin/aarch64/strrchr.S
+++ b/libc/intrin/aarch64/strrchr.S
@@ -26,15 +26,11 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/intrin/aarch64/asmdefs.internal.h"
+.yoink arm_optimized_routines_notice
 
 #define __strrchr_aarch64 strrchr
 #define __rindex_aarch64 rindex
 
-.ident "\n\n\
-Optimized Routines (MIT License)\n\
-Copyright 2022 ARM Limited\n"
-.include "libc/disclaimer.inc"
-
 /* Assumptions:
  *
  * ARMv8-a, AArch64
diff --git a/libc/intrin/armopt.c b/libc/intrin/armopt.c
new file mode 100644
index 000000000..19fc8da12
--- /dev/null
+++ b/libc/intrin/armopt.c
@@ -0,0 +1,3 @@
+__notice(arm_optimized_routines_notice, "\
+Optimized Routines (MIT License)\n\
+Copyright 2022 ARM Limited");
diff --git a/libc/intrin/chromium.c b/libc/intrin/chromium.c
new file mode 100644
index 000000000..543de07d3
--- /dev/null
+++ b/libc/intrin/chromium.c
@@ -0,0 +1,3 @@
+__notice(chromium_notice, "\
+Chromium (BSD-3 License)\n\
+Copyright 2017 The Chromium Authors");
diff --git a/libc/intrin/cxalock.c b/libc/intrin/cxalock.c
index 3ebcf3c8d..4e6de59d0 100644
--- a/libc/intrin/cxalock.c
+++ b/libc/intrin/cxalock.c
@@ -33,10 +33,6 @@ void __cxa_unlock(void) {
   pthread_mutex_unlock(&__cxa_lock_obj);
 }
 
-static textstartup void __cxa_init() {
+__attribute__((__constructor__(60))) static textstartup void __cxa_init() {
   pthread_atfork(__cxa_lock, __cxa_unlock, __cxa_wipe);
 }
-
-const void *const __cxa_ctor[] initarray = {
-    __cxa_init,
-};
diff --git a/libc/intrin/fblibm.c b/libc/intrin/fblibm.c
new file mode 100644
index 000000000..5a5fbd3fa
--- /dev/null
+++ b/libc/intrin/fblibm.c
@@ -0,0 +1,4 @@
+__notice(freebsd_libm_notice, "FreeBSD libm (BSD-2 License)\n\
+Copyright (c) 1992-2024 The FreeBSD Project.\n\
+Copyright (c) 2005-2011 Bruce D. Evans, Steven G. Kargl, David Schultz\n\
+Copyright (c) 2009-2013 Steven G. Kargl");
diff --git a/libc/intrin/fdlibm.c b/libc/intrin/fdlibm.c
new file mode 100644
index 000000000..2ea7510e2
--- /dev/null
+++ b/libc/intrin/fdlibm.c
@@ -0,0 +1,2 @@
+__notice(fdlibm_notice, "fdlibm (fdlibm license)\n\
+Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.");
diff --git a/libc/intrin/folibm.c b/libc/intrin/folibm.c
new file mode 100644
index 000000000..2fe77e001
--- /dev/null
+++ b/libc/intrin/folibm.c
@@ -0,0 +1,2 @@
+__notice(openbsd_libm_notice, "OpenBSD libm (ISC License)\n\
+Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>");
diff --git a/libc/intrin/isrunningundermake.c b/libc/intrin/isrunningundermake.c
index 929ece7f8..53bcf4840 100644
--- a/libc/intrin/isrunningundermake.c
+++ b/libc/intrin/isrunningundermake.c
@@ -30,10 +30,6 @@ bool32 IsRunningUnderMake(void) {
   return g_isrunningundermake;
 }
 
-textstartup void g_isrunningundermake_init(void) {
+__attribute__((__constructor__(30))) textstartup void onmake_init(void) {
   g_isrunningundermake = !!getenv("MAKEFLAGS");
 }
-
-const void *const g_isrunningundermake_ctor[] initarray = {
-    g_isrunningundermake_init,
-};
diff --git a/libc/intrin/memchr.c b/libc/intrin/memchr.c
index b29c6e67a..676cb04be 100644
--- a/libc/intrin/memchr.c
+++ b/libc/intrin/memchr.c
@@ -36,8 +36,8 @@ static inline const unsigned char *memchr_pure(const unsigned char *s,
 }
 
 #if defined(__x86_64__) && !defined(__chibicc__)
-static inline const unsigned char *memchr_sse(const unsigned char *s,
-                                              unsigned char c, size_t n) {
+static __vex const unsigned char *memchr_sse(const unsigned char *s,
+                                             unsigned char c, size_t n) {
   size_t i;
   unsigned m;
   xmm_t v, t = {c, c, c, c, c, c, c, c, c, c, c, c, c, c, c, c};
@@ -67,11 +67,9 @@ static inline const unsigned char *memchr_sse(const unsigned char *s,
  * @return is pointer to first instance of c or NULL if not found
  * @asyncsignalsafe
  */
-__vex void *memchr(const void *s, int c, size_t n) {
+void *memchr(const void *s, int c, size_t n) {
 #if defined(__x86_64__) && !defined(__chibicc__)
-  const void *r;
-  r = memchr_sse(s, c, n);
-  return (void *)r;
+  return (void *)memchr_sse(s, c, n);
 #else
   return (void *)memchr_pure(s, c, n);
 #endif
diff --git a/libc/intrin/memmove.c b/libc/intrin/memmove.c
index f6687c4b5..493210aac 100644
--- a/libc/intrin/memmove.c
+++ b/libc/intrin/memmove.c
@@ -27,66 +27,7 @@
 typedef long long xmm_t __attribute__((__vector_size__(16), __aligned__(1)));
 typedef long long xmm_a __attribute__((__vector_size__(16), __aligned__(16)));
 
-/**
- * Copies memory.
- *
- *     memmove n=0                        661 picoseconds
- *     memmove n=1                        661 ps/byte          1,476 mb/s
- *     memmove n=2                        330 ps/byte          2,952 mb/s
- *     memmove n=3                        330 ps/byte          2,952 mb/s
- *     memmove n=4                        165 ps/byte          5,904 mb/s
- *     memmove n=7                        141 ps/byte          6,888 mb/s
- *     memmove n=8                         82 ps/byte             11 GB/s
- *     memmove n=15                        44 ps/byte             21 GB/s
- *     memmove n=16                        41 ps/byte             23 GB/s
- *     memmove n=31                        32 ps/byte             29 GB/s
- *     memmove n=32                        31 ps/byte             30 GB/s
- *     memmove n=63                        21 ps/byte             45 GB/s
- *     memmove n=64                        15 ps/byte             61 GB/s
- *     memmove n=127                       13 ps/byte             73 GB/s
- *     memmove n=128                       31 ps/byte             30 GB/s
- *     memmove n=255                       20 ps/byte             45 GB/s
- *     memmove n=256                       19 ps/byte             49 GB/s
- *     memmove n=511                       16 ps/byte             56 GB/s
- *     memmove n=512                       17 ps/byte             54 GB/s
- *     memmove n=1023                      18 ps/byte             52 GB/s
- *     memmove n=1024                      13 ps/byte             72 GB/s
- *     memmove n=2047                       9 ps/byte             96 GB/s
- *     memmove n=2048                       9 ps/byte             98 GB/s
- *     memmove n=4095                       8 ps/byte            112 GB/s
- *     memmove n=4096                       8 ps/byte            109 GB/s
- *     memmove n=8191                       7 ps/byte            124 GB/s
- *     memmove n=8192                       7 ps/byte            125 GB/s
- *     memmove n=16383                      7 ps/byte            134 GB/s
- *     memmove n=16384                      7 ps/byte            134 GB/s
- *     memmove n=32767                     13 ps/byte             72 GB/s
- *     memmove n=32768                     13 ps/byte             72 GB/s
- *     memmove n=65535                     13 ps/byte             68 GB/s
- *     memmove n=65536                     14 ps/byte             67 GB/s
- *     memmove n=131071                    14 ps/byte             65 GB/s
- *     memmove n=131072                    14 ps/byte             64 GB/s
- *     memmove n=262143                    15 ps/byte             63 GB/s
- *     memmove n=262144                    15 ps/byte             63 GB/s
- *     memmove n=524287                    15 ps/byte             61 GB/s
- *     memmove n=524288                    15 ps/byte             61 GB/s
- *     memmove n=1048575                   15 ps/byte             61 GB/s
- *     memmove n=1048576                   15 ps/byte             61 GB/s
- *     memmove n=2097151                   19 ps/byte             48 GB/s
- *     memmove n=2097152                   27 ps/byte             35 GB/s
- *     memmove n=4194303                   28 ps/byte             33 GB/s
- *     memmove n=4194304                   28 ps/byte             33 GB/s
- *     memmove n=8388607                   28 ps/byte             33 GB/s
- *     memmove n=8388608                   28 ps/byte             33 GB/s
- *
- * DST and SRC may overlap.
- *
- * @param dst is destination
- * @param src is memory to copy
- * @param n is number of bytes to copy
- * @return dst
- * @asyncsignalsafe
- */
-__vex void *memmove(void *dst, const void *src, size_t n) {
+static __vex void *__memmove(void *dst, const void *src, size_t n) {
   char *d;
   size_t i;
   const char *s;
@@ -338,6 +279,69 @@ __vex void *memmove(void *dst, const void *src, size_t n) {
   }
 }
 
+/**
+ * Copies memory.
+ *
+ *     memmove n=0                        661 picoseconds
+ *     memmove n=1                        661 ps/byte          1,476 mb/s
+ *     memmove n=2                        330 ps/byte          2,952 mb/s
+ *     memmove n=3                        330 ps/byte          2,952 mb/s
+ *     memmove n=4                        165 ps/byte          5,904 mb/s
+ *     memmove n=7                        141 ps/byte          6,888 mb/s
+ *     memmove n=8                         82 ps/byte             11 GB/s
+ *     memmove n=15                        44 ps/byte             21 GB/s
+ *     memmove n=16                        41 ps/byte             23 GB/s
+ *     memmove n=31                        32 ps/byte             29 GB/s
+ *     memmove n=32                        31 ps/byte             30 GB/s
+ *     memmove n=63                        21 ps/byte             45 GB/s
+ *     memmove n=64                        15 ps/byte             61 GB/s
+ *     memmove n=127                       13 ps/byte             73 GB/s
+ *     memmove n=128                       31 ps/byte             30 GB/s
+ *     memmove n=255                       20 ps/byte             45 GB/s
+ *     memmove n=256                       19 ps/byte             49 GB/s
+ *     memmove n=511                       16 ps/byte             56 GB/s
+ *     memmove n=512                       17 ps/byte             54 GB/s
+ *     memmove n=1023                      18 ps/byte             52 GB/s
+ *     memmove n=1024                      13 ps/byte             72 GB/s
+ *     memmove n=2047                       9 ps/byte             96 GB/s
+ *     memmove n=2048                       9 ps/byte             98 GB/s
+ *     memmove n=4095                       8 ps/byte            112 GB/s
+ *     memmove n=4096                       8 ps/byte            109 GB/s
+ *     memmove n=8191                       7 ps/byte            124 GB/s
+ *     memmove n=8192                       7 ps/byte            125 GB/s
+ *     memmove n=16383                      7 ps/byte            134 GB/s
+ *     memmove n=16384                      7 ps/byte            134 GB/s
+ *     memmove n=32767                     13 ps/byte             72 GB/s
+ *     memmove n=32768                     13 ps/byte             72 GB/s
+ *     memmove n=65535                     13 ps/byte             68 GB/s
+ *     memmove n=65536                     14 ps/byte             67 GB/s
+ *     memmove n=131071                    14 ps/byte             65 GB/s
+ *     memmove n=131072                    14 ps/byte             64 GB/s
+ *     memmove n=262143                    15 ps/byte             63 GB/s
+ *     memmove n=262144                    15 ps/byte             63 GB/s
+ *     memmove n=524287                    15 ps/byte             61 GB/s
+ *     memmove n=524288                    15 ps/byte             61 GB/s
+ *     memmove n=1048575                   15 ps/byte             61 GB/s
+ *     memmove n=1048576                   15 ps/byte             61 GB/s
+ *     memmove n=2097151                   19 ps/byte             48 GB/s
+ *     memmove n=2097152                   27 ps/byte             35 GB/s
+ *     memmove n=4194303                   28 ps/byte             33 GB/s
+ *     memmove n=4194304                   28 ps/byte             33 GB/s
+ *     memmove n=8388607                   28 ps/byte             33 GB/s
+ *     memmove n=8388608                   28 ps/byte             33 GB/s
+ *
+ * DST and SRC may overlap.
+ *
+ * @param dst is destination
+ * @param src is memory to copy
+ * @param n is number of bytes to copy
+ * @return dst
+ * @asyncsignalsafe
+ */
+void *memmove(void *dst, const void *src, size_t n) {
+  return __memmove(dst, src, n);
+}
+
 __weak_reference(memmove, memcpy);
 
 #endif /* __aarch64__ */
diff --git a/libc/intrin/memrchr.c b/libc/intrin/memrchr.c
index 59f0d6686..f0f8e8689 100644
--- a/libc/intrin/memrchr.c
+++ b/libc/intrin/memrchr.c
@@ -36,8 +36,8 @@ static inline const unsigned char *memrchr_pure(const unsigned char *s,
 }
 
 #if defined(__x86_64__) && !defined(__chibicc__)
-static inline const unsigned char *memrchr_sse(const unsigned char *s,
-                                               unsigned char c, size_t n) {
+static __vex const unsigned char *memrchr_sse(const unsigned char *s,
+                                              unsigned char c, size_t n) {
   size_t i;
   unsigned m;
   xmm_t v, t = {c, c, c, c, c, c, c, c, c, c, c, c, c, c, c, c};
@@ -67,11 +67,9 @@ static inline const unsigned char *memrchr_sse(const unsigned char *s,
  * @return is pointer to first instance of c or NULL if not found
  * @asyncsignalsafe
  */
-__vex void *memrchr(const void *s, int c, size_t n) {
+void *memrchr(const void *s, int c, size_t n) {
 #if defined(__x86_64__) && !defined(__chibicc__)
-  const void *r;
-  r = memrchr_sse(s, c, n);
-  return (void *)r;
+  return (void *)memrchr_sse(s, c, n);
 #else
   return (void *)memrchr_pure(s, c, n);
 #endif
diff --git a/libc/intrin/musl.c b/libc/intrin/musl.c
new file mode 100644
index 000000000..28f7ed4db
--- /dev/null
+++ b/libc/intrin/musl.c
@@ -0,0 +1,3 @@
+__notice(musl_libc_notice, "\
+Musl libc (MIT License)\n\
+Copyright 2005-2014 Rich Felker, et. al.");
diff --git a/libc/intrin/nocolor.c b/libc/intrin/nocolor.c
index e864b0396..efd2fa7f6 100644
--- a/libc/intrin/nocolor.c
+++ b/libc/intrin/nocolor.c
@@ -48,13 +48,9 @@
  */
 bool __nocolor;
 
-optimizesize textstartup void __nocolor_init(int argc, char **argv, char **envp,
-                                             intptr_t *auxv) {
+__attribute__((__constructor__(20))) optimizesize textstartup void
+__nocolor_init(int argc, char **argv, char **envp, intptr_t *auxv) {
   char *s;
   __nocolor = (IsWindows() && !IsAtLeastWindows10()) ||
               ((s = getenv("TERM")) && IsDumb(s));
 }
-
-const void *const __nocolor_ctor[] initarray = {
-    __nocolor_init,
-};
diff --git a/libc/intrin/scalblnl.c b/libc/intrin/scalblnl.c
index 0150ec683..c93790767 100644
--- a/libc/intrin/scalblnl.c
+++ b/libc/intrin/scalblnl.c
@@ -28,11 +28,8 @@
 #include "libc/limits.h"
 #include "libc/math.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 long double scalblnl(long double x, long n) {
diff --git a/libc/intrin/scalbn.c b/libc/intrin/scalbn.c
index 54178a67c..55f7e6cda 100644
--- a/libc/intrin/scalbn.c
+++ b/libc/intrin/scalbn.c
@@ -26,11 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 /**
diff --git a/libc/intrin/scalbnf.c b/libc/intrin/scalbnf.c
index d5c5d1d02..b900e589d 100644
--- a/libc/intrin/scalbnf.c
+++ b/libc/intrin/scalbnf.c
@@ -26,11 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 /**
diff --git a/libc/intrin/scalbnl.c b/libc/intrin/scalbnl.c
index c87cb76f7..f9739ff36 100644
--- a/libc/intrin/scalbnl.c
+++ b/libc/intrin/scalbnl.c
@@ -28,11 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 /**
diff --git a/libc/intrin/stackchkguard.S b/libc/intrin/stackchkguard.S
index 9f4437ba4..b78117a5d 100644
--- a/libc/intrin/stackchkguard.S
+++ b/libc/intrin/stackchkguard.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 //	Canary for -fstack-protector.
 //
diff --git a/libc/intrin/strchr.c b/libc/intrin/strchr.c
index ab7439239..4d05ffb05 100644
--- a/libc/intrin/strchr.c
+++ b/libc/intrin/strchr.c
@@ -31,7 +31,7 @@ static inline const char *strchr_pure(const char *s, int c) {
 
 #if defined(__x86_64__) && !defined(__chibicc__)
 typedef char xmm_t __attribute__((__vector_size__(16), __aligned__(16)));
-static inline const char *strchr_sse(const char *s, unsigned char c) {
+static __vex const char *strchr_sse(const char *s, unsigned char c) {
   unsigned k;
   unsigned m;
   const xmm_t *p;
@@ -94,7 +94,7 @@ static inline const char *strchr_x64(const char *p, uint64_t c) {
  * @asyncsignalsafe
  * @vforksafe
  */
-__vex char *strchr(const char *s, int c) {
+char *strchr(const char *s, int c) {
 #if defined(__x86_64__) && !defined(__chibicc__)
   const char *r;
   if (X86_HAVE(SSE)) {
@@ -102,7 +102,6 @@ __vex char *strchr(const char *s, int c) {
   } else {
     r = strchr_pure(s, c);
   }
-  unassert(!r || *r || !(c & 255));
   return (char *)r;
 #else
   const char *r;
@@ -110,9 +109,7 @@ __vex char *strchr(const char *s, int c) {
     if ((*s & 255) == c) return (char *)s;
     if (!*s) return NULL;
   }
-  r = strchr_x64(s, c);
-  unassert(!r || *r || !c);
-  return (char *)r;
+  return (char *)strchr_x64(s, c);
 #endif
 }
 
diff --git a/libc/intrin/strlen.c b/libc/intrin/strlen.c
index 89d85e9d2..72a32c33d 100644
--- a/libc/intrin/strlen.c
+++ b/libc/intrin/strlen.c
@@ -20,14 +20,7 @@
 #include "libc/str/str.h"
 #ifndef __aarch64__
 
-/**
- * Returns length of NUL-terminated string.
- *
- * @param s is non-null NUL-terminated string pointer
- * @return number of bytes (excluding NUL)
- * @asyncsignalsafe
- */
-size_t strlen(const char *s) {
+static __vex size_t __strlen(const char *s) {
 #if defined(__x86_64__) && !defined(__chibicc__)
   typedef char xmm_t __attribute__((__vector_size__(16), __aligned__(16)));
   xmm_t z = {0};
@@ -56,4 +49,16 @@ size_t strlen(const char *s) {
   return (const char *)p + (__builtin_ctzl(w) >> 3) - s;
 #endif
 }
+
+/**
+ * Returns length of NUL-terminated string.
+ *
+ * @param s is non-null NUL-terminated string pointer
+ * @return number of bytes (excluding NUL)
+ * @asyncsignalsafe
+ */
+size_t strlen(const char *s) {
+  return __strlen(s);
+}
+
 #endif /* __aarch64__ */
diff --git a/libc/intrin/ubsan.c b/libc/intrin/ubsan.c
index d42f8e20b..5b3051ac5 100644
--- a/libc/intrin/ubsan.c
+++ b/libc/intrin/ubsan.c
@@ -627,7 +627,7 @@ void *__ubsan_get_current_report_data(void) {
   return 0;
 }
 
-static textstartup void ubsan_init() {
+__attribute__((__constructor__(90))) static textstartup void ubsan_init() {
   STRACE(" _   _ ____ ____    _    _   _");
   STRACE("| | | | __ ) ___|  / \\  | \\ | |");
   STRACE("| | | |  _ \\___ \\ / _ \\ |  \\| |");
@@ -635,7 +635,3 @@ static textstartup void ubsan_init() {
   STRACE(" \\___/|____/____/_/   \\_\\_| \\_|");
   STRACE("cosmopolitan behavior module initialized");
 }
-
-const void *const ubsan_ctor[] initarray = {
-    ubsan_init,
-};
diff --git a/libc/intrin/x86.c b/libc/intrin/x86.c
index 3a2d21c77..7912d9de8 100644
--- a/libc/intrin/x86.c
+++ b/libc/intrin/x86.c
@@ -12,8 +12,6 @@
 //
 //===----------------------------------------------------------------------===//
 #if defined(__x86_64__) && (defined(__GNUC__) || defined(__clang__))
-#include "libc/intrin/strace.internal.h"
-#include "third_party/compiler_rt/cpu_model.h"
 
 enum VendorSignatures {
   SIG_INTEL = 0x756e6547,  // Genu
@@ -740,8 +738,6 @@ static void getAvailableFeatures(unsigned ECX, unsigned EDX, unsigned MaxLeaf,
 #undef setFeature
 }
 
-int __cpu_indicator_init(void) CONSTRUCTOR_ATTRIBUTE;
-
 struct __processor_model {
   unsigned int __cpu_vendor;
   unsigned int __cpu_type;
@@ -757,7 +753,7 @@ unsigned __cpu_features2[(CPU_FEATURE_MAX - 1) / 32];
 // the priority set.  However, it still runs after ifunc initializers and
 // needs to be called explicitly there.
 
-int CONSTRUCTOR_ATTRIBUTE __cpu_indicator_init(void) {
+__attribute__((__constructor__(1))) textstartup int __cpu_indicator_init(void) {
   unsigned EAX, EBX, ECX, EDX;
   unsigned MaxLeaf = 5;
   unsigned Vendor;
diff --git a/libc/isystem/sys/syscall.h b/libc/isystem/sys/syscall.h
index 9a508f8c4..a65aa2c65 100644
--- a/libc/isystem/sys/syscall.h
+++ b/libc/isystem/sys/syscall.h
@@ -1,4 +1,4 @@
-#ifndef COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_
-#define COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_
+#ifndef COSMOPOLITAN_LIBC_ISYSTEM_SYS_SYSCALL_H_
+#define COSMOPOLITAN_LIBC_ISYSTEM_SYS_SYSCALL_H_
 #include "libc/stdio/syscall.h"
-#endif /* COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_ */
+#endif /* COSMOPOLITAN_LIBC_ISYSTEM_SYS_SYSCALL_H_ */
diff --git a/libc/isystem/syscall.h b/libc/isystem/syscall.h
new file mode 100644
index 000000000..9a508f8c4
--- /dev/null
+++ b/libc/isystem/syscall.h
@@ -0,0 +1,4 @@
+#ifndef COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_
+#define COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_
+#include "libc/stdio/syscall.h"
+#endif /* COSMOPOLITAN_LIBC_ISYSTEM_SYSCALL_H_ */
diff --git a/libc/log/countbranch_report.c b/libc/log/countbranch_report.c
index 80c5381ad..7616c4e7a 100644
--- a/libc/log/countbranch_report.c
+++ b/libc/log/countbranch_report.c
@@ -96,12 +96,9 @@ void countbranch_report(void) {
   }
 }
 
-static textstartup void countbranch_init() {
+__attribute__((__constructor__(90))) static textstartup void
+countbranch_init() {
   atexit(countbranch_report);
 }
 
-const void *const countbranch_ctor[] initarray = {
-    countbranch_init,
-};
-
 #endif /* __x86_64__ */
diff --git a/libc/log/countexpr_data.S b/libc/log/countexpr_data.S
index c06a2fe9e..72db2252f 100644
--- a/libc/log/countexpr_data.S
+++ b/libc/log/countexpr_data.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 	.yoink	countexpr_report
 
diff --git a/libc/log/countexpr_report.c b/libc/log/countexpr_report.c
index 2fe072624..3d1c50025 100644
--- a/libc/log/countexpr_report.c
+++ b/libc/log/countexpr_report.c
@@ -76,12 +76,8 @@ void countexpr_report(void) {
   }
 }
 
-static textstartup void countexpr_init() {
+__attribute__((__constructor__(90))) static textstartup void countexpr_init() {
   atexit(countexpr_report);
 }
 
-const void *const countexpr_ctor[] initarray = {
-    countexpr_init,
-};
-
 #endif /* __x86_64__ */
diff --git a/libc/log/err.c b/libc/log/err.c
index 455e4ae40..83b459760 100644
--- a/libc/log/err.c
+++ b/libc/log/err.c
@@ -36,12 +36,9 @@
 #include "libc/str/str.h"
 // clang-format off
 
-asm(".ident\t\"\\n\\n\
-FreeBSD Err (BSD-3 License)\\n\
-Copyright (c) 1993\\n\
-\tThe Regents of the University of California.\\n\
-\tAll rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(freebsd_err_notice, "\
+FreeBSD Error Library (BSD-3 License)\n\
+Copyright 1993 The Regents of the University of California");
 
 static FILE *err_file; /* file to use for error output */
 static void (*err_exit)(int);
diff --git a/libc/log/logfile.c b/libc/log/logfile.c
index dc6221c8d..7cc70c86a 100644
--- a/libc/log/logfile.c
+++ b/libc/log/logfile.c
@@ -21,6 +21,6 @@
 
 FILE *__log_file;
 
-__attribute__((__constructor__)) static void init(void) {
+__attribute__((__constructor__(60))) static void loginit(void) {
   __log_file = stderr;
 }
diff --git a/libc/log/memlog.c b/libc/log/memlog.c
index 4c641bb41..62304ea7a 100644
--- a/libc/log/memlog.c
+++ b/libc/log/memlog.c
@@ -66,7 +66,7 @@ static struct Memlog {
     struct Alloc {
       void *addr;
       long size;
-    } * p;
+    } *p;
   } allocs;
   atomic_long usage;
 } __memlog;
@@ -270,7 +270,9 @@ static textexit void __memlog_destroy(void) {
   __memlog_unlock();
 }
 
-static textstartup void __memlog_init(void) {
+__attribute__((__constructor__(90)))  //
+static textstartup void
+__memlog_init(void) {
   GetSymbolTable();
   __memlog_lock();
   __memlog.free = hook_free;
@@ -290,7 +292,3 @@ static textstartup void __memlog_init(void) {
   atexit(__memlog_destroy);
   __memlog_unlock();
 }
-
-const void *const enable_memory_log[] initarray = {
-    __memlog_init,
-};
diff --git a/libc/math.h b/libc/math.h
index 0dd9fe063..1a9ffb611 100644
--- a/libc/math.h
+++ b/libc/math.h
@@ -4,59 +4,69 @@
 │ cosmopolitan § mathematics                                                   │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 
-#define M_E        2.7182818284590452354 /* 𝑒 */
-#define M_LOG2_10  0xd.49a784bcd1b8afep-2 /* log₂10 ≈ 3.3219280948873623478 */
-#define M_LOG2E    0xb.8aa3b295c17f0bcp-3 /* log₂𝑒 ≈ 1.4426950408889634074 */
-#define M_LOG10E   0.43429448190325182765  /* log₁₀𝑒 */
-#define M_LN2      0xb.17217f7d1cf79acp-4  /* logₑ2 ≈ */
-#define M_LN10     2.30258509299404568402  /* logₑ10 */
-#define M_TAU      0x1.921fb54442d1846ap+2 /* τ = 2π */
-#define M_PI       0x1.921fb54442d1846ap+1 /* π ≈ 3.14159265358979323846 */
-#define M_PI_2     1.57079632679489661923  /* π/2 */
-#define M_PI_4     0.78539816339744830962  /* π/4 */
-#define M_1_PI     0.31830988618379067154  /* 1/π */
-#define M_2_PI     0.63661977236758134308  /* 2/π */
-#define M_2_SQRTPI 1.12837916709551257390  /* 2/sqrtπ */
-#define M_SQRT2    1.41421356237309504880  /* sqrt2 */
-#define M_SQRT1_2  0.70710678118654752440  /* 1/sqrt2 */
+#define M_E        2.7182818284590452354  /* 𝑒 */
+#define M_LOG2E    1.4426950408889634074  /* log₂𝑒 */
+#define M_LOG10E   0.43429448190325182765 /* log₁₀𝑒 */
+#define M_LN2      0.69314718055994530942 /* logₑ2 */
+#define M_LN10     2.30258509299404568402 /* logₑ10 */
+#define M_PI       3.14159265358979323846 /* pi */
+#define M_PI_2     1.57079632679489661923 /* pi/2 */
+#define M_PI_4     0.78539816339744830962 /* pi/4 */
+#define M_1_PI     0.31830988618379067154 /* 1/pi */
+#define M_2_PI     0.63661977236758134308 /* 2/pi */
+#define M_2_SQRTPI 1.12837916709551257390 /* 2/sqrt(pi) */
+#define M_SQRT2    1.41421356237309504880 /* sqrt(2) */
+#define M_SQRT1_2  0.70710678118654752440 /* 1/sqrt(2) */
 
-#define DBL_DECIMAL_DIG  __DBL_DECIMAL_DIG__
-#define DBL_DIG          __DBL_DIG__
-#define DBL_EPSILON      __DBL_EPSILON__
-#define DBL_MANT_DIG     __DBL_MANT_DIG__
-#define DBL_MANT_DIG     __DBL_MANT_DIG__
-#define DBL_MAX          __DBL_MAX__
-#define DBL_MAX_10_EXP   __DBL_MAX_10_EXP__
-#define DBL_MAX_EXP      __DBL_MAX_EXP__
-#define DBL_MIN          __DBL_MIN__ /* 2.23e–308 ↔ 1.79e308 */
-#define DBL_MIN_10_EXP   __DBL_MIN_10_EXP__
-#define DBL_MIN_EXP      __DBL_MIN_EXP__
-#define DECIMAL_DIG      __LDBL_DECIMAL_DIG__
-#define FLT_DECIMAL_DIG  __FLT_DECIMAL_DIG__
-#define FLT_RADIX        __FLT_RADIX__
-#define FLT_DIG          __FLT_DIG__
-#define FLT_EPSILON      __FLT_EPSILON__
-#define FLT_MANT_DIG     __FLT_MANT_DIG__
-#define FLT_MANT_DIG     __FLT_MANT_DIG__
-#define FLT_MAX          __FLT_MAX__
-#define FLT_MAX_10_EXP   __FLT_MAX_10_EXP__
-#define FLT_MAX_EXP      __FLT_MAX_EXP__
-#define FLT_MIN          __FLT_MIN__ /* 1.18e–38 ↔ 3.40e38 */
-#define FLT_MIN_10_EXP   __FLT_MIN_10_EXP__
-#define FLT_MIN_EXP      __FLT_MIN_EXP__
-#define HLF_MAX          6.50e4f
-#define HLF_MIN          3.10e-5f
-#define LDBL_DECIMAL_DIG __LDBL_DECIMAL_DIG__
-#define LDBL_DIG         __LDBL_DIG__
-#define LDBL_EPSILON     __LDBL_EPSILON__
-#define LDBL_MANT_DIG    __LDBL_MANT_DIG__
-#define LDBL_MANT_DIG    __LDBL_MANT_DIG__
-#define LDBL_MAX         __LDBL_MAX__
-#define LDBL_MAX_10_EXP  __LDBL_MAX_10_EXP__
-#define LDBL_MAX_EXP     __LDBL_MAX_EXP__
-#define LDBL_MIN         __LDBL_MIN__ /* 3.37e–4932 ↔ 1.18e4932 */
-#define LDBL_MIN_10_EXP  __LDBL_MIN_10_EXP__
-#define LDBL_MIN_EXP     __LDBL_MIN_EXP__
+#define DBL_DECIMAL_DIG   __DBL_DECIMAL_DIG__
+#define DBL_DIG           __DBL_DIG__
+#define DBL_EPSILON       __DBL_EPSILON__
+#define DBL_HAS_SUBNORM   __DBL_HAS_DENORM__
+#define DBL_IS_IEC_60559  __DBL_IS_IEC_60559__
+#define DBL_MANT_DIG      __DBL_MANT_DIG__
+#define DBL_MANT_DIG      __DBL_MANT_DIG__
+#define DBL_MAX           __DBL_MAX__
+#define DBL_MAX_10_EXP    __DBL_MAX_10_EXP__
+#define DBL_MAX_EXP       __DBL_MAX_EXP__
+#define DBL_MIN           __DBL_MIN__ /* 2.23e–308 ↔ 1.79e308 */
+#define DBL_MIN_10_EXP    __DBL_MIN_10_EXP__
+#define DBL_MIN_EXP       __DBL_MIN_EXP__
+#define DBL_NORM_MAX      __DBL_NORM_MAX__
+#define DBL_TRUE_MIN      __DBL_DENORM_MIN__
+#define DECIMAL_DIG       __LDBL_DECIMAL_DIG__
+#define FLT_DECIMAL_DIG   __FLT_DECIMAL_DIG__
+#define FLT_DIG           __FLT_DIG__
+#define FLT_EPSILON       __FLT_EPSILON__
+#define FLT_HAS_SUBNORM   __FLT_HAS_DENORM__
+#define FLT_IS_IEC_60559  __FLT_IS_IEC_60559__
+#define FLT_MANT_DIG      __FLT_MANT_DIG__
+#define FLT_MANT_DIG      __FLT_MANT_DIG__
+#define FLT_MAX           __FLT_MAX__
+#define FLT_MAX_10_EXP    __FLT_MAX_10_EXP__
+#define FLT_MAX_EXP       __FLT_MAX_EXP__
+#define FLT_MIN           __FLT_MIN__ /* 1.18e–38 ↔ 3.40e38 */
+#define FLT_MIN_10_EXP    __FLT_MIN_10_EXP__
+#define FLT_MIN_EXP       __FLT_MIN_EXP__
+#define FLT_NORM_MAX      __FLT_NORM_MAX__
+#define FLT_RADIX         __FLT_RADIX__
+#define FLT_TRUE_MIN      __FLT_DENORM_MIN__
+#define HLF_MAX           6.50e4f
+#define HLF_MIN           3.10e-5f
+#define LDBL_DECIMAL_DIG  __LDBL_DECIMAL_DIG__
+#define LDBL_DIG          __LDBL_DIG__
+#define LDBL_EPSILON      __LDBL_EPSILON__
+#define LDBL_HAS_SUBNORM  __LDBL_HAS_DENORM__
+#define LDBL_IS_IEC_60559 __LDBL_IS_IEC_60559__
+#define LDBL_MANT_DIG     __LDBL_MANT_DIG__
+#define LDBL_MANT_DIG     __LDBL_MANT_DIG__
+#define LDBL_MAX          __LDBL_MAX__
+#define LDBL_MAX_10_EXP   __LDBL_MAX_10_EXP__
+#define LDBL_MAX_EXP      __LDBL_MAX_EXP__
+#define LDBL_MIN          __LDBL_MIN__ /* 3.37e–4932 ↔ 1.18e4932 */
+#define LDBL_MIN_10_EXP   __LDBL_MIN_10_EXP__
+#define LDBL_MIN_EXP      __LDBL_MIN_EXP__
+#define LDBL_NORM_MAX     __LDBL_NORM_MAX__
+#define LDBL_TRUE_MIN     __LDBL_DENORM_MIN__
 
 #define FP_NAN       0
 #define FP_INFINITE  1
diff --git a/libc/mem/gc.c b/libc/mem/gc.c
index b977d0964..07957ba57 100644
--- a/libc/mem/gc.c
+++ b/libc/mem/gc.c
@@ -51,7 +51,7 @@ static void TeardownGc(void) {
   }
 }
 
-__attribute__((__constructor__)) static void InitializeGc(void) {
+__attribute__((__constructor__(51))) static textstartup void InitGc(void) {
   atexit(TeardownGc);
 }
 
diff --git a/libc/mem/heapsort.c b/libc/mem/heapsort.c
index 31645308a..9577c3891 100644
--- a/libc/mem/heapsort.c
+++ b/libc/mem/heapsort.c
@@ -31,6 +31,7 @@
 #include "libc/mem/alg.h"
 #include "libc/mem/mem.h"
 #include "libc/sysv/errfuns.h"
+__static_yoink("openbsd_sorting_notice");
 // clang-format off
 
 /*
diff --git a/libc/mem/mergesort.c b/libc/mem/mergesort.c
index 70344f552..400b6dfe3 100644
--- a/libc/mem/mergesort.c
+++ b/libc/mem/mergesort.c
@@ -33,13 +33,9 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 #include "libc/sysv/errfuns.h"
+__static_yoink("openbsd_sorting_notice");
 // clang-format off
 
-asm(".ident\t\"\\n\\n\
-OpenBSD Sorting (BSD-3)\\n\
-Copyright 1993 The Regents of the University of California\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /*
  * Hybrid exponential search/linear search merge sort with hybrid
  * natural/pairwise first pass.  Requires about .3% more comparisons
diff --git a/libc/mem/opensort.c b/libc/mem/opensort.c
new file mode 100644
index 000000000..36f061dfd
--- /dev/null
+++ b/libc/mem/opensort.c
@@ -0,0 +1,2 @@
+__notice(openbsd_sorting_notice, "OpenBSD Sorting (BSD-3)\n\
+Copyright 1993 The Regents of the University of California");
diff --git a/libc/mem/qsort.c b/libc/mem/qsort.c
index ca96a532a..361f26a86 100644
--- a/libc/mem/qsort.c
+++ b/libc/mem/qsort.c
@@ -31,13 +31,9 @@
 #include "libc/macros.internal.h"
 #include "libc/mem/alg.h"
 #include "libc/str/str.h"
+__static_yoink("openbsd_sorting_notice");
 // clang-format off
 
-asm(".ident\t\"\\n\\n\
-OpenBSD Sorting (BSD-3)\\n\
-Copyright 1993 The Regents of the University of California\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #define SWAPTYPE_BYTEV	1
 #define SWAPTYPE_INTV	2
 #define SWAPTYPE_LONGV	3
diff --git a/libc/mem/realpath.c b/libc/mem/realpath.c
index bba846b7d..3ea1432a5 100644
--- a/libc/mem/realpath.c
+++ b/libc/mem/realpath.c
@@ -28,20 +28,17 @@
 #include "libc/assert.h"
 #include "libc/calls/calls.h"
 #include "libc/errno.h"
-#include "libc/serialize.h"
 #include "libc/intrin/safemacros.internal.h"
 #include "libc/limits.h"
 #include "libc/log/backtrace.internal.h"
 #include "libc/mem/mem.h"
+#include "libc/serialize.h"
 #include "libc/str/str.h"
 #include "libc/sysv/errfuns.h"
+__static_yoink("musl_libc_notice");
 
 #define SYMLOOP_MAX 40
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 static size_t GetSlashLen(const char *s)
diff --git a/libc/nexgen32e/argc.S b/libc/nexgen32e/argc.S
index 3dc3852c2..9e85f4409 100644
--- a/libc/nexgen32e/argc.S
+++ b/libc/nexgen32e/argc.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 	.initbss 300,_init_argc
 //	Global variable holding _start(argc) parameter.
diff --git a/libc/nexgen32e/argv.S b/libc/nexgen32e/argv.S
index a3b95116b..9ee093476 100644
--- a/libc/nexgen32e/argv.S
+++ b/libc/nexgen32e/argv.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 	.initbss 300,_init_argv
 //	Global variable holding _start(argv) parameter.
diff --git a/libc/nexgen32e/auxv.S b/libc/nexgen32e/auxv.S
index bc81ef64c..0c921df7c 100644
--- a/libc/nexgen32e/auxv.S
+++ b/libc/nexgen32e/auxv.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 	.initbss 300,_init_auxv
 //	Global variable holding _start(auxv) parameter.
diff --git a/libc/nexgen32e/environ.S b/libc/nexgen32e/environ.S
index bf368d773..d1419a52c 100644
--- a/libc/nexgen32e/environ.S
+++ b/libc/nexgen32e/environ.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 //	Environment variable pointer list.
 	.bss
diff --git a/libc/nexgen32e/gc.S b/libc/nexgen32e/gc.S
index 037ce8b31..302dcd5a2 100644
--- a/libc/nexgen32e/gc.S
+++ b/libc/nexgen32e/gc.S
@@ -18,7 +18,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
 #include "libc/dce.h"
-#include "libc/notice.inc"
 
 	nop
 
@@ -37,7 +36,7 @@ __gc:	.ftrace2
 
 #ifdef __x86_64__
 
-	mov	%fs:0,%rcx			// __get_tls()
+	mov	%gs:0x30,%rcx			// __get_tls()
 	mov	0x18(%rcx),%rcx			// tls::garbages
 	decl	(%rcx)				// --g->i
 	mov	(%rcx),%r8d			// r8 = g->i
diff --git a/libc/nexgen32e/gclongjmp.S b/libc/nexgen32e/gclongjmp.S
index 18f092000..1fb68131b 100644
--- a/libc/nexgen32e/gclongjmp.S
+++ b/libc/nexgen32e/gclongjmp.S
@@ -35,7 +35,7 @@ gclongjmp:
 #ifdef __x86_64__
 	push	%rbp
 	mov	%rsp,%rbp
-	mov	%fs:0,%r12			// __get_tls()
+	mov	%gs:0x30,%r12			// __get_tls()
 	mov	0x18(%r12),%r12			// Tls::garbages
 	test	%r12,%r12
 	jz	0f
diff --git a/libc/nexgen32e/identity.S b/libc/nexgen32e/identity.S
index af26136a2..7fc23e4d8 100644
--- a/libc/nexgen32e/identity.S
+++ b/libc/nexgen32e/identity.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 //	The identity() function.
 //	@return	first argument
diff --git a/libc/nexgen32e/sha1.S b/libc/nexgen32e/sha1.S
index 24fb18306..1016c0498 100644
--- a/libc/nexgen32e/sha1.S
+++ b/libc/nexgen32e/sha1.S
@@ -33,10 +33,11 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
 
-.ident "\n\
-AVX2 SHA-1 (BSD-3 License)\n\
-Copyright 2014 Intel Corporation\n"
-.include "libc/disclaimer.inc"
+	.section .notice,"aR",@progbits
+	.asciz	"\n\n\
+AVX2 SHA-1 (BSD-3 License)n\
+Copyright 2014 Intel Corporation"
+	.previous
 
 #define	CTX	%rdi	/* arg1 */
 #define BUF	%rsi	/* arg2 */
diff --git a/libc/nexgen32e/sha1ni.S b/libc/nexgen32e/sha1ni.S
index f1e577d22..223f5f25d 100644
--- a/libc/nexgen32e/sha1ni.S
+++ b/libc/nexgen32e/sha1ni.S
@@ -33,14 +33,15 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
 
-.text
-.balign 32
-.ident "\n\
+.section .notice,"aR",@progbits
+.asciz "\n\n\
 Intel SHA-NI (BSD-3 License)\n\
 Copyright 2015 Intel Corporation\n\
 Sean Gulley <sean.m.gulley@intel.com>\n\
-Tim Chen <tim.c.chen@linux.intel.com>\n"
-.include "libc/disclaimer.inc"
+Tim Chen <tim.c.chen@linux.intel.com>"
+
+.text
+.balign 32
 
 #define FRAME_SIZE	32
 #define DIGEST_PTR	%rdi
diff --git a/libc/nexgen32e/sha256.S b/libc/nexgen32e/sha256.S
index 3858101dd..df175bf5b 100644
--- a/libc/nexgen32e/sha256.S
+++ b/libc/nexgen32e/sha256.S
@@ -49,10 +49,11 @@
 /////////////////////////////////////////////////////////////////////////
 #include "libc/macros.internal.h"
 
-.ident "\n\
+.section .notice,"aR",@progbits
+.asciz "\n\n\
 AVX2 SHA2 (BSD-2 License)\n\
-Copyright 2013 Intel Corporation\n"
-.include "libc/disclaimer.inc"
+Copyright 2013 Intel Corporation"
+.previous
 
 ## assume buffers not aligned
 #define	VMOVDQ vmovdqu
diff --git a/libc/nexgen32e/sha256ni.S b/libc/nexgen32e/sha256ni.S
index d04857550..eb020d706 100644
--- a/libc/nexgen32e/sha256ni.S
+++ b/libc/nexgen32e/sha256ni.S
@@ -33,14 +33,15 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
 
-.text
-.balign 32
-.ident "\n\
+.section .notice,"aR",@progbits
+.asciz "\n\n\
 Intel SHA-NI (BSD-3 License)\n\
 Copyright 2015 Intel Corporation\n\
 Sean Gulley <sean.m.gulley@intel.com>\n\
-Tim Chen <tim.c.chen@linux.intel.com>\n"
-.include "libc/disclaimer.inc"
+Tim Chen <tim.c.chen@linux.intel.com>"
+
+.text
+.balign 32
 
 #define DIGEST_PTR	%rdi	/* 1st arg */
 #define DATA_PTR	%rsi	/* 2nd arg */
diff --git a/libc/nexgen32e/sha512.S b/libc/nexgen32e/sha512.S
index b858a44d9..6e36d6d1b 100644
--- a/libc/nexgen32e/sha512.S
+++ b/libc/nexgen32e/sha512.S
@@ -50,10 +50,11 @@
 /////////////////////////////////////////////////////////////////////////
 #include "libc/macros.internal.h"
 
-.ident "\n\
-AVX2 SHA2 (BSD-2 License)\n\
-Copyright 2013 Intel Corporation\n"
-.include "libc/disclaimer.inc"
+.section .notice,"aR",@progbits
+.asciz "\n\n\
+AVX2 SHA512 (BSD-2 License)\n\
+Copyright 2013 Intel Corporation"
+.previous
 
 # Virtual Registers
 Y_0 = %ymm4
diff --git a/libc/notice.inc b/libc/notice.inc
deleted file mode 100644
index 5b9c68964..000000000
--- a/libc/notice.inc
+++ /dev/null
@@ -1,17 +0,0 @@
-.ident "\n\
-Cosmopolitan\n\
-Copyright 2020 Justine Alexandra Roberts Tunney\n\
-\n\
-Permission to use, copy, modify, and/or distribute this software for\n\
-any purpose with or without fee is hereby granted, provided that the\n\
-above copyright notice and this permission notice appear in all copies.\n\
-\n\
-THE SOFTWARE IS PROVIDED \"AS IS\" AND THE AUTHOR DISCLAIMS ALL\n\
-WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED\n\
-WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE\n\
-AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL\n\
-DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR\n\
-PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER\n\
-TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR\n\
-PERFORMANCE OF THIS SOFTWARE.\
-"
diff --git a/libc/notice.internal.h b/libc/notice.internal.h
deleted file mode 100644
index 4bb9ab304..000000000
--- a/libc/notice.internal.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef COSMOPOLITAN_LIBC_INTERNAL_NOTICE_H_
-#define COSMOPOLITAN_LIBC_INTERNAL_NOTICE_H_
-#ifndef __STRICT_ANSI__
-
-#ifdef __ASSEMBLER__
-.include "libc/notice.inc"
-#else
-asm(".include \"libc/notice.inc\"");
-#endif
-
-#endif /* !ANSI */
-#endif /* COSMOPOLITAN_LIBC_INTERNAL_NOTICE_H_ */
diff --git a/libc/nt/privilege.h b/libc/nt/privilege.h
index 95088a34a..d281a330a 100644
--- a/libc/nt/privilege.h
+++ b/libc/nt/privilege.h
@@ -46,5 +46,8 @@ bool32 AdjustTokenPrivileges(int64_t TokenHandle, bool32 DisableAllPrivileges,
 bool32 ImpersonateSelf(int kNtSecurityImpersonationLevel);
 bool32 RevertToSelf(void);
 
+bool32 OpenThreadToken(intptr_t ThreadHandle, uint32_t DesiredAccess,
+                       bool32 OpenAsSelf, intptr_t *TokenHandle);
+
 COSMOPOLITAN_C_END_
 #endif /* COSMOPOLITAN_LIBC_NT_PRIVILEGE_H_ */
diff --git a/libc/proc/execve-sysv.c b/libc/proc/execve-sysv.c
index 04a40f861..aa638aac6 100644
--- a/libc/proc/execve-sysv.c
+++ b/libc/proc/execve-sysv.c
@@ -98,7 +98,7 @@ static void SetupExecve(void) {
   g_execve.tmpdir = getenv("TMPDIR");
 }
 
-__attribute__((__constructor__)) static void InitExecve(void) {
+__attribute__((__constructor__(10))) static textstartup void InitExecve(void) {
   cosmo_once(&g_execve.once, SetupExecve);
 }
 
diff --git a/libc/proc/vfork.S b/libc/proc/vfork.S
index ff701c1c5..d43faf4cf 100644
--- a/libc/proc/vfork.S
+++ b/libc/proc/vfork.S
@@ -61,7 +61,7 @@ vfork:
 #endif
 	pop	%rbp
 #endif
-	mov	%fs:0,%r9		// get thread information block
+	mov	%gs:0x30,%r9		// get thread information block
 	mov	0x3c(%r9),%r8d		// avoid question of @vforksafe errno
 	pop	%rsi			// saves return address in a register
 	mov	__NR_vfork(%rip),%eax
diff --git a/libc/runtime/clone.c b/libc/runtime/clone.c
index cb1d0733f..7a9913ccb 100644
--- a/libc/runtime/clone.c
+++ b/libc/runtime/clone.c
@@ -400,9 +400,7 @@ static wontreturn void FreebsdThreadMain(void *p) {
 #ifdef __aarch64__
   asm volatile("mov\tx28,%0" : /* no outputs */ : "r"(wt->tls));
 #elif defined(__x86_64__)
-  if (__tls_morphed) {
-    sys_set_tls(AMD64_SET_GSBASE, wt->tls);
-  }
+  sys_set_tls(AMD64_SET_GSBASE, wt->tls);
 #endif
   *wt->ctid = wt->tid;
   wt->func(wt->arg, wt->tid);
@@ -575,7 +573,7 @@ static int CloneLinux(int (*func)(void *arg, int rc), char *stk, size_t stksz,
 #endif
   wt = (struct LinuxCloneArgs *)sp;
 #ifdef __x86_64__
-  if ((flags & CLONE_SETTLS) && __tls_morphed) {
+  if (flags & CLONE_SETTLS) {
     flags &= ~CLONE_SETTLS;
     wt->arg = arg;
     wt->tls = tls;
diff --git a/libc/runtime/cosmo.S b/libc/runtime/cosmo.S
index e8d1d9fb1..31ee018a9 100644
--- a/libc/runtime/cosmo.S
+++ b/libc/runtime/cosmo.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 #include "libc/sysv/consts/prot.h"
 #include "libc/sysv/consts/map.h"
 #include "libc/intrin/strace.internal.h"
@@ -85,19 +84,18 @@ cosmo:	push	%rbp
 	call	_init
 
 //	call constructors
-	ezlea	__init_array_end,ax		// static ctors in forward order
-	.weak	__init_array_end		// could be called multiple times
-	ezlea	__init_array_start,cx		// idempotency recommended
-	.weak	__init_array_start		// @see ape/ape.lds
-1:	cmp	%rax,%rcx
+	.weak	__init_array_end
+	.weak	__init_array_start
+	mov	$__init_array_start,%eax
+1:	cmp	$__init_array_end,%eax
 	je	2f
-	sub	$8,%rax
 	push	%rax
-	push	%rcx
+	push	%rax
 	call	.Largs
 	call	*(%rax)
-	pop	%rcx
 	pop	%rax
+	pop	%rax
+	add	$8,%eax
 	jmp	1b
 
 //	call main()
diff --git a/libc/runtime/cosmo2.c b/libc/runtime/cosmo2.c
index f01c63720..8181bb9b3 100644
--- a/libc/runtime/cosmo2.c
+++ b/libc/runtime/cosmo2.c
@@ -59,8 +59,6 @@ extern char syscon_openbsd[];
 extern char syscon_netbsd[];
 extern char syscon_windows[];
 extern init_f __strace_init;
-extern init_f *__preinit_array_start[] __attribute__((__weak__));
-extern init_f *__preinit_array_end[] __attribute__((__weak__));
 extern init_f *__init_array_start[] __attribute__((__weak__));
 extern init_f *__init_array_end[] __attribute__((__weak__));
 extern char ape_stack_prot[] __attribute__((__weak__));
@@ -189,7 +187,7 @@ wontreturn textstartup void cosmo(long *sp, struct Syslib *m1, char *exename,
 #if SYSDEBUG
   argc = __strace_init(argc, argv, envp, auxv);
 #endif
-  for (init_f **fp = __init_array_end; fp-- > __init_array_start;) {
+  for (init_f **fp = __init_array_start; fp < __init_array_end; ++fp) {
     (*fp)(argc, argv, envp, auxv);
   }
 #ifdef FTRACE
diff --git a/libc/runtime/dsohandle.S b/libc/runtime/dsohandle.S
index 6d344c37e..39cc3e989 100644
--- a/libc/runtime/dsohandle.S
+++ b/libc/runtime/dsohandle.S
@@ -17,7 +17,6 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/macros.internal.h"
-#include "libc/notice.inc"
 
 	.underrun
 //	Uniquely identifies each artifact linked in an address space.
diff --git a/libc/runtime/enable_tls.c b/libc/runtime/enable_tls.c
index 5395b6d6f..045740baf 100644
--- a/libc/runtime/enable_tls.c
+++ b/libc/runtime/enable_tls.c
@@ -251,7 +251,7 @@ textstartup void __enable_tls(void) {
 
 #ifdef __x86_64__
   // rewrite the executable tls opcodes in memory
-  if (IsWindows() || IsXnu()) {
+  if (IsWindows() || IsOpenbsd() || IsNetbsd()) {
     __morph_tls();
   }
 #endif
diff --git a/libc/runtime/fenv.h b/libc/runtime/fenv.h
index ac9b33c47..4f25de55a 100644
--- a/libc/runtime/fenv.h
+++ b/libc/runtime/fenv.h
@@ -59,7 +59,9 @@ typedef double fenv_t;
 
 #endif /* __x86_64__ */
 
-#ifdef __FLT_EVAL_METHOD__
+#ifdef __STDC_WANT_IEC_60559_TYPES_EXT__
+#define FLT_EVAL_METHOD __FLT_EVAL_METHOD_TS_18661_3__
+#elif defined(__FLT_EVAL_METHOD__)
 #define FLT_EVAL_METHOD __FLT_EVAL_METHOD__
 #else
 #define FLT_EVAL_METHOD 0
diff --git a/libc/runtime/morph_tls.c b/libc/runtime/morph_tls.c
index b780533ed..aef561580 100644
--- a/libc/runtime/morph_tls.c
+++ b/libc/runtime/morph_tls.c
@@ -17,6 +17,7 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "ape/sections.internal.h"
+#include "libc/intrin/kprintf.h"
 #include "libc/runtime/internal.h"
 #include "libc/runtime/runtime.h"
 #include "libc/serialize.h"
@@ -49,68 +50,105 @@ privileged void __morph_tls(void) {
   unsigned char *p;
   __morph_begin();
 
-  if (IsXnu()) {
-    // Apple is quite straightforward to patch. We basically
-    // just change the segment register, and the linear slot
-    // address 0x30 was promised to us, according to Go team
-    // https://github.com/golang/go/issues/23617
-    dis = 0x30;
-  } else if (IsWindows()) {
+  if (IsWindows()) {
     // MSVC __declspec(thread) generates binary code for this
     // %gs:0x1480 abi. So long as TlsAlloc() isn't called >64
     // times we should be good.
     dis = 0x1480 + __tls_index * 8;
-  } else {
-    dis = 0;
-  }
 
-  // iterate over modifiable code looking for 9 byte instruction
-  // this used to take 30ms with xed to enable tls on python.com
-  for (p = _ereal; p + 9 <= __privileged_start; p += n) {
+    // iterate over modifiable code looking for 9 byte instruction
+    // this used to take 30ms with xed to enable tls on python.com
+    for (p = _ereal; p + 9 <= __privileged_start; p += n) {
 
-    // use sse to zoom zoom to fs register prefixes
-    // that way it'll take 1 ms to morph python.com
-    // we recompiled a 13mb binary in 1 millisecond
-    while (p + 9 + 16 <= __privileged_start) {
-      if ((m = __builtin_ia32_pmovmskb128(
-               *(xmm_t *)p == (xmm_t){0144, 0144, 0144, 0144, 0144, 0144, 0144,
-                                      0144, 0144, 0144, 0144, 0144, 0144, 0144,
-                                      0144, 0144}))) {
-        m = __builtin_ctzll(m);
-        p += m;
-        break;
+      // use sse to zoom zoom to fs register prefixes
+      // that way it'll take 1 ms to morph python.com
+      // we recompiled a 13mb binary in 1 millisecond
+      while (p + 9 + 16 <= __privileged_start) {
+        if ((m = __builtin_ia32_pmovmskb128(
+                 *(xmm_t *)p == (xmm_t){0145, 0145, 0145, 0145, 0145, 0145,
+                                        0145, 0145, 0145, 0145, 0145, 0145,
+                                        0145, 0145, 0145, 0145}))) {
+          m = __builtin_ctzll(m);
+          p += m;
+          break;
+        } else {
+          p += 16;
+        }
+      }
+
+      // we're checking for the following expression:
+      //   0145 == p[0] &&           // %gs
+      //   0110 == (p[1] & 0373) &&  // rex.w (and ignore rex.r)
+      //   (0213 == p[2] ||          // mov reg/mem → reg (word-sized)
+      //   0003 == p[2]) &&          // add reg/mem → reg (word-sized)
+      //   0004 == (p[3] & 0307) &&  // mod/rm (4,reg,0) means sib → reg
+      //   0x30 == p[4] &&           // sib (5,4,0) → (rbp,rsp,0) → disp32
+      //   0000 == p[5] &&           // displacement (von Neumann endian)
+      //   0000 == p[6] &&           // displacement
+      //   0000 == p[7] &&           // displacement
+      //   0000 == p[8]              // displacement
+      w = READ64LE(p) & READ64LE("\377\373\377\307\377\377\377\377");
+      if ((w == READ64LE("\145\110\213\004\045\060\000\000") ||
+           w == READ64LE("\145\110\003\004\045\060\000\000")) &&
+          !p[8]) {
+
+        // now change the code
+        p[5] = (dis & 0x000000ff) >> 000;  // displacement
+        p[6] = (dis & 0x0000ff00) >> 010;  // displacement
+        p[7] = (dis & 0x00ff0000) >> 020;  // displacement
+        p[8] = (dis & 0xff000000) >> 030;  // displacement
+
+        // advance to the next instruction
+        n = 9;
       } else {
-        p += 16;
+        n = 1;
       }
     }
+  } else {
+    // iterate over modifiable code looking for 9 byte instruction
+    // this used to take 30ms with xed to enable tls on python.com
+    for (p = _ereal; p + 9 <= __privileged_start; p += n) {
 
-    // we're checking for the following expression:
-    //   0144 == p[0] &&           // %fs
-    //   0110 == (p[1] & 0373) &&  // rex.w (and ignore rex.r)
-    //   (0213 == p[2] ||          // mov reg/mem → reg (word-sized)
-    //   0003 == p[2]) &&          // add reg/mem → reg (word-sized)
-    //   0004 == (p[3] & 0307) &&  // mod/rm (4,reg,0) means sib → reg
-    //   0045 == p[4] &&           // sib (5,4,0) → (rbp,rsp,0) → disp32
-    //   0000 == p[5] &&           // displacement (von Neumann endian)
-    //   0000 == p[6] &&           // displacement
-    //   0000 == p[7] &&           // displacement
-    //   0000 == p[8]              // displacement
-    w = READ64LE(p) & READ64LE("\377\373\377\307\377\377\377\377");
-    if ((w == READ64LE("\144\110\213\004\045\000\000\000") ||
-         w == READ64LE("\144\110\003\004\045\000\000\000")) &&
-        !p[8]) {
+      // use sse to zoom zoom to fs register prefixes
+      // that way it'll take 1 ms to morph python.com
+      // we recompiled a 13mb binary in 1 millisecond
+      while (p + 9 + 16 <= __privileged_start) {
+        if ((m = __builtin_ia32_pmovmskb128(
+                 *(xmm_t *)p == (xmm_t){0145, 0145, 0145, 0145, 0145, 0145,
+                                        0145, 0145, 0145, 0145, 0145, 0145,
+                                        0145, 0145, 0145, 0145}))) {
+          m = __builtin_ctzll(m);
+          p += m;
+          break;
+        } else {
+          p += 16;
+        }
+      }
 
-      // now change the code
-      p[0] = 0145;                       // change %fs to %gs
-      p[5] = (dis & 0x000000ff) >> 000;  // displacement
-      p[6] = (dis & 0x0000ff00) >> 010;  // displacement
-      p[7] = (dis & 0x00ff0000) >> 020;  // displacement
-      p[8] = (dis & 0xff000000) >> 030;  // displacement
+      // we're checking for the following expression:
+      //   0145 == p[0] &&           // %gs
+      //   0110 == (p[1] & 0373) &&  // rex.w (and ignore rex.r)
+      //   (0213 == p[2] ||          // mov reg/mem → reg (word-sized)
+      //   0003 == p[2]) &&          // add reg/mem → reg (word-sized)
+      //   0004 == (p[3] & 0307) &&  // mod/rm (4,reg,0) means sib → reg
+      //   0x30 == p[4] &&           // sib (5,4,0) → (rbp,rsp,0) → disp32
+      //   0000 == p[5] &&           // displacement (von Neumann endian)
+      //   0000 == p[6] &&           // displacement
+      //   0000 == p[7] &&           // displacement
+      //   0000 == p[8]              // displacement
+      w = READ64LE(p) & READ64LE("\377\373\377\307\377\377\377\377");
+      if ((w == READ64LE("\145\110\213\004\045\060\000\000") ||
+           w == READ64LE("\145\110\003\004\045\060\000\000")) &&
+          !p[8]) {
 
-      // advance to the next instruction
-      n = 9;
-    } else {
-      n = 1;
+        // now change the code
+        p[0] = 0144;  // change %gs to %fs
+
+        // advance to the next instruction
+        n = 9;
+      } else {
+        n = 1;
+      }
     }
   }
 
diff --git a/libc/runtime/set_tls.c b/libc/runtime/set_tls.c
index 29972f484..0f54e8d12 100644
--- a/libc/runtime/set_tls.c
+++ b/libc/runtime/set_tls.c
@@ -39,10 +39,10 @@ dontinstrument textstartup void __set_tls(struct CosmoTib *tib) {
   // ask the operating system to change the x86 segment register
   if (IsWindows()) {
     asm("mov\t%1,%%gs:%0" : "=m"(*((long *)0x1480 + __tls_index)) : "r"(tib));
-  } else if (IsFreebsd()) {
-    sys_set_tls(__tls_morphed ? AMD64_SET_GSBASE : AMD64_SET_FSBASE, tib);
   } else if (IsLinux()) {
-    sys_set_tls(__tls_morphed ? ARCH_SET_GS : ARCH_SET_FS, tib);
+    sys_set_tls(ARCH_SET_GS, tib);
+  } else if (IsFreebsd()) {
+    sys_set_tls(AMD64_SET_GSBASE, tib);
   } else if (IsNetbsd()) {
     // netbsd has sysarch(X86_SET_FSBASE) but we can't use that because
     // signal handlers will cause it to be reset due to not setting the
diff --git a/libc/runtime/zipos-open.c b/libc/runtime/zipos-open.c
index da92c9d37..83b71ed76 100644
--- a/libc/runtime/zipos-open.c
+++ b/libc/runtime/zipos-open.c
@@ -302,7 +302,7 @@ int __zipos_open(struct ZiposUri *name, int flags) {
   return rc;
 }
 
-__attribute__((__constructor__)) static void __zipos_ctor(void) {
+__attribute__((__constructor__(60))) static textstartup void zipos_ctor(void) {
   __zipos_wipe();
   pthread_atfork(__zipos_lock, __zipos_unlock, __zipos_wipe);
 }
diff --git a/libc/sock/epoll.c b/libc/sock/epoll.c
index 73fcf66cd..4c86e35e1 100644
--- a/libc/sock/epoll.c
+++ b/libc/sock/epoll.c
@@ -91,11 +91,10 @@
  * TODO(jart): Polyfill kqueue for XNU/FreeBSD/OpenBSD.
  */
 
-asm(".ident\t\"\\n\\n\
-wepoll (BSD-2)\\n\
-Copyright 2012-2020 Bert Belder\\n\
-https://github.com/piscisaureus/wepoll\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(wepoll_notice, "\
+wepoll (BSD-2)\n\
+Copyright 2012-2020 Bert Belder\n\
+https://github.com/piscisaureus/wepoll");
 
 #define MAX_GROUP_SIZE 32
 
@@ -128,7 +127,7 @@ asm(".include \"libc/disclaimer.inc\"");
   } while (0)
 
 #define CONTAINOF(ptr, type, member) \
-  ((type *)((uintptr_t)(ptr)-offsetof(type, member)))
+  ((type *)((uintptr_t)(ptr) - offsetof(type, member)))
 
 #define TREE__ROTATE(cis, trans)       \
   struct TreeNode *p = node;           \
diff --git a/libc/sock/inet_aton.c b/libc/sock/inet_aton.c
index 98869e463..c826cb955 100644
--- a/libc/sock/inet_aton.c
+++ b/libc/sock/inet_aton.c
@@ -29,11 +29,7 @@
 #include "libc/sock/sock.h"
 #include "libc/sock/struct/sockaddr.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /**
  * Converts dotted IPv4 address string to network order binary.
diff --git a/libc/stdio/demangle.c b/libc/stdio/demangle.c
index 42b532d23..2950ad7ba 100644
--- a/libc/stdio/demangle.c
+++ b/libc/stdio/demangle.c
@@ -34,11 +34,10 @@
 #include "libc/stdio/stdio.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-C++ Demangle (BSD-2)\\n\
-Copyright (c) 2007 Hyogeol Lee <hyogeollee@gmail.com>\\n\
-Copyright (c) 2015-2017 Kai Wang <kaiwang27@gmail.com>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(demangle_notice, "\
+FreeBSD C++ Demangle (BSD-2)\n\
+Copyright (c) 2007 Hyogeol Lee <hyogeollee@gmail.com>\n\
+Copyright (c) 2015-2017 Kai Wang <kaiwang27@gmail.com>");
 // https://github.com/freebsd/freebsd-src/blob/2176c9ab71c85efd90a6c7af4a9e04fe8e3d49ca/contrib/libcxxrt/libelftc_dem_gnu3.c
 // clang-format off
 
diff --git a/libc/stdio/ecvt.c b/libc/stdio/ecvt.c
index cb3324fdc..ebb72075c 100644
--- a/libc/stdio/ecvt.c
+++ b/libc/stdio/ecvt.c
@@ -27,10 +27,9 @@
 #include "libc/str/str.h"
 #include "third_party/gdtoa/gdtoa.h"
 
-asm(".ident\t\"\\n\\n\
-OpenBSD ecvt/gcvt (MIT)\\n\
-Copyright (c) 2002, 2006, 2010 Todd C. Miller <millert@openbsd.org>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(ecvt_notice, "\
+OpenBSD ecvt (MIT)\n\
+Copyright (c) 2002, 2006, 2010 Todd C. Miller <millert@openbsd.org>");
 // clang-format off
 
 static char *s;
@@ -42,7 +41,7 @@ __cvt_atexit(void)
 	s = 0;
 }
 
-static void __attribute__((__constructor__))
+static __attribute__((__constructor__(60))) textstartup void
 __cvt_init(void)
 {
 	atexit(__cvt_atexit);
diff --git a/libc/stdio/fflush_unlocked.c b/libc/stdio/fflush_unlocked.c
index ab889dec9..fd8887d79 100644
--- a/libc/stdio/fflush_unlocked.c
+++ b/libc/stdio/fflush_unlocked.c
@@ -69,7 +69,7 @@ static void __stdio_fork_child(void) {
   pthread_mutex_init(&__fflush_lock_obj, 0);
 }
 
-__attribute__((__constructor__)) static void __stdio_init(void) {
+__attribute__((__constructor__(60))) static textstartup void stdioinit(void) {
   pthread_atfork(__stdio_fork_prepare, __stdio_fork_parent, __stdio_fork_child);
 }
 
diff --git a/libc/stdio/ftw.c b/libc/stdio/ftw.c
index 9440b75f8..ded9820ff 100644
--- a/libc/stdio/ftw.c
+++ b/libc/stdio/ftw.c
@@ -26,11 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/stdio/ftw.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 /**
diff --git a/libc/stdio/gcvt.c b/libc/stdio/gcvt.c
index c0ecbe869..dfc824415 100644
--- a/libc/stdio/gcvt.c
+++ b/libc/stdio/gcvt.c
@@ -28,10 +28,9 @@
 #include "libc/str/unicode.h"
 #include "third_party/gdtoa/gdtoa.h"
 
-asm(".ident\t\"\\n\\n\
-OpenBSD ecvt/gcvt (MIT)\\n\
-Copyright (c) 2002, 2006, 2010 Todd C. Miller <millert@openbsd.org>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(gcvt_notice, "\
+OpenBSD gcvt (MIT)\n\
+Copyright (c) 2002, 2006, 2010 Todd C. Miller <millert@openbsd.org>");
 // clang-format off
 
 #define DEFPREC	6
diff --git a/libc/stdio/iconv.c b/libc/stdio/iconv.c
index 791d73277..3b04b6b8d 100644
--- a/libc/stdio/iconv.c
+++ b/libc/stdio/iconv.c
@@ -32,11 +32,8 @@
 #include "libc/str/str.h"
 #include "libc/thread/tls.h"
 // clang-format off
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 
 #define UTF_32BE    0300
 #define UTF_16LE    0301
diff --git a/libc/stdio/mt19937.c b/libc/stdio/mt19937.c
index 3ef4e8362..8a4b8b8d2 100644
--- a/libc/stdio/mt19937.c
+++ b/libc/stdio/mt19937.c
@@ -38,10 +38,8 @@
 #include "libc/macros.internal.h"
 #include "libc/stdio/rand.h"
 
-asm(".ident\t\"\\n\\n\
-mt19937 (BSD-3)\\n\
-Copyright 1997-2004 Makoto Matsumoto and Takuji Nishimura\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(mt19937_notice, "mt19937 (BSD-3)\n\
+Copyright 1997-2004 Makoto Matsumoto and Takuji Nishimura");
 
 /*
  * A C-program for MT19937-64 (2004/9/29 version).
diff --git a/libc/stdio/nftw.c b/libc/stdio/nftw.c
index 35c84b0cb..53c5ca2da 100644
--- a/libc/stdio/nftw.c
+++ b/libc/stdio/nftw.c
@@ -36,11 +36,8 @@
 #include "libc/sysv/consts/o.h"
 #include "libc/sysv/consts/s.h"
 #include "libc/thread/thread.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 struct history
diff --git a/libc/stdio/random.c b/libc/stdio/random.c
index 513300d76..aa78e31c9 100644
--- a/libc/stdio/random.c
+++ b/libc/stdio/random.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/stdio/rand.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /*
  * this code uses the same lagged fibonacci generator as the
diff --git a/libc/stdio/scandir.c b/libc/stdio/scandir.c
index 07973bcc6..e9be6e664 100644
--- a/libc/stdio/scandir.c
+++ b/libc/stdio/scandir.c
@@ -31,11 +31,8 @@
 #include "libc/mem/alg.h"
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 int scandir(const char *path, struct dirent ***res,
diff --git a/libc/stdio/stderr.c b/libc/stdio/stderr.c
index 9065dd114..444a3ed0c 100644
--- a/libc/stdio/stderr.c
+++ b/libc/stdio/stderr.c
@@ -30,7 +30,7 @@ FILE *stderr;
 
 static FILE __stderr;
 
-__attribute__((__constructor__)) static void __stderr_init(void) {
+__attribute__((__constructor__(60))) static textstartup void errinit(void) {
   stderr = &__stderr;
   stderr->fd = STDERR_FILENO;
   stderr->bufmode = _IONBF;
diff --git a/libc/stdio/stdin.c b/libc/stdio/stdin.c
index 0c08324d8..2407187fe 100644
--- a/libc/stdio/stdin.c
+++ b/libc/stdio/stdin.c
@@ -30,7 +30,7 @@ FILE *stdin;
 
 static FILE __stdin;
 
-__attribute__((__constructor__)) static void __stdin_init(void) {
+__attribute__((__constructor__(60))) static textstartup void initin(void) {
   stdin = &__stdin;
   stdin->fd = STDIN_FILENO;
   stdin->iomode = O_RDONLY;
diff --git a/libc/stdio/stdout.c b/libc/stdio/stdout.c
index 3de0d80f3..06eee4475 100644
--- a/libc/stdio/stdout.c
+++ b/libc/stdio/stdout.c
@@ -31,7 +31,7 @@ FILE *stdout;
 
 static FILE __stdout;
 
-__attribute__((__constructor__)) static void __stdout_init(void) {
+__attribute__((__constructor__(60))) static textstartup void outinit(void) {
   stdout = &__stdout;
 
   stdout->fd = STDOUT_FILENO;
diff --git a/libc/str/blake2.c b/libc/str/blake2.c
index 44e0ebc3b..362a198da 100644
--- a/libc/str/blake2.c
+++ b/libc/str/blake2.c
@@ -23,10 +23,9 @@
 
 #define ROR(v, n) (((v) >> (n)) | ((v) << (64 - (n))))
 
-asm(".ident\t\"\\n\\n\
-boringssl blake2b (ISC License)\\n\
-Copyright 2021 Google LLC\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(blake2b_notice, "\
+boringssl blake2b (ISC License)\n\
+Copyright 2021 Google LLC");
 
 // https://tools.ietf.org/html/rfc7693#section-2.6
 static const uint64_t kIV[8] = {
diff --git a/libc/str/btowc.c b/libc/str/btowc.c
index e0faaf8e0..4e3cb74ab 100644
--- a/libc/str/btowc.c
+++ b/libc/str/btowc.c
@@ -29,11 +29,7 @@
 #include "libc/stdio/stdio.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 wint_t btowc(int c) {
   int b = (unsigned char)c;
diff --git a/libc/str/c16rtomb.c b/libc/str/c16rtomb.c
index 3289411f6..34406401e 100644
--- a/libc/str/c16rtomb.c
+++ b/libc/str/c16rtomb.c
@@ -30,11 +30,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t c16rtomb(char *restrict s, char16_t c16, mbstate_t *restrict ps) {
   static unsigned internal_state;
diff --git a/libc/str/highwayhash64.c b/libc/str/highwayhash64.c
index 7cb0af034..5cd7581a3 100644
--- a/libc/str/highwayhash64.c
+++ b/libc/str/highwayhash64.c
@@ -18,10 +18,9 @@
 #include "libc/str/highwayhash64.h"
 #include "libc/serialize.h"
 
-asm(".ident\t\"\\n\\n\
-HighwayHash (Apache 2.0)\\n\
-Copyright 2017 Google LLC\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(highwayhash_notice, "\
+HighwayHash (Apache 2.0)\n\
+Copyright 2017 Google LLC");
 
 typedef struct {
   uint64_t v0[4];
diff --git a/libc/str/langinfo.c b/libc/str/langinfo.c
index d66386082..fe88cd54a 100644
--- a/libc/str/langinfo.c
+++ b/libc/str/langinfo.c
@@ -29,11 +29,8 @@
 #include "libc/str/locale.h"
 #include "libc/str/nltypes.h"
 #include "libc/thread/tls.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 static const char c_time[] =
diff --git a/libc/str/mb.c b/libc/str/mb.c
index 3838b676c..98cbf47dd 100644
--- a/libc/str/mb.c
+++ b/libc/str/mb.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/str/mb.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 #define C(x) (x < 2 ? -1 : (R(0x80, 0xc0) | x))
 #define D(x) C((x + 16))
diff --git a/libc/str/mbrtoc16.c b/libc/str/mbrtoc16.c
index 1a41aa9f7..40d366118 100644
--- a/libc/str/mbrtoc16.c
+++ b/libc/str/mbrtoc16.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t mbrtoc16(char16_t *pc16, const char *s, size_t n, mbstate_t *ps) {
   static unsigned internal_state;
diff --git a/libc/str/mbrtoc32.c b/libc/str/mbrtoc32.c
index 5cfca124b..00cc13ba7 100644
--- a/libc/str/mbrtoc32.c
+++ b/libc/str/mbrtoc32.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/macros.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t mbrtoc32(char32_t *pc32, const char *s, size_t n, mbstate_t *ps) {
   static unsigned internal_state;
diff --git a/libc/str/mbrtowc.c b/libc/str/mbrtowc.c
index 1d89eb7b9..8a0f996ef 100644
--- a/libc/str/mbrtowc.c
+++ b/libc/str/mbrtowc.c
@@ -30,11 +30,7 @@
 #include "libc/macros.internal.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t mbrtowc(wchar_t *wc, const char *src, size_t n, mbstate_t *st) {
   static unsigned internal_state;
diff --git a/libc/str/mbsnrtowcs.c b/libc/str/mbsnrtowcs.c
index d3dd1499a..504a81c45 100644
--- a/libc/str/mbsnrtowcs.c
+++ b/libc/str/mbsnrtowcs.c
@@ -30,11 +30,7 @@
 #include "libc/macros.internal.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t mbsnrtowcs(wchar_t *wcs, const char **src, size_t n, size_t wn,
                   mbstate_t *st) {
diff --git a/libc/str/mbsrtowcs.c b/libc/str/mbsrtowcs.c
index b3aa5f0ae..986172725 100644
--- a/libc/str/mbsrtowcs.c
+++ b/libc/str/mbsrtowcs.c
@@ -30,11 +30,7 @@
 #include "libc/macros.internal.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t mbsrtowcs(wchar_t *ws, const char **src, size_t wn, mbstate_t *st) {
   const unsigned char *s = (const void *)*src;
diff --git a/libc/str/mbtowc.c b/libc/str/mbtowc.c
index 669dcaf47..155270726 100644
--- a/libc/str/mbtowc.c
+++ b/libc/str/mbtowc.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int mbtowc(wchar_t *restrict wc, const char *restrict src, size_t n) {
   unsigned c;
diff --git a/libc/str/smoothsort.c b/libc/str/smoothsort.c
index e9cf02b26..ba84be2e4 100644
--- a/libc/str/smoothsort.c
+++ b/libc/str/smoothsort.c
@@ -26,11 +26,10 @@
 #include "libc/mem/alg.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-Smoothsort (MIT License)\\n\
-Copyright 2011 Valentin Ochs\\n\
-Discovered by Edsger Dijkstra\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(smoothsort_notice, "\
+Smoothsort (MIT License)\n\
+Copyright 2011 Valentin Ochs\n\
+Discovered by Edsger Dijkstra");
 
 typedef int (*cmpfun)(const void *, const void *, void *);
 
diff --git a/libc/str/strlcat.c b/libc/str/strlcat.c
index 445ec2742..087376cc7 100644
--- a/libc/str/strlcat.c
+++ b/libc/str/strlcat.c
@@ -19,11 +19,7 @@
 #include "libc/str/str.h"
 // clang-format off
 // $OpenBSD: strlcat.c,v 1.19 2019/01/25 00:19:25 millert Exp $
-
-asm(".ident\t\"\\n\\n\
-OpenBSD Strings (ISC)\\n\
-Copyright (c) 1998, 2015 Todd C. Miller <millert@openbsd.org>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("openbsd_strings_notice");
 
 /**
  * Appends string, the BSD way.
diff --git a/libc/str/strlcpy.c b/libc/str/strlcpy.c
index 89bbbd8fa..785eaacaf 100644
--- a/libc/str/strlcpy.c
+++ b/libc/str/strlcpy.c
@@ -19,11 +19,7 @@
 #include "libc/str/str.h"
 // clang-format off
 // $OpenBSD: strlcpy.c,v 1.16 2019/01/25 00:19:25 millert Exp $
-
-asm(".ident\t\"\\n\\n\
-OpenBSD Strings (ISC)\\n\
-Copyright (c) 1998, 2015 Todd C. Miller <millert@openbsd.org>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("openbsd_strings_notice");
 
 /**
  * Copies string, the BSD way.
diff --git a/libc/str/strverscmp.c b/libc/str/strverscmp.c
index 0652be54c..3cc3740ef 100644
--- a/libc/str/strverscmp.c
+++ b/libc/str/strverscmp.c
@@ -26,11 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/str/str.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
 // clang-format off
 
 /**
diff --git a/libc/str/timingsafe_memcmp.c b/libc/str/timingsafe_memcmp.c
index a209fd829..dff98b620 100644
--- a/libc/str/timingsafe_memcmp.c
+++ b/libc/str/timingsafe_memcmp.c
@@ -19,10 +19,9 @@
 #include "libc/limits.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-timingsafe_memcmp (ISC License)\\n\
-Copyright 2014 Google Inc.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(timingsafe_memcmp_notice, "\
+timingsafe_memcmp (ISC License)\n\
+Copyright 2014 Google Inc.");
 
 /**
  * Lexicographically compares the first 𝑛 bytes in 𝑝 and 𝑞.
diff --git a/libc/str/todd.c b/libc/str/todd.c
new file mode 100644
index 000000000..133d948d7
--- /dev/null
+++ b/libc/str/todd.c
@@ -0,0 +1,2 @@
+__notice(openbsd_strings_notice, "OpenBSD Strings (ISC)\n\
+Copyright (c) 1998, 2015 Todd C. Miller <millert@openbsd.org>");
diff --git a/libc/str/wcrtomb.c b/libc/str/wcrtomb.c
index 23116b707..0a8f6b155 100644
--- a/libc/str/wcrtomb.c
+++ b/libc/str/wcrtomb.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t wcrtomb(char *s, wchar_t wc, mbstate_t *st) {
   if (!s) return 1;
diff --git a/libc/str/wcsnrtombs.c b/libc/str/wcsnrtombs.c
index ef684d9b0..4486ce579 100644
--- a/libc/str/wcsnrtombs.c
+++ b/libc/str/wcsnrtombs.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t wcsnrtombs(char *dst, const wchar_t **wcs, size_t wn, size_t n,
                   mbstate_t *st) {
diff --git a/libc/str/wcsrtombs.c b/libc/str/wcsrtombs.c
index 013635e39..f1ad160e9 100644
--- a/libc/str/wcsrtombs.c
+++ b/libc/str/wcsrtombs.c
@@ -29,11 +29,7 @@
 #include "libc/limits.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 size_t wcsrtombs(char *s, const wchar_t **ws, size_t n, mbstate_t *st) {
   const wchar_t *ws2;
diff --git a/libc/str/wctob.c b/libc/str/wctob.c
index 62e3ed8a7..bd5ff66b6 100644
--- a/libc/str/wctob.c
+++ b/libc/str/wctob.c
@@ -29,11 +29,7 @@
 #include "libc/stdio/stdio.h"
 #include "libc/str/mb.internal.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int wctob(wint_t c) {
   if (c < 128U) return c;
diff --git a/libc/sysv/errno.c b/libc/sysv/errno.c
index 570f29d5b..438ee9508 100644
--- a/libc/sysv/errno.c
+++ b/libc/sysv/errno.c
@@ -17,7 +17,7 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/errno.h"
-#include "libc/thread/tls.h"
+#include "libc/thread/tls2.internal.h"
 
 /**
  * Global variable for last error.
diff --git a/libc/sysv/systemfive.S b/libc/sysv/systemfive.S
index 02b237817..e858edc54 100644
--- a/libc/sysv/systemfive.S
+++ b/libc/sysv/systemfive.S
@@ -108,7 +108,7 @@ systemfive_cp:
 systemfive_cancellable:			// our pthread_cancel() miracle code
 	cmpb	$0,__tls_enabled(%rip)	// inspired by the musl libc design!
 	je	1f			// we handle linux and bsd together!
-	mov	%fs:0,%r10		// CosmoTib::tib_self
+	mov	%gs:0x30,%r10		// CosmoTib::tib_self
 	mov	0x28(%r10),%r10		// CosmoTib::tib_pthread
 	test	%r10,%r10		// is it a posix thread?
 	jz	1f			// it's spawn() probably
@@ -135,7 +135,7 @@ systemfive_cancellable_end:		// i/o calls park here for long time
 	jne	systemfive_errno	// werent interrupted by OnSigCancel
 	cmpb	$0,__tls_enabled(%rip)	// make sure it's safe to grab %fs:0
 	je	systemfive_errno	// tls is disabled we can't continue
-	mov	%fs:0,%rcx		// CosmoTib::tib_self
+	mov	%gs:0x30,%rcx		// CosmoTib::tib_self
 	mov	0x28(%rcx),%rcx		// CosmoTib::tib_pthread
 	test	%rcx,%rcx		// is it a posix thread?
 	jz	systemfive_errno	// it's spawn() probably
diff --git a/libc/thread/tls.h b/libc/thread/tls.h
index b5ef8b793..87c0b0bc5 100644
--- a/libc/thread/tls.h
+++ b/libc/thread/tls.h
@@ -70,7 +70,7 @@ forceinline pureconst struct CosmoTib *__get_tls(void) {
   return 0;
 #elif __x86_64__
   struct CosmoTib *__tib;
-  __asm__("mov\t%%fs:0,%0" : "=r"(__tib));
+  __asm__("mov\t%%gs:0x30,%0" : "=r"(__tib));
   return __tib;
 #elif defined(__aarch64__)
   register struct CosmoTib *__tls __asm__("x28");
diff --git a/libc/thread/tls2.internal.h b/libc/thread/tls2.internal.h
index e91ce33d7..e739a0981 100644
--- a/libc/thread/tls2.internal.h
+++ b/libc/thread/tls2.internal.h
@@ -13,12 +13,8 @@ COSMOPOLITAN_C_START_
  */
 forceinline struct CosmoTib *__get_tls_privileged(void) {
   char *tib, *lin = (char *)0x30;
-  if (IsLinux() || IsFreebsd() || IsNetbsd() || IsOpenbsd() || IsMetal()) {
-    if (!__tls_morphed) {
-      asm("mov\t%%fs:(%1),%0" : "=a"(tib) : "r"(lin) : "memory");
-    } else {
-      asm("mov\t%%gs:(%1),%0" : "=a"(tib) : "r"(lin) : "memory");
-    }
+  if (IsNetbsd() || IsOpenbsd()) {
+    asm("mov\t%%fs:(%1),%0" : "=a"(tib) : "r"(lin) : "memory");
   } else {
     asm("mov\t%%gs:(%1),%0" : "=a"(tib) : "r"(lin) : "memory");
     if (IsWindows()) {
diff --git a/libc/time/BUILD.mk b/libc/time/BUILD.mk
index 8705b3347..bb3b136f9 100644
--- a/libc/time/BUILD.mk
+++ b/libc/time/BUILD.mk
@@ -4,7 +4,6 @@
 PKGS += LIBC_TIME
 
 LIBC_TIME_ARTIFACTS += LIBC_TIME_A
-LIBC_TIME_ZONEINFOS = $(wildcard usr/share/zoneinfo/*)
 LIBC_TIME = $(LIBC_TIME_A_DEPS) $(LIBC_TIME_A)
 LIBC_TIME_A = o/$(MODE)/libc/time/time.a
 LIBC_TIME_A_FILES := $(wildcard libc/time/struct/*) $(wildcard libc/time/*)
@@ -12,6 +11,10 @@ LIBC_TIME_A_HDRS := $(filter %.h,$(LIBC_TIME_A_FILES))
 LIBC_TIME_A_SRCS_S = $(filter %.S,$(LIBC_TIME_A_FILES))
 LIBC_TIME_A_SRCS_C = $(filter %.c,$(LIBC_TIME_A_FILES))
 
+LIBC_TIME_ZONEINFOS :=					\
+	$(wildcard usr/share/zoneinfo/*)		\
+	$(wildcard usr/share/zoneinfo/US/*)
+
 LIBC_TIME_A_SRCS =					\
 	$(LIBC_TIME_A_SRCS_S)				\
 	$(LIBC_TIME_A_SRCS_C)
diff --git a/libc/time/localtime.c b/libc/time/localtime.c
index c90686b2c..3c5c567c8 100644
--- a/libc/time/localtime.c
+++ b/libc/time/localtime.c
@@ -5,9 +5,9 @@
 #include "libc/calls/blockcancel.internal.h"
 #include "libc/calls/calls.h"
 #include "libc/cxxabi.h"
-#include "libc/serialize.h"
 #include "libc/mem/gc.h"
 #include "libc/mem/mem.h"
+#include "libc/serialize.h"
 #include "libc/str/str.h"
 #include "libc/sysv/consts/o.h"
 #include "libc/thread/thread.h"
@@ -60,7 +60,8 @@ void localtime_unlock(void) {
 	pthread_mutex_unlock(&locallock);
 }
 
-__attribute__((__constructor__)) static void localtime_init(void) {
+__attribute__((__constructor__(80)))
+static textstartup void localtime_init(void) {
 	localtime_wipe();
 	pthread_atfork(localtime_lock,
 		       localtime_unlock,
diff --git a/libc/time/strftime.c b/libc/time/strftime.c
index 1906cf1db..1a607b35e 100644
--- a/libc/time/strftime.c
+++ b/libc/time/strftime.c
@@ -26,10 +26,8 @@
 
 #define DIVISOR	100
 
-asm(".ident\t\"\\n\\n\
-strftime (BSD-3)\\n\
-Copyright 1989 The Regents of the University of California\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(strftime_notice, "strftime (BSD-3)\n\
+Copyright 1989 The Regents of the University of California");
 
 /*
 ** Based on the UCB version with the copyright notice appearing above.
diff --git a/libc/time/strptime.c b/libc/time/strptime.c
index 74c558466..63d9020ac 100644
--- a/libc/time/strptime.c
+++ b/libc/time/strptime.c
@@ -30,11 +30,7 @@
 #include "libc/str/str.h"
 #include "libc/time/struct/tm.h"
 #include "libc/time/time.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2019 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 char *strptime(const char *s, const char *f, struct tm *tm) {
   int i, w, neg, adj, min, range, itemsize, *dest, dummy;
@@ -44,8 +40,7 @@ char *strptime(const char *s, const char *f, struct tm *tm) {
   while (*f) {
     if (*f != '%') {
       if (isspace(*f)) {
-        for (; *s && isspace(*s); s++)
-          ;
+        for (; *s && isspace(*s); s++);
       } else if (*s != *f) {
         return 0;
       } else {
@@ -138,8 +133,7 @@ char *strptime(const char *s, const char *f, struct tm *tm) {
         goto numeric_range;
       case 'n':
       case 't':
-        for (; *s && isspace(*s); s++)
-          ;
+        for (; *s && isspace(*s); s++);
         break;
       case 'p':
         ex = "AM";
diff --git a/libc/time/tzfile.internal.h b/libc/time/tzfile.internal.h
index 3c892b77f..abab9923d 100644
--- a/libc/time/tzfile.internal.h
+++ b/libc/time/tzfile.internal.h
@@ -29,7 +29,7 @@
 #endif /* !defined TZDEFAULT */
 
 #ifndef TZDEFRULES
-#define TZDEFRULES	"New_York"
+#define TZDEFRULES	"US/Pacific"
 #endif /* !defined TZDEFRULES */
 
 
diff --git a/libc/tinymath/.clang-format b/libc/tinymath/.clang-format
new file mode 100644
index 000000000..47a38a93f
--- /dev/null
+++ b/libc/tinymath/.clang-format
@@ -0,0 +1,2 @@
+DisableFormat: true
+SortIncludes: Never
diff --git a/libc/tinymath/__cexp.c b/libc/tinymath/__cexp.c
index 788edab61..59d3e36fd 100644
--- a/libc/tinymath/__cexp.c
+++ b/libc/tinymath/__cexp.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_exp.c */
 /*-
diff --git a/libc/tinymath/__cexpf.c b/libc/tinymath/__cexpf.c
index 5bf24743f..1840c31e1 100644
--- a/libc/tinymath/__cexpf.c
+++ b/libc/tinymath/__cexpf.c
@@ -28,12 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_expf.c */
 /*-
diff --git a/libc/tinymath/__math_divzero.c b/libc/tinymath/__math_divzero.c
index 6c4fecba0..24b104690 100644
--- a/libc/tinymath/__math_divzero.c
+++ b/libc/tinymath/__math_divzero.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 double __math_divzero(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_divzerof.c b/libc/tinymath/__math_divzerof.c
index aba79e70d..b241eecb0 100644
--- a/libc/tinymath/__math_divzerof.c
+++ b/libc/tinymath/__math_divzerof.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 float __math_divzerof(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_invalid.c b/libc/tinymath/__math_invalid.c
index 77a8a245e..890a8eadf 100644
--- a/libc/tinymath/__math_invalid.c
+++ b/libc/tinymath/__math_invalid.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 double __math_invalid(double x)
 {
diff --git a/libc/tinymath/__math_invalidf.c b/libc/tinymath/__math_invalidf.c
index 5c8bd6951..7bcb5870b 100644
--- a/libc/tinymath/__math_invalidf.c
+++ b/libc/tinymath/__math_invalidf.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 float __math_invalidf(float x)
 {
diff --git a/libc/tinymath/__math_oflow.c b/libc/tinymath/__math_oflow.c
index 3e8a756a0..588c80167 100644
--- a/libc/tinymath/__math_oflow.c
+++ b/libc/tinymath/__math_oflow.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument double __math_oflow(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_oflowf.c b/libc/tinymath/__math_oflowf.c
index c289062d4..a7b8d7f1a 100644
--- a/libc/tinymath/__math_oflowf.c
+++ b/libc/tinymath/__math_oflowf.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument float __math_oflowf(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_uflow.c b/libc/tinymath/__math_uflow.c
index c810da1c6..7ccda32fe 100644
--- a/libc/tinymath/__math_uflow.c
+++ b/libc/tinymath/__math_uflow.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument double __math_uflow(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_uflowf.c b/libc/tinymath/__math_uflowf.c
index de7cf7c06..b355b5a7e 100644
--- a/libc/tinymath/__math_uflowf.c
+++ b/libc/tinymath/__math_uflowf.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument float __math_uflowf(uint32_t sign)
 {
diff --git a/libc/tinymath/__math_xflow.c b/libc/tinymath/__math_xflow.c
index 7e87826dc..df508896c 100644
--- a/libc/tinymath/__math_xflow.c
+++ b/libc/tinymath/__math_xflow.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument double __math_xflow(uint32_t sign, double y)
 {
diff --git a/libc/tinymath/__math_xflowf.c b/libc/tinymath/__math_xflowf.c
index 1ea47b7fe..6b23ee8f3 100644
--- a/libc/tinymath/__math_xflowf.c
+++ b/libc/tinymath/__math_xflowf.c
@@ -26,7 +26,6 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
-// clang-format off
 
 dontinstrument float __math_xflowf(uint32_t sign, float y)
 {
diff --git a/libc/tinymath/acos.c b/libc/tinymath/acos.c
index 7631f77ea..9bd36c7c8 100644
--- a/libc/tinymath/acos.c
+++ b/libc/tinymath/acos.c
@@ -27,11 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_acos.c */
 /*
diff --git a/libc/tinymath/acosf.c b/libc/tinymath/acosf.c
index 969a464b4..ee451d43a 100644
--- a/libc/tinymath/acosf.c
+++ b/libc/tinymath/acosf.c
@@ -28,11 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_acosf.c */
 /*
diff --git a/libc/tinymath/acosh.c b/libc/tinymath/acosh.c
index b1bb9a4fa..cc0c3f0f5 100644
--- a/libc/tinymath/acosh.c
+++ b/libc/tinymath/acosh.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic cosine of 𝑥.
diff --git a/libc/tinymath/acoshf.c b/libc/tinymath/acoshf.c
index f0bc86b5c..d8d4be8e9 100644
--- a/libc/tinymath/acoshf.c
+++ b/libc/tinymath/acoshf.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic cosine of 𝑥.
diff --git a/libc/tinymath/acoshl.c b/libc/tinymath/acoshl.c
index 019d6ae74..a0b16cc66 100644
--- a/libc/tinymath/acoshl.c
+++ b/libc/tinymath/acoshl.c
@@ -39,15 +39,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* EXP_LARGE is the threshold above which we use acosh(x) ~= log(2x). */
 #if LDBL_MANT_DIG == 64
diff --git a/libc/tinymath/acosl.c b/libc/tinymath/acosl.c
index d65ab0a27..eaa8ffb7f 100644
--- a/libc/tinymath/acosl.c
+++ b/libc/tinymath/acosl.c
@@ -29,15 +29,8 @@
 #include "libc/tinymath/invtrigl.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_acosl.c */
 /*
diff --git a/libc/tinymath/asin.c b/libc/tinymath/asin.c
index 0d5afb86f..761f33636 100644
--- a/libc/tinymath/asin.c
+++ b/libc/tinymath/asin.c
@@ -26,12 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_asin.c */
 /*
diff --git a/libc/tinymath/asinf.c b/libc/tinymath/asinf.c
index 36e4320af..ff0eed85a 100644
--- a/libc/tinymath/asinf.c
+++ b/libc/tinymath/asinf.c
@@ -28,14 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_asinf.c */
 /*
diff --git a/libc/tinymath/asinh.c b/libc/tinymath/asinh.c
index 62d70372b..fd9967fff 100644
--- a/libc/tinymath/asinh.c
+++ b/libc/tinymath/asinh.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/freebsd.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic sine of 𝑥.
diff --git a/libc/tinymath/asinhf.c b/libc/tinymath/asinhf.c
index bd66d94cf..840e40852 100644
--- a/libc/tinymath/asinhf.c
+++ b/libc/tinymath/asinhf.c
@@ -27,12 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic sine of 𝑥.
diff --git a/libc/tinymath/asinhl.c b/libc/tinymath/asinhl.c
index 16d2790e6..c39f5f97c 100644
--- a/libc/tinymath/asinhl.c
+++ b/libc/tinymath/asinhl.c
@@ -38,16 +38,11 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 /* EXP_LARGE is the threshold above which we use asinh(x) ~= log(2x). */
 /* EXP_TINY is the threshold below which we use asinh(x) ~= x. */
diff --git a/libc/tinymath/asinl.c b/libc/tinymath/asinl.c
index 07ff3f32c..bec55a702 100644
--- a/libc/tinymath/asinl.c
+++ b/libc/tinymath/asinl.c
@@ -30,15 +30,8 @@
 #include "libc/tinymath/invtrigl.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_asinl.c */
 /*
diff --git a/libc/tinymath/atan.c b/libc/tinymath/atan.c
index 9314d1ba1..bfcf1fb67 100644
--- a/libc/tinymath/atan.c
+++ b/libc/tinymath/atan.c
@@ -27,15 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_atan.c */
 /*
diff --git a/libc/tinymath/atan2.c b/libc/tinymath/atan2.c
index 663d979aa..04b79f109 100644
--- a/libc/tinymath/atan2.c
+++ b/libc/tinymath/atan2.c
@@ -38,14 +38,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* atan2(y,x)
  * Method :
diff --git a/libc/tinymath/atan2f.c b/libc/tinymath/atan2f.c
index e62ead380..3e8ca8194 100644
--- a/libc/tinymath/atan2f.c
+++ b/libc/tinymath/atan2f.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/atanf_common.internal.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 #define Pi (0x1.921fb6p+1f)
 #define PiOver2 (0x1.921fb6p+0f)
diff --git a/libc/tinymath/atan2l.c b/libc/tinymath/atan2l.c
index f2f1cac8c..f8b61f6d8 100644
--- a/libc/tinymath/atan2l.c
+++ b/libc/tinymath/atan2l.c
@@ -30,15 +30,9 @@
 #include "libc/tinymath/invtrigl.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_atan2l.c */
 /*
diff --git a/libc/tinymath/atan_common.internal.h b/libc/tinymath/atan_common.internal.h
index 57c6aaf01..7c50045a6 100644
--- a/libc/tinymath/atan_common.internal.h
+++ b/libc/tinymath/atan_common.internal.h
@@ -4,7 +4,6 @@
 #include "libc/tinymath/estrin_wrap.internal.h"
 #include "libc/tinymath/horner.internal.h"
 COSMOPOLITAN_C_START_
-// clang-format off
 
 /*
  * Double-precision polynomial evaluation function for scalar and vector atan(x)
diff --git a/libc/tinymath/atan_data.c b/libc/tinymath/atan_data.c
index e28e2d5b0..3b465d0c5 100644
--- a/libc/tinymath/atan_data.c
+++ b/libc/tinymath/atan_data.c
@@ -27,12 +27,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/atan_data.internal.h"
 
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
 const struct atan_poly_data __atan_poly_data = {
   .poly = {/* Coefficients of polynomial P such that atan(x)~x+x*P(x^2) on
 	      [2**-1022, 1.0]. See atan.sollya for details of how these were
diff --git a/libc/tinymath/atanf.c b/libc/tinymath/atanf.c
index 59e773020..e2a687bf8 100644
--- a/libc/tinymath/atanf.c
+++ b/libc/tinymath/atanf.c
@@ -27,15 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_atanf.c */
 /*
diff --git a/libc/tinymath/atanf_common.internal.h b/libc/tinymath/atanf_common.internal.h
index 13368d3a1..38a2df65d 100644
--- a/libc/tinymath/atanf_common.internal.h
+++ b/libc/tinymath/atanf_common.internal.h
@@ -4,7 +4,6 @@
 #include "libc/tinymath/estrin_wrap.internal.h"
 #include "libc/tinymath/hornerf.internal.h"
 COSMOPOLITAN_C_START_
-// clang-format off
 
 #if WANT_VMATH
 
diff --git a/libc/tinymath/atanf_data.c b/libc/tinymath/atanf_data.c
index cf22b0506..e44a7631e 100644
--- a/libc/tinymath/atanf_data.c
+++ b/libc/tinymath/atanf_data.c
@@ -27,12 +27,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/atanf_data.internal.h"
 
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
 /* Coefficients of polynomial P such that atan(x)~x+x*P(x^2) on [2**-128, 1.0].
  */
 const struct atanf_poly_data __atanf_poly_data = {
diff --git a/libc/tinymath/atanh.c b/libc/tinymath/atanh.c
index 36bd107dc..f05f0466d 100644
--- a/libc/tinymath/atanh.c
+++ b/libc/tinymath/atanh.c
@@ -27,12 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic tangent of 𝑥.
diff --git a/libc/tinymath/atanhf.c b/libc/tinymath/atanhf.c
index fb00d8452..f2431cc90 100644
--- a/libc/tinymath/atanhf.c
+++ b/libc/tinymath/atanhf.c
@@ -27,12 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic tangent of 𝑥.
diff --git a/libc/tinymath/atanhl.c b/libc/tinymath/atanhl.c
index 208ff9b9c..e10b33c6a 100644
--- a/libc/tinymath/atanhl.c
+++ b/libc/tinymath/atanhl.c
@@ -29,12 +29,7 @@
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns inverse hyperbolic tangent of 𝑥.
diff --git a/libc/tinymath/atanl.c b/libc/tinymath/atanl.c
index ced707ff5..6f3a29942 100644
--- a/libc/tinymath/atanl.c
+++ b/libc/tinymath/atanl.c
@@ -29,15 +29,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_atanl.c */
 /*
diff --git a/libc/tinymath/cacos.c b/libc/tinymath/cacos.c
index 9ad032fd7..6c774f583 100644
--- a/libc/tinymath/cacos.c
+++ b/libc/tinymath/cacos.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 // FIXME: Hull et al. "Implementing the complex arcsine and arccosine functions using exception handling" 1997
 
diff --git a/libc/tinymath/cacosf.c b/libc/tinymath/cacosf.c
index 1b6bfcd3e..0c8add51d 100644
--- a/libc/tinymath/cacosf.c
+++ b/libc/tinymath/cacosf.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 // FIXME
 
diff --git a/libc/tinymath/cacosh.c b/libc/tinymath/cacosh.c
index 4eccd729c..3a663ee16 100644
--- a/libc/tinymath/cacosh.c
+++ b/libc/tinymath/cacosh.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /* acosh(z) = i acos(z) */
 double complex cacosh(double complex z)
diff --git a/libc/tinymath/cacoshf.c b/libc/tinymath/cacoshf.c
index a029f08da..0160691fa 100644
--- a/libc/tinymath/cacoshf.c
+++ b/libc/tinymath/cacoshf.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 float complex cacoshf(float complex z)
 {
diff --git a/libc/tinymath/cacoshl.c b/libc/tinymath/cacoshl.c
index 1c5e96cb4..5e129462a 100644
--- a/libc/tinymath/cacoshl.c
+++ b/libc/tinymath/cacoshl.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 long double complex cacoshl(long double complex z)
 {
diff --git a/libc/tinymath/cacosl.c b/libc/tinymath/cacosl.c
index 0e09d4753..adfff0b5d 100644
--- a/libc/tinymath/cacosl.c
+++ b/libc/tinymath/cacosl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 long double complex cacosl(long double complex z) {
 #if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
diff --git a/libc/tinymath/casin.c b/libc/tinymath/casin.c
index b0e85ca2a..f6478d061 100644
--- a/libc/tinymath/casin.c
+++ b/libc/tinymath/casin.c
@@ -28,12 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 // FIXME
 
diff --git a/libc/tinymath/casinf.c b/libc/tinymath/casinf.c
index 1d0b35414..04766d91d 100644
--- a/libc/tinymath/casinf.c
+++ b/libc/tinymath/casinf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 // FIXME
 
diff --git a/libc/tinymath/casinh.c b/libc/tinymath/casinh.c
index 5a11878a7..56d958807 100644
--- a/libc/tinymath/casinh.c
+++ b/libc/tinymath/casinh.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 /* asinh(z) = -i asin(i z) */
 
diff --git a/libc/tinymath/casinhf.c b/libc/tinymath/casinhf.c
index 503eb3437..8bcc1e726 100644
--- a/libc/tinymath/casinhf.c
+++ b/libc/tinymath/casinhf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 float complex casinhf(float complex z)
 {
diff --git a/libc/tinymath/casinhl.c b/libc/tinymath/casinhl.c
index 5766ff6d9..193d8d53a 100644
--- a/libc/tinymath/casinhl.c
+++ b/libc/tinymath/casinhl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 long double complex casinhl(long double complex z) {
 #if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
diff --git a/libc/tinymath/casinl.c b/libc/tinymath/casinl.c
index 618d236be..78913278d 100644
--- a/libc/tinymath/casinl.c
+++ b/libc/tinymath/casinl.c
@@ -27,12 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 long double complex casinl(long double complex z)
 {
diff --git a/libc/tinymath/catan.c b/libc/tinymath/catan.c
index b9c6b7faa..dd3b4a90e 100644
--- a/libc/tinymath/catan.c
+++ b/libc/tinymath/catan.c
@@ -28,16 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
-
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/s_catan.c */
 /*
@@ -99,7 +91,6 @@ asm(".include \"libc/disclaimer.inc\"");
  * 2.9e-17.  See also clog().
  */
 
-
 #define MAXNUM 1.0e308
 
 static const double DP1 = 3.14159265160560607910E0;
diff --git a/libc/tinymath/catanf.c b/libc/tinymath/catanf.c
index ce5f2cdb0..846c2a7ad 100644
--- a/libc/tinymath/catanf.c
+++ b/libc/tinymath/catanf.c
@@ -28,16 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/s_catanf.c */
 /*
diff --git a/libc/tinymath/catanh.c b/libc/tinymath/catanh.c
index 087239547..12b2c88b1 100644
--- a/libc/tinymath/catanh.c
+++ b/libc/tinymath/catanh.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 /* atanh = -i atan(i z) */
 
diff --git a/libc/tinymath/catanhf.c b/libc/tinymath/catanhf.c
index 90f491207..b7282e92e 100644
--- a/libc/tinymath/catanhf.c
+++ b/libc/tinymath/catanhf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 float complex catanhf(float complex z)
 {
diff --git a/libc/tinymath/catanhl.c b/libc/tinymath/catanhl.c
index 299aadece..59c2a88fc 100644
--- a/libc/tinymath/catanhl.c
+++ b/libc/tinymath/catanhl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 long double complex catanhl(long double complex z) {
 #if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
diff --git a/libc/tinymath/catanl.c b/libc/tinymath/catanl.c
index 3f87d3a3b..633d155ea 100644
--- a/libc/tinymath/catanl.c
+++ b/libc/tinymath/catanl.c
@@ -22,15 +22,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("openbsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /*
  *      Complex circular arc tangent
diff --git a/libc/tinymath/cbrt.c b/libc/tinymath/cbrt.c
index b88b7e297..55e324164 100644
--- a/libc/tinymath/cbrt.c
+++ b/libc/tinymath/cbrt.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cbrt.c */
 /*
diff --git a/libc/tinymath/cbrtf.c b/libc/tinymath/cbrtf.c
index 0ec14e758..088ac83a7 100644
--- a/libc/tinymath/cbrtf.c
+++ b/libc/tinymath/cbrtf.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cbrtf.c */
 /*
diff --git a/libc/tinymath/cbrtl.c b/libc/tinymath/cbrtl.c
index ccd1040f1..4e1e2863c 100644
--- a/libc/tinymath/cbrtl.c
+++ b/libc/tinymath/cbrtl.c
@@ -28,17 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cbrtl.c */
 /*-
diff --git a/libc/tinymath/ccos.c b/libc/tinymath/ccos.c
index 35d3f137f..ccfddd586 100644
--- a/libc/tinymath/ccos.c
+++ b/libc/tinymath/ccos.c
@@ -17,8 +17,11 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-#include "libc/math.h"
 
 double complex ccos(double complex z) {
   return ccosh(CMPLX(-cimag(z), creal(z)));
 }
+
+#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
+__weak_reference(ccos, ccosl);
+#endif
diff --git a/libc/tinymath/ccosf.c b/libc/tinymath/ccosf.c
index 5613099ed..6356d54cb 100644
--- a/libc/tinymath/ccosf.c
+++ b/libc/tinymath/ccosf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 float complex ccosf(float complex z)
 {
diff --git a/libc/tinymath/ccosh.c b/libc/tinymath/ccosh.c
index 28bb3cbc2..c45f39ee3 100644
--- a/libc/tinymath/ccosh.c
+++ b/libc/tinymath/ccosh.c
@@ -29,14 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_ccosh.c */
 /*-
diff --git a/libc/tinymath/ccoshf.c b/libc/tinymath/ccoshf.c
index d2e137044..d7e02c489 100644
--- a/libc/tinymath/ccoshf.c
+++ b/libc/tinymath/ccoshf.c
@@ -29,14 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_ccoshf.c */
 /*-
diff --git a/libc/tinymath/ccoshl.c b/libc/tinymath/ccoshl.c
index 7d22ca69d..bc388ffc3 100644
--- a/libc/tinymath/ccoshl.c
+++ b/libc/tinymath/ccoshl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 //FIXME
 long double complex ccoshl(long double complex z)
diff --git a/libc/tinymath/ccosl.c b/libc/tinymath/ccosl.c
index 76548b9d6..a16144507 100644
--- a/libc/tinymath/ccosl.c
+++ b/libc/tinymath/ccosl.c
@@ -26,17 +26,11 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
 long double complex ccosl(long double complex z) {
-#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
-	return ccos(z);
-#else
 	return ccoshl(CMPLXL(-cimagl(z), creall(z)));
-#endif
 }
+
+#endif
diff --git a/libc/tinymath/ceil.c b/libc/tinymath/ceil.c
index 4ba6ec6da..dd5c4f6eb 100644
--- a/libc/tinymath/ceil.c
+++ b/libc/tinymath/ceil.c
@@ -31,12 +31,7 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 #if FLT_EVAL_METHOD==0 || FLT_EVAL_METHOD==1
 #define EPS DBL_EPSILON
diff --git a/libc/tinymath/ceilf.c b/libc/tinymath/ceilf.c
index 2dad43065..ab34911fd 100644
--- a/libc/tinymath/ceilf.c
+++ b/libc/tinymath/ceilf.c
@@ -30,12 +30,7 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns smallest integral value not less than 𝑥.
diff --git a/libc/tinymath/ceill.c b/libc/tinymath/ceill.c
index 921a93836..42c733662 100644
--- a/libc/tinymath/ceill.c
+++ b/libc/tinymath/ceill.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns smallest integral value not less than 𝑥.
diff --git a/libc/tinymath/cexp.c b/libc/tinymath/cexp.c
index c6fd6e086..5f09c24e8 100644
--- a/libc/tinymath/cexp.c
+++ b/libc/tinymath/cexp.c
@@ -28,13 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cexp.c */
 /*-
diff --git a/libc/tinymath/cexpf.c b/libc/tinymath/cexpf.c
index 200b9de4f..a26b30feb 100644
--- a/libc/tinymath/cexpf.c
+++ b/libc/tinymath/cexpf.c
@@ -28,13 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cexpf.c */
 /*-
diff --git a/libc/tinymath/cexpl.c b/libc/tinymath/cexpl.c
index a58b94d2e..1800be2d5 100644
--- a/libc/tinymath/cexpl.c
+++ b/libc/tinymath/cexpl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 //FIXME
 long double complex cexpl(long double complex z)
diff --git a/libc/tinymath/clog.c b/libc/tinymath/clog.c
index 60c48fb9c..77b4cb284 100644
--- a/libc/tinymath/clog.c
+++ b/libc/tinymath/clog.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 // FIXME
 
@@ -49,3 +42,7 @@ double complex clog(double complex z)
 	phi = carg(z);
 	return CMPLX(log(r), phi);
 }
+
+#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
+__weak_reference(clog, clogl);
+#endif
diff --git a/libc/tinymath/clogf.c b/libc/tinymath/clogf.c
index dcc3a4b29..e658c58f3 100644
--- a/libc/tinymath/clogf.c
+++ b/libc/tinymath/clogf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 // FIXME
 
diff --git a/libc/tinymath/clogl.c b/libc/tinymath/clogl.c
index 868f44eb5..d55a153b8 100644
--- a/libc/tinymath/clogl.c
+++ b/libc/tinymath/clogl.c
@@ -27,21 +27,10 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
 #include "libc/math.h"
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 #include "libc/tinymath/complex.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
-
-#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
-long double complex clogl(long double complex z)
-{
-	return clog(z);
-}
-#else
-// FIXME
 long double complex clogl(long double complex z)
 {
 	long double r, phi;
@@ -49,4 +38,5 @@ long double complex clogl(long double complex z)
 	phi = cargl(z);
 	return CMPLXL(logl(r), phi);
 }
+
 #endif
diff --git a/libc/tinymath/cos.c b/libc/tinymath/cos.c
index 00776746a..d1b2b58f4 100644
--- a/libc/tinymath/cos.c
+++ b/libc/tinymath/cos.c
@@ -28,15 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cos.c */
 /*
diff --git a/libc/tinymath/cosdf.c b/libc/tinymath/cosdf.c
index 0fb5e3da1..3d35a6436 100644
--- a/libc/tinymath/cosdf.c
+++ b/libc/tinymath/cosdf.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_cosf.c */
 /*
diff --git a/libc/tinymath/cosf.c b/libc/tinymath/cosf.c
index 6e12a7b14..36d3ca2db 100644
--- a/libc/tinymath/cosf.c
+++ b/libc/tinymath/cosf.c
@@ -29,15 +29,10 @@
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_cosf.c */
 /*
diff --git a/libc/tinymath/cosh.c b/libc/tinymath/cosh.c
index 3c98b9253..4db136b38 100644
--- a/libc/tinymath/cosh.c
+++ b/libc/tinymath/cosh.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/expo.internal.h"
 #include "libc/tinymath/feval.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns hyperbolic cosine of 𝑥.
diff --git a/libc/tinymath/coshf.c b/libc/tinymath/coshf.c
index 1e352d1c4..e0bef68dc 100644
--- a/libc/tinymath/coshf.c
+++ b/libc/tinymath/coshf.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/expo.internal.h"
 #include "libc/tinymath/feval.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns hyperbolic cosine of 𝑥.
diff --git a/libc/tinymath/coshl.c b/libc/tinymath/coshl.c
index f7cfb1ed3..c2cb21c6e 100644
--- a/libc/tinymath/coshl.c
+++ b/libc/tinymath/coshl.c
@@ -38,16 +38,11 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 #if LDBL_MAX_EXP != 0x4000
 /* We also require the usual expsign encoding. */
diff --git a/libc/tinymath/cosl.c b/libc/tinymath/cosl.c
index 1368cc825..6642af10e 100644
--- a/libc/tinymath/cosl.c
+++ b/libc/tinymath/cosl.c
@@ -30,12 +30,7 @@
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns cosine of 𝑥.
diff --git a/libc/tinymath/cpow.c b/libc/tinymath/cpow.c
index 4a0aa4b10..5d58e14d9 100644
--- a/libc/tinymath/cpow.c
+++ b/libc/tinymath/cpow.c
@@ -26,18 +26,15 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-#include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /* pow(z, c) = exp(c log(z)), See C99 G.6.4.1 */
-
 double complex cpow(double complex z, double complex c)
 {
 	return cexp(c * clog(z));
 }
+
+#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
+__weak_reference(cpow, cpowl);
+#endif
diff --git a/libc/tinymath/cpowf.c b/libc/tinymath/cpowf.c
index 671bad57e..9d3656faa 100644
--- a/libc/tinymath/cpowf.c
+++ b/libc/tinymath/cpowf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 float complex cpowf(float complex z, float complex c)
 {
diff --git a/libc/tinymath/cpowl.c b/libc/tinymath/cpowl.c
index 389642031..1fa76aee6 100644
--- a/libc/tinymath/cpowl.c
+++ b/libc/tinymath/cpowl.c
@@ -26,21 +26,12 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
-
-#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
-long double complex cpowl(long double complex z, long double complex c)
-{
-	return cpow(z, c);
-}
-#else
 long double complex cpowl(long double complex z, long double complex c)
 {
 	return cexpl(c * clogl(z));
 }
+
 #endif
diff --git a/libc/tinymath/csin.c b/libc/tinymath/csin.c
index 23d0f27c2..eeb1a30f4 100644
--- a/libc/tinymath/csin.c
+++ b/libc/tinymath/csin.c
@@ -28,19 +28,15 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 /* sin(z) = -i sinh(i z) */
-
 double complex csin(double complex z)
 {
 	z = csinh(CMPLX(-cimag(z), creal(z)));
 	return CMPLX(cimag(z), -creal(z));
 }
+
+#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
+__weak_reference(csin, csinl);
+#endif
diff --git a/libc/tinymath/csinf.c b/libc/tinymath/csinf.c
index 24511f17b..0883e26cd 100644
--- a/libc/tinymath/csinf.c
+++ b/libc/tinymath/csinf.c
@@ -28,12 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 
 
diff --git a/libc/tinymath/csinh.c b/libc/tinymath/csinh.c
index 8f59253da..22da546e0 100644
--- a/libc/tinymath/csinh.c
+++ b/libc/tinymath/csinh.c
@@ -29,15 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
-
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_csinh.c */
 /*-
diff --git a/libc/tinymath/csinhf.c b/libc/tinymath/csinhf.c
index 07471ce81..d75ac0c07 100644
--- a/libc/tinymath/csinhf.c
+++ b/libc/tinymath/csinhf.c
@@ -29,14 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_csinhf.c */
 /*-
diff --git a/libc/tinymath/csinhl.c b/libc/tinymath/csinhl.c
index 1a83181fc..f85b8b26a 100644
--- a/libc/tinymath/csinhl.c
+++ b/libc/tinymath/csinhl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 //FIXME
 long double complex csinhl(long double complex z)
diff --git a/libc/tinymath/csinl.c b/libc/tinymath/csinl.c
index ddf599d90..02b5d0178 100644
--- a/libc/tinymath/csinl.c
+++ b/libc/tinymath/csinl.c
@@ -26,22 +26,13 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
-
-#if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
-long double complex csinl(long double complex z)
-{
-	return csin(z);
-}
-#else
 long double complex csinl(long double complex z)
 {
 	z = csinhl(CMPLXL(-cimagl(z), creall(z)));
 	return CMPLXL(cimagl(z), -creall(z));
 }
+
 #endif
diff --git a/libc/tinymath/csqrt.c b/libc/tinymath/csqrt.c
index 70583e7a2..d2bccdc38 100644
--- a/libc/tinymath/csqrt.c
+++ b/libc/tinymath/csqrt.c
@@ -28,12 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_csqrt.c */
 /*-
diff --git a/libc/tinymath/csqrtf.c b/libc/tinymath/csqrtf.c
index 27c1e6889..61d142f11 100644
--- a/libc/tinymath/csqrtf.c
+++ b/libc/tinymath/csqrtf.c
@@ -28,13 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_csqrtf.c */
 /*-
diff --git a/libc/tinymath/csqrtl.c b/libc/tinymath/csqrtl.c
index c6b1e7524..df534fd9f 100644
--- a/libc/tinymath/csqrtl.c
+++ b/libc/tinymath/csqrtl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 //FIXME
 long double complex csqrtl(long double complex z)
diff --git a/libc/tinymath/ctan.c b/libc/tinymath/ctan.c
index b3e2de71d..8e20f8757 100644
--- a/libc/tinymath/ctan.c
+++ b/libc/tinymath/ctan.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 /* tan(z) = -i tanh(i z) */
 
diff --git a/libc/tinymath/ctanf.c b/libc/tinymath/ctanf.c
index f584c5ea9..fe65ba01b 100644
--- a/libc/tinymath/ctanf.c
+++ b/libc/tinymath/ctanf.c
@@ -28,14 +28,7 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
-
+__static_yoink("musl_libc_notice");
 
 float complex ctanf(float complex z)
 {
diff --git a/libc/tinymath/ctanh.c b/libc/tinymath/ctanh.c
index 94a1776d8..721e303da 100644
--- a/libc/tinymath/ctanh.c
+++ b/libc/tinymath/ctanh.c
@@ -28,16 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-ctahnh (BSD-2 License)\\n\
-Copyright (c) 2011 David Schultz <das@FreeBSD.ORG>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_ctanh.c */
 /*-
diff --git a/libc/tinymath/ctanhf.c b/libc/tinymath/ctanhf.c
index 29d7e3f1b..b806dcf58 100644
--- a/libc/tinymath/ctanhf.c
+++ b/libc/tinymath/ctanhf.c
@@ -28,13 +28,8 @@
 #include "libc/complex.h"
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_ctanhf.c */
 /*-
diff --git a/libc/tinymath/ctanhl.c b/libc/tinymath/ctanhl.c
index 6aa599c3c..43061dcf9 100644
--- a/libc/tinymath/ctanhl.c
+++ b/libc/tinymath/ctanhl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 //FIXME
 long double complex ctanhl(long double complex z)
diff --git a/libc/tinymath/ctanl.c b/libc/tinymath/ctanl.c
index d8602d9b7..27f0d94ca 100644
--- a/libc/tinymath/ctanl.c
+++ b/libc/tinymath/ctanl.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/complex.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
 
 #if LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024
 long double complex ctanl(long double complex z)
diff --git a/libc/tinymath/erf.c b/libc/tinymath/erf.c
index 029a41840..a3b976358 100644
--- a/libc/tinymath/erf.c
+++ b/libc/tinymath/erf.c
@@ -26,15 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_erf.c */
 /*
diff --git a/libc/tinymath/erff.c b/libc/tinymath/erff.c
index cea72f1a4..8e7d60006 100644
--- a/libc/tinymath/erff.c
+++ b/libc/tinymath/erff.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("freebsd_libm_notice");
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_erff.c */
 /*
diff --git a/libc/tinymath/erfl.c b/libc/tinymath/erfl.c
index 6ac746b30..60bb0ef2a 100644
--- a/libc/tinymath/erfl.c
+++ b/libc/tinymath/erfl.c
@@ -28,15 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("openbsd_libm_notice");
+__static_yoink("musl_libc_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_erfl.c */
 /*
diff --git a/libc/tinymath/estrin_wrap.internal.h b/libc/tinymath/estrin_wrap.internal.h
index c71015fa0..ee77c108e 100644
--- a/libc/tinymath/estrin_wrap.internal.h
+++ b/libc/tinymath/estrin_wrap.internal.h
@@ -8,7 +8,6 @@
  * SPDX-License-Identifier: MIT OR Apache-2.0 WITH LLVM-exception
  */
 
-// clang-format off
 #define  ESTRIN_1_(x,                  c, i) FMA(x,   c(1 + i),                        c(i))
 #define  ESTRIN_2_(x, x2,              c, i) FMA(x2,  c(2 + i),                        ESTRIN_1_(x,              c, i))
 #define  ESTRIN_3_(x, x2,              c, i) FMA(x2,  ESTRIN_1_(x,         c,  2 + i), ESTRIN_1_(x,              c, i))
diff --git a/libc/tinymath/exp.c b/libc/tinymath/exp.c
index e9ea6fddb..4d6cf8541 100644
--- a/libc/tinymath/exp.c
+++ b/libc/tinymath/exp.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/exp_data.internal.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Double-precision e^x function.
diff --git a/libc/tinymath/exp10.c b/libc/tinymath/exp10.c
index 27baff1cd..db15c0c61 100644
--- a/libc/tinymath/exp10.c
+++ b/libc/tinymath/exp10.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns 10ˣ.
diff --git a/libc/tinymath/exp10f.c b/libc/tinymath/exp10f.c
index 16dc2071a..9bbae5350 100644
--- a/libc/tinymath/exp10f.c
+++ b/libc/tinymath/exp10f.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns 10ˣ.
diff --git a/libc/tinymath/exp10l.c b/libc/tinymath/exp10l.c
index e108701d7..1111bde6e 100644
--- a/libc/tinymath/exp10l.c
+++ b/libc/tinymath/exp10l.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns 10ˣ.
diff --git a/libc/tinymath/exp2.c b/libc/tinymath/exp2.c
index 1dc2c9a43..627a823ad 100644
--- a/libc/tinymath/exp2.c
+++ b/libc/tinymath/exp2.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/exp_data.internal.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Double-precision 2^x function.
diff --git a/libc/tinymath/exp2f.c b/libc/tinymath/exp2f.c
index 1687b8b30..ad68d4a70 100644
--- a/libc/tinymath/exp2f.c
+++ b/libc/tinymath/exp2f.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/exp2f_data.internal.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Single-precision 2^x function.
diff --git a/libc/tinymath/exp2f_data.c b/libc/tinymath/exp2f_data.c
index 94442cc73..e2b098251 100644
--- a/libc/tinymath/exp2f_data.c
+++ b/libc/tinymath/exp2f_data.c
@@ -1,9 +1,9 @@
 /*-*- mode:c;indent-tabs-mode:t;c-basic-offset:8;tab-width:8;coding:utf-8   -*-│
-│ vi: set et ft=c ts=8 sw=8 fenc=utf-8                                     :vi │
+│ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/exp2f_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Shared data between expf, exp2f and powf.
diff --git a/libc/tinymath/exp2l.c b/libc/tinymath/exp2l.c
index 28553aed2..a4c8f9a7f 100644
--- a/libc/tinymath/exp2l.c
+++ b/libc/tinymath/exp2l.c
@@ -28,16 +28,11 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
-#if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
 
 /* origin: FreeBSD /usr/src/lib/msun/ld80/s_exp2l.c and /usr/src/lib/msun/ld128/s_exp2l.c */
 /*-
diff --git a/libc/tinymath/exp_data.c b/libc/tinymath/exp_data.c
index 0d4b880a8..df7efc71e 100644
--- a/libc/tinymath/exp_data.c
+++ b/libc/tinymath/exp_data.c
@@ -27,12 +27,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/exp_data.internal.h"
 
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
 /*
  * Shared data between exp, exp2 and pow.
  *
diff --git a/libc/tinymath/expf.c b/libc/tinymath/expf.c
index 4879566ab..43f33219b 100644
--- a/libc/tinymath/expf.c
+++ b/libc/tinymath/expf.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/exp2f_data.internal.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Single-precision e^x function.
diff --git a/libc/tinymath/expl.c b/libc/tinymath/expl.c
index 5ff1010c3..52dffa2f4 100644
--- a/libc/tinymath/expl.c
+++ b/libc/tinymath/expl.c
@@ -17,18 +17,11 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_expl.c */
 /*
@@ -149,15 +142,7 @@ long double expl(long double x)
 
 #elif LDBL_MANT_DIG == 113 && LDBL_MAX_EXP == 16384
 #include "libc/tinymath/freebsd.internal.h"
-
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
 
 /*-
  * SPDX-License-Identifier: BSD-2-Clause-FreeBSD
diff --git a/libc/tinymath/expm1.c b/libc/tinymath/expm1.c
index 12600b69b..2fc2d66aa 100644
--- a/libc/tinymath/expm1.c
+++ b/libc/tinymath/expm1.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_expm1.c */
 /*
diff --git a/libc/tinymath/expm1f.c b/libc/tinymath/expm1f.c
index 6500aec27..097aa40e4 100644
--- a/libc/tinymath/expm1f.c
+++ b/libc/tinymath/expm1f.c
@@ -38,14 +38,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 static const float
 one		= 1.0,
diff --git a/libc/tinymath/expm1l.c b/libc/tinymath/expm1l.c
index bd585bca8..45dae3426 100644
--- a/libc/tinymath/expm1l.c
+++ b/libc/tinymath/expm1l.c
@@ -1,12 +1,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
-// clang-format off
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_expm1l.c */
 /*
diff --git a/libc/tinymath/expo2.c b/libc/tinymath/expo2.c
index 3f8eda928..befac99ad 100644
--- a/libc/tinymath/expo2.c
+++ b/libc/tinymath/expo2.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/expo.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define asdouble(i) ((union{uint64_t _i; double _f;}){i})._f
 #define INSERT_WORDS(d,hi,lo)                     \
diff --git a/libc/tinymath/expo2f.c b/libc/tinymath/expo2f.c
index 6d149db93..fc5c66245 100644
--- a/libc/tinymath/expo2f.c
+++ b/libc/tinymath/expo2f.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define asfloat(i) ((union{uint32_t _i; float _f;}){i})._f
 #define SET_FLOAT_WORD(d,w)                       \
diff --git a/libc/tinymath/floor.c b/libc/tinymath/floor.c
index c492fc05c..a19b2e8f4 100644
--- a/libc/tinymath/floor.c
+++ b/libc/tinymath/floor.c
@@ -31,12 +31,8 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if FLT_EVAL_METHOD==0 || FLT_EVAL_METHOD==1
 #define EPS DBL_EPSILON
diff --git a/libc/tinymath/floorf.c b/libc/tinymath/floorf.c
index de575ee27..c83aae1a7 100644
--- a/libc/tinymath/floorf.c
+++ b/libc/tinymath/floorf.c
@@ -30,12 +30,8 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns largest integral value not greater than 𝑥.
diff --git a/libc/tinymath/floorl.c b/libc/tinymath/floorl.c
index 499012d8f..37d17d9f1 100644
--- a/libc/tinymath/floorl.c
+++ b/libc/tinymath/floorl.c
@@ -29,12 +29,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns largest integral value not greater than 𝑥.
diff --git a/libc/tinymath/fma.c b/libc/tinymath/fma.c
index 72a0ba9a0..308e31a77 100644
--- a/libc/tinymath/fma.c
+++ b/libc/tinymath/fma.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/nexgen32e/x86feature.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define ASUINT64(x) ((union {double f; uint64_t i;}){x}).i
 #define ZEROINFNAN (0x7ff-0x3ff-52-1)
diff --git a/libc/tinymath/fmaf.c b/libc/tinymath/fmaf.c
index c65a3d481..544301ca5 100644
--- a/libc/tinymath/fmaf.c
+++ b/libc/tinymath/fmaf.c
@@ -28,15 +28,8 @@
 #include "libc/math.h"
 #include "libc/nexgen32e/x86feature.h"
 #include "libc/runtime/fenv.h"
-
-asm(".ident\t\"\\n\\n\
-Fused Multiply Add (MIT License)\\n\
-Copyright (c) 2005-2011 David Schultz <das@FreeBSD.ORG>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_fmaf.c */
 /*-
diff --git a/libc/tinymath/fmal.c b/libc/tinymath/fmal.c
index cefdca2d6..53b4db60f 100644
--- a/libc/tinymath/fmal.c
+++ b/libc/tinymath/fmal.c
@@ -31,13 +31,10 @@
 #include "libc/runtime/fenv.h"
 #include "libc/tinymath/freebsd.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
-#if (LDBL_MANT_DIG == 64 || LDBL_MANT_DIG == 113) && LDBL_MAX_EXP == 16384
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+
+#if (LDBL_MANT_DIG == 64 || LDBL_MANT_DIG == 113) && LDBL_MAX_EXP == 16384
 
 #if LDBL_MANT_DIG == 64
 #define LASTBIT(u) (u.i.m & 1)
diff --git a/libc/tinymath/fmod.c b/libc/tinymath/fmod.c
index e08205299..b9956e9ef 100644
--- a/libc/tinymath/fmod.c
+++ b/libc/tinymath/fmod.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Does (𝑥 rem 𝑦) w/ round()-style rounding.
diff --git a/libc/tinymath/fmodf.c b/libc/tinymath/fmodf.c
index 6e01ef28c..5e74153ef 100644
--- a/libc/tinymath/fmodf.c
+++ b/libc/tinymath/fmodf.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float fmodf(float x, float y)
 {
diff --git a/libc/tinymath/fmodl.c b/libc/tinymath/fmodl.c
index 2b7ecf532..ffe9fdb66 100644
--- a/libc/tinymath/fmodl.c
+++ b/libc/tinymath/fmodl.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Does (𝑥 rem 𝑦) w/ round()-style rounding.
diff --git a/libc/tinymath/freebsd.internal.h b/libc/tinymath/freebsd.internal.h
index 0aca0a1ac..dff035568 100644
--- a/libc/tinymath/freebsd.internal.h
+++ b/libc/tinymath/freebsd.internal.h
@@ -5,7 +5,6 @@
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
 COSMOPOLITAN_C_START_
-// clang-format off
 
 #define	__CONCAT1(x,y)	x ## y
 #define	__CONCAT(x,y)	__CONCAT1(x,y)
diff --git a/libc/tinymath/frexp.c b/libc/tinymath/frexp.c
index 4ba2e1b7a..78a8e80b3 100644
--- a/libc/tinymath/frexp.c
+++ b/libc/tinymath/frexp.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Splits number normalized fraction and exponent.
diff --git a/libc/tinymath/frexpf.c b/libc/tinymath/frexpf.c
index 4666ec4e5..759a87006 100644
--- a/libc/tinymath/frexpf.c
+++ b/libc/tinymath/frexpf.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 /**
  * Splits number normalized fraction and exponent.
diff --git a/libc/tinymath/frexpl.c b/libc/tinymath/frexpl.c
index 3d4fd6b41..45b5bbcac 100644
--- a/libc/tinymath/frexpl.c
+++ b/libc/tinymath/frexpl.c
@@ -29,13 +29,10 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 /**
  * Splits number normalized fraction and exponent.
diff --git a/libc/tinymath/horner_wrap.internal.h b/libc/tinymath/horner_wrap.internal.h
index 984c728c7..0c213a4a0 100644
--- a/libc/tinymath/horner_wrap.internal.h
+++ b/libc/tinymath/horner_wrap.internal.h
@@ -8,7 +8,6 @@
  * SPDX-License-Identifier: MIT OR Apache-2.0 WITH LLVM-exception
  */
 
-// clang-format off
 #define  HORNER_1_(x, c, i) FMA(c(i + 1), x, c(i))
 #define  HORNER_2_(x, c, i) FMA(HORNER_1_ (x, c, i + 1), x, c(i))
 #define  HORNER_3_(x, c, i) FMA(HORNER_2_ (x, c, i + 1), x, c(i))
diff --git a/libc/tinymath/hypot.c b/libc/tinymath/hypot.c
index cd8cac22b..37ba5b507 100644
--- a/libc/tinymath/hypot.c
+++ b/libc/tinymath/hypot.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if FLT_EVAL_METHOD > 1U && LDBL_MANT_DIG == 64
 #define SPLIT (0x1p32 + 1)
diff --git a/libc/tinymath/hypotf.c b/libc/tinymath/hypotf.c
index 5dd9c8eef..4692b97b7 100644
--- a/libc/tinymath/hypotf.c
+++ b/libc/tinymath/hypotf.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float hypotf(float x, float y)
 {
diff --git a/libc/tinymath/hypotl.c b/libc/tinymath/hypotl.c
index b524fd52b..a45fd7ed4 100644
--- a/libc/tinymath/hypotl.c
+++ b/libc/tinymath/hypotl.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if LDBL_MANT_DIG == 64
 #define SPLIT (0x1p32L+1)
diff --git a/libc/tinymath/ilogb.c b/libc/tinymath/ilogb.c
index fb327610d..595da8c28 100644
--- a/libc/tinymath/ilogb.c
+++ b/libc/tinymath/ilogb.c
@@ -28,12 +28,8 @@
 #include "libc/limits.h"
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns log₂𝑥 exponent part of double.
diff --git a/libc/tinymath/ilogbf.c b/libc/tinymath/ilogbf.c
index 6e3297604..744594eb1 100644
--- a/libc/tinymath/ilogbf.c
+++ b/libc/tinymath/ilogbf.c
@@ -28,12 +28,8 @@
 #include "libc/limits.h"
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns log₂𝑥 exponent part of double.
diff --git a/libc/tinymath/internal.h b/libc/tinymath/internal.h
index cc121fed0..5d4b097b1 100644
--- a/libc/tinymath/internal.h
+++ b/libc/tinymath/internal.h
@@ -8,7 +8,6 @@ COSMOPOLITAN_C_START_
 #define issignalingf_inline(x) 0
 #define issignaling_inline(x)  0
 
-// clang-format off
 #define asuint(f) ((union{float _f; uint32_t _i;}){f})._i
 #define asfloat(i) ((union{uint32_t _i; float _f;}){i})._f
 #define asuint64(f) ((union{double _f; uint64_t _i;}){f})._i
diff --git a/libc/tinymath/invtrigl.c b/libc/tinymath/invtrigl.c
index 67d984089..c03ff89da 100644
--- a/libc/tinymath/invtrigl.c
+++ b/libc/tinymath/invtrigl.c
@@ -31,15 +31,9 @@
 #include "libc/tinymath/invtrigl.internal.h"
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
 static const long double
diff --git a/libc/tinymath/j0.c b/libc/tinymath/j0.c
index 6d6eda6d5..2c4746213 100644
--- a/libc/tinymath/j0.c
+++ b/libc/tinymath/j0.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_j0.c */
 /*
diff --git a/libc/tinymath/j0f.c b/libc/tinymath/j0f.c
index 6723ab68b..80c663da0 100644
--- a/libc/tinymath/j0f.c
+++ b/libc/tinymath/j0f.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_j0f.c */
 /*
diff --git a/libc/tinymath/j1.c b/libc/tinymath/j1.c
index bd3240687..25b04ce1c 100644
--- a/libc/tinymath/j1.c
+++ b/libc/tinymath/j1.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_j1.c */
 /*
diff --git a/libc/tinymath/j1f.c b/libc/tinymath/j1f.c
index 8b86031f3..0b016b683 100644
--- a/libc/tinymath/j1f.c
+++ b/libc/tinymath/j1f.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_j1f.c */
 /*
diff --git a/libc/tinymath/jn.c b/libc/tinymath/jn.c
index 5df41b18b..2b5213e21 100644
--- a/libc/tinymath/jn.c
+++ b/libc/tinymath/jn.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_jn.c */
 /*
diff --git a/libc/tinymath/jnf.c b/libc/tinymath/jnf.c
index ff33eb8ac..09f67338d 100644
--- a/libc/tinymath/jnf.c
+++ b/libc/tinymath/jnf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_jnf.c */
 /*
diff --git a/libc/tinymath/kcos.c b/libc/tinymath/kcos.c
index f2467a0b7..a80529aec 100644
--- a/libc/tinymath/kcos.c
+++ b/libc/tinymath/kcos.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_cos.c */
 /*
diff --git a/libc/tinymath/kcosl.c b/libc/tinymath/kcosl.c
index 8d54f5fcb..6563de59b 100644
--- a/libc/tinymath/kcosl.c
+++ b/libc/tinymath/kcosl.c
@@ -27,18 +27,10 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/ld80/k_cosl.c */
 /* origin: FreeBSD /usr/src/lib/msun/ld128/k_cosl.c */
diff --git a/libc/tinymath/kexpl.c b/libc/tinymath/kexpl.c
index 5dc19b613..d02bf2aaa 100644
--- a/libc/tinymath/kexpl.c
+++ b/libc/tinymath/kexpl.c
@@ -37,11 +37,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/freebsd.internal.h"
+__static_yoink("freebsd_libm_notice");
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-// clang-format off
 
 /*
  * ld128 version of k_expl.h.  See ../ld80/s_expl.c for most comments.
diff --git a/libc/tinymath/ksin.c b/libc/tinymath/ksin.c
index 2a61bcbac..080382c19 100644
--- a/libc/tinymath/ksin.c
+++ b/libc/tinymath/ksin.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_sin.c */
 /*
diff --git a/libc/tinymath/ksinl.c b/libc/tinymath/ksinl.c
index c430c0322..bce105867 100644
--- a/libc/tinymath/ksinl.c
+++ b/libc/tinymath/ksinl.c
@@ -28,17 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/ld80/k_sinl.c */
 /* origin: FreeBSD /usr/src/lib/msun/ld128/k_sinl.c */
diff --git a/libc/tinymath/ktan.c b/libc/tinymath/ktan.c
index 695c6b70e..02ca7daa6 100644
--- a/libc/tinymath/ktan.c
+++ b/libc/tinymath/ktan.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_tan.c */
 /*
diff --git a/libc/tinymath/ktanl.c b/libc/tinymath/ktanl.c
index cbe9228b8..f7b618bc2 100644
--- a/libc/tinymath/ktanl.c
+++ b/libc/tinymath/ktanl.c
@@ -27,19 +27,12 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 /* origin: FreeBSD /usr/src/lib/msun/ld80/k_tanl.c */
 /* origin: FreeBSD /usr/src/lib/msun/ld128/k_tanl.c */
diff --git a/libc/tinymath/lgamma_r.c b/libc/tinymath/lgamma_r.c
index 356155bca..4632b144c 100644
--- a/libc/tinymath/lgamma_r.c
+++ b/libc/tinymath/lgamma_r.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_lgamma_r.c */
 /*
diff --git a/libc/tinymath/lgammaf_r.c b/libc/tinymath/lgammaf_r.c
index f51eeea00..ee6b7a286 100644
--- a/libc/tinymath/lgammaf_r.c
+++ b/libc/tinymath/lgammaf_r.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_lgammaf_r.c */
 /*
diff --git a/libc/tinymath/lgammal.c b/libc/tinymath/lgammal.c
index de54194b7..ec3309acb 100644
--- a/libc/tinymath/lgammal.c
+++ b/libc/tinymath/lgammal.c
@@ -29,15 +29,9 @@
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_lgammal.c */
 /*
diff --git a/libc/tinymath/log.c b/libc/tinymath/log.c
index 61b13ef62..e793dd923 100644
--- a/libc/tinymath/log.c
+++ b/libc/tinymath/log.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Double-precision log(x) function.
diff --git a/libc/tinymath/log10.c b/libc/tinymath/log10.c
index 3c3f19650..e073c1e38 100644
--- a/libc/tinymath/log10.c
+++ b/libc/tinymath/log10.c
@@ -30,15 +30,9 @@
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log2_data.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_log10.c */
 /*
diff --git a/libc/tinymath/log10f.c b/libc/tinymath/log10f.c
index e078f8982..90ecb8cea 100644
--- a/libc/tinymath/log10f.c
+++ b/libc/tinymath/log10f.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_log10f.c */
 /*
diff --git a/libc/tinymath/log10l.c b/libc/tinymath/log10l.c
index 04f658b86..35de4cc8b 100644
--- a/libc/tinymath/log10l.c
+++ b/libc/tinymath/log10l.c
@@ -28,15 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_log10l.c */
 /*
diff --git a/libc/tinymath/log1p.c b/libc/tinymath/log1p.c
index 3d6883422..97e9b51c6 100644
--- a/libc/tinymath/log1p.c
+++ b/libc/tinymath/log1p.c
@@ -29,12 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("freebsd_libm_notice");
+__static_yoink("fdlibm_notice");
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_log1p.c */
 /*
diff --git a/libc/tinymath/log1pf.c b/libc/tinymath/log1pf.c
index 0f1b867f9..59fe2aace 100644
--- a/libc/tinymath/log1pf.c
+++ b/libc/tinymath/log1pf.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log1pf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 #define Ln2 (0x1.62e43p-1f)
 #define SignMask (0x80000000)
diff --git a/libc/tinymath/log1pf_data.c b/libc/tinymath/log1pf_data.c
index eea85b405..abcc21bef 100644
--- a/libc/tinymath/log1pf_data.c
+++ b/libc/tinymath/log1pf_data.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/log1pf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /* Polynomial coefficients generated using floating-point minimax
    algorithm, see tools/log1pf.sollya for details.  */
diff --git a/libc/tinymath/log1pl.c b/libc/tinymath/log1pl.c
index 42f5a3b17..784feab56 100644
--- a/libc/tinymath/log1pl.c
+++ b/libc/tinymath/log1pl.c
@@ -29,15 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/s_log1pl.c */
 /*
diff --git a/libc/tinymath/log2.c b/libc/tinymath/log2.c
index d1d4d40f9..29a774c56 100644
--- a/libc/tinymath/log2.c
+++ b/libc/tinymath/log2.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -30,12 +30,7 @@
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log2_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Double-precision log2(x) function.
diff --git a/libc/tinymath/log2_data.c b/libc/tinymath/log2_data.c
index 259ffd0db..2ecd1da72 100644
--- a/libc/tinymath/log2_data.c
+++ b/libc/tinymath/log2_data.c
@@ -27,12 +27,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/log2_data.internal.h"
 
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
-
 /*
  * Data for log2.
  *
diff --git a/libc/tinymath/log2f.c b/libc/tinymath/log2f.c
index 118a02d29..d7cc7c88e 100644
--- a/libc/tinymath/log2f.c
+++ b/libc/tinymath/log2f.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -30,12 +30,7 @@
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/log2f_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Single-precision log2 function.
diff --git a/libc/tinymath/log2f_data.c b/libc/tinymath/log2f_data.c
index 783213a91..bd04b407d 100644
--- a/libc/tinymath/log2f_data.c
+++ b/libc/tinymath/log2f_data.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/log2f_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Data definition for log2f.
diff --git a/libc/tinymath/log2l.c b/libc/tinymath/log2l.c
index fd1489688..6ad25ef6c 100644
--- a/libc/tinymath/log2l.c
+++ b/libc/tinymath/log2l.c
@@ -29,15 +29,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_log2l.c */
 /*
diff --git a/libc/tinymath/log_data.c b/libc/tinymath/log_data.c
index be87c4105..0dca131ae 100644
--- a/libc/tinymath/log_data.c
+++ b/libc/tinymath/log_data.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/log_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Data for log.
diff --git a/libc/tinymath/logf.c b/libc/tinymath/logf.c
index 2bce5c3a3..c25d898b4 100644
--- a/libc/tinymath/logf.c
+++ b/libc/tinymath/logf.c
@@ -29,12 +29,7 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/logf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Single-precision log function.
diff --git a/libc/tinymath/logf_data.c b/libc/tinymath/logf_data.c
index 504d50a26..cd8f944fe 100644
--- a/libc/tinymath/logf_data.c
+++ b/libc/tinymath/logf_data.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/logf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Data definition for logf.
diff --git a/libc/tinymath/logl.c b/libc/tinymath/logl.c
index b6f269c45..6f4c36d16 100644
--- a/libc/tinymath/logl.c
+++ b/libc/tinymath/logl.c
@@ -28,15 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_logl.c */
 /*
diff --git a/libc/tinymath/loglq.c b/libc/tinymath/loglq.c
index 0d2dbdcfb..5af871c9c 100644
--- a/libc/tinymath/loglq.c
+++ b/libc/tinymath/loglq.c
@@ -29,13 +29,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
+__static_yoink("freebsd_libm_notice");
 #if LDBL_MANT_DIG == 113
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Implementation of the natural logarithm of x for 128-bit format.
diff --git a/libc/tinymath/modf.c b/libc/tinymath/modf.c
index 3047590c4..2701093c1 100644
--- a/libc/tinymath/modf.c
+++ b/libc/tinymath/modf.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns fractional part of 𝑥.
diff --git a/libc/tinymath/modff.c b/libc/tinymath/modff.c
index 99e60840c..720f05e80 100644
--- a/libc/tinymath/modff.c
+++ b/libc/tinymath/modff.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float modff(float x, float *iptr)
 {
diff --git a/libc/tinymath/modfl.c b/libc/tinymath/modfl.c
index 9b430c646..0d3f585ec 100644
--- a/libc/tinymath/modfl.c
+++ b/libc/tinymath/modfl.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 static const long double toint = 1/LDBL_EPSILON;
 
diff --git a/libc/tinymath/nextafter.c b/libc/tinymath/nextafter.c
index f041bf0f3..a1da2b40a 100644
--- a/libc/tinymath/nextafter.c
+++ b/libc/tinymath/nextafter.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 double nextafter(double x, double y)
 {
diff --git a/libc/tinymath/nextafterf.c b/libc/tinymath/nextafterf.c
index 80e884413..82bb5bd48 100644
--- a/libc/tinymath/nextafterf.c
+++ b/libc/tinymath/nextafterf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float nextafterf(float x, float y)
 {
diff --git a/libc/tinymath/nextafterl.c b/libc/tinymath/nextafterl.c
index e85824851..c213e93ff 100644
--- a/libc/tinymath/nextafterl.c
+++ b/libc/tinymath/nextafterl.c
@@ -30,12 +30,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 long double nextafterl(long double x, long double y)
 {
diff --git a/libc/tinymath/nexttoward.c b/libc/tinymath/nexttoward.c
index c7a16bd50..d451122de 100644
--- a/libc/tinymath/nexttoward.c
+++ b/libc/tinymath/nexttoward.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 double nexttoward(double x, long double y)
 {
diff --git a/libc/tinymath/nexttowardf.c b/libc/tinymath/nexttowardf.c
index da5eb59ba..470c27edd 100644
--- a/libc/tinymath/nexttowardf.c
+++ b/libc/tinymath/nexttowardf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float nexttowardf(float x, long double y)
 {
diff --git a/libc/tinymath/polevll.c b/libc/tinymath/polevll.c
index aac4505d8..92356ca36 100644
--- a/libc/tinymath/polevll.c
+++ b/libc/tinymath/polevll.c
@@ -26,15 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/polevll.c */
 /*
diff --git a/libc/tinymath/pow.c b/libc/tinymath/pow.c
index 4e223fb54..abca8c6d5 100644
--- a/libc/tinymath/pow.c
+++ b/libc/tinymath/pow.c
@@ -30,12 +30,7 @@
 #include "libc/tinymath/exp_data.internal.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/pow_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Double-precision x^y function.
diff --git a/libc/tinymath/pow_data.c b/libc/tinymath/pow_data.c
index 43dc5d008..ce804633e 100644
--- a/libc/tinymath/pow_data.c
+++ b/libc/tinymath/pow_data.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/pow_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Data for the log part of pow.
diff --git a/libc/tinymath/powf.c b/libc/tinymath/powf.c
index 2ad8f7f63..aeab1bbb7 100644
--- a/libc/tinymath/powf.c
+++ b/libc/tinymath/powf.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -31,12 +31,7 @@
 #include "libc/tinymath/exp_data.internal.h"
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/powf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Copyright (c) 2017-2018, Arm Limited.
diff --git a/libc/tinymath/powf_data.c b/libc/tinymath/powf_data.c
index 0ef109ad6..77cd461f1 100644
--- a/libc/tinymath/powf_data.c
+++ b/libc/tinymath/powf_data.c
@@ -2,8 +2,8 @@
 │ vi: set noet ft=c ts=8 sw=8 fenc=utf-8                                   :vi │
 ╚──────────────────────────────────────────────────────────────────────────────╝
 │                                                                              │
-│  Musl Libc                                                                   │
-│  Copyright © 2005-2014 Rich Felker, et al.                                   │
+│  Optimized Routines                                                          │
+│  Copyright (c) 1999-2022, Arm Limited.                                       │
 │                                                                              │
 │  Permission is hereby granted, free of charge, to any person obtaining       │
 │  a copy of this software and associated documentation files (the             │
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/powf_data.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Double-precision math functions (MIT License)\\n\
-Copyright 2018 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /*
  * Data definition for powf.
diff --git a/libc/tinymath/powl.c b/libc/tinymath/powl.c
index cb53b0e2c..6de2ae99a 100644
--- a/libc/tinymath/powl.c
+++ b/libc/tinymath/powl.c
@@ -96,15 +96,8 @@ long double powl(long double x, long double y) {
 }
 
 #elif LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
-
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_powl.c */
 /*
@@ -616,10 +609,6 @@ static long double powil(long double x, int nn)
 #elif LDBL_MANT_DIG == 113 && LDBL_MAX_EXP == 16384
 #include "libc/tinymath/freebsd.internal.h"
 
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-
 /*-
  * ====================================================
  * Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.
diff --git a/libc/tinymath/poz.c b/libc/tinymath/poz.c
index 4599ab71a..09c4a0476 100644
--- a/libc/tinymath/poz.c
+++ b/libc/tinymath/poz.c
@@ -1,7 +1,6 @@
 /*-*- mode:c;indent-tabs-mode:t;c-basic-offset:4;tab-width:4;coding:utf-8   -*-│
 │ vi: set noet ft=c ts=4 sts=4 sw=4 fenc=utf-8                             :vi │
 ╚─────────────────────────────────────────────────────────────────────────────*/
-/* clang-format off */
 /*
 
     Compute probability of measured Chi Square value.
diff --git a/libc/tinymath/randtest.c b/libc/tinymath/randtest.c
index ae96db067..10dc8d5c4 100644
--- a/libc/tinymath/randtest.c
+++ b/libc/tinymath/randtest.c
@@ -1,4 +1,3 @@
-/* clang-format off */
 /*
 
 	 Apply various randomness tests to a stream of bytes
diff --git a/libc/tinymath/remainderf.c b/libc/tinymath/remainderf.c
index b02a9da9c..f403db37e 100644
--- a/libc/tinymath/remainderf.c
+++ b/libc/tinymath/remainderf.c
@@ -19,7 +19,6 @@
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
 
-// clang-format off
 
 /**
  * Returns remainder of dividing 𝑥 by 𝑦.
diff --git a/libc/tinymath/rempio2.c b/libc/tinymath/rempio2.c
index 7337b0684..09565bcd7 100644
--- a/libc/tinymath/rempio2.c
+++ b/libc/tinymath/rempio2.c
@@ -29,15 +29,9 @@
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_rem_pio2.c */
 /*
diff --git a/libc/tinymath/rempio2f.c b/libc/tinymath/rempio2f.c
index 86ed07f49..b7ea0658b 100644
--- a/libc/tinymath/rempio2f.c
+++ b/libc/tinymath/rempio2f.c
@@ -29,15 +29,9 @@
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_rem_pio2f.c */
 /*
diff --git a/libc/tinymath/rempio2l.c b/libc/tinymath/rempio2l.c
index 5648b89aa..818989d80 100644
--- a/libc/tinymath/rempio2l.c
+++ b/libc/tinymath/rempio2l.c
@@ -29,19 +29,13 @@
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
+
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/ld80/e_rem_pio2.c */
 /*
diff --git a/libc/tinymath/rempio2large.c b/libc/tinymath/rempio2large.c
index 4f2b2ad22..7dce93dab 100644
--- a/libc/tinymath/rempio2large.c
+++ b/libc/tinymath/rempio2large.c
@@ -27,17 +27,11 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
 #pragma GCC diagnostic ignored "-Wmaybe-uninitialized"
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_rem_pio2.c */
 /*
diff --git a/libc/tinymath/remquo.c b/libc/tinymath/remquo.c
index 975451a21..156a1e320 100644
--- a/libc/tinymath/remquo.c
+++ b/libc/tinymath/remquo.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Computes remainder and part of quotient.
diff --git a/libc/tinymath/remquof.c b/libc/tinymath/remquof.c
index ca4fabcf4..a5eddfdb3 100644
--- a/libc/tinymath/remquof.c
+++ b/libc/tinymath/remquof.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Computes remainder and part of quotient.
diff --git a/libc/tinymath/remquol.c b/libc/tinymath/remquol.c
index e5c196fa4..4e86521ee 100644
--- a/libc/tinymath/remquol.c
+++ b/libc/tinymath/remquol.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Computes remainder and part of quotient.
diff --git a/libc/tinymath/rint.c b/libc/tinymath/rint.c
index dd1732d18..f8cfc91f5 100644
--- a/libc/tinymath/rint.c
+++ b/libc/tinymath/rint.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if FLT_EVAL_METHOD==0 || FLT_EVAL_METHOD==1
 #define EPS DBL_EPSILON
diff --git a/libc/tinymath/rintf.c b/libc/tinymath/rintf.c
index d9270e6c4..abef412d7 100644
--- a/libc/tinymath/rintf.c
+++ b/libc/tinymath/rintf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if FLT_EVAL_METHOD==0
 #define EPS FLT_EPSILON
diff --git a/libc/tinymath/rintl.c b/libc/tinymath/rintl.c
index b9c2e9bce..ba6324740 100644
--- a/libc/tinymath/rintl.c
+++ b/libc/tinymath/rintl.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Rounds to integer in current rounding mode.
diff --git a/libc/tinymath/round.c b/libc/tinymath/round.c
index d5966f9eb..800df248f 100644
--- a/libc/tinymath/round.c
+++ b/libc/tinymath/round.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #if FLT_EVAL_METHOD==0 || FLT_EVAL_METHOD==1
 #define EPS DBL_EPSILON
diff --git a/libc/tinymath/roundf.c b/libc/tinymath/roundf.c
index e84568cee..4eac941c3 100644
--- a/libc/tinymath/roundf.c
+++ b/libc/tinymath/roundf.c
@@ -28,12 +28,7 @@
 #include "libc/math.h"
 #include "libc/runtime/fenv.h"
 #include "libc/tinymath/internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("musl_libc_notice");
 
 #if FLT_EVAL_METHOD==0
 #define EPS FLT_EPSILON
diff --git a/libc/tinymath/roundl.c b/libc/tinymath/roundl.c
index 2eb2bd7d0..3f2518945 100644
--- a/libc/tinymath/roundl.c
+++ b/libc/tinymath/roundl.c
@@ -29,12 +29,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Rounds 𝑥 to nearest integer, away from zero.
diff --git a/libc/tinymath/scalb.c b/libc/tinymath/scalb.c
index f8137e3a2..73048c389 100644
--- a/libc/tinymath/scalb.c
+++ b/libc/tinymath/scalb.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/e_scalb.c */
 /*
diff --git a/libc/tinymath/scalbf.c b/libc/tinymath/scalbf.c
index e71d1e719..e3f0eb8f9 100644
--- a/libc/tinymath/scalbf.c
+++ b/libc/tinymath/scalbf.c
@@ -26,15 +26,9 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 float scalbf(float x, float fn)
 {
diff --git a/libc/tinymath/sin.c b/libc/tinymath/sin.c
index 362e1a5bb..468e93128 100644
--- a/libc/tinymath/sin.c
+++ b/libc/tinymath/sin.c
@@ -28,15 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_sin.c */
 /*
diff --git a/libc/tinymath/sincos.c b/libc/tinymath/sincos.c
index 69fcec219..934123dc2 100644
--- a/libc/tinymath/sincos.c
+++ b/libc/tinymath/sincos.c
@@ -29,15 +29,9 @@
 #include "libc/runtime/runtime.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_sin.c */
 /*
diff --git a/libc/tinymath/sincosf.c b/libc/tinymath/sincosf.c
index 33371598d..cc0a52d55 100644
--- a/libc/tinymath/sincosf.c
+++ b/libc/tinymath/sincosf.c
@@ -28,12 +28,7 @@
 #include "libc/intrin/likely.h"
 #include "libc/math.h"
 #include "libc/tinymath/sincosf.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("arm_optimized_routines_notice");
 
 /* Fast sincosf implementation.  Worst-case ULP is 0.5607, maximum relative
    error is 0.5303 * 2^-23.  A single-step range reduction is used for
diff --git a/libc/tinymath/sincosf.internal.h b/libc/tinymath/sincosf.internal.h
index e07239828..57a596c98 100644
--- a/libc/tinymath/sincosf.internal.h
+++ b/libc/tinymath/sincosf.internal.h
@@ -2,7 +2,6 @@
 #define COSMOPOLITAN_LIBC_TINYMATH_SINCOSF_INTERNAL_H_
 #include "libc/tinymath/internal.h"
 COSMOPOLITAN_C_START_
-// clang-format off
 
 /*
  * Header for sinf, cosf and sincosf.
diff --git a/libc/tinymath/sincosf_data.c b/libc/tinymath/sincosf_data.c
index d90de5455..61a9ff067 100644
--- a/libc/tinymath/sincosf_data.c
+++ b/libc/tinymath/sincosf_data.c
@@ -26,12 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/sincosf.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Optimized Routines (MIT License)\\n\
-Copyright 2022 ARM Limited\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
+__static_yoink("arm_optimized_routines_notice");
 
 /* The constants and polynomials for sine and cosine.  The 2nd entry
    computes -cos (x) rather than cos (x) to get negation for free.  */
diff --git a/libc/tinymath/sincosl.c b/libc/tinymath/sincosl.c
index 8cc1c3d2b..cb4d03a49 100644
--- a/libc/tinymath/sincosl.c
+++ b/libc/tinymath/sincosl.c
@@ -30,12 +30,8 @@
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns sine and cosine of 𝑥.
diff --git a/libc/tinymath/sindf.c b/libc/tinymath/sindf.c
index 06a4196f1..60608e0c1 100644
--- a/libc/tinymath/sindf.c
+++ b/libc/tinymath/sindf.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_sinf.c */
 /*
diff --git a/libc/tinymath/sinf.c b/libc/tinymath/sinf.c
index 81d3218d2..14f8de8a5 100644
--- a/libc/tinymath/sinf.c
+++ b/libc/tinymath/sinf.c
@@ -29,15 +29,9 @@
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_sinf.c */
 /*
diff --git a/libc/tinymath/sinh.c b/libc/tinymath/sinh.c
index d740da683..6c3b9975c 100644
--- a/libc/tinymath/sinh.c
+++ b/libc/tinymath/sinh.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/expo.internal.h"
 #include "libc/tinymath/freebsd.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns hyperbolic sine of 𝑥.
diff --git a/libc/tinymath/sinhf.c b/libc/tinymath/sinhf.c
index 75fcb49c6..1f1e359f3 100644
--- a/libc/tinymath/sinhf.c
+++ b/libc/tinymath/sinhf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/expo.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns hyperbolic sine of 𝑥.
diff --git a/libc/tinymath/sinhl.c b/libc/tinymath/sinhl.c
index 372038ada..bb642a6df 100644
--- a/libc/tinymath/sinhl.c
+++ b/libc/tinymath/sinhl.c
@@ -39,16 +39,11 @@
 #include "libc/intrin/likely.h"
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 #if LDBL_MAX_EXP != 0x4000
 /* We also require the usual expsign encoding. */
diff --git a/libc/tinymath/sinl.c b/libc/tinymath/sinl.c
index b65e21e4b..2d87d6b3f 100644
--- a/libc/tinymath/sinl.c
+++ b/libc/tinymath/sinl.c
@@ -30,12 +30,8 @@
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns sine of 𝑥.
diff --git a/libc/tinymath/sqrt.c b/libc/tinymath/sqrt.c
index b6e1cd215..1d23a9fd7 100644
--- a/libc/tinymath/sqrt.c
+++ b/libc/tinymath/sqrt.c
@@ -28,12 +28,8 @@
 #include "libc/intrin/likely.h"
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define FENV_SUPPORT 1
 
diff --git a/libc/tinymath/sqrt_data.c b/libc/tinymath/sqrt_data.c
index 9850440cd..c0e9e6f75 100644
--- a/libc/tinymath/sqrt_data.c
+++ b/libc/tinymath/sqrt_data.c
@@ -26,12 +26,8 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 const uint16_t __rsqrt_tab[128] = {
 0xb451,0xb2f0,0xb196,0xb044,0xaef9,0xadb6,0xac79,0xab43,
diff --git a/libc/tinymath/sqrtf.c b/libc/tinymath/sqrtf.c
index 21374d03f..f37ecbc12 100644
--- a/libc/tinymath/sqrtf.c
+++ b/libc/tinymath/sqrtf.c
@@ -28,12 +28,8 @@
 #include "libc/intrin/likely.h"
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define FENV_SUPPORT 1
 
diff --git a/libc/tinymath/sqrtl.c b/libc/tinymath/sqrtl.c
index 4bf9d7df2..335d40f4d 100644
--- a/libc/tinymath/sqrtl.c
+++ b/libc/tinymath/sqrtl.c
@@ -30,12 +30,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 #define FENV_SUPPORT 1
 
diff --git a/libc/tinymath/tan.c b/libc/tinymath/tan.c
index 0d5a45f73..92d072404 100644
--- a/libc/tinymath/tan.c
+++ b/libc/tinymath/tan.c
@@ -28,15 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_tan.c */
 /*
diff --git a/libc/tinymath/tandf.c b/libc/tinymath/tandf.c
index 93b79289b..bff4e4adb 100644
--- a/libc/tinymath/tandf.c
+++ b/libc/tinymath/tandf.c
@@ -27,15 +27,9 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
+__static_yoink("fdlibm_notice");
 
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-/* clang-format off */
 
 /* origin: FreeBSD /usr/src/lib/msun/src/k_tanf.c */
 /*
diff --git a/libc/tinymath/tanf.c b/libc/tinymath/tanf.c
index 5fb7c4453..508fec260 100644
--- a/libc/tinymath/tanf.c
+++ b/libc/tinymath/tanf.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/complex.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: FreeBSD /usr/src/lib/msun/src/s_tanf.c */
 /*
diff --git a/libc/tinymath/tanh.c b/libc/tinymath/tanh.c
index 043d788f7..760440a1f 100644
--- a/libc/tinymath/tanh.c
+++ b/libc/tinymath/tanh.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns hyperbolic tangent of 𝑥.
diff --git a/libc/tinymath/tanhf.c b/libc/tinymath/tanhf.c
index 2b735ff2d..628a4860e 100644
--- a/libc/tinymath/tanhf.c
+++ b/libc/tinymath/tanhf.c
@@ -27,12 +27,8 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns hyperbolic tangent of 𝑥.
diff --git a/libc/tinymath/tanhl.c b/libc/tinymath/tanhl.c
index 295f72de7..d1e8ffb78 100644
--- a/libc/tinymath/tanhl.c
+++ b/libc/tinymath/tanhl.c
@@ -39,16 +39,11 @@
 #include "libc/intrin/likely.h"
 #include "libc/math.h"
 #include "libc/tinymath/freebsd.internal.h"
-#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
-asm(".ident\t\"\\n\\n\
-FreeBSD libm (BSD-2 License)\\n\
-Copyright (c) 2005-2011, Bruce D. Evans, Steven G. Kargl, David Schultz.\"");
-asm(".ident\t\"\\n\\n\
-fdlibm (fdlibm license)\\n\
-Copyright (C) 1993 by Sun Microsystems, Inc. All rights reserved.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
+__static_yoink("fdlibm_notice");
+__static_yoink("freebsd_libm_notice");
+
+#if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
 
 #if LDBL_MAX_EXP != 0x4000
 /* We also require the usual expsign encoding. */
diff --git a/libc/tinymath/tanl.c b/libc/tinymath/tanl.c
index 9ef8475a7..90b6d69e4 100644
--- a/libc/tinymath/tanl.c
+++ b/libc/tinymath/tanl.c
@@ -30,12 +30,8 @@
 #include "libc/tinymath/kernel.internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Returns tangent of x.
diff --git a/libc/tinymath/tgamma.c b/libc/tinymath/tgamma.c
index 35369f448..822f43249 100644
--- a/libc/tinymath/tgamma.c
+++ b/libc/tinymath/tgamma.c
@@ -28,12 +28,8 @@
 #include "libc/math.h"
 #include "libc/tinymath/feval.internal.h"
 #include "libc/tinymath/kernel.internal.h"
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /*
 "A Precision Approximation of the Gamma Function" - Cornelius Lanczos (1964)
diff --git a/libc/tinymath/tgammal.c b/libc/tinymath/tgammal.c
index ae73f31dd..ee5ffafd1 100644
--- a/libc/tinymath/tgammal.c
+++ b/libc/tinymath/tgammal.c
@@ -28,15 +28,9 @@
 #include "libc/math.h"
 #include "libc/tinymath/internal.h"
 #if LDBL_MANT_DIG == 64 && LDBL_MAX_EXP == 16384
+__static_yoink("musl_libc_notice");
+__static_yoink("openbsd_libm_notice");
 
-asm(".ident\t\"\\n\\n\
-OpenBSD libm (ISC License)\\n\
-Copyright (c) 2008 Stephen L. Moshier <steve@moshier.net>\"");
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /* origin: OpenBSD /usr/src/lib/libm/src/ld80/e_tgammal.c */
 /*
diff --git a/libc/tinymath/trunc.c b/libc/tinymath/trunc.c
index 0262f4028..3dc404c6d 100644
--- a/libc/tinymath/trunc.c
+++ b/libc/tinymath/trunc.c
@@ -30,12 +30,8 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Rounds to integer, towards zero.
diff --git a/libc/tinymath/truncf.c b/libc/tinymath/truncf.c
index 3f85aef61..906e00b0a 100644
--- a/libc/tinymath/truncf.c
+++ b/libc/tinymath/truncf.c
@@ -30,12 +30,8 @@
 #ifndef __llvm__
 #include "third_party/intel/smmintrin.internal.h"
 #endif
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Rounds to integer, towards zero.
diff --git a/libc/tinymath/truncl.c b/libc/tinymath/truncl.c
index 52f8586ca..55b305294 100644
--- a/libc/tinymath/truncl.c
+++ b/libc/tinymath/truncl.c
@@ -29,12 +29,8 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
+__static_yoink("musl_libc_notice");
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-// clang-format off
 
 /**
  * Rounds to integer, towards zero.
diff --git a/net/http/base32.c b/net/http/base32.c
index 56c56c9ef..6e7963d38 100644
--- a/net/http/base32.c
+++ b/net/http/base32.c
@@ -21,15 +21,11 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-Apache License, Version 2.0\\n\
-Copyright 2010 Google Inc.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 const char base32def[] = "0123456789abcdefghjkmnpqrstvwxyz";
 
 int tobits(int b) {
-  int bits = 0; while (b && (b >>= 1)) bits++;
+  int bits = 0;
+  while (b && (b >>= 1)) bits++;
   return bits;
 }
 
@@ -48,20 +44,19 @@ int tobits(int b) {
  * @param ol if non-NULL receives output length
  * @return allocated NUL-terminated buffer, or NULL w/ errno
  */
-char* EncodeBase32(const char *s, size_t sl,
-                   const char *a, size_t al,
+char *EncodeBase32(const char *s, size_t sl, const char *a, size_t al,
                    size_t *ol) {
   size_t count = 0;
   char *r = NULL;
   if (sl == -1) sl = s ? strlen(s) : 0;
   if (al == 0) {
     a = base32def;
-    al = sizeof(base32def)/sizeof(a[0]);
+    al = sizeof(base32def) / sizeof(a[0]);
   }
   unassert(2 <= al && al <= 128);
   int bl = tobits(al);
   int mask = (1 << bl) - 1;
-  size_t n = (sl * 8 + bl - 1) / bl; // calculate output length
+  size_t n = (sl * 8 + bl - 1) / bl;  // calculate output length
   if ((r = malloc(n + 1))) {
     int buffer = s[0];
     size_t next = 1;
@@ -92,11 +87,11 @@ static signed char kBase32[256] = {
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -2, -2, -1, -1, -2, -1, -1,  // 0x00
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  // 0x10
     -2, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -2, -1, -1,  // 0x20
-     0,  1,  2,  3,  4,  5,  6,  7,  8,  9, -1, -1, -1, -1, -1, -1,  // 0x30
-    -1, 10, 11, 12, 13, 14, 15, 16, 17,  1, 18, 19,  1, 20, 21, -1,  // 0x40
-    22, 23, 24, 25, 26,  0, 27, 28, 29, 30, 31, -1, -1, -1, -1, -1,  // 0x50
-    -1, 10, 11, 12, 13, 14, 15, 16, 17,  1, 18, 19,  1, 20, 21, -1,  // 0x60
-    22, 23, 24, 25, 26,  0, 27, 28, 29, 30, 31, -1, -1, -1, -1, -1,  // 0x70
+    0,  1,  2,  3,  4,  5,  6,  7,  8,  9,  -1, -1, -1, -1, -1, -1,  // 0x30
+    -1, 10, 11, 12, 13, 14, 15, 16, 17, 1,  18, 19, 1,  20, 21, -1,  // 0x40
+    22, 23, 24, 25, 26, 0,  27, 28, 29, 30, 31, -1, -1, -1, -1, -1,  // 0x50
+    -1, 10, 11, 12, 13, 14, 15, 16, 17, 1,  18, 19, 1,  20, 21, -1,  // 0x60
+    22, 23, 24, 25, 26, 0,  27, 28, 29, 30, 31, -1, -1, -1, -1, -1,  // 0x70
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  // 0x80
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  // 0x90
     -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  // 0xa0
@@ -120,19 +115,18 @@ static signed char kBase32[256] = {
  * @param ol if non-NULL receives output length
  * @return allocated NUL-terminated buffer, or NULL w/ errno
  */
-char* DecodeBase32(const char *s, size_t sl,
-                   const char *a, size_t al,
+char *DecodeBase32(const char *s, size_t sl, const char *a, size_t al,
                    size_t *ol) {
   size_t count = 0;
   char *r = NULL;
   if (sl == -1) sl = s ? strlen(s) : 0;
   if (al == 0) {
     a = base32def;
-    al = sizeof(base32def)/sizeof(a[0]);
+    al = sizeof(base32def) / sizeof(a[0]);
   }
   unassert(2 <= al && al <= 128);
   int bl = tobits(al);
-  size_t n = (sl * bl + 1) / 8 + 1; // calculate output length
+  size_t n = (sl * bl + 1) / 8 + 1;  // calculate output length
   // process input
   if ((r = malloc(n + 1))) {
     unsigned int buffer = 0;
diff --git a/test/libc/calls/dup_test.c b/test/libc/calls/dup_test.c
index c8cb7098a..cad66f18e 100644
--- a/test/libc/calls/dup_test.c
+++ b/test/libc/calls/dup_test.c
@@ -40,7 +40,8 @@ void SetUpOnce(void) {
   testlib_enable_tmp_setup_teardown();
 }
 
-static textstartup void TestInit(int argc, char **argv) {
+__attribute__((__constructor__)) static textstartup void TestInit(int argc,
+                                                                  char **argv) {
   int fd;
   if (argc == 2 && !strcmp(argv[1], "boop")) {
     if ((fd = open("/dev/null", O_RDWR | O_CLOEXEC)) == 3) {
@@ -51,8 +52,6 @@ static textstartup void TestInit(int argc, char **argv) {
   }
 }
 
-const void *const TestCtor[] initarray = {TestInit};
-
 TEST(dup, ebadf) {
   ASSERT_SYS(EBADF, -1, dup(-1));
   ASSERT_SYS(EBADF, -1, dup2(-1, 0));
diff --git a/test/libc/calls/pledge_test.c b/test/libc/calls/pledge_test.c
index 36cdfe0e7..a9a160392 100644
--- a/test/libc/calls/pledge_test.c
+++ b/test/libc/calls/pledge_test.c
@@ -550,6 +550,7 @@ TEST(pledge, execpromises_reducesAtExecOnLinux) {
 }
 
 TEST(pledge_openbsd, execpromisesIsNull_letsItDoAnything) {
+  if (IsOpenbsd()) return;  // mimmutable() ugh
   if (!IsOpenbsd()) return;
   int ws, pid;
   ASSERT_NE(-1, (pid = fork()));
@@ -566,6 +567,7 @@ TEST(pledge_openbsd, execpromisesIsNull_letsItDoAnything) {
 }
 
 TEST(pledge_openbsd, execpromisesIsSuperset_letsItDoAnything) {
+  if (IsOpenbsd()) return;  // mimmutable() ugh
   if (!IsOpenbsd()) return;
   int ws, pid;
   ASSERT_NE(-1, (pid = fork()));
@@ -585,6 +587,7 @@ TEST(pledge_linux, execpromisesIsSuperset_notPossible) {
 }
 
 TEST(pledge_openbsd, execpromises_notok) {
+  if (IsOpenbsd()) return;  // mimmutable() ugh
   int ws, pid;
   ASSERT_NE(-1, (pid = fork()));
   if (!pid) {
diff --git a/test/libc/calls/readlinkat_test.c b/test/libc/calls/readlinkat_test.c
index 8da142e5f..0d3d5a4a3 100644
--- a/test/libc/calls/readlinkat_test.c
+++ b/test/libc/calls/readlinkat_test.c
@@ -23,7 +23,6 @@
 #include "libc/limits.h"
 #include "libc/log/log.h"
 #include "libc/mem/gc.h"
-#include "libc/mem/gc.h"
 #include "libc/runtime/symbols.internal.h"
 #include "libc/str/str.h"
 #include "libc/sysv/consts/at.h"
diff --git a/test/libc/calls/unveil_test.c b/test/libc/calls/unveil_test.c
index 1c5767a79..b252fd8ce 100644
--- a/test/libc/calls/unveil_test.c
+++ b/test/libc/calls/unveil_test.c
@@ -94,6 +94,7 @@ TEST(unveil, api_differences) {
 }
 
 TEST(unveil, rx_readOnlyPreexistingExecutable_worksFine) {
+  if (IsOpenbsd()) return;  // TOOD(jart): why pledge violation?
   SPAWN(fork);
   ASSERT_SYS(0, 0, mkdir("folder", 0755));
   testlib_extract("/zip/life.elf", "folder/life.elf", 0755);
@@ -150,6 +151,7 @@ TEST(unveil, rwc_createExecutableFile_isAllowedButCantBeRun) {
 }
 
 TEST(unveil, rwcx_createExecutableFile_canAlsoBeRun) {
+  if (IsOpenbsd()) return;  // TOOD(jart): why pledge violation?
   SPAWN(fork);
   ASSERT_SYS(0, 0, mkdir("folder", 0755));
   ASSERT_SYS(0, 0, unveil("folder", "rwcx"));
diff --git a/test/libc/mem/realpath_test.c b/test/libc/mem/realpath_test.c
index e24060045..a4796d091 100644
--- a/test/libc/mem/realpath_test.c
+++ b/test/libc/mem/realpath_test.c
@@ -19,6 +19,7 @@
 #include "libc/calls/calls.h"
 #include "libc/dce.h"
 #include "libc/errno.h"
+#include "libc/intrin/strace.internal.h"
 #include "libc/mem/gc.h"
 #include "libc/mem/mem.h"
 #include "libc/runtime/runtime.h"
@@ -45,6 +46,8 @@ TEST(realpath, test2) {
 }
 
 TEST(realpath, test3) {
+  strace_enabled(+1);
+  ASSERT_TRUE(fileexists("conftest.a"));
   char *name = gc(realpath("conftest.l/../conftest.a", NULL));
   if (IsWindows()) {
     // WIN32 acts as a flat namespace, rather than linear inode crawl.
@@ -54,6 +57,7 @@ TEST(realpath, test3) {
     // Every other OS FS is a UNIX inode crawl.
     ASSERT_SYS(ENOTDIR, NULL, name);
   }
+  strace_enabled(-1);
 }
 
 TEST(realpath, test4) {
diff --git a/test/libc/proc/posix_spawn_test.c b/test/libc/proc/posix_spawn_test.c
index d4fe8aa7f..046710146 100644
--- a/test/libc/proc/posix_spawn_test.c
+++ b/test/libc/proc/posix_spawn_test.c
@@ -36,7 +36,6 @@
 #include "libc/intrin/safemacros.internal.h"
 #include "libc/limits.h"
 #include "libc/mem/gc.h"
-#include "libc/mem/gc.h"
 #include "libc/mem/mem.h"
 #include "libc/proc/proc.internal.h"
 #include "libc/runtime/internal.h"
@@ -130,6 +129,7 @@ TEST(posix_spawn, ape) {
 }
 
 TEST(posix_spawn, elf) {
+  if (IsOpenbsd()) return;  // mimmutable() ugh
   if (IsXnu() || IsWindows() || IsMetal()) return;
   int ws, pid;
   char *prog = "./life.elf";  // assimilate -bcef
diff --git a/test/libc/release/BUILD.mk b/test/libc/release/BUILD.mk
index e98174d41..85da0a377 100644
--- a/test/libc/release/BUILD.mk
+++ b/test/libc/release/BUILD.mk
@@ -38,10 +38,12 @@ o/$(MODE)/test/libc/release/smoke.o:					\
 		-nostdinc						\
 		-D_COSMO_SOURCE						\
 		-Wl,--gc-sections					\
+		-z noexecstack						\
 		-fno-omit-frame-pointer					\
 		-include o/cosmopolitan.h				\
 		-Wl,-z,max-page-size=0x1000				\
 		-Wl,-z,common-page-size=0x1000				\
+		-Wl,-z,noexecstack					\
 		$<
 
 o/$(MODE)/test/libc/release/smoke.com.dbg:				\
@@ -55,6 +57,7 @@ o/$(MODE)/test/libc/release/smoke.com.dbg:				\
 		-no-pie							\
 		-nostdlib						\
 		--gc-sections						\
+		-z noexecstack						\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
 		-T o/$(MODE)/ape/ape.lds				\
@@ -75,6 +78,7 @@ o/$(MODE)/test/libc/release/smoke-nms.com.dbg:				\
 		-no-pie							\
 		-nostdlib						\
 		--gc-sections						\
+		-z noexecstack						\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
 		-T o/$(MODE)/ape/ape.lds				\
@@ -96,6 +100,7 @@ o/$(MODE)/test/libc/release/smoke-chibicc.com.dbg:			\
 		-no-pie							\
 		-nostdlib						\
 		--gc-sections						\
+		-z noexecstack						\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
 		-T o/$(MODE)/ape/ape.lds				\
@@ -138,6 +143,7 @@ o/$(MODE)/test/libc/release/smokecxx.com.dbg:				\
 		-no-pie							\
 		-nostdlib						\
 		--gc-sections						\
+		-z noexecstack						\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
 		-T o/$(MODE)/ape/ape.lds				\
@@ -158,6 +164,7 @@ o/$(MODE)/test/libc/release/smokecxx.o:					\
 		-fno-pie						\
 		-nostdinc						\
 		-Wl,--gc-sections					\
+		-Wl,-z,noexecstack					\
 		-fno-omit-frame-pointer					\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
@@ -175,6 +182,7 @@ o/$(MODE)/test/libc/release/smokeansi.com.dbg:				\
 		-no-pie							\
 		-nostdlib						\
 		--gc-sections						\
+		-z noexecstack						\
 		-z max-page-size=0x1000					\
 		-z common-page-size=0x1000				\
 		-T o/$(MODE)/ape/ape.lds				\
@@ -198,6 +206,7 @@ o/$(MODE)/test/libc/release/smokeansi.o:				\
 		-nostdinc						\
 		-D_COSMO_SOURCE						\
 		-Wl,--gc-sections					\
+		-Wl,-z,noexecstack					\
 		-fno-omit-frame-pointer					\
 		-include o/cosmopolitan.h				\
 		-Wl,-z,max-page-size=0x1000				\
diff --git a/test/libc/runtime/initorder_test.c b/test/libc/runtime/initorder_test.c
new file mode 100644
index 000000000..d8f5a70a5
--- /dev/null
+++ b/test/libc/runtime/initorder_test.c
@@ -0,0 +1,260 @@
+/*-*- mode:c;indent-tabs-mode:nil;c-basic-offset:2;tab-width:8;coding:utf-8 -*-│
+│ vi: set et ft=c ts=2 sts=2 sw=2 fenc=utf-8                               :vi │
+╞══════════════════════════════════════════════════════════════════════════════╡
+│ Copyright 2024 Justine Alexandra Roberts Tunney                              │
+│                                                                              │
+│ Permission to use, copy, modify, and/or distribute this software for         │
+│ any purpose with or without fee is hereby granted, provided that the         │
+│ above copyright notice and this permission notice appear in all copies.      │
+│                                                                              │
+│ THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL                │
+│ WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED                │
+│ WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE             │
+│ AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL         │
+│ DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR        │
+│ PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER               │
+│ TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR             │
+│ PERFORMANCE OF THIS SOFTWARE.                                                │
+╚─────────────────────────────────────────────────────────────────────────────*/
+#include "libc/stdio/stdio.h"
+
+// orders of events
+//  -4         pc() bios
+//  -4  ApeLoader() unix
+//  -3    WinMain() win32
+//  -3     _start() unix
+//  -2      cosmo() x86
+//  -2     cosmo2() aarch64
+//  -1      _init() x86
+//   0   ASMPREINIT ok
+//   1   ASMINIT105 ok
+//   2      CTOR110 ok
+//   3   ASMINIT115 ok
+//   4      CTOR120 ok
+//   5     CTOR1000 ok
+//   6    CTOR10000 ok
+//   7    CTOR40000 ok
+//   8    CTOR60000 ok
+//   9   ASMCTOR113 ok
+//  10   ASMCTOR103 ok
+//  11    CTOR65500 ok
+//  12    CTOR65534 ok
+//  13         CTOR ok
+//  14    CTOR65535 ok
+//  15      ASMINIT ok
+//  16      ASMCTOR ok
+//   n       main()
+
+#define N 18
+
+#define COSMOINIT -1
+
+#define CTOR      1
+#define CTOR110   110
+#define CTOR120   120
+#define CTOR1000  1000
+#define CTOR10000 10000
+#define CTOR40000 40000
+#define CTOR60000 60000
+#define CTOR65500 65500
+#define CTOR65534 65534
+#define CTOR65535 65535
+
+#define ASMINIT    2
+#define ASMINIT105 105
+#define ASMINIT115 115
+
+#define ASMCTOR    3
+#define ASMCTOR103 103
+#define ASMCTOR113 113
+
+#define ASMPREINIT    4
+#define ASMPREINIT107 107
+#define ASMPREINIT117 117
+
+#ifdef __x86__
+#define GUFF " #"
+#else
+#define GUFF " //"
+#endif
+
+int event;
+int events[N];
+
+const char *getname(int x) {
+  switch (x) {
+    case 0:
+      return "0";
+    case CTOR:
+      return "CTOR";
+    case CTOR110:
+      return "CTOR110";
+    case CTOR120:
+      return "CTOR120";
+    case CTOR1000:
+      return "CTOR1000";
+    case CTOR10000:
+      return "CTOR10000";
+    case CTOR40000:
+      return "CTOR40000";
+    case CTOR60000:
+      return "CTOR60000";
+    case CTOR65500:
+      return "CTOR65500";
+    case CTOR65534:
+      return "CTOR65534";
+    case CTOR65535:
+      return "CTOR65535";
+    case ASMINIT:
+      return "ASMINIT";
+    case ASMINIT105:
+      return "ASMINIT105";
+    case ASMINIT115:
+      return "ASMINIT115";
+    case ASMCTOR:
+      return "ASMCTOR";
+    case ASMCTOR103:
+      return "ASMCTOR103";
+    case ASMCTOR113:
+      return "ASMCTOR113";
+    case ASMPREINIT:
+      return "ASMPREINIT";
+    case ASMPREINIT107:
+      return "ASMPREINIT107";
+    case ASMPREINIT117:
+      return "ASMPREINIT117";
+    default:
+      return "???";
+  }
+}
+
+__attribute__((__constructor__)) void ctor(void) {
+  events[event++] = CTOR;
+}
+
+__attribute__((__constructor__(110))) void ctor110(void) {
+  events[event++] = CTOR110;
+}
+
+__attribute__((__constructor__(120))) void ctor120(void) {
+  events[event++] = CTOR120;
+}
+
+__attribute__((__constructor__(1000))) void ctor1000(void) {
+  events[event++] = CTOR1000;
+}
+
+__attribute__((__constructor__(10000))) void ctor10000(void) {
+  events[event++] = CTOR10000;
+}
+
+__attribute__((__constructor__(40000))) void ctor40000(void) {
+  events[event++] = CTOR40000;
+}
+
+__attribute__((__constructor__(60000))) void ctor60000(void) {
+  events[event++] = CTOR60000;
+}
+
+__attribute__((__constructor__(65500))) void ctor65500(void) {
+  events[event++] = CTOR65500;
+}
+
+__attribute__((__constructor__(65534))) void ctor65534(void) {
+  events[event++] = CTOR65534;
+}
+
+__attribute__((__constructor__(65535))) void ctor65535(void) {
+  events[event++] = CTOR65535;
+}
+
+void asminit(void) {
+  events[event++] = ASMINIT;
+}
+__attribute__((__section__(
+    ".init_array,\"aw\",@init_array" GUFF))) void *const kAsminit[] = {asminit};
+
+void asminit105(void) {
+  events[event++] = ASMINIT105;
+}
+__attribute__((__section__(
+    ".init_array.105,\"aw\",@init_array" GUFF))) void *const kAsminit105[] = {
+    asminit105};
+
+void asminit115(void) {
+  events[event++] = ASMINIT115;
+}
+__attribute__((__section__(
+    ".init_array.115,\"aw\",@init_array" GUFF))) void *const kAsminit115[] = {
+    asminit115};
+
+void asmpreinit(void) {
+  events[event++] = ASMPREINIT;
+}
+__attribute__((__section__(
+    ".preinit_array,\"a\",@preinit_array" GUFF))) void *const kAsmpreinit[] = {
+    asmpreinit};
+
+void asmpreinit107(void) {
+  events[event++] = ASMPREINIT107;
+}
+__attribute__((
+    __section__(".preinit_array.107,\"a\",@preinit_array" GUFF))) void
+    *const kAsmpreinit107[] = {asmpreinit107};
+
+void asmctor(void) {
+  events[event++] = ASMCTOR;
+}
+__attribute__((__section__(
+    ".ctors,\"aw\",@init_array" GUFF))) void *const kAsmctor[] = {asmctor};
+
+void asmctor103(void) {
+  events[event++] = ASMCTOR103;
+}
+__attribute__((__section__(
+    ".ctors.103,\"aw\",@init_array" GUFF))) void *const kAsmctor103[] = {
+    asmctor103};
+
+void asmctor113(void) {
+  events[event++] = ASMCTOR113;
+}
+__attribute__((__section__(
+    ".ctors.113,\"aw\",@init_array" GUFF))) void *const kAsmctor113[] = {
+    asmctor113};
+
+const int want[N] = {
+    ASMPREINIT,  //
+    ASMINIT105,  //
+    CTOR110,     //
+    ASMINIT115,  //
+    CTOR120,     //
+    CTOR1000,    //
+    CTOR10000,   //
+    CTOR40000,   //
+    CTOR60000,   //
+    ASMCTOR113,  //
+    ASMCTOR103,  //
+    CTOR65500,   //
+    CTOR65534,   //
+    CTOR,        //
+    CTOR65535,   //
+    ASMINIT,     //
+    ASMCTOR,     //
+};
+
+int main() {
+  int fails = 0;
+  printf("\nevents:\n");
+  for (int i = 0; i < N; ++i) {
+    printf("%3d %12s ", i, getname(events[i]));
+    if (events[i] == want[i]) {
+      printf("ok");
+    } else {
+      printf("should be %s", getname(want[i]));
+      ++fails;
+    }
+    printf("\n");
+  }
+  printf("\n");
+  return fails;
+}
diff --git a/test/libc/time/strftime_test.c b/test/libc/time/strftime_test.c
index 2e37463c0..275c0fb7c 100644
--- a/test/libc/time/strftime_test.c
+++ b/test/libc/time/strftime_test.c
@@ -23,10 +23,9 @@
 #include "libc/time/struct/tm.h"
 #include "libc/time/time.h"
 
-textstartup static void strftime_test_init(void) {
+__attribute__((__constructor__)) void init(void) {
   setenv("TZ", "GST", true);
 }
-const void *const strftime_test_ctor[] initarray = {strftime_test_init};
 
 char *FormatTime(const char *fmt, struct tm *tm) {
   static char buf[64];
diff --git a/test/libcxx/openmp_test.cc b/test/libcxx/openmp_test.cc
index 7a8e7782c..ecea814ea 100644
--- a/test/libcxx/openmp_test.cc
+++ b/test/libcxx/openmp_test.cc
@@ -22,6 +22,7 @@
 #include <cstdio>
 #include <cstring>
 #include <ctime>
+#include "libc/stdio/rand.h"
 
 #define PRECISION 2e-6
 #define LV1DCACHE 49152
@@ -63,36 +64,36 @@ void transpose(long m, long n, const TA *A, long lda, TB *B, long ldb) {
 }
 
 // m×k * k×n → m×n
-// k×m * k×n → m×n if aT
-// m×k * n×k → m×n if bT
-// k×m * n×k → m×n if aT and bT
+// k×m * k×n → m×n if aᵀ
+// m×k * n×k → m×n if bᵀ
+// k×m * n×k → m×n if aᵀ and bᵀ
 template <typename TC, typename TA, typename TB>
-void dgemm(bool aT, bool bT, long m, long n, long k, float alpha, const TA *A,
-           long lda, const TB *B, long ldb, float beta, TC *C, long ldc) {
+void dgemm(bool aᵀ, bool bᵀ, long m, long n, long k, float α, const TA *A,
+           long lda, const TB *B, long ldb, float β, TC *C, long ldc) {
 #pragma omp parallel for collapse(2) if (m * n * k > THRESHOLD)
   for (long i = 0; i < m; ++i)
     for (long j = 0; j < n; ++j) {
       double sum = 0;
       for (long l = 0; l < k; ++l)
-        sum = std::fma((aT ? A[lda * l + i] : A[lda * i + l]) * alpha,
-                       (bT ? B[ldb * j + l] : B[ldb * l + j]), sum);
-      C[ldc * i + j] = beta * C[ldc * i + j] + sum;
+        sum = std::fma((aᵀ ? A[lda * l + i] : A[lda * i + l]) * α,
+                       (bᵀ ? B[ldb * j + l] : B[ldb * l + j]), sum);
+      C[ldc * i + j] = C[ldc * i + j] * β + sum;
     }
 }
 
 template <typename T, typename TC, typename TA, typename TB>
 struct Gemmlin {
  public:
-  Gemmlin(bool aT, bool bT, float alpha, const TA *A, long lda, const TB *B,
-          long ldb, float beta, TC *C, long ldc)
+  Gemmlin(bool aT, bool bT, float α, const TA *A, long lda, const TB *B,
+          long ldb, float β, TC *C, long ldc)
       : aT(aT),
         bT(bT),
-        alpha(alpha),
+        α(α),
         A(A),
         lda(lda),
         B(B),
         ldb(ldb),
-        beta(beta),
+        β(β),
         C(C),
         ldc(ldc) {
   }
@@ -101,7 +102,7 @@ struct Gemmlin {
     if (!m || !n) return;
     for (long i = 0; i < m; ++i)
       for (long j = 0; j < n; ++j) {
-        C[ldc * i + j] *= beta;
+        C[ldc * i + j] *= β;
       }
     if (!k) return;
     cub = sqrt(LV1DCACHE) / sqrt(sizeof(T) * 3);
@@ -168,8 +169,8 @@ struct Gemmlin {
     T Ac[mc / mr][kc][mr];
     for (long i = 0; i < mc; ++i)
       for (long j = 0; j < kc; ++j)
-        Ac[i / mr][j][i % mr] = alpha * (aT ? A[lda * (pc + j) + (ic + i)]
-                                            : A[lda * (ic + i) + (pc + j)]);
+        Ac[i / mr][j][i % mr] = α * (aT ? A[lda * (pc + j) + (ic + i)]
+                                        : A[lda * (ic + i) + (pc + j)]);
     for (long jc = n0; jc < n; jc += nc) {
       T Bc[nc / nr][nr][kc];
       for (long j = 0; j < nc; ++j)
@@ -220,12 +221,12 @@ struct Gemmlin {
 
   bool aT;
   bool bT;
-  float alpha;
+  float α;
   const TA *A;
   long lda;
   const TB *B;
   long ldb;
-  float beta;
+  float β;
   TC *C;
   long ldc;
   long ops;
@@ -236,9 +237,9 @@ struct Gemmlin {
 };
 
 template <typename TC, typename TA, typename TB>
-void sgemm(bool aT, bool bT, long m, long n, long k, float alpha, const TA *A,
-           long lda, const TB *B, long ldb, float beta, TC *C, long ldc) {
-  Gemmlin<float, TC, TA, TB> g{aT, bT, alpha, A, lda, B, ldb, beta, C, ldc};
+void sgemm(bool aT, bool bT, long m, long n, long k, float α, const TA *A,
+           long lda, const TB *B, long ldb, float β, TC *C, long ldc) {
+  Gemmlin<float, TC, TA, TB> g{aT, bT, α, A, lda, B, ldb, β, C, ldc};
   g.gemm(m, n, k);
 }
 
@@ -360,20 +361,12 @@ long micros(void) {
            #x);                                                             \
   } while (0)
 
-unsigned long rando(void) {
-  static unsigned long s;
-  unsigned long z = (s += 0x9e3779b97f4a7c15);
-  z = (z ^ (z >> 30)) * 0xbf58476d1ce4e5b9;
-  z = (z ^ (z >> 27)) * 0x94d049bb133111eb;
-  return z ^ (z >> 31);
-}
-
 double real01(unsigned long x) {  // (0,1)
   return 1. / 4503599627370496. * ((x >> 12) + .5);
 }
 
 double numba(void) {  // (-1,1)
-  return real01(rando()) * 2 - 1;
+  return real01(lemur64()) * 2 - 1;
 }
 
 template <typename T>
@@ -390,20 +383,20 @@ void test_gemm(long m, long n, long k) {
   float *Bt = new float[n * k];
   float *C = new float[m * n];
   float *GOLD = new float[m * n];
-  float alpha = 1;
-  float beta = 0;
+  float α = 1;
+  float β = 0;
   fill(A, m * k);
   fill(B, k * n);
   dgemm(0, 0, m, n, k, 1, A, k, B, n, 0, GOLD, n);
   transpose(m, k, A, k, At, m);
   transpose(k, n, B, n, Bt, k);
-  sgemm(0, 0, m, n, k, alpha, A, k, B, n, beta, C, n);
+  sgemm(0, 0, m, n, k, α, A, k, B, n, β, C, n);
   check(PRECISION, m, n, GOLD, n, C, n);
-  sgemm(1, 0, m, n, k, alpha, At, m, B, n, beta, C, n);
+  sgemm(1, 0, m, n, k, α, At, m, B, n, β, C, n);
   check(PRECISION, m, n, GOLD, n, C, n);
-  sgemm(0, 1, m, n, k, alpha, A, k, Bt, k, beta, C, n);
+  sgemm(0, 1, m, n, k, α, A, k, Bt, k, β, C, n);
   check(PRECISION, m, n, GOLD, n, C, n);
-  sgemm(1, 1, m, n, k, alpha, At, m, Bt, k, beta, C, n);
+  sgemm(1, 1, m, n, k, α, At, m, Bt, k, β, C, n);
   check(PRECISION, m, n, GOLD, n, C, n);
   delete[] GOLD;
   delete[] C;
diff --git a/third_party/aarch64/arm_acle.internal.h b/third_party/aarch64/arm_acle.internal.h
index 9c0bd0be6..687b133d2 100644
--- a/third_party/aarch64/arm_acle.internal.h
+++ b/third_party/aarch64/arm_acle.internal.h
@@ -1,12 +1,45 @@
 #if defined(__aarch64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
 #ifndef _GCC_ARM_ACLE_H
 #define _GCC_ARM_ACLE_H
+#pragma GCC aarch64 "arm_acle.h"
 #ifdef __cplusplus
 extern "C" {
 #endif
+#define _GCC_ARM_ACLE_ROR_FN(NAME, TYPE) __extension__ extern __inline TYPE __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) NAME (TYPE __value, uint32_t __rotate) { size_t __size = sizeof (TYPE) * __CHAR_BIT__; __rotate = __rotate % __size; return __value >> __rotate | __value << ((__size - __rotate) % __size); }
+_GCC_ARM_ACLE_ROR_FN (__ror, uint32_t)
+_GCC_ARM_ACLE_ROR_FN (__rorl, unsigned long)
+_GCC_ARM_ACLE_ROR_FN (__rorll, uint64_t)
+#undef _GCC_ARM_ACLE_ROR_FN
+#define _GCC_ARM_ACLE_DATA_FN(NAME, BUILTIN, ITYPE, RTYPE) __extension__ extern __inline RTYPE __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) __##NAME (ITYPE __value) { return __builtin_##BUILTIN (__value); }
+_GCC_ARM_ACLE_DATA_FN (clz, clz, uint32_t, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (clzl, clzl, unsigned long, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (clzll, clzll, uint64_t, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (cls, clrsb, uint32_t, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (clsl, clrsbl, unsigned long, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (clsll, clrsbll, uint64_t, unsigned int)
+_GCC_ARM_ACLE_DATA_FN (rev16, aarch64_rev16, uint32_t, uint32_t)
+_GCC_ARM_ACLE_DATA_FN (rev16l, aarch64_rev16l, unsigned long, unsigned long)
+_GCC_ARM_ACLE_DATA_FN (rev16ll, aarch64_rev16ll, uint64_t, uint64_t)
+_GCC_ARM_ACLE_DATA_FN (rbit, aarch64_rbit, uint32_t, uint32_t)
+_GCC_ARM_ACLE_DATA_FN (rbitl, aarch64_rbitl, unsigned long, unsigned long)
+_GCC_ARM_ACLE_DATA_FN (rbitll, aarch64_rbitll, uint64_t, uint64_t)
+_GCC_ARM_ACLE_DATA_FN (revsh, bswap16, int16_t, int16_t)
+_GCC_ARM_ACLE_DATA_FN (rev, bswap32, uint32_t, uint32_t)
+_GCC_ARM_ACLE_DATA_FN (revll, bswap64, uint64_t, uint64_t)
+#undef _GCC_ARM_ACLE_DATA_FN
+__extension__ extern __inline unsigned long
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__revl (unsigned long __value)
+{
+  if (sizeof (unsigned long) == 8)
+    return __revll (__value);
+  else
+    return __rev (__value);
+}
 #pragma GCC push_options
 #pragma GCC target ("arch=armv8.3-a")
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __jcvt (double __a)
 {
   return __builtin_aarch64_jcvtzs (__a);
@@ -14,42 +47,50 @@ __jcvt (double __a)
 #pragma GCC pop_options
 #pragma GCC push_options
 #pragma GCC target ("arch=armv8.5-a")
-__funline float
+__extension__ extern __inline float
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint32zf (float __a)
 {
   return __builtin_aarch64_frint32zsf (__a);
 }
-__funline double
+__extension__ extern __inline double
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint32z (double __a)
 {
   return __builtin_aarch64_frint32zdf (__a);
 }
-__funline float
+__extension__ extern __inline float
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint64zf (float __a)
 {
   return __builtin_aarch64_frint64zsf (__a);
 }
-__funline double
+__extension__ extern __inline double
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint64z (double __a)
 {
   return __builtin_aarch64_frint64zdf (__a);
 }
-__funline float
+__extension__ extern __inline float
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint32xf (float __a)
 {
   return __builtin_aarch64_frint32xsf (__a);
 }
-__funline double
+__extension__ extern __inline double
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint32x (double __a)
 {
   return __builtin_aarch64_frint32xdf (__a);
 }
-__funline float
+__extension__ extern __inline float
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint64xf (float __a)
 {
   return __builtin_aarch64_frint64xsf (__a);
 }
-__funline double
+__extension__ extern __inline double
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rint64x (double __a)
 {
   return __builtin_aarch64_frint64xdf (__a);
@@ -57,42 +98,50 @@ __rint64x (double __a)
 #pragma GCC pop_options
 #pragma GCC push_options
 #pragma GCC target ("+nothing+crc")
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32b (uint32_t __a, uint8_t __b)
 {
   return __builtin_aarch64_crc32b (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32cb (uint32_t __a, uint8_t __b)
 {
   return __builtin_aarch64_crc32cb (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32ch (uint32_t __a, uint16_t __b)
 {
   return __builtin_aarch64_crc32ch (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32cw (uint32_t __a, uint32_t __b)
 {
   return __builtin_aarch64_crc32cw (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32cd (uint32_t __a, uint64_t __b)
 {
   return __builtin_aarch64_crc32cx (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32h (uint32_t __a, uint16_t __b)
 {
   return __builtin_aarch64_crc32h (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32w (uint32_t __a, uint32_t __b)
 {
   return __builtin_aarch64_crc32w (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __crc32d (uint32_t __a, uint64_t __b)
 {
   return __builtin_aarch64_crc32x (__a, __b);
@@ -112,36 +161,72 @@ __crc32d (uint32_t __a, uint64_t __b)
 #define _TMFAILURE_DBG 0x00400000u
 #define _TMFAILURE_INT 0x00800000u
 #define _TMFAILURE_TRIVIAL 0x01000000u
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __tstart (void)
 {
   return __builtin_aarch64_tstart ();
 }
-__funline void
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __tcommit (void)
 {
   __builtin_aarch64_tcommit ();
 }
-__funline void
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __tcancel (const uint64_t __reason)
 {
   __builtin_aarch64_tcancel (__reason);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __ttest (void)
 {
   return __builtin_aarch64_ttest ();
 }
 #pragma GCC pop_options
 #endif
+#ifdef __ARM_FEATURE_LS64
+#pragma GCC push_options
+#pragma GCC target ("+nothing+ls64")
+typedef __arm_data512_t data512_t;
+__extension__ extern __inline data512_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_ld64b (const void *__addr)
+{
+  return __builtin_aarch64_ld64b (__addr);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_st64b (void *__addr, data512_t __value)
+{
+  __builtin_aarch64_st64b (__addr, __value);
+}
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_st64bv (void *__addr, data512_t __value)
+{
+  return __builtin_aarch64_st64bv (__addr, __value);
+}
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+__arm_st64bv0 (void *__addr, data512_t __value)
+{
+  return __builtin_aarch64_st64bv0 (__addr, __value);
+}
+#pragma GCC pop_options
+#endif
 #pragma GCC push_options
 #pragma GCC target ("+nothing+rng")
-__funline int
+__extension__ extern __inline int
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rndr (uint64_t *__res)
 {
   return __builtin_aarch64_rndr (__res);
 }
-__funline int
+__extension__ extern __inline int
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 __rndrrs (uint64_t *__res)
 {
   return __builtin_aarch64_rndrrs (__res);
diff --git a/third_party/aarch64/arm_fp16.internal.h b/third_party/aarch64/arm_fp16.internal.h
index ddc72f764..84185a620 100644
--- a/third_party/aarch64/arm_fp16.internal.h
+++ b/third_party/aarch64/arm_fp16.internal.h
@@ -4,447 +4,536 @@
 #pragma GCC push_options
 #pragma GCC target ("arch=armv8.2-a+fp16")
 typedef __fp16 float16_t;
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vabsh_f16 (float16_t __a)
 {
   return __builtin_aarch64_abshf (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vceqzh_f16 (float16_t __a)
 {
   return __builtin_aarch64_cmeqhf_uss (__a, 0.0f);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcgezh_f16 (float16_t __a)
 {
   return __builtin_aarch64_cmgehf_uss (__a, 0.0f);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcgtzh_f16 (float16_t __a)
 {
   return __builtin_aarch64_cmgthf_uss (__a, 0.0f);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vclezh_f16 (float16_t __a)
 {
   return __builtin_aarch64_cmlehf_uss (__a, 0.0f);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcltzh_f16 (float16_t __a)
 {
   return __builtin_aarch64_cmlthf_uss (__a, 0.0f);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_s16 (int16_t __a)
 {
   return __builtin_aarch64_floathihf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_s32 (int32_t __a)
 {
   return __builtin_aarch64_floatsihf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_s64 (int64_t __a)
 {
   return __builtin_aarch64_floatdihf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_u16 (uint16_t __a)
 {
   return __builtin_aarch64_floatunshihf_us (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_u32 (uint32_t __a)
 {
   return __builtin_aarch64_floatunssihf_us (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_f16_u64 (uint64_t __a)
 {
   return __builtin_aarch64_floatunsdihf_us (__a);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_s16_f16 (float16_t __a)
 {
   return __builtin_aarch64_fix_trunchfhi (__a);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_s32_f16 (float16_t __a)
 {
   return __builtin_aarch64_fix_trunchfsi (__a);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_s64_f16 (float16_t __a)
 {
   return __builtin_aarch64_fix_trunchfdi (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_u16_f16 (float16_t __a)
 {
   return __builtin_aarch64_fixuns_trunchfhi_us (__a);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_u32_f16 (float16_t __a)
 {
   return __builtin_aarch64_fixuns_trunchfsi_us (__a);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_u64_f16 (float16_t __a)
 {
   return __builtin_aarch64_fixuns_trunchfdi_us (__a);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_s16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lroundhfhi (__a);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_s32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lroundhfsi (__a);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_s64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lroundhfdi (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_u16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lrounduhfhi_us (__a);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_u32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lrounduhfsi_us (__a);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtah_u64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lrounduhfdi_us (__a);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_s16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfloorhfhi (__a);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_s32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfloorhfsi (__a);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_s64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfloorhfdi (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_u16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lflooruhfhi_us (__a);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_u32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lflooruhfsi_us (__a);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtmh_u64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lflooruhfdi_us (__a);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_s16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnhfhi (__a);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_s32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnhfsi (__a);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_s64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnhfdi (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_u16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnuhfhi_us (__a);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_u32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnuhfsi_us (__a);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtnh_u64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lfrintnuhfdi_us (__a);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_s16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceilhfhi (__a);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_s32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceilhfsi (__a);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_s64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceilhfdi (__a);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_u16_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceiluhfhi_us (__a);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_u32_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceiluhfsi_us (__a);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtph_u64_f16 (float16_t __a)
 {
   return __builtin_aarch64_lceiluhfdi_us (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vnegh_f16 (float16_t __a)
 {
   return __builtin_aarch64_neghf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrecpeh_f16 (float16_t __a)
 {
   return __builtin_aarch64_frecpehf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrecpxh_f16 (float16_t __a)
 {
   return __builtin_aarch64_frecpxhf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndh_f16 (float16_t __a)
 {
   return __builtin_aarch64_btrunchf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndah_f16 (float16_t __a)
 {
   return __builtin_aarch64_roundhf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndih_f16 (float16_t __a)
 {
   return __builtin_aarch64_nearbyinthf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndmh_f16 (float16_t __a)
 {
   return __builtin_aarch64_floorhf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndnh_f16 (float16_t __a)
 {
-  return __builtin_aarch64_frintnhf (__a);
+  return __builtin_aarch64_roundevenhf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndph_f16 (float16_t __a)
 {
   return __builtin_aarch64_ceilhf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndxh_f16 (float16_t __a)
 {
   return __builtin_aarch64_rinthf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsqrteh_f16 (float16_t __a)
 {
   return __builtin_aarch64_rsqrtehf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsqrth_f16 (float16_t __a)
 {
   return __builtin_aarch64_sqrthf (__a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddh_f16 (float16_t __a, float16_t __b)
 {
   return __a + __b;
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vabdh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fabdhf (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcageh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_facgehf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcagth_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_facgthf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcaleh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_faclehf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcalth_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_faclthf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vceqh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_cmeqhf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcgeh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_cmgehf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcgth_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_cmgthf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcleh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_cmlehf_uss (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vclth_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_cmlthf_uss (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_s16 (int16_t __a, const int __b)
 {
   return __builtin_aarch64_scvtfhi (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_s32 (int32_t __a, const int __b)
 {
   return __builtin_aarch64_scvtfsihf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_s64 (int64_t __a, const int __b)
 {
   return __builtin_aarch64_scvtfdihf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_u16 (uint16_t __a, const int __b)
 {
   return __builtin_aarch64_ucvtfhi_sus (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_u32 (uint32_t __a, const int __b)
 {
   return __builtin_aarch64_ucvtfsihf_sus (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_f16_u64 (uint64_t __a, const int __b)
 {
   return __builtin_aarch64_ucvtfdihf_sus (__a, __b);
 }
-__funline int16_t
+__extension__ extern __inline int16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_s16_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzshf (__a, __b);
 }
-__funline int32_t
+__extension__ extern __inline int32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_s32_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzshfsi (__a, __b);
 }
-__funline int64_t
+__extension__ extern __inline int64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_s64_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzshfdi (__a, __b);
 }
-__funline uint16_t
+__extension__ extern __inline uint16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_u16_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzuhf_uss (__a, __b);
 }
-__funline uint32_t
+__extension__ extern __inline uint32_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_u32_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzuhfsi_uss (__a, __b);
 }
-__funline uint64_t
+__extension__ extern __inline uint64_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvth_n_u64_f16 (float16_t __a, const int __b)
 {
   return __builtin_aarch64_fcvtzuhfdi_uss (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vdivh_f16 (float16_t __a, float16_t __b)
 {
   return __a / __b;
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmaxh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fmaxhf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmaxnmh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fmaxhf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vminh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fminhf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vminnmh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fminhf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmulh_f16 (float16_t __a, float16_t __b)
 {
   return __a * __b;
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmulxh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_fmulxhf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrecpsh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_frecpshf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsqrtsh_f16 (float16_t __a, float16_t __b)
 {
   return __builtin_aarch64_rsqrtshf (__a, __b);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubh_f16 (float16_t __a, float16_t __b)
 {
   return __a - __b;
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vfmah_f16 (float16_t __a, float16_t __b, float16_t __c)
 {
   return __builtin_aarch64_fmahf (__b, __c, __a);
 }
-__funline float16_t
+__extension__ extern __inline float16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vfmsh_f16 (float16_t __a, float16_t __b, float16_t __c)
 {
   return __builtin_aarch64_fnmahf (__b, __c, __a);
diff --git a/third_party/aarch64/arm_neon.internal.h b/third_party/aarch64/arm_neon.internal.h
index 78560ec85..1cbe2db72 100644
--- a/third_party/aarch64/arm_neon.internal.h
+++ b/third_party/aarch64/arm_neon.internal.h
@@ -3,6 +3,7 @@
 #define _AARCH64_NEON_H_
 #pragma GCC push_options
 #pragma GCC target ("+nothing+simd")
+#pragma GCC aarch64 "arm_neon.h"
 #pragma GCC diagnostic ignored "-Wmissing-braces"
 #define __AARCH64_UINT64_C(__C) ((uint64_t) __C)
 #define __AARCH64_INT64_C(__C) ((int64_t) __C)
@@ -43,366 +44,6 @@ typedef float float32_t;
 typedef double float64_t;
 typedef __Bfloat16x4_t bfloat16x4_t;
 typedef __Bfloat16x8_t bfloat16x8_t;
-typedef struct bfloat16x4x2_t
-{
-  bfloat16x4_t val[2];
-} bfloat16x4x2_t;
-typedef struct bfloat16x8x2_t
-{
-  bfloat16x8_t val[2];
-} bfloat16x8x2_t;
-typedef struct bfloat16x4x3_t
-{
-  bfloat16x4_t val[3];
-} bfloat16x4x3_t;
-typedef struct bfloat16x8x3_t
-{
-  bfloat16x8_t val[3];
-} bfloat16x8x3_t;
-typedef struct bfloat16x4x4_t
-{
-  bfloat16x4_t val[4];
-} bfloat16x4x4_t;
-typedef struct bfloat16x8x4_t
-{
-  bfloat16x8_t val[4];
-} bfloat16x8x4_t;
-typedef struct int8x8x2_t
-{
-  int8x8_t val[2];
-} int8x8x2_t;
-typedef struct int8x16x2_t
-{
-  int8x16_t val[2];
-} int8x16x2_t;
-typedef struct int16x4x2_t
-{
-  int16x4_t val[2];
-} int16x4x2_t;
-typedef struct int16x8x2_t
-{
-  int16x8_t val[2];
-} int16x8x2_t;
-typedef struct int32x2x2_t
-{
-  int32x2_t val[2];
-} int32x2x2_t;
-typedef struct int32x4x2_t
-{
-  int32x4_t val[2];
-} int32x4x2_t;
-typedef struct int64x1x2_t
-{
-  int64x1_t val[2];
-} int64x1x2_t;
-typedef struct int64x2x2_t
-{
-  int64x2_t val[2];
-} int64x2x2_t;
-typedef struct uint8x8x2_t
-{
-  uint8x8_t val[2];
-} uint8x8x2_t;
-typedef struct uint8x16x2_t
-{
-  uint8x16_t val[2];
-} uint8x16x2_t;
-typedef struct uint16x4x2_t
-{
-  uint16x4_t val[2];
-} uint16x4x2_t;
-typedef struct uint16x8x2_t
-{
-  uint16x8_t val[2];
-} uint16x8x2_t;
-typedef struct uint32x2x2_t
-{
-  uint32x2_t val[2];
-} uint32x2x2_t;
-typedef struct uint32x4x2_t
-{
-  uint32x4_t val[2];
-} uint32x4x2_t;
-typedef struct uint64x1x2_t
-{
-  uint64x1_t val[2];
-} uint64x1x2_t;
-typedef struct uint64x2x2_t
-{
-  uint64x2_t val[2];
-} uint64x2x2_t;
-typedef struct float16x4x2_t
-{
-  float16x4_t val[2];
-} float16x4x2_t;
-typedef struct float16x8x2_t
-{
-  float16x8_t val[2];
-} float16x8x2_t;
-typedef struct float32x2x2_t
-{
-  float32x2_t val[2];
-} float32x2x2_t;
-typedef struct float32x4x2_t
-{
-  float32x4_t val[2];
-} float32x4x2_t;
-typedef struct float64x2x2_t
-{
-  float64x2_t val[2];
-} float64x2x2_t;
-typedef struct float64x1x2_t
-{
-  float64x1_t val[2];
-} float64x1x2_t;
-typedef struct poly8x8x2_t
-{
-  poly8x8_t val[2];
-} poly8x8x2_t;
-typedef struct poly8x16x2_t
-{
-  poly8x16_t val[2];
-} poly8x16x2_t;
-typedef struct poly16x4x2_t
-{
-  poly16x4_t val[2];
-} poly16x4x2_t;
-typedef struct poly16x8x2_t
-{
-  poly16x8_t val[2];
-} poly16x8x2_t;
-typedef struct poly64x1x2_t
-{
-  poly64x1_t val[2];
-} poly64x1x2_t;
-typedef struct poly64x1x3_t
-{
-  poly64x1_t val[3];
-} poly64x1x3_t;
-typedef struct poly64x1x4_t
-{
-  poly64x1_t val[4];
-} poly64x1x4_t;
-typedef struct poly64x2x2_t
-{
-  poly64x2_t val[2];
-} poly64x2x2_t;
-typedef struct poly64x2x3_t
-{
-  poly64x2_t val[3];
-} poly64x2x3_t;
-typedef struct poly64x2x4_t
-{
-  poly64x2_t val[4];
-} poly64x2x4_t;
-typedef struct int8x8x3_t
-{
-  int8x8_t val[3];
-} int8x8x3_t;
-typedef struct int8x16x3_t
-{
-  int8x16_t val[3];
-} int8x16x3_t;
-typedef struct int16x4x3_t
-{
-  int16x4_t val[3];
-} int16x4x3_t;
-typedef struct int16x8x3_t
-{
-  int16x8_t val[3];
-} int16x8x3_t;
-typedef struct int32x2x3_t
-{
-  int32x2_t val[3];
-} int32x2x3_t;
-typedef struct int32x4x3_t
-{
-  int32x4_t val[3];
-} int32x4x3_t;
-typedef struct int64x1x3_t
-{
-  int64x1_t val[3];
-} int64x1x3_t;
-typedef struct int64x2x3_t
-{
-  int64x2_t val[3];
-} int64x2x3_t;
-typedef struct uint8x8x3_t
-{
-  uint8x8_t val[3];
-} uint8x8x3_t;
-typedef struct uint8x16x3_t
-{
-  uint8x16_t val[3];
-} uint8x16x3_t;
-typedef struct uint16x4x3_t
-{
-  uint16x4_t val[3];
-} uint16x4x3_t;
-typedef struct uint16x8x3_t
-{
-  uint16x8_t val[3];
-} uint16x8x3_t;
-typedef struct uint32x2x3_t
-{
-  uint32x2_t val[3];
-} uint32x2x3_t;
-typedef struct uint32x4x3_t
-{
-  uint32x4_t val[3];
-} uint32x4x3_t;
-typedef struct uint64x1x3_t
-{
-  uint64x1_t val[3];
-} uint64x1x3_t;
-typedef struct uint64x2x3_t
-{
-  uint64x2_t val[3];
-} uint64x2x3_t;
-typedef struct float16x4x3_t
-{
-  float16x4_t val[3];
-} float16x4x3_t;
-typedef struct float16x8x3_t
-{
-  float16x8_t val[3];
-} float16x8x3_t;
-typedef struct float32x2x3_t
-{
-  float32x2_t val[3];
-} float32x2x3_t;
-typedef struct float32x4x3_t
-{
-  float32x4_t val[3];
-} float32x4x3_t;
-typedef struct float64x2x3_t
-{
-  float64x2_t val[3];
-} float64x2x3_t;
-typedef struct float64x1x3_t
-{
-  float64x1_t val[3];
-} float64x1x3_t;
-typedef struct poly8x8x3_t
-{
-  poly8x8_t val[3];
-} poly8x8x3_t;
-typedef struct poly8x16x3_t
-{
-  poly8x16_t val[3];
-} poly8x16x3_t;
-typedef struct poly16x4x3_t
-{
-  poly16x4_t val[3];
-} poly16x4x3_t;
-typedef struct poly16x8x3_t
-{
-  poly16x8_t val[3];
-} poly16x8x3_t;
-typedef struct int8x8x4_t
-{
-  int8x8_t val[4];
-} int8x8x4_t;
-typedef struct int8x16x4_t
-{
-  int8x16_t val[4];
-} int8x16x4_t;
-typedef struct int16x4x4_t
-{
-  int16x4_t val[4];
-} int16x4x4_t;
-typedef struct int16x8x4_t
-{
-  int16x8_t val[4];
-} int16x8x4_t;
-typedef struct int32x2x4_t
-{
-  int32x2_t val[4];
-} int32x2x4_t;
-typedef struct int32x4x4_t
-{
-  int32x4_t val[4];
-} int32x4x4_t;
-typedef struct int64x1x4_t
-{
-  int64x1_t val[4];
-} int64x1x4_t;
-typedef struct int64x2x4_t
-{
-  int64x2_t val[4];
-} int64x2x4_t;
-typedef struct uint8x8x4_t
-{
-  uint8x8_t val[4];
-} uint8x8x4_t;
-typedef struct uint8x16x4_t
-{
-  uint8x16_t val[4];
-} uint8x16x4_t;
-typedef struct uint16x4x4_t
-{
-  uint16x4_t val[4];
-} uint16x4x4_t;
-typedef struct uint16x8x4_t
-{
-  uint16x8_t val[4];
-} uint16x8x4_t;
-typedef struct uint32x2x4_t
-{
-  uint32x2_t val[4];
-} uint32x2x4_t;
-typedef struct uint32x4x4_t
-{
-  uint32x4_t val[4];
-} uint32x4x4_t;
-typedef struct uint64x1x4_t
-{
-  uint64x1_t val[4];
-} uint64x1x4_t;
-typedef struct uint64x2x4_t
-{
-  uint64x2_t val[4];
-} uint64x2x4_t;
-typedef struct float16x4x4_t
-{
-  float16x4_t val[4];
-} float16x4x4_t;
-typedef struct float16x8x4_t
-{
-  float16x8_t val[4];
-} float16x8x4_t;
-typedef struct float32x2x4_t
-{
-  float32x2_t val[4];
-} float32x2x4_t;
-typedef struct float32x4x4_t
-{
-  float32x4_t val[4];
-} float32x4x4_t;
-typedef struct float64x2x4_t
-{
-  float64x2_t val[4];
-} float64x2x4_t;
-typedef struct float64x1x4_t
-{
-  float64x1_t val[4];
-} float64x1x4_t;
-typedef struct poly8x8x4_t
-{
-  poly8x8_t val[4];
-} poly8x8x4_t;
-typedef struct poly8x16x4_t
-{
-  poly8x16_t val[4];
-} poly8x16x4_t;
-typedef struct poly16x4x4_t
-{
-  poly16x4_t val[4];
-} poly16x4x4_t;
-typedef struct poly16x8x4_t
-{
-  poly16x8_t val[4];
-} poly16x8x4_t;
 #define __aarch64_vdup_lane_any(__size, __q, __a, __b) vdup##__q##_n_##__size (__aarch64_vget_lane_any (__a, __b))
 #define __aarch64_vdup_lane_f16(__a, __b) __aarch64_vdup_lane_any (f16, , __a, __b)
 #define __aarch64_vdup_lane_f32(__a, __b) __aarch64_vdup_lane_any (f32, , __a, __b)
@@ -593,475 +234,433 @@ __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_saddlv8qi (__a, __b);
+  return __builtin_aarch64_saddlv8qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_saddlv4hi (__a, __b);
+  return __builtin_aarch64_saddlv4hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_saddlv2si (__a, __b);
+  return __builtin_aarch64_saddlv2si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uaddlv8qi ((int8x8_t) __a,
-         (int8x8_t) __b);
+  return __builtin_aarch64_uaddlv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uaddlv4hi ((int16x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_uaddlv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_uaddlv2si ((int32x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_uaddlv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_s8 (int8x16_t __a, int8x16_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_saddl2v16qi (__a, __b);
+  return __builtin_aarch64_saddl2v16qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_saddl2v8hi (__a, __b);
+  return __builtin_aarch64_saddl2v8hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_saddl2v4si (__a, __b);
+  return __builtin_aarch64_saddl2v4si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uaddl2v16qi ((int8x16_t) __a,
-           (int8x16_t) __b);
+  return __builtin_aarch64_uaddl2v16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uaddl2v8hi ((int16x8_t) __a,
-          (int16x8_t) __b);
+  return __builtin_aarch64_uaddl2v8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddl_high_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_uaddl2v4si ((int32x4_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_uaddl2v4si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_s8 (int16x8_t __a, int8x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_saddwv8qi (__a, __b);
+  return __builtin_aarch64_saddwv8qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_s16 (int32x4_t __a, int16x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_saddwv4hi (__a, __b);
+  return __builtin_aarch64_saddwv4hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_s32 (int64x2_t __a, int32x2_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_saddwv2si (__a, __b);
+  return __builtin_aarch64_saddwv2si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_u8 (uint16x8_t __a, uint8x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uaddwv8qi ((int16x8_t) __a,
-         (int8x8_t) __b);
+  return __builtin_aarch64_uaddwv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_u16 (uint32x4_t __a, uint16x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uaddwv4hi ((int32x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_uaddwv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_u32 (uint64x2_t __a, uint32x2_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_uaddwv2si ((int64x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_uaddwv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_s8 (int16x8_t __a, int8x16_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_saddw2v16qi (__a, __b);
+  return __builtin_aarch64_saddw2v16qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_s16 (int32x4_t __a, int16x8_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_saddw2v8hi (__a, __b);
+  return __builtin_aarch64_saddw2v8hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_s32 (int64x2_t __a, int32x4_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_saddw2v4si (__a, __b);
+  return __builtin_aarch64_saddw2v4si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_u8 (uint16x8_t __a, uint8x16_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uaddw2v16qi ((int16x8_t) __a,
-           (int8x16_t) __b);
+  return __builtin_aarch64_uaddw2v16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_u16 (uint32x4_t __a, uint16x8_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uaddw2v8hi ((int32x4_t) __a,
-          (int16x8_t) __b);
+  return __builtin_aarch64_uaddw2v8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddw_high_u32 (uint64x2_t __a, uint32x4_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_uaddw2v4si ((int64x2_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_uaddw2v4si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_shaddv8qi (__a, __b);
+  return __builtin_aarch64_shaddv8qi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_shaddv4hi (__a, __b);
+  return __builtin_aarch64_shaddv4hi (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_shaddv2si (__a, __b);
+  return __builtin_aarch64_shaddv2si (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_uhaddv8qi ((int8x8_t) __a,
-        (int8x8_t) __b);
+  return __builtin_aarch64_uhaddv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_uhaddv4hi ((int16x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_uhaddv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhadd_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_uhaddv2si ((int32x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_uhaddv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_s8 (int8x16_t __a, int8x16_t __b)
 {
-  return (int8x16_t) __builtin_aarch64_shaddv16qi (__a, __b);
+  return __builtin_aarch64_shaddv16qi (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_shaddv8hi (__a, __b);
+  return __builtin_aarch64_shaddv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_shaddv4si (__a, __b);
+  return __builtin_aarch64_shaddv4si (__a, __b);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  return (uint8x16_t) __builtin_aarch64_uhaddv16qi ((int8x16_t) __a,
-          (int8x16_t) __b);
+  return __builtin_aarch64_uhaddv16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uhaddv8hi ((int16x8_t) __a,
-         (int16x8_t) __b);
+  return __builtin_aarch64_uhaddv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhaddq_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uhaddv4si ((int32x4_t) __a,
-         (int32x4_t) __b);
+  return __builtin_aarch64_uhaddv4si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_srhaddv8qi (__a, __b);
+  return __builtin_aarch64_srhaddv8qi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_srhaddv4hi (__a, __b);
+  return __builtin_aarch64_srhaddv4hi (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_srhaddv2si (__a, __b);
+  return __builtin_aarch64_srhaddv2si (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_urhaddv8qi ((int8x8_t) __a,
-         (int8x8_t) __b);
+  return __builtin_aarch64_urhaddv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_urhaddv4hi ((int16x4_t) __a,
-          (int16x4_t) __b);
+  return __builtin_aarch64_urhaddv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhadd_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_urhaddv2si ((int32x2_t) __a,
-          (int32x2_t) __b);
+  return __builtin_aarch64_urhaddv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_s8 (int8x16_t __a, int8x16_t __b)
 {
-  return (int8x16_t) __builtin_aarch64_srhaddv16qi (__a, __b);
+  return __builtin_aarch64_srhaddv16qi (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_srhaddv8hi (__a, __b);
+  return __builtin_aarch64_srhaddv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_srhaddv4si (__a, __b);
+  return __builtin_aarch64_srhaddv4si (__a, __b);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  return (uint8x16_t) __builtin_aarch64_urhaddv16qi ((int8x16_t) __a,
-           (int8x16_t) __b);
+  return __builtin_aarch64_urhaddv16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_urhaddv8hi ((int16x8_t) __a,
-          (int16x8_t) __b);
+  return __builtin_aarch64_urhaddv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrhaddq_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_urhaddv4si ((int32x4_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_urhaddv4si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_addhnv8hi (__a, __b);
+  return __builtin_aarch64_addhnv8hi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_addhnv4si (__a, __b);
+  return __builtin_aarch64_addhnv4si (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_addhnv2di (__a, __b);
+  return __builtin_aarch64_addhnv2di (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_addhnv8hi ((int16x8_t) __a,
-        (int16x8_t) __b);
+  return __builtin_aarch64_addhnv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_addhnv4si ((int32x4_t) __a,
-         (int32x4_t) __b);
+  return __builtin_aarch64_addhnv4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_addhnv2di ((int64x2_t) __a,
-         (int64x2_t) __b);
+  return __builtin_aarch64_addhnv2di_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_raddhnv8hi (__a, __b);
+  return __builtin_aarch64_raddhnv8hi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_raddhnv4si (__a, __b);
+  return __builtin_aarch64_raddhnv4si (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_raddhnv2di (__a, __b);
+  return __builtin_aarch64_raddhnv2di (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_raddhnv8hi ((int16x8_t) __a,
-         (int16x8_t) __b);
+  return __builtin_aarch64_raddhnv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_raddhnv4si ((int32x4_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_raddhnv4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_raddhnv2di ((int64x2_t) __a,
-          (int64x2_t) __b);
+  return __builtin_aarch64_raddhnv2di_uuu (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_addhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_addhn2v8hi (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_addhn2v4si (__a, __b, __c);
+  return __builtin_aarch64_addhn2v4si (__a, __b, __c);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_addhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_addhn2v2di (__a, __b, __c);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_addhn2v8hi ((int8x8_t) __a,
-          (int16x8_t) __b,
-          (int16x8_t) __c);
+  return __builtin_aarch64_addhn2v8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_addhn2v4si ((int16x4_t) __a,
-          (int32x4_t) __b,
-          (int32x4_t) __c);
+  return __builtin_aarch64_addhn2v4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_addhn2v2di ((int32x2_t) __a,
-          (int64x2_t) __b,
-          (int64x2_t) __c);
+  return __builtin_aarch64_addhn2v2di_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_raddhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_raddhn2v8hi (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_raddhn2v4si (__a, __b, __c);
+  return __builtin_aarch64_raddhn2v4si (__a, __b, __c);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_raddhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_raddhn2v2di (__a, __b, __c);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_raddhn2v8hi ((int8x8_t) __a,
-           (int16x8_t) __b,
-           (int16x8_t) __c);
+  return __builtin_aarch64_raddhn2v8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_raddhn2v4si ((int16x4_t) __a,
-           (int32x4_t) __b,
-           (int32x4_t) __c);
+  return __builtin_aarch64_raddhn2v4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vraddhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_raddhn2v2di ((int32x2_t) __a,
-           (int64x2_t) __b,
-           (int64x2_t) __c);
+  return __builtin_aarch64_raddhn2v2di_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -1139,8 +738,7 @@ __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmul_p8 (poly8x8_t __a, poly8x8_t __b)
 {
-  return (poly8x8_t) __builtin_aarch64_pmulv8qi ((int8x8_t) __a,
-       (int8x8_t) __b);
+  return __builtin_aarch64_pmulv8qi_ppp (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -1194,8 +792,7 @@ __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmulq_p8 (poly8x16_t __a, poly8x16_t __b)
 {
-  return (poly8x16_t) __builtin_aarch64_pmulv16qi ((int8x16_t) __a,
-         (int8x16_t) __b);
+  return __builtin_aarch64_pmulv16qi_ppp (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -1801,157 +1398,145 @@ __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_ssublv8qi (__a, __b);
+  return __builtin_aarch64_ssublv8qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_ssublv4hi (__a, __b);
+  return __builtin_aarch64_ssublv4hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_ssublv2si (__a, __b);
+  return __builtin_aarch64_ssublv2si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_usublv8qi ((int8x8_t) __a,
-         (int8x8_t) __b);
+  return __builtin_aarch64_usublv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_usublv4hi ((int16x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_usublv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_usublv2si ((int32x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_usublv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_s8 (int8x16_t __a, int8x16_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_ssubl2v16qi (__a, __b);
+  return __builtin_aarch64_ssubl2v16qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_ssubl2v8hi (__a, __b);
+  return __builtin_aarch64_ssubl2v8hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_ssubl2v4si (__a, __b);
+  return __builtin_aarch64_ssubl2v4si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_usubl2v16qi ((int8x16_t) __a,
-           (int8x16_t) __b);
+  return __builtin_aarch64_usubl2v16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_usubl2v8hi ((int16x8_t) __a,
-          (int16x8_t) __b);
+  return __builtin_aarch64_usubl2v8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubl_high_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_usubl2v4si ((int32x4_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_usubl2v4si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_s8 (int16x8_t __a, int8x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_ssubwv8qi (__a, __b);
+  return __builtin_aarch64_ssubwv8qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_s16 (int32x4_t __a, int16x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_ssubwv4hi (__a, __b);
+  return __builtin_aarch64_ssubwv4hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_s32 (int64x2_t __a, int32x2_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_ssubwv2si (__a, __b);
+  return __builtin_aarch64_ssubwv2si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_u8 (uint16x8_t __a, uint8x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_usubwv8qi ((int16x8_t) __a,
-         (int8x8_t) __b);
+  return __builtin_aarch64_usubwv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_u16 (uint32x4_t __a, uint16x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_usubwv4hi ((int32x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_usubwv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_u32 (uint64x2_t __a, uint32x2_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_usubwv2si ((int64x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_usubwv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_s8 (int16x8_t __a, int8x16_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_ssubw2v16qi (__a, __b);
+  return __builtin_aarch64_ssubw2v16qi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_s16 (int32x4_t __a, int16x8_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_ssubw2v8hi (__a, __b);
+  return __builtin_aarch64_ssubw2v8hi (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_s32 (int64x2_t __a, int32x4_t __b)
 {
-  return (int64x2_t) __builtin_aarch64_ssubw2v4si (__a, __b);
+  return __builtin_aarch64_ssubw2v4si (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_u8 (uint16x8_t __a, uint8x16_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_usubw2v16qi ((int16x8_t) __a,
-           (int8x16_t) __b);
+  return __builtin_aarch64_usubw2v16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_u16 (uint32x4_t __a, uint16x8_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_usubw2v8hi ((int32x4_t) __a,
-          (int16x8_t) __b);
+  return __builtin_aarch64_usubw2v8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubw_high_u32 (uint64x2_t __a, uint32x4_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_usubw2v4si ((int64x2_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_usubw2v4si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -1987,241 +1572,217 @@ __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int8x8_t)__builtin_aarch64_shsubv8qi (__a, __b);
+  return __builtin_aarch64_shsubv8qi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_shsubv4hi (__a, __b);
+  return __builtin_aarch64_shsubv4hi (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_shsubv2si (__a, __b);
+  return __builtin_aarch64_shsubv2si (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_uhsubv8qi ((int8x8_t) __a,
-        (int8x8_t) __b);
+  return __builtin_aarch64_uhsubv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_uhsubv4hi ((int16x4_t) __a,
-         (int16x4_t) __b);
+  return __builtin_aarch64_uhsubv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsub_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_uhsubv2si ((int32x2_t) __a,
-         (int32x2_t) __b);
+  return __builtin_aarch64_uhsubv2si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_s8 (int8x16_t __a, int8x16_t __b)
 {
-  return (int8x16_t) __builtin_aarch64_shsubv16qi (__a, __b);
+  return __builtin_aarch64_shsubv16qi (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_shsubv8hi (__a, __b);
+  return __builtin_aarch64_shsubv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_shsubv4si (__a, __b);
+  return __builtin_aarch64_shsubv4si (__a, __b);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  return (uint8x16_t) __builtin_aarch64_uhsubv16qi ((int8x16_t) __a,
-          (int8x16_t) __b);
+  return __builtin_aarch64_uhsubv16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_uhsubv8hi ((int16x8_t) __a,
-         (int16x8_t) __b);
+  return __builtin_aarch64_uhsubv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vhsubq_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_uhsubv4si ((int32x4_t) __a,
-         (int32x4_t) __b);
+  return __builtin_aarch64_uhsubv4si_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_subhnv8hi (__a, __b);
+  return __builtin_aarch64_subhnv8hi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_subhnv4si (__a, __b);
+  return __builtin_aarch64_subhnv4si (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_subhnv2di (__a, __b);
+  return __builtin_aarch64_subhnv2di (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_subhnv8hi ((int16x8_t) __a,
-        (int16x8_t) __b);
+  return __builtin_aarch64_subhnv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_subhnv4si ((int32x4_t) __a,
-         (int32x4_t) __b);
+  return __builtin_aarch64_subhnv4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_subhnv2di ((int64x2_t) __a,
-         (int64x2_t) __b);
+  return __builtin_aarch64_subhnv2di_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s16 (int16x8_t __a, int16x8_t __b)
 {
-  return (int8x8_t) __builtin_aarch64_rsubhnv8hi (__a, __b);
+  return __builtin_aarch64_rsubhnv8hi (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s32 (int32x4_t __a, int32x4_t __b)
 {
-  return (int16x4_t) __builtin_aarch64_rsubhnv4si (__a, __b);
+  return __builtin_aarch64_rsubhnv4si (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_s64 (int64x2_t __a, int64x2_t __b)
 {
-  return (int32x2_t) __builtin_aarch64_rsubhnv2di (__a, __b);
+  return __builtin_aarch64_rsubhnv2di (__a, __b);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_rsubhnv8hi ((int16x8_t) __a,
-         (int16x8_t) __b);
+  return __builtin_aarch64_rsubhnv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_rsubhnv4si ((int32x4_t) __a,
-          (int32x4_t) __b);
+  return __builtin_aarch64_rsubhnv4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_rsubhnv2di ((int64x2_t) __a,
-          (int64x2_t) __b);
+  return __builtin_aarch64_rsubhnv2di_uuu (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_rsubhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v8hi (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_rsubhn2v4si (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v4si (__a, __b, __c);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_rsubhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_rsubhn2v2di (__a, __b, __c);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_rsubhn2v8hi ((int8x8_t) __a,
-           (int16x8_t) __b,
-           (int16x8_t) __c);
+  return __builtin_aarch64_rsubhn2v8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_rsubhn2v4si ((int16x4_t) __a,
-           (int32x4_t) __b,
-           (int32x4_t) __c);
+  return __builtin_aarch64_rsubhn2v4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrsubhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_rsubhn2v2di ((int32x2_t) __a,
-           (int64x2_t) __b,
-           (int64x2_t) __c);
+  return __builtin_aarch64_rsubhn2v2di_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s16 (int8x8_t __a, int16x8_t __b, int16x8_t __c)
 {
-  return (int8x16_t) __builtin_aarch64_subhn2v8hi (__a, __b, __c);
+  return __builtin_aarch64_subhn2v8hi (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s32 (int16x4_t __a, int32x4_t __b, int32x4_t __c)
 {
-  return (int16x8_t) __builtin_aarch64_subhn2v4si (__a, __b, __c);;
+  return __builtin_aarch64_subhn2v4si (__a, __b, __c);;
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_s64 (int32x2_t __a, int64x2_t __b, int64x2_t __c)
 {
-  return (int32x4_t) __builtin_aarch64_subhn2v2di (__a, __b, __c);
+  return __builtin_aarch64_subhn2v2di (__a, __b, __c);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u16 (uint8x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_subhn2v8hi ((int8x8_t) __a,
-          (int16x8_t) __b,
-          (int16x8_t) __c);
+  return __builtin_aarch64_subhn2v8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u32 (uint16x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_subhn2v4si ((int16x4_t) __a,
-          (int32x4_t) __b,
-          (int32x4_t) __c);
+  return __builtin_aarch64_subhn2v4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vsubhn_high_u64 (uint32x2_t __a, uint64x2_t __b, uint64x2_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_subhn2v2di ((int32x2_t) __a,
-          (int64x2_t) __b,
-          (int64x2_t) __c);
+  return __builtin_aarch64_subhn2v2di_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5461,19 +5022,19 @@ __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_s8 (int8x8_t __a, int8x8_t __b)
 {
-  return (int8x16_t) __builtin_aarch64_combinev8qi (__a, __b);
+  return __builtin_aarch64_combinev8qi (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_s16 (int16x4_t __a, int16x4_t __b)
 {
-  return (int16x8_t) __builtin_aarch64_combinev4hi (__a, __b);
+  return __builtin_aarch64_combinev4hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_s32 (int32x2_t __a, int32x2_t __b)
 {
-  return (int32x4_t) __builtin_aarch64_combinev2si (__a, __b);
+  return __builtin_aarch64_combinev2si (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5491,34 +5052,31 @@ __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_f32 (float32x2_t __a, float32x2_t __b)
 {
-  return (float32x4_t) __builtin_aarch64_combinev2sf (__a, __b);
+  return __builtin_aarch64_combinev2sf (__a, __b);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint8x16_t) __builtin_aarch64_combinev8qi ((int8x8_t) __a,
-           (int8x8_t) __b);
+  return __builtin_aarch64_combinev8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint16x8_t) __builtin_aarch64_combinev4hi ((int16x4_t) __a,
-           (int16x4_t) __b);
+  return __builtin_aarch64_combinev4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint32x4_t) __builtin_aarch64_combinev2si ((int32x2_t) __a,
-           (int32x2_t) __b);
+  return __builtin_aarch64_combinev2si_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_u64 (uint64x1_t __a, uint64x1_t __b)
 {
-  return (uint64x2_t) __builtin_aarch64_combinedi (__a[0], __b[0]);
+  return __builtin_aarch64_combinedi_uuu (__a[0], __b[0]);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5530,21 +5088,19 @@ __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_p8 (poly8x8_t __a, poly8x8_t __b)
 {
-  return (poly8x16_t) __builtin_aarch64_combinev8qi ((int8x8_t) __a,
-           (int8x8_t) __b);
+  return __builtin_aarch64_combinev8qi_ppp (__a, __b);
 }
 __extension__ extern __inline poly16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_p16 (poly16x4_t __a, poly16x4_t __b)
 {
-  return (poly16x8_t) __builtin_aarch64_combinev4hi ((int16x4_t) __a,
-           (int16x4_t) __b);
+  return __builtin_aarch64_combinev4hi_ppp (__a, __b);
 }
 __extension__ extern __inline poly64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_p64 (poly64x1_t __a, poly64x1_t __b)
 {
-  return (poly64x2_t) __builtin_aarch64_combinedi_ppp (__a[0], __b[0]);
+  return __builtin_aarch64_combinedi_ppp (__a[0], __b[0]);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5898,46 +5454,25 @@ __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtx_f32_f64 (float64x2_t __a)
 {
-  float32x2_t __result;
-  __asm__ ("fcvtxn %0.2s,%1.2d"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_float_trunc_rodd_lo_v2sf (__a);
 }
 __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtx_high_f32_f64 (float32x2_t __a, float64x2_t __b)
 {
-  float32x4_t __result;
-  __asm__ ("fcvtxn2 %0.4s,%1.2d"
-           : "=w"(__result)
-           : "w" (__b), "0"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_float_trunc_rodd_hi_v4sf (__a, __b);
 }
 __extension__ extern __inline float32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcvtxd_f32_f64 (float64_t __a)
 {
-  float32_t __result;
-  __asm__ ("fcvtxn %s0,%d1"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_float_trunc_rodd_df (__a);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_n_f32 (float32x2_t __a, float32x2_t __b, float32_t __c)
 {
-  float32x2_t __result;
-  float32x2_t __t1;
-  __asm__ ("fmul %1.2s, %3.2s, %4.s[0]; fadd %0.2s, %0.2s, %1.2s"
-           : "=w"(__result), "=w"(__t1)
-           : "0"(__a), "w"(__b), "w"(__c)
-           : );
-  return __result;
+  return __builtin_aarch64_float_mla_nv2sf (__a, __b, __c);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5955,17 +5490,13 @@ __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_n_u16 (uint16x4_t __a, uint16x4_t __b, uint16_t __c)
 {
-  return (uint16x4_t) __builtin_aarch64_mla_nv4hi ((int16x4_t) __a,
-                                                   (int16x4_t) __b,
-                                                   (int16_t) __c);
+  return __builtin_aarch64_mla_nv4hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_n_u32 (uint32x2_t __a, uint32x2_t __b, uint32_t __c)
 {
-  return (uint32x2_t) __builtin_aarch64_mla_nv2si ((int32x2_t) __a,
-                                                   (int32x2_t) __b,
-                                                   (int32_t) __c);
+  return __builtin_aarch64_mla_nv2si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -5989,25 +5520,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_u8 (uint8x8_t __a, uint8x8_t __b, uint8x8_t __c)
 {
-  return (uint8x8_t) __builtin_aarch64_mlav8qi ((int8x8_t) __a,
-                                                (int8x8_t) __b,
-                                                (int8x8_t) __c);
+  return __builtin_aarch64_mlav8qi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_u16 (uint16x4_t __a, uint16x4_t __b, uint16x4_t __c)
 {
-  return (uint16x4_t) __builtin_aarch64_mlav4hi ((int16x4_t) __a,
-                                                 (int16x4_t) __b,
-                                                 (int16x4_t) __c);
+  return __builtin_aarch64_mlav4hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_u32 (uint32x2_t __a, uint32x2_t __b, uint32x2_t __c)
 {
-  return (uint32x2_t) __builtin_aarch64_mlav2si ((int32x2_t) __a,
-                                                 (int32x2_t) __b,
-                                                 (int32x2_t) __c);
+  return __builtin_aarch64_mlav2si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6237,13 +5762,7 @@ __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_n_f32 (float32x4_t __a, float32x4_t __b, float32_t __c)
 {
-  float32x4_t __result;
-  float32x4_t __t1;
-  __asm__ ("fmul %1.4s, %3.4s, %4.s[0]; fadd %0.4s, %0.4s, %1.4s"
-           : "=w"(__result), "=w"(__t1)
-           : "0"(__a), "w"(__b), "w"(__c)
-           : );
-  return __result;
+  return __builtin_aarch64_float_mla_nv4sf (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6261,17 +5780,13 @@ __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_n_u16 (uint16x8_t __a, uint16x8_t __b, uint16_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_mla_nv8hi ((int16x8_t) __a,
-                                                   (int16x8_t) __b,
-                                                   (int16_t) __c);
+  return __builtin_aarch64_mla_nv8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_n_u32 (uint32x4_t __a, uint32x4_t __b, uint32_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_mla_nv4si ((int32x4_t) __a,
-                                                   (int32x4_t) __b,
-                                                   (int32_t) __c);
+  return __builtin_aarch64_mla_nv4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6295,37 +5810,25 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_u8 (uint8x16_t __a, uint8x16_t __b, uint8x16_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_mlav16qi ((int8x16_t) __a,
-                                                  (int8x16_t) __b,
-                                                  (int8x16_t) __c);
+  return __builtin_aarch64_mlav16qi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_u16 (uint16x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_mlav8hi ((int16x8_t) __a,
-                                                 (int16x8_t) __b,
-                                                 (int16x8_t) __c);
+  return __builtin_aarch64_mlav8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_u32 (uint32x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_mlav4si ((int32x4_t) __a,
-                                                 (int32x4_t) __b,
-                                                 (int32x4_t) __c);
+  return __builtin_aarch64_mlav4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_n_f32 (float32x2_t __a, float32x2_t __b, float32_t __c)
 {
-  float32x2_t __result;
-  float32x2_t __t1;
-  __asm__ ("fmul %1.2s, %3.2s, %4.s[0]; fsub %0.2s, %0.2s, %1.2s"
-           : "=w"(__result), "=w"(__t1)
-           : "0"(__a), "w"(__b), "w"(__c)
-           : );
-  return __result;
+  return __builtin_aarch64_float_mls_nv2sf (__a, __b, __c);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6343,17 +5846,13 @@ __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_n_u16 (uint16x4_t __a, uint16x4_t __b, uint16_t __c)
 {
-  return (uint16x4_t) __builtin_aarch64_mls_nv4hi ((int16x4_t) __a,
-                                                   (int16x4_t) __b,
-                                                   (int16_t) __c);
+  return __builtin_aarch64_mls_nv4hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_n_u32 (uint32x2_t __a, uint32x2_t __b, uint32_t __c)
 {
-  return (uint32x2_t) __builtin_aarch64_mls_nv2si ((int32x2_t) __a,
-                                                   (int32x2_t) __b,
-                                                   (int32_t) __c);
+  return __builtin_aarch64_mls_nv2si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6377,25 +5876,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_u8 (uint8x8_t __a, uint8x8_t __b, uint8x8_t __c)
 {
-  return (uint8x8_t) __builtin_aarch64_mlsv8qi ((int8x8_t) __a,
-                                                (int8x8_t) __b,
-                                                (int8x8_t) __c);
+  return __builtin_aarch64_mlsv8qi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_u16 (uint16x4_t __a, uint16x4_t __b, uint16x4_t __c)
 {
-  return (uint16x4_t) __builtin_aarch64_mlsv4hi ((int16x4_t) __a,
-                                                 (int16x4_t) __b,
-                                                 (int16x4_t) __c);
+  return __builtin_aarch64_mlsv4hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_u32 (uint32x2_t __a, uint32x2_t __b, uint32x2_t __c)
 {
-  return (uint32x2_t) __builtin_aarch64_mlsv2si ((int32x2_t) __a,
-                                                 (int32x2_t) __b,
-                                                 (int32x2_t) __c);
+  return __builtin_aarch64_mlsv2si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6629,13 +6122,7 @@ __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_n_f32 (float32x4_t __a, float32x4_t __b, float32_t __c)
 {
-  float32x4_t __result;
-  float32x4_t __t1;
-  __asm__ ("fmul %1.4s, %3.4s, %4.s[0]; fsub %0.4s, %0.4s, %1.4s"
-           : "=w"(__result), "=w"(__t1)
-           : "0"(__a), "w"(__b), "w"(__c)
-           : );
-  return __result;
+  return __builtin_aarch64_float_mls_nv4sf (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6653,17 +6140,13 @@ __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_n_u16 (uint16x8_t __a, uint16x8_t __b, uint16_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_mls_nv8hi ((int16x8_t) __a,
-                                                   (int16x8_t) __b,
-                                                   (int16_t) __c);
+  return __builtin_aarch64_mls_nv8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_n_u32 (uint32x4_t __a, uint32x4_t __b, uint32_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_mls_nv4si ((int32x4_t) __a,
-                                                   (int32x4_t) __b,
-                                                   (int32_t) __c);
+  return __builtin_aarch64_mls_nv4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6687,25 +6170,19 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_u8 (uint8x16_t __a, uint8x16_t __b, uint8x16_t __c)
 {
-  return (uint8x16_t) __builtin_aarch64_mlsv16qi ((int8x16_t) __a,
-                                                  (int8x16_t) __b,
-                                                  (int8x16_t) __c);
+  return __builtin_aarch64_mlsv16qi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_u16 (uint16x8_t __a, uint16x8_t __b, uint16x8_t __c)
 {
-  return (uint16x8_t) __builtin_aarch64_mlsv8hi ((int16x8_t) __a,
-                                                 (int16x8_t) __b,
-                                                 (int16x8_t) __c);
+  return __builtin_aarch64_mlsv8hi_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_u32 (uint32x4_t __a, uint32x4_t __b, uint32x4_t __c)
 {
-  return (uint32x4_t) __builtin_aarch64_mlsv4si ((int32x4_t) __a,
-                                                 (int32x4_t) __b,
-                                                 (int32x4_t) __c);
+  return __builtin_aarch64_mlsv4si_uuuu (__a, __b, __c);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6801,22 +6278,19 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_high_u16 (uint8x8_t __a, uint16x8_t __b)
 {
-  return (uint8x16_t)
-    __builtin_aarch64_xtn2v8hi ((int8x8_t) __a, (int16x8_t) __b);
+  return __builtin_aarch64_xtn2v8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_high_u32 (uint16x4_t __a, uint32x4_t __b)
 {
-  return (uint16x8_t)
-    __builtin_aarch64_xtn2v4si ((int16x4_t) __a, (int32x4_t) __b);
+  return __builtin_aarch64_xtn2v4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_high_u64 (uint32x2_t __a, uint64x2_t __b)
 {
-  return (uint32x4_t)
-    __builtin_aarch64_xtn2v2di ((int32x2_t) __a, (int64x2_t) __b);
+  return __builtin_aarch64_xtn2v2di_uuu (__a, __b);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6840,19 +6314,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_u16 (uint16x8_t __a)
 {
-  return (uint8x8_t)__builtin_aarch64_xtnv8hi ((int16x8_t) __a);
+  return __builtin_aarch64_xtnv8hi_uu (__a);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_u32 (uint32x4_t __a)
 {
-  return (uint16x4_t) __builtin_aarch64_xtnv4si ((int32x4_t )__a);
+  return __builtin_aarch64_xtnv4si_uu (__a);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmovn_u64 (uint64x2_t __a)
 {
-  return (uint32x2_t) __builtin_aarch64_xtnv2di ((int64x2_t) __a);
+  return __builtin_aarch64_xtnv2di_uu (__a);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6876,19 +6350,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_n_u16 (uint16x8_t __a, const int __b)
 {
-  return (uint8x8_t)__builtin_aarch64_shrnv8hi ((int16x8_t)__a, __b);
+  return __builtin_aarch64_shrnv8hi_uus (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_n_u32 (uint32x4_t __a, const int __b)
 {
-  return (uint16x4_t)__builtin_aarch64_shrnv4si ((int32x4_t)__a, __b);
+  return __builtin_aarch64_shrnv4si_uus (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_n_u64 (uint64x2_t __a, const int __b)
 {
-  return (uint32x2_t)__builtin_aarch64_shrnv2di ((int64x2_t)__a, __b);
+  return __builtin_aarch64_shrnv2di_uus (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -6966,12 +6440,7 @@ __extension__ extern __inline poly16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmull_high_p8 (poly8x16_t __a, poly8x16_t __b)
 {
-  poly16x8_t __result;
-  __asm__ ("pmull2 %0.8h,%1.16b,%2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_pmull_hiv16qi_ppp (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7085,12 +6554,7 @@ __extension__ extern __inline poly16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmull_p8 (poly8x8_t __a, poly8x8_t __b)
 {
-  poly16x8_t __result;
-  __asm__ ("pmull %0.8h, %1.8b, %2.8b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_pmullv8qi_ppp (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7144,12 +6608,7 @@ __extension__ extern __inline int64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpadal_s32 (int64x1_t __a, int32x2_t __b)
 {
-  int64x1_t __result;
-  __asm__ ("sadalp %0.1d,%2.2s"
-           : "=w"(__result)
-           : "0"(__a), "w"(__b)
-           : );
-  return __result;
+  return (int64x1_t) __builtin_aarch64_sadalpv2si (__a[0], __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7167,12 +6626,7 @@ __extension__ extern __inline uint64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpadal_u32 (uint64x1_t __a, uint32x2_t __b)
 {
-  uint64x1_t __result;
-  __asm__ ("uadalp %0.1d,%2.2s"
-           : "=w"(__result)
-           : "0"(__a), "w"(__b)
-           : );
-  return __result;
+  return (uint64x1_t) __builtin_aarch64_uadalpv2si_uuu (__a[0], __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7214,265 +6668,145 @@ __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_s8 (int8x8_t __a)
 {
-  int16x4_t __result;
-  __asm__ ("saddlp %0.4h,%1.8b"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_saddlpv8qi (__a);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_s16 (int16x4_t __a)
 {
-  int32x2_t __result;
-  __asm__ ("saddlp %0.2s,%1.4h"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_saddlpv4hi (__a);
 }
 __extension__ extern __inline int64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_s32 (int32x2_t __a)
 {
-  int64x1_t __result;
-  __asm__ ("saddlp %0.1d,%1.2s"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return (int64x1_t) __builtin_aarch64_saddlpv2si (__a);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_u8 (uint8x8_t __a)
 {
-  uint16x4_t __result;
-  __asm__ ("uaddlp %0.4h,%1.8b"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_uaddlpv8qi_uu (__a);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_u16 (uint16x4_t __a)
 {
-  uint32x2_t __result;
-  __asm__ ("uaddlp %0.2s,%1.4h"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_uaddlpv4hi_uu (__a);
 }
 __extension__ extern __inline uint64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddl_u32 (uint32x2_t __a)
 {
-  uint64x1_t __result;
-  __asm__ ("uaddlp %0.1d,%1.2s"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return (uint64x1_t) __builtin_aarch64_uaddlpv2si_uu (__a);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_s8 (int8x16_t __a)
 {
-  int16x8_t __result;
-  __asm__ ("saddlp %0.8h,%1.16b"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_saddlpv16qi (__a);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_s16 (int16x8_t __a)
 {
-  int32x4_t __result;
-  __asm__ ("saddlp %0.4s,%1.8h"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_saddlpv8hi (__a);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_s32 (int32x4_t __a)
 {
-  int64x2_t __result;
-  __asm__ ("saddlp %0.2d,%1.4s"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_saddlpv4si (__a);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_u8 (uint8x16_t __a)
 {
-  uint16x8_t __result;
-  __asm__ ("uaddlp %0.8h,%1.16b"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_uaddlpv16qi_uu (__a);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_u16 (uint16x8_t __a)
 {
-  uint32x4_t __result;
-  __asm__ ("uaddlp %0.4s,%1.8h"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_uaddlpv8hi_uu (__a);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddlq_u32 (uint32x4_t __a)
 {
-  uint64x2_t __result;
-  __asm__ ("uaddlp %0.2d,%1.4s"
-           : "=w"(__result)
-           : "w"(__a)
-           : );
-  return __result;
+  return __builtin_aarch64_uaddlpv4si_uu (__a);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_s8 (int8x16_t __a, int8x16_t __b)
 {
-  int8x16_t __result;
-  __asm__ ("addp %0.16b,%1.16b,%2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv16qi (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_s16 (int16x8_t __a, int16x8_t __b)
 {
-  int16x8_t __result;
-  __asm__ ("addp %0.8h,%1.8h,%2.8h"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_s32 (int32x4_t __a, int32x4_t __b)
 {
-  int32x4_t __result;
-  __asm__ ("addp %0.4s,%1.4s,%2.4s"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv4si (__a, __b);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_s64 (int64x2_t __a, int64x2_t __b)
 {
-  int64x2_t __result;
-  __asm__ ("addp %0.2d,%1.2d,%2.2d"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv2di (__a, __b);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_u8 (uint8x16_t __a, uint8x16_t __b)
 {
-  uint8x16_t __result;
-  __asm__ ("addp %0.16b,%1.16b,%2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv16qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_u16 (uint16x8_t __a, uint16x8_t __b)
 {
-  uint16x8_t __result;
-  __asm__ ("addp %0.8h,%1.8h,%2.8h"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv8hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_u32 (uint32x4_t __a, uint32x4_t __b)
 {
-  uint32x4_t __result;
-  __asm__ ("addp %0.4s,%1.4s,%2.4s"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv4si_uuu (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddq_u64 (uint64x2_t __a, uint64x2_t __b)
 {
-  uint64x2_t __result;
-  __asm__ ("addp %0.2d,%1.2d,%2.2d"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_addpv2di_uuu (__a, __b);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqdmulh_n_s16 (int16x4_t __a, int16_t __b)
 {
-  int16x4_t __result;
-  __asm__ ("sqdmulh %0.4h,%1.4h,%2.h[0]"
-           : "=w"(__result)
-           : "w"(__a), "x"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqdmulh_nv4hi (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqdmulh_n_s32 (int32x2_t __a, int32_t __b)
 {
-  int32x2_t __result;
-  __asm__ ("sqdmulh %0.2s,%1.2s,%2.s[0]"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqdmulh_nv2si (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqdmulhq_n_s16 (int16x8_t __a, int16_t __b)
 {
-  int16x8_t __result;
-  __asm__ ("sqdmulh %0.8h,%1.8h,%2.h[0]"
-           : "=w"(__result)
-           : "w"(__a), "x"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqdmulh_nv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqdmulhq_n_s32 (int32x4_t __a, int32_t __b)
 {
-  int32x4_t __result;
-  __asm__ ("sqdmulh %0.4s,%1.4s,%2.s[0]"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqdmulh_nv4si (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7532,45 +6866,25 @@ __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqrdmulh_n_s16 (int16x4_t __a, int16_t __b)
 {
-  int16x4_t __result;
-  __asm__ ("sqrdmulh %0.4h,%1.4h,%2.h[0]"
-           : "=w"(__result)
-           : "w"(__a), "x"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqrdmulh_nv4hi (__a, __b);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqrdmulh_n_s32 (int32x2_t __a, int32_t __b)
 {
-  int32x2_t __result;
-  __asm__ ("sqrdmulh %0.2s,%1.2s,%2.s[0]"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqrdmulh_nv2si (__a, __b);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqrdmulhq_n_s16 (int16x8_t __a, int16_t __b)
 {
-  int16x8_t __result;
-  __asm__ ("sqrdmulh %0.8h,%1.8h,%2.h[0]"
-           : "=w"(__result)
-           : "w"(__a), "x"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqrdmulh_nv8hi (__a, __b);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqrdmulhq_n_s32 (int32x4_t __a, int32_t __b)
 {
-  int32x4_t __result;
-  __asm__ ("sqrdmulh %0.4s,%1.4s,%2.s[0]"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_sqrdmulh_nv4si (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7702,22 +7016,19 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_high_n_u16 (uint8x8_t __a, uint16x8_t __b, const int __c)
 {
-  return (uint8x16_t) __builtin_aarch64_rshrn2v8hi ((int8x8_t) __a,
-          (int16x8_t) __b, __c);
+  return __builtin_aarch64_rshrn2v8hi_uuus (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_high_n_u32 (uint16x4_t __a, uint32x4_t __b, const int __c)
 {
-  return (uint16x8_t) __builtin_aarch64_rshrn2v4si ((int16x4_t) __a,
-          (int32x4_t) __b, __c);
+  return __builtin_aarch64_rshrn2v4si_uuus (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_high_n_u64 (uint32x2_t __a, uint64x2_t __b, const int __c)
 {
-  return (uint32x4_t) __builtin_aarch64_rshrn2v2di ((int32x2_t)__a,
-          (int64x2_t)__b, __c);
+  return __builtin_aarch64_rshrn2v2di_uuus (__a, __b, __c);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7741,19 +7052,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_n_u16 (uint16x8_t __a, const int __b)
 {
-  return (uint8x8_t) __builtin_aarch64_rshrnv8hi ((int16x8_t) __a, __b);
+  return __builtin_aarch64_rshrnv8hi_uus (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_n_u32 (uint32x4_t __a, const int __b)
 {
-  return (uint16x4_t) __builtin_aarch64_rshrnv4si ((int32x4_t) __a, __b);
+  return __builtin_aarch64_rshrnv4si_uus (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrshrn_n_u64 (uint64x2_t __a, const int __b)
 {
-  return (uint32x2_t) __builtin_aarch64_rshrnv2di ((int64x2_t) __a, __b);
+  return __builtin_aarch64_rshrnv2di_uus (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -7789,33 +7100,80 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_high_n_u16 (uint8x8_t __a, uint16x8_t __b, const int __c)
 {
-  return (uint8x16_t)
-    __builtin_aarch64_shrn2v8hi ((int8x8_t) __a, (int16x8_t) __b, __c);
+  return __builtin_aarch64_shrn2v8hi_uuus (__a, __b, __c);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_high_n_u32 (uint16x4_t __a, uint32x4_t __b, const int __c)
 {
-  return (uint16x8_t)
-    __builtin_aarch64_shrn2v4si ((int16x4_t) __a, (int32x4_t) __b, __c);
+  return __builtin_aarch64_shrn2v4si_uuus (__a, __b, __c);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrn_high_n_u64 (uint32x2_t __a, uint64x2_t __b, const int __c)
 {
-  return (uint32x4_t)
-    __builtin_aarch64_shrn2v2di ((int32x2_t) __a, (int64x2_t) __b, __c);
+  return __builtin_aarch64_shrn2v2di_uuus (__a, __b, __c);
+}
+__extension__ extern __inline poly8x8_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsli_n_p8 (poly8x8_t __a, poly8x8_t __b, const int __c)
+{
+  return __builtin_aarch64_ssli_nv8qi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly16x4_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsli_n_p16 (poly16x4_t __a, poly16x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ssli_nv4hi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly8x16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsliq_n_p8 (poly8x16_t __a, poly8x16_t __b, const int __c)
+{
+  return __builtin_aarch64_ssli_nv16qi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly16x8_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsliq_n_p16 (poly16x8_t __a, poly16x8_t __b, const int __c)
+{
+  return __builtin_aarch64_ssli_nv8hi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly8x8_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsri_n_p8 (poly8x8_t __a, poly8x8_t __b, const int __c)
+{
+  return __builtin_aarch64_ssri_nv8qi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly16x4_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsri_n_p16 (poly16x4_t __a, poly16x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ssri_nv4hi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly64x1_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsri_n_p64 (poly64x1_t __a, poly64x1_t __b, const int __c)
+{
+  return (poly64x1_t) __builtin_aarch64_ssri_ndi_ppps (__a[0], __b[0], __c);
+}
+__extension__ extern __inline poly8x16_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsriq_n_p8 (poly8x16_t __a, poly8x16_t __b, const int __c)
+{
+  return __builtin_aarch64_ssri_nv16qi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly16x8_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsriq_n_p16 (poly16x8_t __a, poly16x8_t __b, const int __c)
+{
+  return __builtin_aarch64_ssri_nv8hi_ppps (__a, __b, __c);
+}
+__extension__ extern __inline poly64x2_t
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vsriq_n_p64 (poly64x2_t __a, poly64x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ssri_nv2di_ppps (__a, __b, __c);
 }
-#define vsli_n_p8(a, b, c) __extension__ ({ poly8x8_t b_ = (b); poly8x8_t a_ = (a); poly8x8_t result; __asm__ ("sli %0.8b,%2.8b,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsli_n_p16(a, b, c) __extension__ ({ poly16x4_t b_ = (b); poly16x4_t a_ = (a); poly16x4_t result; __asm__ ("sli %0.4h,%2.4h,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsliq_n_p8(a, b, c) __extension__ ({ poly8x16_t b_ = (b); poly8x16_t a_ = (a); poly8x16_t result; __asm__ ("sli %0.16b,%2.16b,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsliq_n_p16(a, b, c) __extension__ ({ poly16x8_t b_ = (b); poly16x8_t a_ = (a); poly16x8_t result; __asm__ ("sli %0.8h,%2.8h,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsri_n_p8(a, b, c) __extension__ ({ poly8x8_t b_ = (b); poly8x8_t a_ = (a); poly8x8_t result; __asm__ ("sri %0.8b,%2.8b,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsri_n_p16(a, b, c) __extension__ ({ poly16x4_t b_ = (b); poly16x4_t a_ = (a); poly16x4_t result; __asm__ ("sri %0.4h,%2.4h,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsri_n_p64(a, b, c) __extension__ ({ poly64x1_t b_ = (b); poly64x1_t a_ = (a); poly64x1_t result; __asm__ ("sri %d0,%d2,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsriq_n_p8(a, b, c) __extension__ ({ poly8x16_t b_ = (b); poly8x16_t a_ = (a); poly8x16_t result; __asm__ ("sri %0.16b,%2.16b,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsriq_n_p16(a, b, c) __extension__ ({ poly16x8_t b_ = (b); poly16x8_t a_ = (a); poly16x8_t result; __asm__ ("sri %0.8h,%2.8h,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
-#define vsriq_n_p64(a, b, c) __extension__ ({ poly64x2_t b_ = (b); poly64x2_t a_ = (a); poly64x2_t result; __asm__ ("sri %0.2d,%2.2d,%3" : "=w"(result) : "0"(a_), "w"(b_), "i"(c) : ); result; })
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtst_p8 (poly8x8_t __a, poly8x8_t __b)
@@ -7857,166 +7215,594 @@ vtstq_p64 (poly64x2_t __a, poly64x2_t __b)
   return (uint64x2_t) ((((uint64x2_t) __a) & ((uint64x2_t) __b))
          != __AARCH64_INT64_C (0));
 }
-#define __STRUCTN(t, sz, nelem) typedef struct t ## sz ## x ## nelem ## _t { t ## sz ## _t val[nelem]; } t ## sz ## x ## nelem ## _t;
-__STRUCTN (int, 8, 2)
-__STRUCTN (int, 16, 2)
-__STRUCTN (uint, 8, 2)
-__STRUCTN (uint, 16, 2)
-__STRUCTN (float, 16, 2)
-__STRUCTN (poly, 8, 2)
-__STRUCTN (poly, 16, 2)
-__STRUCTN (int, 8, 3)
-__STRUCTN (int, 16, 3)
-__STRUCTN (int, 32, 3)
-__STRUCTN (int, 64, 3)
-__STRUCTN (uint, 8, 3)
-__STRUCTN (uint, 16, 3)
-__STRUCTN (uint, 32, 3)
-__STRUCTN (uint, 64, 3)
-__STRUCTN (float, 16, 3)
-__STRUCTN (float, 32, 3)
-__STRUCTN (float, 64, 3)
-__STRUCTN (poly, 8, 3)
-__STRUCTN (poly, 16, 3)
-__STRUCTN (int, 8, 4)
-__STRUCTN (int, 64, 4)
-__STRUCTN (uint, 8, 4)
-__STRUCTN (uint, 64, 4)
-__STRUCTN (poly, 8, 4)
-__STRUCTN (float, 64, 4)
-#undef __STRUCTN
-#define __ST2_LANE_FUNC(intype, largetype, ptrtype, mode, qmode, ptr_mode, funcsuffix, signedtype) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst2_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { __builtin_aarch64_simd_oi __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __o = __builtin_aarch64_set_qregoi##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregoi##qmode (__o, (signedtype) __temp.val[1], 1); __builtin_aarch64_st2_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __o, __c); }
-__ST2_LANE_FUNC (float16x4x2_t, float16x8x2_t, float16_t, v4hf, v8hf, hf, f16,
-   float16x8_t)
-__ST2_LANE_FUNC (float32x2x2_t, float32x4x2_t, float32_t, v2sf, v4sf, sf, f32,
-   float32x4_t)
-__ST2_LANE_FUNC (float64x1x2_t, float64x2x2_t, float64_t, df, v2df, df, f64,
-   float64x2_t)
-__ST2_LANE_FUNC (poly8x8x2_t, poly8x16x2_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__ST2_LANE_FUNC (poly16x4x2_t, poly16x8x2_t, poly16_t, v4hi, v8hi, hi, p16,
-   int16x8_t)
-__ST2_LANE_FUNC (poly64x1x2_t, poly64x2x2_t, poly64_t, di, v2di_ssps, di, p64,
-   poly64x2_t)
-__ST2_LANE_FUNC (int8x8x2_t, int8x16x2_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__ST2_LANE_FUNC (int16x4x2_t, int16x8x2_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__ST2_LANE_FUNC (int32x2x2_t, int32x4x2_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__ST2_LANE_FUNC (int64x1x2_t, int64x2x2_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__ST2_LANE_FUNC (uint8x8x2_t, uint8x16x2_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__ST2_LANE_FUNC (uint16x4x2_t, uint16x8x2_t, uint16_t, v4hi, v8hi, hi, u16,
-   int16x8_t)
-__ST2_LANE_FUNC (uint32x2x2_t, uint32x4x2_t, uint32_t, v2si, v4si, si, u32,
-   int32x4_t)
-__ST2_LANE_FUNC (uint64x1x2_t, uint64x2x2_t, uint64_t, di, v2di, di, u64,
-   int64x2_t)
-#define __ST2Q_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst2q_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { union { intype __i; __builtin_aarch64_simd_oi __o; } __temp = { __b }; __builtin_aarch64_st2_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __temp.__o, __c); }
-__ST2Q_LANE_FUNC (float16x8x2_t, float16_t, v8hf, hf, f16)
-__ST2Q_LANE_FUNC (float32x4x2_t, float32_t, v4sf, sf, f32)
-__ST2Q_LANE_FUNC (float64x2x2_t, float64_t, v2df, df, f64)
-__ST2Q_LANE_FUNC (poly8x16x2_t, poly8_t, v16qi, qi, p8)
-__ST2Q_LANE_FUNC (poly16x8x2_t, poly16_t, v8hi, hi, p16)
-__ST2Q_LANE_FUNC (poly64x2x2_t, poly64_t, v2di, di, p64)
-__ST2Q_LANE_FUNC (int8x16x2_t, int8_t, v16qi, qi, s8)
-__ST2Q_LANE_FUNC (int16x8x2_t, int16_t, v8hi, hi, s16)
-__ST2Q_LANE_FUNC (int32x4x2_t, int32_t, v4si, si, s32)
-__ST2Q_LANE_FUNC (int64x2x2_t, int64_t, v2di, di, s64)
-__ST2Q_LANE_FUNC (uint8x16x2_t, uint8_t, v16qi, qi, u8)
-__ST2Q_LANE_FUNC (uint16x8x2_t, uint16_t, v8hi, hi, u16)
-__ST2Q_LANE_FUNC (uint32x4x2_t, uint32_t, v4si, si, u32)
-__ST2Q_LANE_FUNC (uint64x2x2_t, uint64_t, v2di, di, u64)
-#define __ST3_LANE_FUNC(intype, largetype, ptrtype, mode, qmode, ptr_mode, funcsuffix, signedtype) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst3_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { __builtin_aarch64_simd_ci __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[2] = vcombine_##funcsuffix (__b.val[2], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[1], 1); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[2], 2); __builtin_aarch64_st3_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __o, __c); }
-__ST3_LANE_FUNC (float16x4x3_t, float16x8x3_t, float16_t, v4hf, v8hf, hf, f16,
-   float16x8_t)
-__ST3_LANE_FUNC (float32x2x3_t, float32x4x3_t, float32_t, v2sf, v4sf, sf, f32,
-   float32x4_t)
-__ST3_LANE_FUNC (float64x1x3_t, float64x2x3_t, float64_t, df, v2df, df, f64,
-   float64x2_t)
-__ST3_LANE_FUNC (poly8x8x3_t, poly8x16x3_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__ST3_LANE_FUNC (poly16x4x3_t, poly16x8x3_t, poly16_t, v4hi, v8hi, hi, p16,
-   int16x8_t)
-__ST3_LANE_FUNC (poly64x1x3_t, poly64x2x3_t, poly64_t, di, v2di_ssps, di, p64,
-   poly64x2_t)
-__ST3_LANE_FUNC (int8x8x3_t, int8x16x3_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__ST3_LANE_FUNC (int16x4x3_t, int16x8x3_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__ST3_LANE_FUNC (int32x2x3_t, int32x4x3_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__ST3_LANE_FUNC (int64x1x3_t, int64x2x3_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__ST3_LANE_FUNC (uint8x8x3_t, uint8x16x3_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__ST3_LANE_FUNC (uint16x4x3_t, uint16x8x3_t, uint16_t, v4hi, v8hi, hi, u16,
-   int16x8_t)
-__ST3_LANE_FUNC (uint32x2x3_t, uint32x4x3_t, uint32_t, v2si, v4si, si, u32,
-   int32x4_t)
-__ST3_LANE_FUNC (uint64x1x3_t, uint64x2x3_t, uint64_t, di, v2di, di, u64,
-   int64x2_t)
-#define __ST3Q_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst3q_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { union { intype __i; __builtin_aarch64_simd_ci __o; } __temp = { __b }; __builtin_aarch64_st3_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __temp.__o, __c); }
-__ST3Q_LANE_FUNC (float16x8x3_t, float16_t, v8hf, hf, f16)
-__ST3Q_LANE_FUNC (float32x4x3_t, float32_t, v4sf, sf, f32)
-__ST3Q_LANE_FUNC (float64x2x3_t, float64_t, v2df, df, f64)
-__ST3Q_LANE_FUNC (poly8x16x3_t, poly8_t, v16qi, qi, p8)
-__ST3Q_LANE_FUNC (poly16x8x3_t, poly16_t, v8hi, hi, p16)
-__ST3Q_LANE_FUNC (poly64x2x3_t, poly64_t, v2di, di, p64)
-__ST3Q_LANE_FUNC (int8x16x3_t, int8_t, v16qi, qi, s8)
-__ST3Q_LANE_FUNC (int16x8x3_t, int16_t, v8hi, hi, s16)
-__ST3Q_LANE_FUNC (int32x4x3_t, int32_t, v4si, si, s32)
-__ST3Q_LANE_FUNC (int64x2x3_t, int64_t, v2di, di, s64)
-__ST3Q_LANE_FUNC (uint8x16x3_t, uint8_t, v16qi, qi, u8)
-__ST3Q_LANE_FUNC (uint16x8x3_t, uint16_t, v8hi, hi, u16)
-__ST3Q_LANE_FUNC (uint32x4x3_t, uint32_t, v4si, si, u32)
-__ST3Q_LANE_FUNC (uint64x2x3_t, uint64_t, v2di, di, u64)
-#define __ST4_LANE_FUNC(intype, largetype, ptrtype, mode, qmode, ptr_mode, funcsuffix, signedtype) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst4_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { __builtin_aarch64_simd_xi __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[2] = vcombine_##funcsuffix (__b.val[2], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __temp.val[3] = vcombine_##funcsuffix (__b.val[3], vcreate_##funcsuffix (__AARCH64_UINT64_C (0))); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[1], 1); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[2], 2); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[3], 3); __builtin_aarch64_st4_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __o, __c); }
-__ST4_LANE_FUNC (float16x4x4_t, float16x8x4_t, float16_t, v4hf, v8hf, hf, f16,
-   float16x8_t)
-__ST4_LANE_FUNC (float32x2x4_t, float32x4x4_t, float32_t, v2sf, v4sf, sf, f32,
-   float32x4_t)
-__ST4_LANE_FUNC (float64x1x4_t, float64x2x4_t, float64_t, df, v2df, df, f64,
-   float64x2_t)
-__ST4_LANE_FUNC (poly8x8x4_t, poly8x16x4_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__ST4_LANE_FUNC (poly16x4x4_t, poly16x8x4_t, poly16_t, v4hi, v8hi, hi, p16,
-   int16x8_t)
-__ST4_LANE_FUNC (poly64x1x4_t, poly64x2x4_t, poly64_t, di, v2di_ssps, di, p64,
-   poly64x2_t)
-__ST4_LANE_FUNC (int8x8x4_t, int8x16x4_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__ST4_LANE_FUNC (int16x4x4_t, int16x8x4_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__ST4_LANE_FUNC (int32x2x4_t, int32x4x4_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__ST4_LANE_FUNC (int64x1x4_t, int64x2x4_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__ST4_LANE_FUNC (uint8x8x4_t, uint8x16x4_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__ST4_LANE_FUNC (uint16x4x4_t, uint16x8x4_t, uint16_t, v4hi, v8hi, hi, u16,
-   int16x8_t)
-__ST4_LANE_FUNC (uint32x2x4_t, uint32x4x4_t, uint32_t, v2si, v4si, si, u32,
-   int32x4_t)
-__ST4_LANE_FUNC (uint64x1x4_t, uint64x2x4_t, uint64_t, di, v2di, di, u64,
-   int64x2_t)
-#define __ST4Q_LANE_FUNC(intype, ptrtype, mode, ptr_mode, funcsuffix) __extension__ extern __inline void __attribute__ ((__always_inline__, __gnu_inline__, __artificial__)) vst4q_lane_ ## funcsuffix (ptrtype *__ptr, intype __b, const int __c) { union { intype __i; __builtin_aarch64_simd_xi __o; } __temp = { __b }; __builtin_aarch64_st4_lane##mode ((__builtin_aarch64_simd_ ## ptr_mode *) __ptr, __temp.__o, __c); }
-__ST4Q_LANE_FUNC (float16x8x4_t, float16_t, v8hf, hf, f16)
-__ST4Q_LANE_FUNC (float32x4x4_t, float32_t, v4sf, sf, f32)
-__ST4Q_LANE_FUNC (float64x2x4_t, float64_t, v2df, df, f64)
-__ST4Q_LANE_FUNC (poly8x16x4_t, poly8_t, v16qi, qi, p8)
-__ST4Q_LANE_FUNC (poly16x8x4_t, poly16_t, v8hi, hi, p16)
-__ST4Q_LANE_FUNC (poly64x2x4_t, poly64_t, v2di, di, p64)
-__ST4Q_LANE_FUNC (int8x16x4_t, int8_t, v16qi, qi, s8)
-__ST4Q_LANE_FUNC (int16x8x4_t, int16_t, v8hi, hi, s16)
-__ST4Q_LANE_FUNC (int32x4x4_t, int32_t, v4si, si, s32)
-__ST4Q_LANE_FUNC (int64x2x4_t, int64_t, v2di, di, s64)
-__ST4Q_LANE_FUNC (uint8x16x4_t, uint8_t, v16qi, qi, u8)
-__ST4Q_LANE_FUNC (uint16x8x4_t, uint16_t, v8hi, hi, u16)
-__ST4Q_LANE_FUNC (uint32x4x4_t, uint32_t, v4si, si, u32)
-__ST4Q_LANE_FUNC (uint64x2x4_t, uint64_t, v2di, di, u64)
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_f16 (float16_t *__ptr, float16x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_f32 (float32_t *__ptr, float32x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_f64 (float64_t *__ptr, float64x1x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanedf ((__builtin_aarch64_simd_df *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_p8 (poly8_t *__ptr, poly8x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_p16 (poly16_t *__ptr, poly16x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_p64 (poly64_t *__ptr, poly64x1x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanedi_sps ((__builtin_aarch64_simd_di *) __ptr,
+        __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_s8 (int8_t *__ptr, int8x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_s16 (int16_t *__ptr, int16x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_s32 (int32_t *__ptr, int32x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_s64 (int64_t *__ptr, int64x1x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanedi ((__builtin_aarch64_simd_di *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_u8 (uint8_t *__ptr, uint8x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_u16 (uint16_t *__ptr, uint16x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_u32 (uint32_t *__ptr, uint32x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_u64 (uint64_t *__ptr, uint64x1x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanedi_sus ((__builtin_aarch64_simd_di *) __ptr, __val,
+        __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_f16 (float16_t *__ptr, float16x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_f32 (float32_t *__ptr, float32x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_f64 (float64_t *__ptr, float64x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2df ((__builtin_aarch64_simd_df *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_p8 (poly8_t *__ptr, poly8x16x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev16qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_p16 (poly16_t *__ptr, poly16x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_p64 (poly64_t *__ptr, poly64x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2di_sps ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_s8 (int8_t *__ptr, int8x16x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev16qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+       __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_s16 (int16_t *__ptr, int16x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_s32 (int32_t *__ptr, int32x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_s64 (int64_t *__ptr, int64x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2di ((__builtin_aarch64_simd_di *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_u8 (uint8_t *__ptr, uint8x16x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev16qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_u16 (uint16_t *__ptr, uint16x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_u32 (uint32_t *__ptr, uint32x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_u64 (uint64_t *__ptr, uint64x2x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev2di_sus ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_f16 (float16_t *__ptr, float16x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_f32 (float32_t *__ptr, float32x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_f64 (float64_t *__ptr, float64x1x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanedf ((__builtin_aarch64_simd_df *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_p8 (poly8_t *__ptr, poly8x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_p16 (poly16_t *__ptr, poly16x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_p64 (poly64_t *__ptr, poly64x1x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanedi_sps ((__builtin_aarch64_simd_di *) __ptr, __val,
+        __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_s8 (int8_t *__ptr, int8x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_s16 (int16_t *__ptr, int16x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_s32 (int32_t *__ptr, int32x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_s64 (int64_t *__ptr, int64x1x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanedi ((__builtin_aarch64_simd_di *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_u8 (uint8_t *__ptr, uint8x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_u16 (uint16_t *__ptr, uint16x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_u32 (uint32_t *__ptr, uint32x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_u64 (uint64_t *__ptr, uint64x1x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanedi_sus ((__builtin_aarch64_simd_di *) __ptr, __val,
+        __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_f16 (float16_t *__ptr, float16x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_f32 (float32_t *__ptr, float32x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_f64 (float64_t *__ptr, float64x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2df ((__builtin_aarch64_simd_df *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_p8 (poly8_t *__ptr, poly8x16x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev16qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_p16 (poly16_t *__ptr, poly16x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_p64 (poly64_t *__ptr, poly64x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2di_sps ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_s8 (int8_t *__ptr, int8x16x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev16qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+       __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_s16 (int16_t *__ptr, int16x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_s32 (int32_t *__ptr, int32x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_s64 (int64_t *__ptr, int64x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2di ((__builtin_aarch64_simd_di *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_u8 (uint8_t *__ptr, uint8x16x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev16qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_u16 (uint16_t *__ptr, uint16x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_u32 (uint32_t *__ptr, uint32x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_u64 (uint64_t *__ptr, uint64x2x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev2di_sus ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_f16 (float16_t *__ptr, float16x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_f32 (float32_t *__ptr, float32x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_f64 (float64_t *__ptr, float64x1x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanedf ((__builtin_aarch64_simd_df *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_p8 (poly8_t *__ptr, poly8x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_p16 (poly16_t *__ptr, poly16x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_p64 (poly64_t *__ptr, poly64x1x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanedi_sps ((__builtin_aarch64_simd_di *) __ptr, __val,
+        __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_s8 (int8_t *__ptr, int8x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_s16 (int16_t *__ptr, int16x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_s32 (int32_t *__ptr, int32x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_s64 (int64_t *__ptr, int64x1x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanedi ((__builtin_aarch64_simd_di *) __ptr, __val,
+    __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_u8 (uint8_t *__ptr, uint8x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_u16 (uint16_t *__ptr, uint16x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_u32 (uint32_t *__ptr, uint32x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_u64 (uint64_t *__ptr, uint64x1x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanedi_sus ((__builtin_aarch64_simd_di *) __ptr, __val,
+        __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_f16 (float16_t *__ptr, float16x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8hf ((__builtin_aarch64_simd_hf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_f32 (float32_t *__ptr, float32x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4sf ((__builtin_aarch64_simd_sf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_f64 (float64_t *__ptr, float64x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2df ((__builtin_aarch64_simd_df *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_p8 (poly8_t *__ptr, poly8x16x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev16qi_sps ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_p16 (poly16_t *__ptr, poly16x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8hi_sps ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_p64 (poly64_t *__ptr, poly64x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2di_sps ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_s8 (int8_t *__ptr, int8x16x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev16qi ((__builtin_aarch64_simd_qi *) __ptr, __val,
+       __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_s16 (int16_t *__ptr, int16x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8hi ((__builtin_aarch64_simd_hi *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_s32 (int32_t *__ptr, int32x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4si ((__builtin_aarch64_simd_si *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_s64 (int64_t *__ptr, int64x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2di ((__builtin_aarch64_simd_di *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_u8 (uint8_t *__ptr, uint8x16x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev16qi_sus ((__builtin_aarch64_simd_qi *) __ptr,
+           __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_u16 (uint16_t *__ptr, uint16x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8hi_sus ((__builtin_aarch64_simd_hi *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_u32 (uint32_t *__ptr, uint32x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4si_sus ((__builtin_aarch64_simd_si *) __ptr,
+          __val, __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_u64 (uint64_t *__ptr, uint64x2x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev2di_sus ((__builtin_aarch64_simd_di *) __ptr,
+          __val, __lane);
+}
 __extension__ extern __inline int64_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddlv_s32 (int32x2_t __a)
@@ -8079,339 +7865,198 @@ vqrdmulhq_laneq_s32 (int32x4_t __a, int32x4_t __b, const int __c)
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1_p8 (poly8x16_t __a, uint8x8_t __b)
+vqtbl1_p8 (poly8x16_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1_s8 (int8x16_t __a, uint8x8_t __b)
+vqtbl1_s8 (int8x16_t __tab, uint8x8_t __idx)
 {
-  int8x8_t __result;
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1_u8 (uint8x16_t __a, uint8x8_t __b)
+vqtbl1_u8 (uint8x16_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1q_p8 (poly8x16_t __a, uint8x16_t __b)
+vqtbl1q_p8 (poly8x16_t __tab, uint8x16_t __idx)
 {
-  poly8x16_t __result;
-  __asm__ ("tbl %0.16b, {%1.16b}, %2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v16qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1q_s8 (int8x16_t __a, uint8x16_t __b)
+vqtbl1q_s8 (int8x16_t __tab, uint8x16_t __idx)
 {
-  int8x16_t __result;
-  __asm__ ("tbl %0.16b, {%1.16b}, %2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v16qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vqtbl1q_u8 (uint8x16_t __a, uint8x16_t __b)
+vqtbl1q_u8 (uint8x16_t __tab, uint8x16_t __idx)
 {
-  uint8x16_t __result;
-  __asm__ ("tbl %0.16b, {%1.16b}, %2.16b"
-           : "=w"(__result)
-           : "w"(__a), "w"(__b)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v16qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1_s8 (int8x8_t __r, int8x16_t __tab, uint8x8_t __idx)
 {
-  int8x8_t __result = __r;
-  __asm__ ("tbx %0.8b,{%1.16b},%2.8b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1_u8 (uint8x8_t __r, uint8x16_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result = __r;
-  __asm__ ("tbx %0.8b,{%1.16b},%2.8b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1_p8 (poly8x8_t __r, poly8x16_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result = __r;
-  __asm__ ("tbx %0.8b,{%1.16b},%2.8b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1q_s8 (int8x16_t __r, int8x16_t __tab, uint8x16_t __idx)
 {
-  int8x16_t __result = __r;
-  __asm__ ("tbx %0.16b,{%1.16b},%2.16b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v16qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1q_u8 (uint8x16_t __r, uint8x16_t __tab, uint8x16_t __idx)
 {
-  uint8x16_t __result = __r;
-  __asm__ ("tbx %0.16b,{%1.16b},%2.16b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v16qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx1q_p8 (poly8x16_t __r, poly8x16_t __tab, uint8x16_t __idx)
 {
-  poly8x16_t __result = __r;
-  __asm__ ("tbx %0.16b,{%1.16b},%2.16b"
-           : "+w"(__result)
-           : "w"(__tab), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v16qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl1_s8 (int8x8_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result;
-  int8x16_t __temp = vcombine_s8 (__tab, vcreate_s8 (__AARCH64_UINT64_C (0x0)));
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  int8x16_t __temp = vcombine_s8 (__tab,
+      vcreate_s8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl1v8qi (__temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl1_u8 (uint8x8_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
-  uint8x16_t __temp = vcombine_u8 (__tab, vcreate_u8 (__AARCH64_UINT64_C (0x0)));
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  uint8x16_t __temp = vcombine_u8 (__tab,
+       vcreate_u8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl1v8qi_uuu (__temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl1_p8 (poly8x8_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
-  poly8x16_t __temp = vcombine_p8 (__tab, vcreate_p8 (__AARCH64_UINT64_C (0x0)));
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  poly8x16_t __temp = vcombine_p8 (__tab,
+       vcreate_p8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl1v8qi_ppu (__temp, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl2_s8 (int8x8x2_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result;
   int8x16_t __temp = vcombine_s8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi (__temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl2_u8 (uint8x8x2_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
   uint8x16_t __temp = vcombine_u8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi_uuu (__temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl2_p8 (poly8x8x2_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
   poly8x16_t __temp = vcombine_p8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbl %0.8b, {%1.16b}, %2.8b"
-           : "=w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbl1v8qi_ppu (__temp, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl3_s8 (int8x8x3_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result;
   int8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_s8 (__tab.val[0], __tab.val[1]);
-  __temp.val[1] = vcombine_s8 (__tab.val[2], vcreate_s8 (__AARCH64_UINT64_C (0x0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = __builtin_aarch64_tbl3v8qi (__o, __idx);
-  return __result;
+  __temp.val[1] = vcombine_s8 (__tab.val[2],
+          vcreate_s8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl2v8qi (__temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl3_u8 (uint8x8x3_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
   uint8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_u8 (__tab.val[0], __tab.val[1]);
-  __temp.val[1] = vcombine_u8 (__tab.val[2], vcreate_u8 (__AARCH64_UINT64_C (0x0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (uint8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
-  return __result;
+  __temp.val[1] = vcombine_u8 (__tab.val[2],
+          vcreate_u8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl2v8qi_uuu (__temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl3_p8 (poly8x8x3_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
   poly8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_p8 (__tab.val[0], __tab.val[1]);
-  __temp.val[1] = vcombine_p8 (__tab.val[2], vcreate_p8 (__AARCH64_UINT64_C (0x0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (poly8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
-  return __result;
+  __temp.val[1] = vcombine_p8 (__tab.val[2],
+          vcreate_p8 (__AARCH64_UINT64_C (0x0)));
+  return __builtin_aarch64_qtbl2v8qi_ppu (__temp, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl4_s8 (int8x8x4_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result;
   int8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_s8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_s8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = __builtin_aarch64_tbl3v8qi (__o, __idx);
-  return __result;
+  return __builtin_aarch64_qtbl2v8qi (__temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl4_u8 (uint8x8x4_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
   uint8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_u8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_u8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (uint8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
-  return __result;
+  return __builtin_aarch64_qtbl2v8qi_uuu (__temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbl4_p8 (poly8x8x4_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
   poly8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_p8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_p8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (poly8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
-  return __result;
+  return __builtin_aarch64_qtbl2v8qi_ppu (__temp, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx2_s8 (int8x8_t __r, int8x8x2_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result = __r;
   int8x16_t __temp = vcombine_s8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbx %0.8b, {%1.16b}, %2.8b"
-           : "+w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi (__r, __temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx2_u8 (uint8x8_t __r, uint8x8x2_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result = __r;
   uint8x16_t __temp = vcombine_u8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbx %0.8b, {%1.16b}, %2.8b"
-           : "+w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi_uuuu (__r, __temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx2_p8 (poly8x8_t __r, poly8x8x2_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result = __r;
   poly8x16_t __temp = vcombine_p8 (__tab.val[0], __tab.val[1]);
-  __asm__ ("tbx %0.8b, {%1.16b}, %2.8b"
-           : "+w"(__result)
-           : "w"(__temp), "w"(__idx)
-           : );
-  return __result;
+  return __builtin_aarch64_qtbx1v8qi_pppu (__r, __temp, __idx);
 }
 __extension__ extern __inline float32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -8562,19 +8207,19 @@ __extension__ extern __inline uint8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddv_u8 (uint8x8_t __a)
 {
-  return (uint8_t) __builtin_aarch64_reduc_plus_scal_v8qi ((int8x8_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v8qi_uu (__a);
 }
 __extension__ extern __inline uint16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddv_u16 (uint16x4_t __a)
 {
-  return (uint16_t) __builtin_aarch64_reduc_plus_scal_v4hi ((int16x4_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v4hi_uu (__a);
 }
 __extension__ extern __inline uint32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddv_u32 (uint32x2_t __a)
 {
-  return (int32_t) __builtin_aarch64_reduc_plus_scal_v2si ((int32x2_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v2si_uu (__a);
 }
 __extension__ extern __inline int8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -8604,25 +8249,25 @@ __extension__ extern __inline uint8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddvq_u8 (uint8x16_t __a)
 {
-  return (uint8_t) __builtin_aarch64_reduc_plus_scal_v16qi ((int8x16_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v16qi_uu (__a);
 }
 __extension__ extern __inline uint16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddvq_u16 (uint16x8_t __a)
 {
-  return (uint16_t) __builtin_aarch64_reduc_plus_scal_v8hi ((int16x8_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v8hi_uu (__a);
 }
 __extension__ extern __inline uint32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddvq_u32 (uint32x4_t __a)
 {
-  return (uint32_t) __builtin_aarch64_reduc_plus_scal_v4si ((int32x4_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v4si_uu (__a);
 }
 __extension__ extern __inline uint64_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vaddvq_u64 (uint64x2_t __a)
 {
-  return (uint64_t) __builtin_aarch64_reduc_plus_scal_v2di ((int64x2_t) __a);
+  return __builtin_aarch64_reduc_plus_scal_v2di_uu (__a);
 }
 __extension__ extern __inline float32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -13003,15 +12648,15 @@ __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p8 (const poly8_t *__a)
 {
-  return (poly8x8_t)
-    __builtin_aarch64_ld1v8qi ((const __builtin_aarch64_simd_qi *) __a);
+  return __builtin_aarch64_ld1v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p16 (const poly16_t *__a)
 {
-  return (poly16x4_t)
-    __builtin_aarch64_ld1v4hi ((const __builtin_aarch64_simd_hi *) __a);
+  return __builtin_aarch64_ld1v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -13047,22 +12692,22 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u8 (const uint8_t *__a)
 {
-  return (uint8x8_t)
-    __builtin_aarch64_ld1v8qi ((const __builtin_aarch64_simd_qi *) __a);
+  return __builtin_aarch64_ld1v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u16 (const uint16_t *__a)
 {
-  return (uint16x4_t)
-    __builtin_aarch64_ld1v4hi ((const __builtin_aarch64_simd_hi *) __a);
+  return __builtin_aarch64_ld1v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u32 (const uint32_t *__a)
 {
-  return (uint32x2_t)
-    __builtin_aarch64_ld1v2si ((const __builtin_aarch64_simd_si *) __a);
+  return __builtin_aarch64_ld1v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -13074,337 +12719,184 @@ __extension__ extern __inline uint8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u8_x3 (const uint8_t *__a)
 {
-  uint8x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = (__builtin_aarch64_simd_ci)__builtin_aarch64_ld1x3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  __i.val[1] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  __i.val[2] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vld1_s8_x3 (const uint8_t *__a)
+vld1_s8_x3 (const int8_t *__a)
 {
-  int8x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  __i.val[1] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  __i.val[2] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u16_x3 (const uint16_t *__a)
 {
-  uint16x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  __i.val[1] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  __i.val[2] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s16_x3 (const int16_t *__a)
 {
-  int16x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  __i.val[1] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  __i.val[2] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u32_x3 (const uint32_t *__a)
 {
-  uint32x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2si ((const __builtin_aarch64_simd_si *) __a);
-  __i.val[0] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  __i.val[1] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  __i.val[2] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vld1_s32_x3 (const uint32_t *__a)
+vld1_s32_x3 (const int32_t *__a)
 {
-  int32x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2si ((const __builtin_aarch64_simd_si *) __a);
-  __i.val[0] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  __i.val[1] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  __i.val[2] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u64_x3 (const uint64_t *__a)
 {
-  uint64x1x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  __i.val[1] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  __i.val[2] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s64_x3 (const int64_t *__a)
 {
-  int64x1x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  __i.val[1] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  __i.val[2] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f16_x3 (const float16_t *__a)
 {
-  float16x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4hf ((const __builtin_aarch64_simd_hf *) __a);
-  __i.val[0] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 0);
-  __i.val[1] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 1);
-  __i.val[2] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f32_x3 (const float32_t *__a)
 {
-  float32x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  __i.val[0] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 0);
-  __i.val[1] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 1);
-  __i.val[2] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f64_x3 (const float64_t *__a)
 {
-  float64x1x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3df ((const __builtin_aarch64_simd_df *) __a);
-  __i.val[0] = (float64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  __i.val[1] = (float64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  __i.val[2] = (float64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p8_x3 (const poly8_t *__a)
 {
-  poly8x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  __i.val[1] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  __i.val[2] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p16_x3 (const poly16_t *__a)
 {
-  poly16x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  __i.val[1] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  __i.val[2] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p64_x3 (const poly64_t *__a)
 {
-  poly64x1x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (poly64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  __i.val[1] = (poly64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  __i.val[2] = (poly64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-return __i;
+  return __builtin_aarch64_ld1x3di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u8_x3 (const uint8_t *__a)
 {
-  uint8x16x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  __i.val[1] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  __i.val[2] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s8_x3 (const int8_t *__a)
 {
-  int8x16x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  __i.val[1] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  __i.val[2] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v16qi (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u16_x3 (const uint16_t *__a)
 {
-  uint16x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  __i.val[1] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  __i.val[2] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s16_x3 (const int16_t *__a)
 {
-  int16x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  __i.val[1] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  __i.val[2] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u32_x3 (const uint32_t *__a)
 {
-  uint32x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4si ((const __builtin_aarch64_simd_si *) __a);
-  __i.val[0] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  __i.val[1] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  __i.val[2] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s32_x3 (const int32_t *__a)
 {
-  int32x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4si ((const __builtin_aarch64_simd_si *) __a);
-  __i.val[0] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  __i.val[1] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  __i.val[2] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u64_x3 (const uint64_t *__a)
 {
-  uint64x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  __i.val[1] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  __i.val[2] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s64_x3 (const int64_t *__a)
 {
-  int64x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  __i.val[1] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  __i.val[2] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f16_x3 (const float16_t *__a)
 {
-  float16x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8hf ((const __builtin_aarch64_simd_hf *) __a);
-  __i.val[0] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 0);
-  __i.val[1] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 1);
-  __i.val[2] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f32_x3 (const float32_t *__a)
 {
-  float32x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  __i.val[0] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 0);
-  __i.val[1] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 1);
-  __i.val[2] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f64_x3 (const float64_t *__a)
 {
-  float64x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2df ((const __builtin_aarch64_simd_df *) __a);
-  __i.val[0] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 0);
-  __i.val[1] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 1);
-  __i.val[2] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p8_x3 (const poly8_t *__a)
 {
-  poly8x16x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  __i.val[0] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  __i.val[1] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  __i.val[2] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p16_x3 (const poly16_t *__a)
 {
-  poly16x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  __i.val[0] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  __i.val[1] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  __i.val[2] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p64_x3 (const poly64_t *__a)
 {
-  poly64x2x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v2di ((const __builtin_aarch64_simd_di *) __a);
-  __i.val[0] = (poly64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  __i.val[1] = (poly64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  __i.val[2] = (poly64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -13428,22 +12920,22 @@ __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p8 (const poly8_t *__a)
 {
-  return (poly8x16_t)
-    __builtin_aarch64_ld1v16qi ((const __builtin_aarch64_simd_qi *) __a);
+  return __builtin_aarch64_ld1v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p16 (const poly16_t *__a)
 {
-  return (poly16x8_t)
-    __builtin_aarch64_ld1v8hi ((const __builtin_aarch64_simd_hi *) __a);
+  return __builtin_aarch64_ld1v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p64 (const poly64_t *__a)
 {
-  return (poly64x2_t)
-    __builtin_aarch64_ld1v2di ((const __builtin_aarch64_simd_di *) __a);
+  return __builtin_aarch64_ld1v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -13473,589 +12965,395 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u8 (const uint8_t *__a)
 {
-  return (uint8x16_t)
-    __builtin_aarch64_ld1v16qi ((const __builtin_aarch64_simd_qi *) __a);
+  return __builtin_aarch64_ld1v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u8_x2 (const uint8_t *__a)
 {
-  uint8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s8_x2 (const int8_t *__a)
 {
-  int8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u16_x2 (const uint16_t *__a)
 {
-  uint16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s16_x2 (const int16_t *__a)
 {
-  int16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u32_x2 (const uint32_t *__a)
 {
-  uint32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s32_x2 (const int32_t *__a)
 {
-  int32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u64_x2 (const uint64_t *__a)
 {
-  uint64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s64_x2 (const int64_t *__a)
 {
-  int64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f16_x2 (const float16_t *__a)
 {
-  float16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x4_t) __builtin_aarch64_get_dregoiv4hf (__o, 0);
-  ret.val[1] = (float16x4_t) __builtin_aarch64_get_dregoiv4hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f32_x2 (const float32_t *__a)
 {
-  float32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f64_x2 (const float64_t *__a)
 {
-  float64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 1)};
-  return ret;
+  return __builtin_aarch64_ld1x2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p8_x2 (const poly8_t *__a)
 {
-  poly8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p16_x2 (const poly16_t *__a)
 {
-  poly16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p64_x2 (const poly64_t *__a)
 {
-  poly64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u8_x2 (const uint8_t *__a)
 {
-  uint8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s8_x2 (const int8_t *__a)
 {
-  int8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v16qi (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u16_x2 (const uint16_t *__a)
 {
-  uint16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s16_x2 (const int16_t *__a)
 {
-  int16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u32_x2 (const uint32_t *__a)
 {
-  uint32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s32_x2 (const int32_t *__a)
 {
-  int32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u64_x2 (const uint64_t *__a)
 {
-  uint64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s64_x2 (const int64_t *__a)
 {
-  int64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f16_x2 (const float16_t *__a)
 {
-  float16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x8_t) __builtin_aarch64_get_qregoiv8hf (__o, 0);
-  ret.val[1] = (float16x8_t) __builtin_aarch64_get_qregoiv8hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f32_x2 (const float32_t *__a)
 {
-  float32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f64_x2 (const float64_t *__a)
 {
-  float64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p8_x2 (const poly8_t *__a)
 {
-  poly8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p16_x2 (const poly16_t *__a)
 {
-  poly16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p64_x2 (const poly64_t *__a)
 {
-  poly64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u16 (const uint16_t *__a)
 {
-  return (uint16x8_t)
-    __builtin_aarch64_ld1v8hi ((const __builtin_aarch64_simd_hi *) __a);
+  return __builtin_aarch64_ld1v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u32 (const uint32_t *__a)
 {
-  return (uint32x4_t)
-    __builtin_aarch64_ld1v4si ((const __builtin_aarch64_simd_si *) __a);
+  return __builtin_aarch64_ld1v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u64 (const uint64_t *__a)
 {
-  return (uint64x2_t)
-    __builtin_aarch64_ld1v2di ((const __builtin_aarch64_simd_di *) __a);
+  return __builtin_aarch64_ld1v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s8_x4 (const int8_t *__a)
 {
-  union { int8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s8_x4 (const int8_t *__a)
 {
-  union { int8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v16qi (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s16_x4 (const int16_t *__a)
 {
-  union { int16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s16_x4 (const int16_t *__a)
 {
-  union { int16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s32_x4 (const int32_t *__a)
 {
-  union { int32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-  = __builtin_aarch64_ld1x4v2si ((const __builtin_aarch64_simd_si *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s32_x4 (const int32_t *__a)
 {
-  union { int32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-  = __builtin_aarch64_ld1x4v4si ((const __builtin_aarch64_simd_si *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u8_x4 (const uint8_t *__a)
 {
-  union { uint8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u8_x4 (const uint8_t *__a)
 {
-  union { uint8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u16_x4 (const uint16_t *__a)
 {
-  union { uint16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u16_x4 (const uint16_t *__a)
 {
-  union { uint16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u32_x4 (const uint32_t *__a)
 {
-  union { uint32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2si ((const __builtin_aarch64_simd_si *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u32_x4 (const uint32_t *__a)
 {
-  union { uint32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4si ((const __builtin_aarch64_simd_si *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f16_x4 (const float16_t *__a)
 {
-  union { float16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4hf ((const __builtin_aarch64_simd_hf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f16_x4 (const float16_t *__a)
 {
-  union { float16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8hf ((const __builtin_aarch64_simd_hf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f32_x4 (const float32_t *__a)
 {
-  union { float32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f32_x4 (const float32_t *__a)
 {
-  union { float32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline poly8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p8_x4 (const poly8_t *__a)
 {
-  union { poly8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p8_x4 (const poly8_t *__a)
 {
-  union { poly8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p16_x4 (const poly16_t *__a)
 {
-  union { poly16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p16_x4 (const poly16_t *__a)
 {
-  union { poly16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_s64_x4 (const int64_t *__a)
 {
-  union { int64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_u64_x4 (const uint64_t *__a)
 {
-  union { uint64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline poly64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_p64_x4 (const poly64_t *__a)
 {
-  union { poly64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_s64_x4 (const int64_t *__a)
 {
-  union { int64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_u64_x4 (const uint64_t *__a)
 {
-  union { uint64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline poly64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_p64_x4 (const poly64_t *__a)
 {
-  union { poly64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2di ((const __builtin_aarch64_simd_di *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_f64_x4 (const float64_t *__a)
 {
-  union { float64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4df ((const __builtin_aarch64_simd_df *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline float64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_f64_x4 (const float64_t *__a)
 {
-  union { float64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v2df ((const __builtin_aarch64_simd_df *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -14397,1009 +13695,539 @@ __extension__ extern __inline int64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_s64 (const int64_t * __a)
 {
-  int64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_u64 (const uint64_t * __a)
 {
-  uint64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2di_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_f64 (const float64_t * __a)
 {
-  float64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 1)};
-  return ret;
+  return __builtin_aarch64_ld2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_s8 (const int8_t * __a)
 {
-  int8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_p8 (const poly8_t * __a)
 {
-  poly8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_p64 (const poly64_t * __a)
 {
-  poly64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregoidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregoidi_pss (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2di_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_s16 (const int16_t * __a)
 {
-  int16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_p16 (const poly16_t * __a)
 {
-  poly16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_s32 (const int32_t * __a)
 {
-  int32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_u8 (const uint8_t * __a)
 {
-  uint8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_u16 (const uint16_t * __a)
 {
-  uint16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_u32 (const uint32_t * __a)
 {
-  uint32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_f16 (const float16_t * __a)
 {
-  float16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4hf (__a);
-  ret.val[0] = __builtin_aarch64_get_dregoiv4hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_dregoiv4hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4hf (__a);
 }
 __extension__ extern __inline float32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_f32 (const float32_t * __a)
 {
-  float32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline int8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_s8 (const int8_t * __a)
 {
-  int8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_p8 (const poly8_t * __a)
 {
-  poly8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_s16 (const int16_t * __a)
 {
-  int16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_p16 (const poly16_t * __a)
 {
-  poly16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_p64 (const poly64_t * __a)
 {
-  poly64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di_pss (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_s32 (const int32_t * __a)
 {
-  int32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_s64 (const int64_t * __a)
 {
-  int64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_u8 (const uint8_t * __a)
 {
-  uint8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_u16 (const uint16_t * __a)
 {
-  uint16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_u32 (const uint32_t * __a)
 {
-  uint32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_u64 (const uint64_t * __a)
 {
-  uint64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_f16 (const float16_t * __a)
 {
-  float16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8hf (__a);
-  ret.val[0] = __builtin_aarch64_get_qregoiv8hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_qregoiv8hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8hf (__a);
 }
 __extension__ extern __inline float32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_f32 (const float32_t * __a)
 {
-  float32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_f64 (const float64_t * __a)
 {
-  float64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_s64 (const int64_t * __a)
 {
-  int64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  ret.val[2] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_u64 (const uint64_t * __a)
 {
-  uint64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  ret.val[2] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3di_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_f64 (const float64_t * __a)
 {
-  float64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 1)};
-  ret.val[2] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 2)};
-  return ret;
+  return __builtin_aarch64_ld3df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_s8 (const int8_t * __a)
 {
-  int8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_p8 (const poly8_t * __a)
 {
-  poly8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_s16 (const int16_t * __a)
 {
-  int16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_p16 (const poly16_t * __a)
 {
-  poly16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_s32 (const int32_t * __a)
 {
-  int32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  ret.val[2] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_u8 (const uint8_t * __a)
 {
-  uint8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_u16 (const uint16_t * __a)
 {
-  uint16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_u32 (const uint32_t * __a)
 {
-  uint32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  ret.val[2] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_f16 (const float16_t * __a)
 {
-  float16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4hf (__a);
-  ret.val[0] = __builtin_aarch64_get_dregciv4hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_dregciv4hf (__o, 1);
-  ret.val[2] = __builtin_aarch64_get_dregciv4hf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4hf (__a);
 }
 __extension__ extern __inline float32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_f32 (const float32_t * __a)
 {
-  float32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 1);
-  ret.val[2] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline poly64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_p64 (const poly64_t * __a)
 {
-  poly64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 1);
-  ret.val[2] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3di_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_s8 (const int8_t * __a)
 {
-  int8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_p8 (const poly8_t * __a)
 {
-  poly8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v16qi_ps ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_s16 (const int16_t * __a)
 {
-  int16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_p16 (const poly16_t * __a)
 {
-  poly16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8hi_ps ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_s32 (const int32_t * __a)
 {
-  int32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  ret.val[2] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_s64 (const int64_t * __a)
 {
-  int64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  ret.val[2] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_u8 (const uint8_t * __a)
 {
-  uint8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_u16 (const uint16_t * __a)
 {
-  uint16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_u32 (const uint32_t * __a)
 {
-  uint32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  ret.val[2] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_u64 (const uint64_t * __a)
 {
-  uint64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  ret.val[2] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_f16 (const float16_t * __a)
 {
-  float16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8hf (__a);
-  ret.val[0] = __builtin_aarch64_get_qregciv8hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_qregciv8hf (__o, 1);
-  ret.val[2] = __builtin_aarch64_get_qregciv8hf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8hf (__a);
 }
 __extension__ extern __inline float32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_f32 (const float32_t * __a)
 {
-  float32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 1);
-  ret.val[2] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_f64 (const float64_t * __a)
 {
-  float64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 1);
-  ret.val[2] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_p64 (const poly64_t * __a)
 {
-  poly64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 1);
-  ret.val[2] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_s64 (const int64_t * __a)
 {
-  int64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 1);
-  ret.val[2] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 2);
-  ret.val[3] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_u64 (const uint64_t * __a)
 {
-  uint64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 1);
-  ret.val[2] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 2);
-  ret.val[3] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4di_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_f64 (const float64_t * __a)
 {
-  float64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 1)};
-  ret.val[2] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 2)};
-  ret.val[3] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 3)};
-  return ret;
+  return __builtin_aarch64_ld4df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_s8 (const int8_t * __a)
 {
-  int8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_p8 (const poly8_t * __a)
 {
-  poly8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_s16 (const int16_t * __a)
 {
-  int16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_p16 (const poly16_t * __a)
 {
-  poly16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_s32 (const int32_t * __a)
 {
-  int32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 1);
-  ret.val[2] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 2);
-  ret.val[3] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_u8 (const uint8_t * __a)
 {
-  uint8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_u16 (const uint16_t * __a)
 {
-  uint16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_u32 (const uint32_t * __a)
 {
-  uint32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 1);
-  ret.val[2] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 2);
-  ret.val[3] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_f16 (const float16_t * __a)
 {
-  float16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4hf (__a);
-  ret.val[0] = __builtin_aarch64_get_dregxiv4hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_dregxiv4hf (__o, 1);
-  ret.val[2] = __builtin_aarch64_get_dregxiv4hf (__o, 2);
-  ret.val[3] = __builtin_aarch64_get_dregxiv4hf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4hf (__a);
 }
 __extension__ extern __inline float32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_f32 (const float32_t * __a)
 {
-  float32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 1);
-  ret.val[2] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 2);
-  ret.val[3] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline poly64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_p64 (const poly64_t * __a)
 {
-  poly64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 1);
-  ret.val[2] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 2);
-  ret.val[3] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4di_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_s8 (const int8_t * __a)
 {
-  int8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_p8 (const poly8_t * __a)
 {
-  poly8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_s16 (const int16_t * __a)
 {
-  int16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_p16 (const poly16_t * __a)
 {
-  poly16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_s32 (const int32_t * __a)
 {
-  int32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 1);
-  ret.val[2] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 2);
-  ret.val[3] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_s64 (const int64_t * __a)
 {
-  int64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 1);
-  ret.val[2] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 2);
-  ret.val[3] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_u8 (const uint8_t * __a)
 {
-  uint8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_u16 (const uint16_t * __a)
 {
-  uint16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_u32 (const uint32_t * __a)
 {
-  uint32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 1);
-  ret.val[2] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 2);
-  ret.val[3] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_u64 (const uint64_t * __a)
 {
-  uint64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 1);
-  ret.val[2] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 2);
-  ret.val[3] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_f16 (const float16_t * __a)
 {
-  float16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8hf (__a);
-  ret.val[0] = __builtin_aarch64_get_qregxiv8hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_qregxiv8hf (__o, 1);
-  ret.val[2] = __builtin_aarch64_get_qregxiv8hf (__o, 2);
-  ret.val[3] = __builtin_aarch64_get_qregxiv8hf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8hf (__a);
 }
 __extension__ extern __inline float32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_f32 (const float32_t * __a)
 {
-  float32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 1);
-  ret.val[2] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 2);
-  ret.val[3] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_f64 (const float64_t * __a)
 {
-  float64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 1);
-  ret.val[2] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 2);
-  ret.val[3] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_p64 (const poly64_t * __a)
 {
-  poly64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 1);
-  ret.val[2] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 2);
-  ret.val[3] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline poly128_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -15411,1154 +14239,1142 @@ __extension__ extern __inline int8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_s8 (const int8_t * __a)
 {
-  int8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_s16 (const int16_t * __a)
 {
-  int16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_s32 (const int32_t * __a)
 {
-  int32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_f16 (const float16_t * __a)
 {
-  float16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = __builtin_aarch64_get_dregoiv4hf (__o, 0);
-  ret.val[1] = (float16x4_t) __builtin_aarch64_get_dregoiv4hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_f32 (const float32_t * __a)
 {
-  float32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregoiv2sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_f64 (const float64_t * __a)
 {
-  float64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rdf ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregoidf (__o, 1)};
-  return ret;
+  return __builtin_aarch64_ld2rdf ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline uint8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_u8 (const uint8_t * __a)
 {
-  uint8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_u16 (const uint16_t * __a)
 {
-  uint16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_u32 (const uint32_t * __a)
 {
-  uint32x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregoiv2si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline poly8x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_p8 (const poly8_t * __a)
 {
-  poly8x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregoiv8qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_p16 (const poly16_t * __a)
 {
-  poly16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregoiv4hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_p64 (const poly64_t * __a)
 {
-  poly64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregoidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregoidi_pss (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rdi_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_s64 (const int64_t * __a)
 {
-  int64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rdi ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_u64 (const uint64_t * __a)
 {
-  uint64x1x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregoidi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rdi_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_s8 (const int8_t * __a)
 {
-  int8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_p8 (const poly8_t * __a)
 {
-  poly8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_s16 (const int16_t * __a)
 {
-  int16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_p16 (const poly16_t * __a)
 {
-  poly16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_s32 (const int32_t * __a)
 {
-  int32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_s64 (const int64_t * __a)
 {
-  int64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_u8 (const uint8_t * __a)
 {
-  uint8x16x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregoiv16qi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_u16 (const uint16_t * __a)
 {
-  uint16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregoiv8hi (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_u32 (const uint32_t * __a)
 {
-  uint32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregoiv4si (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_u64 (const uint64_t * __a)
 {
-  uint64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregoiv2di (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_f16 (const float16_t * __a)
 {
-  float16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x8_t) __builtin_aarch64_get_qregoiv8hf (__o, 0);
-  ret.val[1] = __builtin_aarch64_get_qregoiv8hf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_f32 (const float32_t * __a)
 {
-  float32x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregoiv4sf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_f64 (const float64_t * __a)
 {
-  float64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregoiv2df (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly64x2x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_p64 (const poly64_t * __a)
 {
-  poly64x2x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregoiv2di_pss (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_s64 (const int64_t * __a)
 {
-  int64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  ret.val[2] = (int64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rdi ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_u64 (const uint64_t * __a)
 {
-  uint64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 1);
-  ret.val[2] = (uint64x1_t) __builtin_aarch64_get_dregcidi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rdi_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_f64 (const float64_t * __a)
 {
-  float64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rdf ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 1)};
-  ret.val[2] = (float64x1_t) {__builtin_aarch64_get_dregcidf (__o, 2)};
-  return ret;
+  return __builtin_aarch64_ld3rdf ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_s8 (const int8_t * __a)
 {
-  int8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (int8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_p8 (const poly8_t * __a)
 {
-  poly8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (poly8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_s16 (const int16_t * __a)
 {
-  int16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (int16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_p16 (const poly16_t * __a)
 {
-  poly16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (poly16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_s32 (const int32_t * __a)
 {
-  int32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  ret.val[2] = (int32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_u8 (const uint8_t * __a)
 {
-  uint8x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 1);
-  ret.val[2] = (uint8x8_t) __builtin_aarch64_get_dregciv8qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_u16 (const uint16_t * __a)
 {
-  uint16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 1);
-  ret.val[2] = (uint16x4_t) __builtin_aarch64_get_dregciv4hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_u32 (const uint32_t * __a)
 {
-  uint32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 1);
-  ret.val[2] = (uint32x2_t) __builtin_aarch64_get_dregciv2si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_f16 (const float16_t * __a)
 {
-  float16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 0);
-  ret.val[1] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 1);
-  ret.val[2] = (float16x4_t) __builtin_aarch64_get_dregciv4hf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_f32 (const float32_t * __a)
 {
-  float32x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 1);
-  ret.val[2] = (float32x2_t) __builtin_aarch64_get_dregciv2sf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline poly64x1x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_p64 (const poly64_t * __a)
 {
-  poly64x1x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 1);
-  ret.val[2] = (poly64x1_t) __builtin_aarch64_get_dregcidi_pss (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rdi_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_s8 (const int8_t * __a)
 {
-  int8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (int8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_p8 (const poly8_t * __a)
 {
-  poly8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (poly8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_s16 (const int16_t * __a)
 {
-  int16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (int16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_p16 (const poly16_t * __a)
 {
-  poly16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (poly16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_s32 (const int32_t * __a)
 {
-  int32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  ret.val[2] = (int32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_s64 (const int64_t * __a)
 {
-  int64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  ret.val[2] = (int64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_u8 (const uint8_t * __a)
 {
-  uint8x16x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 1);
-  ret.val[2] = (uint8x16_t) __builtin_aarch64_get_qregciv16qi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_u16 (const uint16_t * __a)
 {
-  uint16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 1);
-  ret.val[2] = (uint16x8_t) __builtin_aarch64_get_qregciv8hi (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_u32 (const uint32_t * __a)
 {
-  uint32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 1);
-  ret.val[2] = (uint32x4_t) __builtin_aarch64_get_qregciv4si (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_u64 (const uint64_t * __a)
 {
-  uint64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 1);
-  ret.val[2] = (uint64x2_t) __builtin_aarch64_get_qregciv2di (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_f16 (const float16_t * __a)
 {
-  float16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 0);
-  ret.val[1] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 1);
-  ret.val[2] = (float16x8_t) __builtin_aarch64_get_qregciv8hf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_f32 (const float32_t * __a)
 {
-  float32x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 1);
-  ret.val[2] = (float32x4_t) __builtin_aarch64_get_qregciv4sf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_f64 (const float64_t * __a)
 {
-  float64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 1);
-  ret.val[2] = (float64x2_t) __builtin_aarch64_get_qregciv2df (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly64x2x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_p64 (const poly64_t * __a)
 {
-  poly64x2x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 1);
-  ret.val[2] = (poly64x2_t) __builtin_aarch64_get_qregciv2di_pss (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_s64 (const int64_t * __a)
 {
-  int64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 0);
-  ret.val[1] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 1);
-  ret.val[2] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 2);
-  ret.val[3] = (int64x1_t) __builtin_aarch64_get_dregxidi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rdi ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_u64 (const uint64_t * __a)
 {
-  uint64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rdi ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 0);
-  ret.val[1] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 1);
-  ret.val[2] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 2);
-  ret.val[3] = (uint64x1_t) __builtin_aarch64_get_dregxidi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rdi_us ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_f64 (const float64_t * __a)
 {
-  float64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rdf ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 0)};
-  ret.val[1] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 1)};
-  ret.val[2] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 2)};
-  ret.val[3] = (float64x1_t) {__builtin_aarch64_get_dregxidf (__o, 3)};
-  return ret;
+  return __builtin_aarch64_ld4rdf ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline int8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_s8 (const int8_t * __a)
 {
-  int8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (int8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_p8 (const poly8_t * __a)
 {
-  poly8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (poly8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_s16 (const int16_t * __a)
 {
-  int16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (int16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_p16 (const poly16_t * __a)
 {
-  poly16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (poly16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_s32 (const int32_t * __a)
 {
-  int32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 0);
-  ret.val[1] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 1);
-  ret.val[2] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 2);
-  ret.val[3] = (int32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint8x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_u8 (const uint8_t * __a)
 {
-  uint8x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 0);
-  ret.val[1] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 1);
-  ret.val[2] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 2);
-  ret.val[3] = (uint8x8_t) __builtin_aarch64_get_dregxiv8qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_u16 (const uint16_t * __a)
 {
-  uint16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 0);
-  ret.val[1] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 1);
-  ret.val[2] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 2);
-  ret.val[3] = (uint16x4_t) __builtin_aarch64_get_dregxiv4hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_u32 (const uint32_t * __a)
 {
-  uint32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 0);
-  ret.val[1] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 1);
-  ret.val[2] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 2);
-  ret.val[3] = (uint32x2_t) __builtin_aarch64_get_dregxiv2si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline float16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_f16 (const float16_t * __a)
 {
-  float16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x4_t) __builtin_aarch64_get_dregxiv4hf (__o, 0);
-  ret.val[1] = (float16x4_t) __builtin_aarch64_get_dregxiv4hf (__o, 1);
-  ret.val[2] = (float16x4_t) __builtin_aarch64_get_dregxiv4hf (__o, 2);
-  ret.val[3] = (float16x4_t) __builtin_aarch64_get_dregxiv4hf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_f32 (const float32_t * __a)
 {
-  float32x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 0);
-  ret.val[1] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 1);
-  ret.val[2] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 2);
-  ret.val[3] = (float32x2_t) __builtin_aarch64_get_dregxiv2sf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline poly64x1x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_p64 (const poly64_t * __a)
 {
-  poly64x1x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 0);
-  ret.val[1] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 1);
-  ret.val[2] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 2);
-  ret.val[3] = (poly64x1_t) __builtin_aarch64_get_dregxidi_pss (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rdi_ps ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline int8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_s8 (const int8_t * __a)
 {
-  int8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (int8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv16qi ((const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline poly8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_p8 (const poly8_t * __a)
 {
-  poly8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (poly8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv16qi_ps (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline int16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_s16 (const int16_t * __a)
 {
-  int16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (int16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8hi ((const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline poly16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_p16 (const poly16_t * __a)
 {
-  poly16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (poly16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8hi_ps (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline int32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_s32 (const int32_t * __a)
 {
-  int32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 0);
-  ret.val[1] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 1);
-  ret.val[2] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 2);
-  ret.val[3] = (int32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4si ((const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline int64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_s64 (const int64_t * __a)
 {
-  int64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 0);
-  ret.val[1] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 1);
-  ret.val[2] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 2);
-  ret.val[3] = (int64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2di ((const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline uint8x16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_u8 (const uint8_t * __a)
 {
-  uint8x16x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv16qi ((const __builtin_aarch64_simd_qi *) __a);
-  ret.val[0] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 0);
-  ret.val[1] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 1);
-  ret.val[2] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 2);
-  ret.val[3] = (uint8x16_t) __builtin_aarch64_get_qregxiv16qi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv16qi_us (
+    (const __builtin_aarch64_simd_qi *) __a);
 }
 __extension__ extern __inline uint16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_u16 (const uint16_t * __a)
 {
-  uint16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8hi ((const __builtin_aarch64_simd_hi *) __a);
-  ret.val[0] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 0);
-  ret.val[1] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 1);
-  ret.val[2] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 2);
-  ret.val[3] = (uint16x8_t) __builtin_aarch64_get_qregxiv8hi (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8hi_us (
+    (const __builtin_aarch64_simd_hi *) __a);
 }
 __extension__ extern __inline uint32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_u32 (const uint32_t * __a)
 {
-  uint32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4si ((const __builtin_aarch64_simd_si *) __a);
-  ret.val[0] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 0);
-  ret.val[1] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 1);
-  ret.val[2] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 2);
-  ret.val[3] = (uint32x4_t) __builtin_aarch64_get_qregxiv4si (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4si_us (
+    (const __builtin_aarch64_simd_si *) __a);
 }
 __extension__ extern __inline uint64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_u64 (const uint64_t * __a)
 {
-  uint64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 0);
-  ret.val[1] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 1);
-  ret.val[2] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 2);
-  ret.val[3] = (uint64x2_t) __builtin_aarch64_get_qregxiv2di (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2di_us (
+    (const __builtin_aarch64_simd_di *) __a);
 }
 __extension__ extern __inline float16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_f16 (const float16_t * __a)
 {
-  float16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8hf ((const __builtin_aarch64_simd_hf *) __a);
-  ret.val[0] = (float16x8_t) __builtin_aarch64_get_qregxiv8hf (__o, 0);
-  ret.val[1] = (float16x8_t) __builtin_aarch64_get_qregxiv8hf (__o, 1);
-  ret.val[2] = (float16x8_t) __builtin_aarch64_get_qregxiv8hf (__o, 2);
-  ret.val[3] = (float16x8_t) __builtin_aarch64_get_qregxiv8hf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8hf ((const __builtin_aarch64_simd_hf *) __a);
 }
 __extension__ extern __inline float32x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_f32 (const float32_t * __a)
 {
-  float32x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4sf ((const __builtin_aarch64_simd_sf *) __a);
-  ret.val[0] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 0);
-  ret.val[1] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 1);
-  ret.val[2] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 2);
-  ret.val[3] = (float32x4_t) __builtin_aarch64_get_qregxiv4sf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4sf ((const __builtin_aarch64_simd_sf *) __a);
 }
 __extension__ extern __inline float64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_f64 (const float64_t * __a)
 {
-  float64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2df ((const __builtin_aarch64_simd_df *) __a);
-  ret.val[0] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 0);
-  ret.val[1] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 1);
-  ret.val[2] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 2);
-  ret.val[3] = (float64x2_t) __builtin_aarch64_get_qregxiv2df (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2df ((const __builtin_aarch64_simd_df *) __a);
 }
 __extension__ extern __inline poly64x2x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_p64 (const poly64_t * __a)
 {
-  poly64x2x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv2di ((const __builtin_aarch64_simd_di *) __a);
-  ret.val[0] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 0);
-  ret.val[1] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 1);
-  ret.val[2] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 2);
-  ret.val[3] = (poly64x2_t) __builtin_aarch64_get_qregxiv2di_pss (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv2di_ps (
+    (const __builtin_aarch64_simd_di *) __a);
+}
+__extension__ extern __inline uint8x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_u8 (const uint8_t * __ptr, uint8x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8qi_usus (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_u16 (const uint16_t * __ptr, uint16x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4hi_usus (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_u32 (const uint32_t * __ptr, uint32x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2si_usus (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x1x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_u64 (const uint64_t * __ptr, uint64x1x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanedi_usus (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_s8 (const int8_t * __ptr, int8x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8qi (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_s16 (const int16_t * __ptr, int16x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4hi (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_s32 (const int32_t * __ptr, int32x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2si (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x1x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_s64 (const int64_t * __ptr, int64x1x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanedi (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_f16 (const float16_t * __ptr, float16x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4hf (
+   (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_f32 (const float32_t * __ptr, float32x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2sf (
+   (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x1x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_f64 (const float64_t * __ptr, float64x1x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanedf (
+   (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_p8 (const poly8_t * __ptr, poly8x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8qi_psps (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_p16 (const poly16_t * __ptr, poly16x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4hi_psps (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x1x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_p64 (const poly64_t * __ptr, poly64x1x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanedi_psps (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint8x16x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_u8 (const uint8_t * __ptr, uint8x16x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev16qi_usus (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_u16 (const uint16_t * __ptr, uint16x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8hi_usus (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_u32 (const uint32_t * __ptr, uint32x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4si_usus (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_u64 (const uint64_t * __ptr, uint64x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2di_usus (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x16x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_s8 (const int8_t * __ptr, int8x16x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev16qi (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_s16 (const int16_t * __ptr, int16x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8hi (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_s32 (const int32_t * __ptr, int32x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4si (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_s64 (const int64_t * __ptr, int64x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2di (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_f16 (const float16_t * __ptr, float16x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8hf (
+ (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_f32 (const float32_t * __ptr, float32x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4sf (
+ (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_f64 (const float64_t * __ptr, float64x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2df (
+ (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x16x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_p8 (const poly8_t * __ptr, poly8x16x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev16qi_psps (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_p16 (const poly16_t * __ptr, poly16x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8hi_psps (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x2x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_p64 (const poly64_t * __ptr, poly64x2x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev2di_psps (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint8x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_u8 (const uint8_t * __ptr, uint8x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8qi_usus (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_u16 (const uint16_t * __ptr, uint16x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4hi_usus (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_u32 (const uint32_t * __ptr, uint32x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2si_usus (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x1x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_u64 (const uint64_t * __ptr, uint64x1x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanedi_usus (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_s8 (const int8_t * __ptr, int8x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8qi (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_s16 (const int16_t * __ptr, int16x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4hi (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_s32 (const int32_t * __ptr, int32x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2si (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x1x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_s64 (const int64_t * __ptr, int64x1x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanedi (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_f16 (const float16_t * __ptr, float16x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4hf (
+   (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_f32 (const float32_t * __ptr, float32x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2sf (
+   (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x1x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_f64 (const float64_t * __ptr, float64x1x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanedf (
+   (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_p8 (const poly8_t * __ptr, poly8x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8qi_psps (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_p16 (const poly16_t * __ptr, poly16x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4hi_psps (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x1x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_p64 (const poly64_t * __ptr, poly64x1x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanedi_psps (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint8x16x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_u8 (const uint8_t * __ptr, uint8x16x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev16qi_usus (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_u16 (const uint16_t * __ptr, uint16x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8hi_usus (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_u32 (const uint32_t * __ptr, uint32x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4si_usus (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_u64 (const uint64_t * __ptr, uint64x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2di_usus (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x16x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_s8 (const int8_t * __ptr, int8x16x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev16qi (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_s16 (const int16_t * __ptr, int16x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8hi (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_s32 (const int32_t * __ptr, int32x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4si (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_s64 (const int64_t * __ptr, int64x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2di (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_f16 (const float16_t * __ptr, float16x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8hf (
+ (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_f32 (const float32_t * __ptr, float32x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4sf (
+ (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_f64 (const float64_t * __ptr, float64x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2df (
+ (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x16x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_p8 (const poly8_t * __ptr, poly8x16x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev16qi_psps (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_p16 (const poly16_t * __ptr, poly16x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8hi_psps (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x2x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_p64 (const poly64_t * __ptr, poly64x2x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev2di_psps (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint8x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_u8 (const uint8_t * __ptr, uint8x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8qi_usus (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_u16 (const uint16_t * __ptr, uint16x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4hi_usus (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_u32 (const uint32_t * __ptr, uint32x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2si_usus (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x1x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_u64 (const uint64_t * __ptr, uint64x1x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanedi_usus (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_s8 (const int8_t * __ptr, int8x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8qi (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_s16 (const int16_t * __ptr, int16x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4hi (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_s32 (const int32_t * __ptr, int32x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2si (
+   (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x1x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_s64 (const int64_t * __ptr, int64x1x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanedi (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_f16 (const float16_t * __ptr, float16x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4hf (
+   (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_f32 (const float32_t * __ptr, float32x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2sf (
+   (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x1x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_f64 (const float64_t * __ptr, float64x1x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanedf (
+   (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_p8 (const poly8_t * __ptr, poly8x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8qi_psps (
+   (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_p16 (const poly16_t * __ptr, poly16x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4hi_psps (
+   (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x1x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_p64 (const poly64_t * __ptr, poly64x1x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanedi_psps (
+   (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint8x16x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_u8 (const uint8_t * __ptr, uint8x16x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev16qi_usus (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint16x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_u16 (const uint16_t * __ptr, uint16x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8hi_usus (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint32x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_u32 (const uint32_t * __ptr, uint32x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4si_usus (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline uint64x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_u64 (const uint64_t * __ptr, uint64x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2di_usus (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline int8x16x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_s8 (const int8_t * __ptr, int8x16x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev16qi (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int16x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_s16 (const int16_t * __ptr, int16x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8hi (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline int32x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_s32 (const int32_t * __ptr, int32x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4si (
+ (__builtin_aarch64_simd_si *) __ptr, __b, __c);
+}
+__extension__ extern __inline int64x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_s64 (const int64_t * __ptr, int64x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2di (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
+}
+__extension__ extern __inline float16x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_f16 (const float16_t * __ptr, float16x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8hf (
+ (__builtin_aarch64_simd_hf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float32x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_f32 (const float32_t * __ptr, float32x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4sf (
+ (__builtin_aarch64_simd_sf *) __ptr, __b, __c);
+}
+__extension__ extern __inline float64x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_f64 (const float64_t * __ptr, float64x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2df (
+ (__builtin_aarch64_simd_df *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly8x16x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_p8 (const poly8_t * __ptr, poly8x16x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev16qi_psps (
+ (__builtin_aarch64_simd_qi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly16x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_p16 (const poly16_t * __ptr, poly16x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8hi_psps (
+ (__builtin_aarch64_simd_hi *) __ptr, __b, __c);
+}
+__extension__ extern __inline poly64x2x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_p64 (const poly64_t * __ptr, poly64x2x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev2di_psps (
+ (__builtin_aarch64_simd_di *) __ptr, __b, __c);
 }
-#define __LD2_LANE_FUNC(intype, vectype, largetype, ptrtype, mode, qmode, ptrmode, funcsuffix, signedtype) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld2_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_oi __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (0)); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (0)); __o = __builtin_aarch64_set_qregoi##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregoi##qmode (__o, (signedtype) __temp.val[1], 1); __o = __builtin_aarch64_ld2_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); __b.val[0] = (vectype) __builtin_aarch64_get_dregoidi (__o, 0); __b.val[1] = (vectype) __builtin_aarch64_get_dregoidi (__o, 1); return __b; }
-__LD2_LANE_FUNC (float16x4x2_t, float16x4_t, float16x8x2_t, float16_t, v4hf,
-   v8hf, hf, f16, float16x8_t)
-__LD2_LANE_FUNC (float32x2x2_t, float32x2_t, float32x4x2_t, float32_t, v2sf, v4sf,
-   sf, f32, float32x4_t)
-__LD2_LANE_FUNC (float64x1x2_t, float64x1_t, float64x2x2_t, float64_t, df, v2df,
-   df, f64, float64x2_t)
-__LD2_LANE_FUNC (poly8x8x2_t, poly8x8_t, poly8x16x2_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__LD2_LANE_FUNC (poly16x4x2_t, poly16x4_t, poly16x8x2_t, poly16_t, v4hi, v8hi, hi,
-   p16, int16x8_t)
-__LD2_LANE_FUNC (poly64x1x2_t, poly64x1_t, poly64x2x2_t, poly64_t, di,
-   v2di_ssps, di, p64, poly64x2_t)
-__LD2_LANE_FUNC (int8x8x2_t, int8x8_t, int8x16x2_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__LD2_LANE_FUNC (int16x4x2_t, int16x4_t, int16x8x2_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__LD2_LANE_FUNC (int32x2x2_t, int32x2_t, int32x4x2_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__LD2_LANE_FUNC (int64x1x2_t, int64x1_t, int64x2x2_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__LD2_LANE_FUNC (uint8x8x2_t, uint8x8_t, uint8x16x2_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__LD2_LANE_FUNC (uint16x4x2_t, uint16x4_t, uint16x8x2_t, uint16_t, v4hi, v8hi, hi,
-   u16, int16x8_t)
-__LD2_LANE_FUNC (uint32x2x2_t, uint32x2_t, uint32x4x2_t, uint32_t, v2si, v4si, si,
-   u32, int32x4_t)
-__LD2_LANE_FUNC (uint64x1x2_t, uint64x1_t, uint64x2x2_t, uint64_t, di, v2di, di,
-   u64, int64x2_t)
-#define __LD2Q_LANE_FUNC(intype, vtype, ptrtype, mode, ptrmode, funcsuffix) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld2q_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_oi __o; intype ret; __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __b.val[0], 0); __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __b.val[1], 1); __o = __builtin_aarch64_ld2_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); ret.val[0] = (vtype) __builtin_aarch64_get_qregoiv4si (__o, 0); ret.val[1] = (vtype) __builtin_aarch64_get_qregoiv4si (__o, 1); return ret; }
-__LD2Q_LANE_FUNC (float16x8x2_t, float16x8_t, float16_t, v8hf, hf, f16)
-__LD2Q_LANE_FUNC (float32x4x2_t, float32x4_t, float32_t, v4sf, sf, f32)
-__LD2Q_LANE_FUNC (float64x2x2_t, float64x2_t, float64_t, v2df, df, f64)
-__LD2Q_LANE_FUNC (poly8x16x2_t, poly8x16_t, poly8_t, v16qi, qi, p8)
-__LD2Q_LANE_FUNC (poly16x8x2_t, poly16x8_t, poly16_t, v8hi, hi, p16)
-__LD2Q_LANE_FUNC (poly64x2x2_t, poly64x2_t, poly64_t, v2di, di, p64)
-__LD2Q_LANE_FUNC (int8x16x2_t, int8x16_t, int8_t, v16qi, qi, s8)
-__LD2Q_LANE_FUNC (int16x8x2_t, int16x8_t, int16_t, v8hi, hi, s16)
-__LD2Q_LANE_FUNC (int32x4x2_t, int32x4_t, int32_t, v4si, si, s32)
-__LD2Q_LANE_FUNC (int64x2x2_t, int64x2_t, int64_t, v2di, di, s64)
-__LD2Q_LANE_FUNC (uint8x16x2_t, uint8x16_t, uint8_t, v16qi, qi, u8)
-__LD2Q_LANE_FUNC (uint16x8x2_t, uint16x8_t, uint16_t, v8hi, hi, u16)
-__LD2Q_LANE_FUNC (uint32x4x2_t, uint32x4_t, uint32_t, v4si, si, u32)
-__LD2Q_LANE_FUNC (uint64x2x2_t, uint64x2_t, uint64_t, v2di, di, u64)
-#define __LD3_LANE_FUNC(intype, vectype, largetype, ptrtype, mode, qmode, ptrmode, funcsuffix, signedtype) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld3_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_ci __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (0)); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (0)); __temp.val[2] = vcombine_##funcsuffix (__b.val[2], vcreate_##funcsuffix (0)); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[1], 1); __o = __builtin_aarch64_set_qregci##qmode (__o, (signedtype) __temp.val[2], 2); __o = __builtin_aarch64_ld3_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); __b.val[0] = (vectype) __builtin_aarch64_get_dregcidi (__o, 0); __b.val[1] = (vectype) __builtin_aarch64_get_dregcidi (__o, 1); __b.val[2] = (vectype) __builtin_aarch64_get_dregcidi (__o, 2); return __b; }
-__LD3_LANE_FUNC (float16x4x3_t, float16x4_t, float16x8x3_t, float16_t, v4hf,
-   v8hf, hf, f16, float16x8_t)
-__LD3_LANE_FUNC (float32x2x3_t, float32x2_t, float32x4x3_t, float32_t, v2sf, v4sf,
-   sf, f32, float32x4_t)
-__LD3_LANE_FUNC (float64x1x3_t, float64x1_t, float64x2x3_t, float64_t, df, v2df,
-   df, f64, float64x2_t)
-__LD3_LANE_FUNC (poly8x8x3_t, poly8x8_t, poly8x16x3_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__LD3_LANE_FUNC (poly16x4x3_t, poly16x4_t, poly16x8x3_t, poly16_t, v4hi, v8hi, hi,
-   p16, int16x8_t)
-__LD3_LANE_FUNC (poly64x1x3_t, poly64x1_t, poly64x2x3_t, poly64_t, di,
-   v2di_ssps, di, p64, poly64x2_t)
-__LD3_LANE_FUNC (int8x8x3_t, int8x8_t, int8x16x3_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__LD3_LANE_FUNC (int16x4x3_t, int16x4_t, int16x8x3_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__LD3_LANE_FUNC (int32x2x3_t, int32x2_t, int32x4x3_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__LD3_LANE_FUNC (int64x1x3_t, int64x1_t, int64x2x3_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__LD3_LANE_FUNC (uint8x8x3_t, uint8x8_t, uint8x16x3_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__LD3_LANE_FUNC (uint16x4x3_t, uint16x4_t, uint16x8x3_t, uint16_t, v4hi, v8hi, hi,
-   u16, int16x8_t)
-__LD3_LANE_FUNC (uint32x2x3_t, uint32x2_t, uint32x4x3_t, uint32_t, v2si, v4si, si,
-   u32, int32x4_t)
-__LD3_LANE_FUNC (uint64x1x3_t, uint64x1_t, uint64x2x3_t, uint64_t, di, v2di, di,
-   u64, int64x2_t)
-#define __LD3Q_LANE_FUNC(intype, vtype, ptrtype, mode, ptrmode, funcsuffix) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld3q_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_ci __o; intype ret; __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __b.val[0], 0); __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __b.val[1], 1); __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __b.val[2], 2); __o = __builtin_aarch64_ld3_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); ret.val[0] = (vtype) __builtin_aarch64_get_qregciv4si (__o, 0); ret.val[1] = (vtype) __builtin_aarch64_get_qregciv4si (__o, 1); ret.val[2] = (vtype) __builtin_aarch64_get_qregciv4si (__o, 2); return ret; }
-__LD3Q_LANE_FUNC (float16x8x3_t, float16x8_t, float16_t, v8hf, hf, f16)
-__LD3Q_LANE_FUNC (float32x4x3_t, float32x4_t, float32_t, v4sf, sf, f32)
-__LD3Q_LANE_FUNC (float64x2x3_t, float64x2_t, float64_t, v2df, df, f64)
-__LD3Q_LANE_FUNC (poly8x16x3_t, poly8x16_t, poly8_t, v16qi, qi, p8)
-__LD3Q_LANE_FUNC (poly16x8x3_t, poly16x8_t, poly16_t, v8hi, hi, p16)
-__LD3Q_LANE_FUNC (poly64x2x3_t, poly64x2_t, poly64_t, v2di, di, p64)
-__LD3Q_LANE_FUNC (int8x16x3_t, int8x16_t, int8_t, v16qi, qi, s8)
-__LD3Q_LANE_FUNC (int16x8x3_t, int16x8_t, int16_t, v8hi, hi, s16)
-__LD3Q_LANE_FUNC (int32x4x3_t, int32x4_t, int32_t, v4si, si, s32)
-__LD3Q_LANE_FUNC (int64x2x3_t, int64x2_t, int64_t, v2di, di, s64)
-__LD3Q_LANE_FUNC (uint8x16x3_t, uint8x16_t, uint8_t, v16qi, qi, u8)
-__LD3Q_LANE_FUNC (uint16x8x3_t, uint16x8_t, uint16_t, v8hi, hi, u16)
-__LD3Q_LANE_FUNC (uint32x4x3_t, uint32x4_t, uint32_t, v4si, si, u32)
-__LD3Q_LANE_FUNC (uint64x2x3_t, uint64x2_t, uint64_t, v2di, di, u64)
-#define __LD4_LANE_FUNC(intype, vectype, largetype, ptrtype, mode, qmode, ptrmode, funcsuffix, signedtype) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld4_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_xi __o; largetype __temp; __temp.val[0] = vcombine_##funcsuffix (__b.val[0], vcreate_##funcsuffix (0)); __temp.val[1] = vcombine_##funcsuffix (__b.val[1], vcreate_##funcsuffix (0)); __temp.val[2] = vcombine_##funcsuffix (__b.val[2], vcreate_##funcsuffix (0)); __temp.val[3] = vcombine_##funcsuffix (__b.val[3], vcreate_##funcsuffix (0)); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[0], 0); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[1], 1); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[2], 2); __o = __builtin_aarch64_set_qregxi##qmode (__o, (signedtype) __temp.val[3], 3); __o = __builtin_aarch64_ld4_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); __b.val[0] = (vectype) __builtin_aarch64_get_dregxidi (__o, 0); __b.val[1] = (vectype) __builtin_aarch64_get_dregxidi (__o, 1); __b.val[2] = (vectype) __builtin_aarch64_get_dregxidi (__o, 2); __b.val[3] = (vectype) __builtin_aarch64_get_dregxidi (__o, 3); return __b; }
-__LD4_LANE_FUNC (float16x4x4_t, float16x4_t, float16x8x4_t, float16_t, v4hf,
-   v8hf, hf, f16, float16x8_t)
-__LD4_LANE_FUNC (float32x2x4_t, float32x2_t, float32x4x4_t, float32_t, v2sf, v4sf,
-   sf, f32, float32x4_t)
-__LD4_LANE_FUNC (float64x1x4_t, float64x1_t, float64x2x4_t, float64_t, df, v2df,
-   df, f64, float64x2_t)
-__LD4_LANE_FUNC (poly8x8x4_t, poly8x8_t, poly8x16x4_t, poly8_t, v8qi, v16qi, qi, p8,
-   int8x16_t)
-__LD4_LANE_FUNC (poly16x4x4_t, poly16x4_t, poly16x8x4_t, poly16_t, v4hi, v8hi, hi,
-   p16, int16x8_t)
-__LD4_LANE_FUNC (poly64x1x4_t, poly64x1_t, poly64x2x4_t, poly64_t, di,
-   v2di_ssps, di, p64, poly64x2_t)
-__LD4_LANE_FUNC (int8x8x4_t, int8x8_t, int8x16x4_t, int8_t, v8qi, v16qi, qi, s8,
-   int8x16_t)
-__LD4_LANE_FUNC (int16x4x4_t, int16x4_t, int16x8x4_t, int16_t, v4hi, v8hi, hi, s16,
-   int16x8_t)
-__LD4_LANE_FUNC (int32x2x4_t, int32x2_t, int32x4x4_t, int32_t, v2si, v4si, si, s32,
-   int32x4_t)
-__LD4_LANE_FUNC (int64x1x4_t, int64x1_t, int64x2x4_t, int64_t, di, v2di, di, s64,
-   int64x2_t)
-__LD4_LANE_FUNC (uint8x8x4_t, uint8x8_t, uint8x16x4_t, uint8_t, v8qi, v16qi, qi, u8,
-   int8x16_t)
-__LD4_LANE_FUNC (uint16x4x4_t, uint16x4_t, uint16x8x4_t, uint16_t, v4hi, v8hi, hi,
-   u16, int16x8_t)
-__LD4_LANE_FUNC (uint32x2x4_t, uint32x2_t, uint32x4x4_t, uint32_t, v2si, v4si, si,
-   u32, int32x4_t)
-__LD4_LANE_FUNC (uint64x1x4_t, uint64x1_t, uint64x2x4_t, uint64_t, di, v2di, di,
-   u64, int64x2_t)
-#define __LD4Q_LANE_FUNC(intype, vtype, ptrtype, mode, ptrmode, funcsuffix) __extension__ extern __inline intype __attribute__ ((__always_inline__, __gnu_inline__,__artificial__)) vld4q_lane_##funcsuffix (const ptrtype * __ptr, intype __b, const int __c) { __builtin_aarch64_simd_xi __o; intype ret; __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __b.val[0], 0); __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __b.val[1], 1); __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __b.val[2], 2); __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __b.val[3], 3); __o = __builtin_aarch64_ld4_lane##mode ( (__builtin_aarch64_simd_##ptrmode *) __ptr, __o, __c); ret.val[0] = (vtype) __builtin_aarch64_get_qregxiv4si (__o, 0); ret.val[1] = (vtype) __builtin_aarch64_get_qregxiv4si (__o, 1); ret.val[2] = (vtype) __builtin_aarch64_get_qregxiv4si (__o, 2); ret.val[3] = (vtype) __builtin_aarch64_get_qregxiv4si (__o, 3); return ret; }
-__LD4Q_LANE_FUNC (float16x8x4_t, float16x8_t, float16_t, v8hf, hf, f16)
-__LD4Q_LANE_FUNC (float32x4x4_t, float32x4_t, float32_t, v4sf, sf, f32)
-__LD4Q_LANE_FUNC (float64x2x4_t, float64x2_t, float64_t, v2df, df, f64)
-__LD4Q_LANE_FUNC (poly8x16x4_t, poly8x16_t, poly8_t, v16qi, qi, p8)
-__LD4Q_LANE_FUNC (poly16x8x4_t, poly16x8_t, poly16_t, v8hi, hi, p16)
-__LD4Q_LANE_FUNC (poly64x2x4_t, poly64x2_t, poly64_t, v2di, di, p64)
-__LD4Q_LANE_FUNC (int8x16x4_t, int8x16_t, int8_t, v16qi, qi, s8)
-__LD4Q_LANE_FUNC (int16x8x4_t, int16x8_t, int16_t, v8hi, hi, s16)
-__LD4Q_LANE_FUNC (int32x4x4_t, int32x4_t, int32_t, v4si, si, s32)
-__LD4Q_LANE_FUNC (int64x2x4_t, int64x2_t, int64_t, v2di, di, s64)
-__LD4Q_LANE_FUNC (uint8x16x4_t, uint8x16_t, uint8_t, v16qi, qi, u8)
-__LD4Q_LANE_FUNC (uint16x8x4_t, uint16x8_t, uint16_t, v8hi, hi, u16)
-__LD4Q_LANE_FUNC (uint32x4x4_t, uint32x4_t, uint32_t, v4si, si, u32)
-__LD4Q_LANE_FUNC (uint64x2x4_t, uint64x2_t, uint64_t, v2di, di, u64)
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmax_f32 (float32x2_t __a, float32x2_t __b)
 {
-  return __builtin_aarch64_smax_nanv2sf (__a, __b);
+  return __builtin_aarch64_fmax_nanv2sf (__a, __b);
 }
 __extension__ extern __inline float64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmax_f64 (float64x1_t __a, float64x1_t __b)
 {
     return (float64x1_t)
-      { __builtin_aarch64_smax_nandf (vget_lane_f64 (__a, 0),
+      { __builtin_aarch64_fmax_nandf (vget_lane_f64 (__a, 0),
           vget_lane_f64 (__b, 0)) };
 }
 __extension__ extern __inline int8x8_t
@@ -16604,13 +15420,13 @@ __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmaxq_f32 (float32x4_t __a, float32x4_t __b)
 {
-  return __builtin_aarch64_smax_nanv4sf (__a, __b);
+  return __builtin_aarch64_fmax_nanv4sf (__a, __b);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmaxq_f64 (float64x2_t __a, float64x2_t __b)
 {
-  return __builtin_aarch64_smax_nanv2df (__a, __b);
+  return __builtin_aarch64_fmax_nanv2df (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17173,14 +15989,14 @@ __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmin_f32 (float32x2_t __a, float32x2_t __b)
 {
-  return __builtin_aarch64_smin_nanv2sf (__a, __b);
+  return __builtin_aarch64_fmin_nanv2sf (__a, __b);
 }
 __extension__ extern __inline float64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmin_f64 (float64x1_t __a, float64x1_t __b)
 {
     return (float64x1_t)
-   { __builtin_aarch64_smin_nandf (vget_lane_f64 (__a, 0),
+   { __builtin_aarch64_fmin_nandf (vget_lane_f64 (__a, 0),
        vget_lane_f64 (__b, 0)) };
 }
 __extension__ extern __inline int8x8_t
@@ -17226,13 +16042,13 @@ __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vminq_f32 (float32x4_t __a, float32x4_t __b)
 {
-  return __builtin_aarch64_smin_nanv4sf (__a, __b);
+  return __builtin_aarch64_fmin_nanv4sf (__a, __b);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vminq_f64 (float64x2_t __a, float64x2_t __b)
 {
-  return __builtin_aarch64_smin_nanv2df (__a, __b);
+  return __builtin_aarch64_fmin_nanv2df (__a, __b);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17411,32 +16227,32 @@ __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_f32 (float32x2_t __a, float32x2_t __b, float32x2_t __c)
 {
-  return __a + __b * __c;
+  return __builtin_aarch64_float_mlav2sf (__a, __b, __c);
 }
 __extension__ extern __inline float64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_f64 (float64x1_t __a, float64x1_t __b, float64x1_t __c)
 {
-  return __a + __b * __c;
+  return (float64x1_t) {__builtin_aarch64_float_mladf (__a[0], __b[0], __c[0])};
 }
 __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
 {
-  return __a + __b * __c;
+  return __builtin_aarch64_float_mlav4sf (__a, __b, __c);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_f64 (float64x2_t __a, float64x2_t __b, float64x2_t __c)
 {
-  return __a + __b * __c;
+  return __builtin_aarch64_float_mlav2df (__a, __b, __c);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_lane_f32 (float32x2_t __a, float32x2_t __b,
         float32x2_t __c, const int __lane)
 {
-  return (__a + (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mla_lanev2sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17471,7 +16287,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmla_laneq_f32 (float32x2_t __a, float32x2_t __b,
          float32x4_t __c, const int __lane)
 {
-  return (__a + (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mla_laneqv2sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17506,7 +16322,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_lane_f32 (float32x4_t __a, float32x4_t __b,
   float32x2_t __c, const int __lane)
 {
-  return (__a + (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mla_lanev4sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17541,7 +16357,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlaq_laneq_f32 (float32x4_t __a, float32x4_t __b,
    float32x4_t __c, const int __lane)
 {
-  return (__a + (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mla_laneqv4sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17575,32 +16391,32 @@ __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_f32 (float32x2_t __a, float32x2_t __b, float32x2_t __c)
 {
-  return __a - __b * __c;
+  return __builtin_aarch64_float_mlsv2sf (__a, __b, __c);
 }
 __extension__ extern __inline float64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_f64 (float64x1_t __a, float64x1_t __b, float64x1_t __c)
 {
-  return __a - __b * __c;
+  return (float64x1_t) {__builtin_aarch64_float_mlsdf (__a[0], __b[0], __c[0])};
 }
 __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_f32 (float32x4_t __a, float32x4_t __b, float32x4_t __c)
 {
-  return __a - __b * __c;
+  return __builtin_aarch64_float_mlsv4sf (__a, __b, __c);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_f64 (float64x2_t __a, float64x2_t __b, float64x2_t __c)
 {
-  return __a - __b * __c;
+  return __builtin_aarch64_float_mlsv2df (__a, __b, __c);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_lane_f32 (float32x2_t __a, float32x2_t __b,
         float32x2_t __c, const int __lane)
 {
-  return (__a - (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mls_lanev2sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17635,7 +16451,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmls_laneq_f32 (float32x2_t __a, float32x2_t __b,
         float32x4_t __c, const int __lane)
 {
-  return (__a - (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mls_laneqv2sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17670,7 +16486,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_lane_f32 (float32x4_t __a, float32x4_t __b,
   float32x2_t __c, const int __lane)
 {
-  return (__a - (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mls_lanev4sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -17705,7 +16521,7 @@ __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmlsq_laneq_f32 (float32x4_t __a, float32x4_t __b,
   float32x4_t __c, const int __lane)
 {
-  return (__a - (__b * __aarch64_vget_lane_any (__c, __lane)));
+  return __builtin_aarch64_float_mls_laneqv4sf (__a, __b, __c, __lane);
 }
 __extension__ extern __inline int16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -18346,22 +17162,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpadd_u8 (uint8x8_t __a, uint8x8_t __b)
 {
-  return (uint8x8_t) __builtin_aarch64_addpv8qi ((int8x8_t) __a,
-       (int8x8_t) __b);
+  return __builtin_aarch64_addpv8qi_uuu (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpadd_u16 (uint16x4_t __a, uint16x4_t __b)
 {
-  return (uint16x4_t) __builtin_aarch64_addpv4hi ((int16x4_t) __a,
-        (int16x4_t) __b);
+  return __builtin_aarch64_addpv4hi_uuu (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpadd_u32 (uint32x2_t __a, uint32x2_t __b)
 {
-  return (uint32x2_t) __builtin_aarch64_addpv2si ((int32x2_t) __a,
-        (int32x2_t) __b);
+  return __builtin_aarch64_addpv2si_uuu (__a, __b);
 }
 __extension__ extern __inline float32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -18385,7 +17198,7 @@ __extension__ extern __inline uint64_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vpaddd_u64 (uint64x2_t __a)
 {
-  return __builtin_aarch64_addpdi ((int64x2_t) __a);
+  return __builtin_aarch64_addpdi_uu (__a);
 }
 __extension__ extern __inline int64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -19899,373 +18712,217 @@ __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2_s8 (int8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[1], 1);
-  return __builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl2v8qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2_u8 (uint8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (uint8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl2v8qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2_p8 (poly8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (poly8x8_t)__builtin_aarch64_tbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl2v8qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2q_s8 (int8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return __builtin_aarch64_tbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl2v16qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2q_u8 (uint8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (uint8x16_t)__builtin_aarch64_tbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl2v16qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl2q_p8 (poly8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (poly8x16_t)__builtin_aarch64_tbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl2v16qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3_s8 (int8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return __builtin_aarch64_qtbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl3v8qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3_u8 (uint8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (uint8x8_t)__builtin_aarch64_qtbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl3v8qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3_p8 (poly8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (poly8x8_t)__builtin_aarch64_qtbl3v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl3v8qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3q_s8 (int8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return __builtin_aarch64_qtbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl3v16qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3q_u8 (uint8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (uint8x16_t)__builtin_aarch64_qtbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl3v16qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl3q_p8 (poly8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (poly8x16_t)__builtin_aarch64_qtbl3v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl3v16qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4_s8 (int8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return __builtin_aarch64_qtbl4v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl4v8qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4_u8 (uint8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (uint8x8_t)__builtin_aarch64_qtbl4v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl4v8qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4_p8 (poly8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (poly8x8_t)__builtin_aarch64_qtbl4v8qi (__o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbl4v8qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4q_s8 (int8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return __builtin_aarch64_qtbl4v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl4v16qi_ssu (__tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4q_u8 (uint8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (uint8x16_t)__builtin_aarch64_qtbl4v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl4v16qi_uuu (__tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbl4q_p8 (poly8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (poly8x16_t)__builtin_aarch64_qtbl4v16qi (__o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbl4v16qi_ppu (__tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2_s8 (int8x8_t __r, int8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[1], 1);
-  return __builtin_aarch64_tbx4v8qi (__r, __o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx2v8qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2_u8 (uint8x8_t __r, uint8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (uint8x8_t)__builtin_aarch64_tbx4v8qi ((int8x8_t)__r, __o,
-      (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx2v8qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2_p8 (poly8x8_t __r, poly8x16x2_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (poly8x8_t)__builtin_aarch64_tbx4v8qi ((int8x8_t)__r, __o,
-      (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx2v8qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2q_s8 (int8x16_t __r, int8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, __tab.val[1], 1);
-  return __builtin_aarch64_tbx4v16qi (__r, __o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx2v16qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2q_u8 (uint8x16_t __r, uint8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (uint8x16_t)__builtin_aarch64_tbx4v16qi ((int8x16_t)__r, __o,
-        (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx2v16qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx2q_p8 (poly8x16_t __r, poly8x16x2_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  return (poly8x16_t)__builtin_aarch64_tbx4v16qi ((int8x16_t)__r, __o,
-        (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx2v16qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3_s8 (int8x8_t __r, int8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[2], 2);
-  return __builtin_aarch64_qtbx3v8qi (__r, __o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx3v8qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3_u8 (uint8x8_t __r, uint8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (uint8x8_t)__builtin_aarch64_qtbx3v8qi ((int8x8_t)__r, __o,
-       (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx3v8qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3_p8 (poly8x8_t __r, poly8x16x3_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (poly8x8_t)__builtin_aarch64_qtbx3v8qi ((int8x8_t)__r, __o,
-       (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx3v8qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3q_s8 (int8x16_t __r, int8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, __tab.val[2], 2);
-  return __builtin_aarch64_qtbx3v16qi (__r, __o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx3v16qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3q_u8 (uint8x16_t __r, uint8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (uint8x16_t)__builtin_aarch64_qtbx3v16qi ((int8x16_t)__r, __o,
-         (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx3v16qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx3q_p8 (poly8x16_t __r, poly8x16x3_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  return (poly8x16_t)__builtin_aarch64_qtbx3v16qi ((int8x16_t)__r, __o,
-         (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx3v16qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4_s8 (int8x8_t __r, int8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[3], 3);
-  return __builtin_aarch64_qtbx4v8qi (__r, __o, (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx4v8qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4_u8 (uint8x8_t __r, uint8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (uint8x8_t)__builtin_aarch64_qtbx4v8qi ((int8x8_t)__r, __o,
-       (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx4v8qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4_p8 (poly8x8_t __r, poly8x16x4_t __tab, uint8x8_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (poly8x8_t)__builtin_aarch64_qtbx4v8qi ((int8x8_t)__r, __o,
-       (int8x8_t)__idx);
+  return __builtin_aarch64_qtbx4v8qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline int8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4q_s8 (int8x16_t __r, int8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, __tab.val[3], 3);
-  return __builtin_aarch64_qtbx4v16qi (__r, __o, (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx4v16qi_sssu (__r, __tab, __idx);
 }
 __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4q_u8 (uint8x16_t __r, uint8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (uint8x16_t)__builtin_aarch64_qtbx4v16qi ((int8x16_t)__r, __o,
-         (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx4v16qi_uuuu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vqtbx4q_p8 (poly8x16_t __r, poly8x16x4_t __tab, uint8x16_t __idx)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t)__tab.val[3], 3);
-  return (poly8x16_t)__builtin_aarch64_qtbx4v16qi ((int8x16_t)__r, __o,
-         (int8x16_t)__idx);
+  return __builtin_aarch64_qtbx4v16qi_pppu (__r, __tab, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -20737,31 +19394,31 @@ __extension__ extern __inline float32_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndns_f32 (float32_t __a)
 {
-  return __builtin_aarch64_frintnsf (__a);
+  return __builtin_aarch64_roundevensf (__a);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndn_f32 (float32x2_t __a)
 {
-  return __builtin_aarch64_frintnv2sf (__a);
+  return __builtin_aarch64_roundevenv2sf (__a);
 }
 __extension__ extern __inline float64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndn_f64 (float64x1_t __a)
 {
-  return (float64x1_t) {__builtin_aarch64_frintndf (__a[0])};
+  return (float64x1_t) {__builtin_aarch64_roundevendf (__a[0])};
 }
 __extension__ extern __inline float32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndnq_f32 (float32x4_t __a)
 {
-  return __builtin_aarch64_frintnv4sf (__a);
+  return __builtin_aarch64_roundevenv4sf (__a);
 }
 __extension__ extern __inline float64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndnq_f64 (float64x2_t __a)
 {
-  return __builtin_aarch64_frintnv2df (__a);
+  return __builtin_aarch64_roundevenv2df (__a);
 }
 __extension__ extern __inline float32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -21603,19 +20260,19 @@ __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshr_n_u8 (uint8x8_t __a, const int __b)
 {
-  return (uint8x8_t) __builtin_aarch64_lshrv8qi ((int8x8_t) __a, __b);
+  return __builtin_aarch64_lshrv8qi_uus (__a, __b);
 }
 __extension__ extern __inline uint16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshr_n_u16 (uint16x4_t __a, const int __b)
 {
-  return (uint16x4_t) __builtin_aarch64_lshrv4hi ((int16x4_t) __a, __b);
+  return __builtin_aarch64_lshrv4hi_uus (__a, __b);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshr_n_u32 (uint32x2_t __a, const int __b)
 {
-  return (uint32x2_t) __builtin_aarch64_lshrv2si ((int32x2_t) __a, __b);
+  return __builtin_aarch64_lshrv2si_uus (__a, __b);
 }
 __extension__ extern __inline uint64x1_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -21651,25 +20308,25 @@ __extension__ extern __inline uint8x16_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrq_n_u8 (uint8x16_t __a, const int __b)
 {
-  return (uint8x16_t) __builtin_aarch64_lshrv16qi ((int8x16_t) __a, __b);
+  return __builtin_aarch64_lshrv16qi_uus (__a, __b);
 }
 __extension__ extern __inline uint16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrq_n_u16 (uint16x8_t __a, const int __b)
 {
-  return (uint16x8_t) __builtin_aarch64_lshrv8hi ((int16x8_t) __a, __b);
+  return __builtin_aarch64_lshrv8hi_uus (__a, __b);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrq_n_u32 (uint32x4_t __a, const int __b)
 {
-  return (uint32x4_t) __builtin_aarch64_lshrv4si ((int32x4_t) __a, __b);
+  return __builtin_aarch64_lshrv4si_uus (__a, __b);
 }
 __extension__ extern __inline uint64x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vshrq_n_u64 (uint64x2_t __a, const int __b)
 {
-  return (uint64x2_t) __builtin_aarch64_lshrv2di ((int64x2_t) __a, __b);
+  return __builtin_aarch64_lshrv2di_uus (__a, __b);
 }
 __extension__ extern __inline int64_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -22137,15 +20794,13 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p8 (poly8_t *__a, poly8x8_t __b)
 {
-  __builtin_aarch64_st1v8qi ((__builtin_aarch64_simd_qi *) __a,
-        (int8x8_t) __b);
+  __builtin_aarch64_st1v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p16 (poly16_t *__a, poly16x4_t __b)
 {
-  __builtin_aarch64_st1v4hi ((__builtin_aarch64_simd_hi *) __a,
-        (int16x4_t) __b);
+  __builtin_aarch64_st1v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -22181,22 +20836,19 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u8 (uint8_t *__a, uint8x8_t __b)
 {
-  __builtin_aarch64_st1v8qi ((__builtin_aarch64_simd_qi *) __a,
-        (int8x8_t) __b);
+  __builtin_aarch64_st1v8qi_su ((__builtin_aarch64_simd_qi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u16 (uint16_t *__a, uint16x4_t __b)
 {
-  __builtin_aarch64_st1v4hi ((__builtin_aarch64_simd_hi *) __a,
-        (int16x4_t) __b);
+  __builtin_aarch64_st1v4hi_su ((__builtin_aarch64_simd_hi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u32 (uint32_t *__a, uint32x2_t __b)
 {
-  __builtin_aarch64_st1v2si ((__builtin_aarch64_simd_si *) __a,
-        (int32x2_t) __b);
+  __builtin_aarch64_st1v2si_su ((__builtin_aarch64_simd_si *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -22226,22 +20878,19 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p8 (poly8_t *__a, poly8x16_t __b)
 {
-  __builtin_aarch64_st1v16qi ((__builtin_aarch64_simd_qi *) __a,
-         (int8x16_t) __b);
+  __builtin_aarch64_st1v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p16 (poly16_t *__a, poly16x8_t __b)
 {
-  __builtin_aarch64_st1v8hi ((__builtin_aarch64_simd_hi *) __a,
-        (int16x8_t) __b);
+  __builtin_aarch64_st1v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p64 (poly64_t *__a, poly64x2_t __b)
 {
-  __builtin_aarch64_st1v2di_sp ((__builtin_aarch64_simd_di *) __a,
-    (poly64x2_t) __b);
+  __builtin_aarch64_st1v2di_sp ((__builtin_aarch64_simd_di *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -22271,29 +20920,25 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u8 (uint8_t *__a, uint8x16_t __b)
 {
-  __builtin_aarch64_st1v16qi ((__builtin_aarch64_simd_qi *) __a,
-         (int8x16_t) __b);
+  __builtin_aarch64_st1v16qi_su ((__builtin_aarch64_simd_qi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u16 (uint16_t *__a, uint16x8_t __b)
 {
-  __builtin_aarch64_st1v8hi ((__builtin_aarch64_simd_hi *) __a,
-        (int16x8_t) __b);
+  __builtin_aarch64_st1v8hi_su ((__builtin_aarch64_simd_hi *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u32 (uint32_t *__a, uint32x4_t __b)
 {
-  __builtin_aarch64_st1v4si ((__builtin_aarch64_simd_si *) __a,
-        (int32x4_t) __b);
+  __builtin_aarch64_st1v4si_su ((__builtin_aarch64_simd_si *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u64 (uint64_t *__a, uint64x2_t __b)
 {
-  __builtin_aarch64_st1v2di ((__builtin_aarch64_simd_di *) __a,
-        (int64x2_t) __b);
+  __builtin_aarch64_st1v2di_su ((__builtin_aarch64_simd_di *) __a, __b);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -22467,1879 +21112,1010 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s64_x2 (int64_t * __a, int64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int64x2x2_t __temp;
-  __temp.val[0]
-    = vcombine_s64 (__val.val[0], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[1]
-    = vcombine_s64 (__val.val[1], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u64_x2 (uint64_t * __a, uint64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint64x2x2_t __temp;
-  __temp.val[0]
-    = vcombine_u64 (__val.val[0], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1]
-    = vcombine_u64 (__val.val[1], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f64_x2 (float64_t * __a, float64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float64x2x2_t __temp;
-  __temp.val[0]
-    = vcombine_f64 (__val.val[0], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1]
-    = vcombine_f64 (__val.val[1], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st1x2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s8_x2 (int8_t * __a, int8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int8x16x2_t __temp;
-  __temp.val[0]
-    = vcombine_s8 (__val.val[0], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[1]
-    = vcombine_s8 (__val.val[1], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p8_x2 (poly8_t * __a, poly8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly8x16x2_t __temp;
-  __temp.val[0]
-    = vcombine_p8 (__val.val[0], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1]
-    = vcombine_p8 (__val.val[1], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s16_x2 (int16_t * __a, int16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int16x8x2_t __temp;
-  __temp.val[0]
-    = vcombine_s16 (__val.val[0], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[1]
-    = vcombine_s16 (__val.val[1], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p16_x2 (poly16_t * __a, poly16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly16x8x2_t __temp;
-  __temp.val[0]
-    = vcombine_p16 (__val.val[0], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1]
-    = vcombine_p16 (__val.val[1], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s32_x2 (int32_t * __a, int32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int32x4x2_t __temp;
-  __temp.val[0]
-    = vcombine_s32 (__val.val[0], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[1]
-    = vcombine_s32 (__val.val[1], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x2v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u8_x2 (uint8_t * __a, uint8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint8x16x2_t __temp;
-  __temp.val[0] = vcombine_u8 (__val.val[0], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u8 (__val.val[1], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u16_x2 (uint16_t * __a, uint16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint16x8x2_t __temp;
-  __temp.val[0] = vcombine_u16 (__val.val[0], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u16 (__val.val[1], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u32_x2 (uint32_t * __a, uint32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint32x4x2_t __temp;
-  __temp.val[0] = vcombine_u32 (__val.val[0], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u32 (__val.val[1], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x2v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f16_x2 (float16_t * __a, float16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float16x8x2_t __temp;
-  __temp.val[0] = vcombine_f16 (__val.val[0], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f16 (__val.val[1], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __temp.val[1], 1);
-  __builtin_aarch64_st1x2v4hf (__a, __o);
+  __builtin_aarch64_st1x2v4hf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f32_x2 (float32_t * __a, float32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float32x4x2_t __temp;
-  __temp.val[0] = vcombine_f32 (__val.val[0], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f32 (__val.val[1], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2v2sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st1x2v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p64_x2 (poly64_t * __a, poly64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly64x2x2_t __temp;
-  __temp.val[0] = vcombine_p64 (__val.val[0], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p64 (__val.val[1], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st1x2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s8_x2 (int8_t * __a, int8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p8_x2 (poly8_t * __a, poly8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s16_x2 (int16_t * __a, int16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p16_x2 (poly16_t * __a, poly16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s32_x2 (int32_t * __a, int32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x2v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s64_x2 (int64_t * __a, int64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2v2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u8_x2 (uint8_t * __a, uint8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x2v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u16_x2 (uint16_t * __a, uint16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x2v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u32_x2 (uint32_t * __a, uint32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x2v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u64_x2 (uint64_t * __a, uint64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f16_x2 (float16_t * __a, float16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __val.val[1], 1);
-  __builtin_aarch64_st1x2v8hf (__a, __o);
+  __builtin_aarch64_st1x2v8hf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f32_x2 (float32_t * __a, float32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v4sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st1x2v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f64_x2 (float64_t * __a, float64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st1x2v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p64_x2 (poly64_t * __a, poly64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st1x2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x2v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s64_x3 (int64_t * __a, int64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int64x2x3_t __temp;
-  __temp.val[0] = vcombine_s64 (__val.val[0], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s64 (__val.val[1], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s64 (__val.val[2], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u64_x3 (uint64_t * __a, uint64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint64x2x3_t __temp;
-  __temp.val[0] = vcombine_u64 (__val.val[0], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u64 (__val.val[1], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u64 (__val.val[2], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f64_x3 (float64_t * __a, float64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float64x2x3_t __temp;
-  __temp.val[0] = vcombine_f64 (__val.val[0], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f64 (__val.val[1], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f64 (__val.val[2], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st1x3df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s8_x3 (int8_t * __a, int8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int8x16x3_t __temp;
-  __temp.val[0] = vcombine_s8 (__val.val[0], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s8 (__val.val[1], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s8 (__val.val[2], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p8_x3 (poly8_t * __a, poly8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly8x16x3_t __temp;
-  __temp.val[0] = vcombine_p8 (__val.val[0], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p8 (__val.val[1], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p8 (__val.val[2], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s16_x3 (int16_t * __a, int16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int16x8x3_t __temp;
-  __temp.val[0] = vcombine_s16 (__val.val[0], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s16 (__val.val[1], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s16 (__val.val[2], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p16_x3 (poly16_t * __a, poly16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly16x8x3_t __temp;
-  __temp.val[0] = vcombine_p16 (__val.val[0], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p16 (__val.val[1], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p16 (__val.val[2], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_s32_x3 (int32_t * __a, int32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int32x4x3_t __temp;
-  __temp.val[0] = vcombine_s32 (__val.val[0], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s32 (__val.val[1], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s32 (__val.val[2], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x3v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u8_x3 (uint8_t * __a, uint8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint8x16x3_t __temp;
-  __temp.val[0] = vcombine_u8 (__val.val[0], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u8 (__val.val[1], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u8 (__val.val[2], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u16_x3 (uint16_t * __a, uint16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint16x8x3_t __temp;
-  __temp.val[0] = vcombine_u16 (__val.val[0], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u16 (__val.val[1], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u16 (__val.val[2], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_u32_x3 (uint32_t * __a, uint32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint32x4x3_t __temp;
-  __temp.val[0] = vcombine_u32 (__val.val[0], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u32 (__val.val[1], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u32 (__val.val[2], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x3v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f16_x3 (float16_t * __a, float16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float16x8x3_t __temp;
-  __temp.val[0] = vcombine_f16 (__val.val[0], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f16 (__val.val[1], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f16 (__val.val[2], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v4hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st1x3v4hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_f32_x3 (float32_t * __a, float32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float32x4x3_t __temp;
-  __temp.val[0] = vcombine_f32 (__val.val[0], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f32 (__val.val[1], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f32 (__val.val[2], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v2sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st1x3v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_p64_x3 (poly64_t * __a, poly64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly64x2x3_t __temp;
-  __temp.val[0] = vcombine_p64 (__val.val[0], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p64 (__val.val[1], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p64 (__val.val[2], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s8_x3 (int8_t * __a, int8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p8_x3 (poly8_t * __a, poly8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s16_x3 (int16_t * __a, int16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p16_x3 (poly16_t * __a, poly16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s32_x3 (int32_t * __a, int32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x3v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_s64_x3 (int64_t * __a, int64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3v2di ((__builtin_aarch64_simd_di *) __a,
+          (int64x2x3_t) __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u8_x3 (uint8_t * __a, uint8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st1x3v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u16_x3 (uint16_t * __a, uint16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st1x3v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u32_x3 (uint32_t * __a, uint32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st1x3v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_u64_x3 (uint64_t * __a, uint64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f16_x3 (float16_t * __a, float16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v8hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st1x3v8hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f32_x3 (float32_t * __a, float32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v4sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st1x3v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_f64_x3 (float64_t * __a, float64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st1x3v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_p64_x3 (poly64_t * __a, poly64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st1x3v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_s8_x4 (int8_t * __a, int8x8x4_t val)
+vst1_s8_x4 (int8_t * __a, int8x8x4_t __val)
 {
-  union { int8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_s8_x4 (int8_t * __a, int8x16x4_t val)
+vst1q_s8_x4 (int8_t * __a, int8x16x4_t __val)
 {
-  union { int8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v16qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_s16_x4 (int16_t * __a, int16x4x4_t val)
+vst1_s16_x4 (int16_t * __a, int16x4x4_t __val)
 {
-  union { int16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_s16_x4 (int16_t * __a, int16x8x4_t val)
+vst1q_s16_x4 (int16_t * __a, int16x8x4_t __val)
 {
-  union { int16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_s32_x4 (int32_t * __a, int32x2x4_t val)
+vst1_s32_x4 (int32_t * __a, int32x2x4_t __val)
 {
-  union { int32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2si ((__builtin_aarch64_simd_si *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_s32_x4 (int32_t * __a, int32x4x4_t val)
+vst1q_s32_x4 (int32_t * __a, int32x4x4_t __val)
 {
-  union { int32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4si ((__builtin_aarch64_simd_si *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_u8_x4 (uint8_t * __a, uint8x8x4_t val)
+vst1_u8_x4 (uint8_t * __a, uint8x8x4_t __val)
 {
-  union { uint8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_u8_x4 (uint8_t * __a, uint8x16x4_t val)
+vst1q_u8_x4 (uint8_t * __a, uint8x16x4_t __val)
 {
-  union { uint8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v16qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_u16_x4 (uint16_t * __a, uint16x4x4_t val)
+vst1_u16_x4 (uint16_t * __a, uint16x4x4_t __val)
 {
-  union { uint16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_u16_x4 (uint16_t * __a, uint16x8x4_t val)
+vst1q_u16_x4 (uint16_t * __a, uint16x8x4_t __val)
 {
-  union { uint16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_u32_x4 (uint32_t * __a, uint32x2x4_t val)
+vst1_u32_x4 (uint32_t * __a, uint32x2x4_t __val)
 {
-  union { uint32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2si ((__builtin_aarch64_simd_si *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_u32_x4 (uint32_t * __a, uint32x4x4_t val)
+vst1q_u32_x4 (uint32_t * __a, uint32x4x4_t __val)
 {
-  union { uint32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4si ((__builtin_aarch64_simd_si *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_f16_x4 (float16_t * __a, float16x4x4_t val)
+vst1_f16_x4 (float16_t * __a, float16x4x4_t __val)
 {
-  union { float16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4hf ((__builtin_aarch64_simd_hf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_f16_x4 (float16_t * __a, float16x8x4_t val)
+vst1q_f16_x4 (float16_t * __a, float16x8x4_t __val)
 {
-  union { float16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8hf ((__builtin_aarch64_simd_hf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_f32_x4 (float32_t * __a, float32x2x4_t val)
+vst1_f32_x4 (float32_t * __a, float32x2x4_t __val)
 {
-  union { float32x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2sf ((__builtin_aarch64_simd_sf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_f32_x4 (float32_t * __a, float32x4x4_t val)
+vst1q_f32_x4 (float32_t * __a, float32x4x4_t __val)
 {
-  union { float32x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4sf ((__builtin_aarch64_simd_sf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_p8_x4 (poly8_t * __a, poly8x8x4_t val)
+vst1_p8_x4 (poly8_t * __a, poly8x8x4_t __val)
 {
-  union { poly8x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_p8_x4 (poly8_t * __a, poly8x16x4_t val)
+vst1q_p8_x4 (poly8_t * __a, poly8x16x4_t __val)
 {
-  union { poly8x16x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v16qi ((__builtin_aarch64_simd_qi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_p16_x4 (poly16_t * __a, poly16x4x4_t val)
+vst1_p16_x4 (poly16_t * __a, poly16x4x4_t __val)
 {
-  union { poly16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_p16_x4 (poly16_t * __a, poly16x8x4_t val)
+vst1q_p16_x4 (poly16_t * __a, poly16x8x4_t __val)
 {
-  union { poly16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8hi ((__builtin_aarch64_simd_hi *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_s64_x4 (int64_t * __a, int64x1x4_t val)
+vst1_s64_x4 (int64_t * __a, int64x1x4_t __val)
 {
-  union { int64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_u64_x4 (uint64_t * __a, uint64x1x4_t val)
+vst1_u64_x4 (uint64_t * __a, uint64x1x4_t __val)
 {
-  union { uint64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_p64_x4 (poly64_t * __a, poly64x1x4_t val)
+vst1_p64_x4 (poly64_t * __a, poly64x1x4_t __val)
 {
-  union { poly64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_s64_x4 (int64_t * __a, int64x2x4_t val)
+vst1q_s64_x4 (int64_t * __a, int64x2x4_t __val)
 {
-  union { int64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_u64_x4 (uint64_t * __a, uint64x2x4_t val)
+vst1q_u64_x4 (uint64_t * __a, uint64x2x4_t __val)
 {
-  union { uint64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_p64_x4 (poly64_t * __a, poly64x2x4_t val)
+vst1q_p64_x4 (poly64_t * __a, poly64x2x4_t __val)
 {
-  union { poly64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2di ((__builtin_aarch64_simd_di *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_f64_x4 (float64_t * __a, float64x1x4_t val)
+vst1_f64_x4 (float64_t * __a, float64x1x4_t __val)
 {
-  union { float64x1x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4df ((__builtin_aarch64_simd_df *) __a, __u.__o);
+  __builtin_aarch64_st1x4df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_f64_x4 (float64_t * __a, float64x2x4_t val)
+vst1q_f64_x4 (float64_t * __a, float64x2x4_t __val)
 {
-  union { float64x2x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v2df ((__builtin_aarch64_simd_df *) __a, __u.__o);
+  __builtin_aarch64_st1x4v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_s64 (int64_t * __a, int64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int64x2x2_t __temp;
-  __temp.val[0] = vcombine_s64 (__val.val[0], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s64 (__val.val[1], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_u64 (uint64_t * __a, uint64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint64x2x2_t __temp;
-  __temp.val[0] = vcombine_u64 (__val.val[0], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u64 (__val.val[1], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_f64 (float64_t * __a, float64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float64x2x2_t __temp;
-  __temp.val[0] = vcombine_f64 (__val.val[0], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f64 (__val.val[1], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_s8 (int8_t * __a, int8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int8x16x2_t __temp;
-  __temp.val[0] = vcombine_s8 (__val.val[0], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s8 (__val.val[1], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_p8 (poly8_t * __a, poly8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly8x16x2_t __temp;
-  __temp.val[0] = vcombine_p8 (__val.val[0], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p8 (__val.val[1], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_s16 (int16_t * __a, int16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int16x8x2_t __temp;
-  __temp.val[0] = vcombine_s16 (__val.val[0], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s16 (__val.val[1], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_p16 (poly16_t * __a, poly16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly16x8x2_t __temp;
-  __temp.val[0] = vcombine_p16 (__val.val[0], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p16 (__val.val[1], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_s32 (int32_t * __a, int32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  int32x4x2_t __temp;
-  __temp.val[0] = vcombine_s32 (__val.val[0], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s32 (__val.val[1], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st2v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_u8 (uint8_t * __a, uint8x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint8x16x2_t __temp;
-  __temp.val[0] = vcombine_u8 (__val.val[0], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u8 (__val.val[1], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_u16 (uint16_t * __a, uint16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint16x8x2_t __temp;
-  __temp.val[0] = vcombine_u16 (__val.val[0], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u16 (__val.val[1], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_u32 (uint32_t * __a, uint32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  uint32x4x2_t __temp;
-  __temp.val[0] = vcombine_u32 (__val.val[0], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u32 (__val.val[1], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st2v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_f16 (float16_t * __a, float16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float16x8x2_t __temp;
-  __temp.val[0] = vcombine_f16 (__val.val[0], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f16 (__val.val[1], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __temp.val[1], 1);
-  __builtin_aarch64_st2v4hf (__a, __o);
+  __builtin_aarch64_st2v4hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_f32 (float32_t * __a, float32x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  float32x4x2_t __temp;
-  __temp.val[0] = vcombine_f32 (__val.val[0], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f32 (__val.val[1], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __temp.val[1], 1);
-  __builtin_aarch64_st2v2sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st2v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_p64 (poly64_t * __a, poly64x1x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  poly64x2x2_t __temp;
-  __temp.val[0] = vcombine_p64 (__val.val[0], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p64 (__val.val[1], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[1], 1);
-  __builtin_aarch64_st2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_s8 (int8_t * __a, int8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_p8 (poly8_t * __a, poly8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_s16 (int16_t * __a, int16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_p16 (poly16_t * __a, poly16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_s32 (int32_t * __a, int32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st2v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st2v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_s64 (int64_t * __a, int64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2v2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_u8 (uint8_t * __a, uint8x16x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __builtin_aarch64_st2v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st2v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_u16 (uint16_t * __a, uint16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __builtin_aarch64_st2v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st2v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_u32 (uint32_t * __a, uint32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st2v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st2v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_u64 (uint64_t * __a, uint64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_f16 (float16_t * __a, float16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8hf (__o, __val.val[1], 1);
-  __builtin_aarch64_st2v8hf (__a, __o);
+  __builtin_aarch64_st2v8hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_f32 (float32_t * __a, float32x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv4sf (__o, (float32x4_t) __val.val[1], 1);
-  __builtin_aarch64_st2v4sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st2v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_f64 (float64_t * __a, float64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2df (__o, (float64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st2v2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st2v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_p64 (poly64_t * __a, poly64x2x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv2di_ssps (__o,
-            (poly64x2_t) __val.val[1], 1);
-  __builtin_aarch64_st2v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st2v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_s64 (int64_t * __a, int64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int64x2x3_t __temp;
-  __temp.val[0] = vcombine_s64 (__val.val[0], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s64 (__val.val[1], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s64 (__val.val[2], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_u64 (uint64_t * __a, uint64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint64x2x3_t __temp;
-  __temp.val[0] = vcombine_u64 (__val.val[0], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u64 (__val.val[1], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u64 (__val.val[2], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_f64 (float64_t * __a, float64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float64x2x3_t __temp;
-  __temp.val[0] = vcombine_f64 (__val.val[0], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f64 (__val.val[1], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f64 (__val.val[2], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st3df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st3df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_s8 (int8_t * __a, int8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int8x16x3_t __temp;
-  __temp.val[0] = vcombine_s8 (__val.val[0], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s8 (__val.val[1], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s8 (__val.val[2], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_p8 (poly8_t * __a, poly8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly8x16x3_t __temp;
-  __temp.val[0] = vcombine_p8 (__val.val[0], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p8 (__val.val[1], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p8 (__val.val[2], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_s16 (int16_t * __a, int16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int16x8x3_t __temp;
-  __temp.val[0] = vcombine_s16 (__val.val[0], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s16 (__val.val[1], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s16 (__val.val[2], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_p16 (poly16_t * __a, poly16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly16x8x3_t __temp;
-  __temp.val[0] = vcombine_p16 (__val.val[0], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p16 (__val.val[1], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p16 (__val.val[2], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_s32 (int32_t * __a, int32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  int32x4x3_t __temp;
-  __temp.val[0] = vcombine_s32 (__val.val[0], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s32 (__val.val[1], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s32 (__val.val[2], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st3v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_u8 (uint8_t * __a, uint8x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint8x16x3_t __temp;
-  __temp.val[0] = vcombine_u8 (__val.val[0], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u8 (__val.val[1], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u8 (__val.val[2], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_u16 (uint16_t * __a, uint16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint16x8x3_t __temp;
-  __temp.val[0] = vcombine_u16 (__val.val[0], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u16 (__val.val[1], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u16 (__val.val[2], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_u32 (uint32_t * __a, uint32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  uint32x4x3_t __temp;
-  __temp.val[0] = vcombine_u32 (__val.val[0], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u32 (__val.val[1], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u32 (__val.val[2], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st3v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_f16 (float16_t * __a, float16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float16x8x3_t __temp;
-  __temp.val[0] = vcombine_f16 (__val.val[0], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f16 (__val.val[1], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f16 (__val.val[2], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v4hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st3v4hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_f32 (float32_t * __a, float32x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  float32x4x3_t __temp;
-  __temp.val[0] = vcombine_f32 (__val.val[0], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f32 (__val.val[1], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f32 (__val.val[2], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v2sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st3v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_p64 (poly64_t * __a, poly64x1x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  poly64x2x3_t __temp;
-  __temp.val[0] = vcombine_p64 (__val.val[0], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p64 (__val.val[1], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p64 (__val.val[2], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __temp.val[2], 2);
-  __builtin_aarch64_st3di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_s8 (int8_t * __a, int8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_p8 (poly8_t * __a, poly8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_s16 (int16_t * __a, int16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_p16 (poly16_t * __a, poly16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_s32 (int32_t * __a, int32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st3v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st3v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_s64 (int64_t * __a, int64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3v2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_u8 (uint8_t * __a, uint8x16x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __builtin_aarch64_st3v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st3v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_u16 (uint16_t * __a, uint16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st3v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st3v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_u32 (uint32_t * __a, uint32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4si (__o, (int32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st3v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st3v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_u64 (uint64_t * __a, uint64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di (__o, (int64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_f16 (float16_t * __a, float16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8hf (__o, (float16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st3v8hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st3v8hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_f32 (float32_t * __a, float32x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv4sf (__o, (float32x4_t) __val.val[2], 2);
-  __builtin_aarch64_st3v4sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st3v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_f64 (float64_t * __a, float64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2df (__o, (float64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st3v2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st3v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_p64 (poly64_t * __a, poly64x2x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv2di_ssps (__o,
-            (poly64x2_t) __val.val[2], 2);
-  __builtin_aarch64_st3v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st3v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_s64 (int64_t * __a, int64x1x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  int64x2x4_t __temp;
-  __temp.val[0] = vcombine_s64 (__val.val[0], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s64 (__val.val[1], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s64 (__val.val[2], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __temp.val[3] = vcombine_s64 (__val.val[3], vcreate_s64 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[3], 3);
-  __builtin_aarch64_st4di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_u64 (uint64_t * __a, uint64x1x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  uint64x2x4_t __temp;
-  __temp.val[0] = vcombine_u64 (__val.val[0], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u64 (__val.val[1], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u64 (__val.val[2], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_u64 (__val.val[3], vcreate_u64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __temp.val[3], 3);
-  __builtin_aarch64_st4di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_f64 (float64_t * __a, float64x1x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  float64x2x4_t __temp;
-  __temp.val[0] = vcombine_f64 (__val.val[0], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f64 (__val.val[1], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f64 (__val.val[2], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_f64 (__val.val[3], vcreate_f64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __temp.val[3], 3);
-  __builtin_aarch64_st4df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st4df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_s8 (int8_t * __a, int8x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  int8x16x4_t __temp;
-  __temp.val[0] = vcombine_s8 (__val.val[0], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s8 (__val.val[1], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s8 (__val.val[2], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __temp.val[3] = vcombine_s8 (__val.val[3], vcreate_s8 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v8qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_p8 (poly8_t * __a, poly8x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  poly8x16x4_t __temp;
-  __temp.val[0] = vcombine_p8 (__val.val[0], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p8 (__val.val[1], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p8 (__val.val[2], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_p8 (__val.val[3], vcreate_p8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v8qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_s16 (int16_t * __a, int16x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  int16x8x4_t __temp;
-  __temp.val[0] = vcombine_s16 (__val.val[0], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s16 (__val.val[1], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s16 (__val.val[2], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __temp.val[3] = vcombine_s16 (__val.val[3], vcreate_s16 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v4hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_p16 (poly16_t * __a, poly16x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  poly16x8x4_t __temp;
-  __temp.val[0] = vcombine_p16 (__val.val[0], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p16 (__val.val[1], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p16 (__val.val[2], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_p16 (__val.val[3], vcreate_p16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v4hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_s32 (int32_t * __a, int32x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  int32x4x4_t __temp;
-  __temp.val[0] = vcombine_s32 (__val.val[0], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[1] = vcombine_s32 (__val.val[1], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[2] = vcombine_s32 (__val.val[2], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __temp.val[3] = vcombine_s32 (__val.val[3], vcreate_s32 (__AARCH64_INT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st4v2si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_u8 (uint8_t * __a, uint8x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  uint8x16x4_t __temp;
-  __temp.val[0] = vcombine_u8 (__val.val[0], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u8 (__val.val[1], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u8 (__val.val[2], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_u8 (__val.val[3], vcreate_u8 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v8qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v8qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_u16 (uint16_t * __a, uint16x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  uint16x8x4_t __temp;
-  __temp.val[0] = vcombine_u16 (__val.val[0], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u16 (__val.val[1], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u16 (__val.val[2], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_u16 (__val.val[3], vcreate_u16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v4hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v4hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_u32 (uint32_t * __a, uint32x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  uint32x4x4_t __temp;
-  __temp.val[0] = vcombine_u32 (__val.val[0], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_u32 (__val.val[1], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_u32 (__val.val[2], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_u32 (__val.val[3], vcreate_u32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v2si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st4v2si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_f16 (float16_t * __a, float16x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  float16x8x4_t __temp;
-  __temp.val[0] = vcombine_f16 (__val.val[0], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f16 (__val.val[1], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f16 (__val.val[2], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_f16 (__val.val[3], vcreate_f16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v4hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st4v4hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_f32 (float32_t * __a, float32x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  float32x4x4_t __temp;
-  __temp.val[0] = vcombine_f32 (__val.val[0], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_f32 (__val.val[1], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_f32 (__val.val[2], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_f32 (__val.val[3], vcreate_f32 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v2sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st4v2sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_p64 (poly64_t * __a, poly64x1x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  poly64x2x4_t __temp;
-  __temp.val[0] = vcombine_p64 (__val.val[0], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_p64 (__val.val[1], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_p64 (__val.val[2], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_p64 (__val.val[3], vcreate_p64 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __temp.val[3], 3);
-  __builtin_aarch64_st4di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_s8 (int8_t * __a, int8x16x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[3], 3);
-  __builtin_aarch64_st4v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v16qi ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_p8 (poly8_t * __a, poly8x16x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[3], 3);
-  __builtin_aarch64_st4v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v16qi_sp ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_s16 (int16_t * __a, int16x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[3], 3);
-  __builtin_aarch64_st4v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v8hi ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_p16 (poly16_t * __a, poly16x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[3], 3);
-  __builtin_aarch64_st4v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v8hi_sp ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_s32 (int32_t * __a, int32x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[3], 3);
-  __builtin_aarch64_st4v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st4v4si ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_s64 (int64_t * __a, int64x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[3], 3);
-  __builtin_aarch64_st4v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4v2di ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_u8 (uint8_t * __a, uint8x16x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv16qi (__o, (int8x16_t) __val.val[3], 3);
-  __builtin_aarch64_st4v16qi ((__builtin_aarch64_simd_qi *) __a, __o);
+  __builtin_aarch64_st4v16qi_su ((__builtin_aarch64_simd_qi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_u16 (uint16_t * __a, uint16x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hi (__o, (int16x8_t) __val.val[3], 3);
-  __builtin_aarch64_st4v8hi ((__builtin_aarch64_simd_hi *) __a, __o);
+  __builtin_aarch64_st4v8hi_su ((__builtin_aarch64_simd_hi *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_u32 (uint32_t * __a, uint32x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4si (__o, (int32x4_t) __val.val[3], 3);
-  __builtin_aarch64_st4v4si ((__builtin_aarch64_simd_si *) __a, __o);
+  __builtin_aarch64_st4v4si_su ((__builtin_aarch64_simd_si *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_u64 (uint64_t * __a, uint64x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di (__o, (int64x2_t) __val.val[3], 3);
-  __builtin_aarch64_st4v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4v2di_su ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_f16 (float16_t * __a, float16x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8hf (__o, (float16x8_t) __val.val[3], 3);
-  __builtin_aarch64_st4v8hf ((__builtin_aarch64_simd_hf *) __a, __o);
+  __builtin_aarch64_st4v8hf ((__builtin_aarch64_simd_hf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_f32 (float32_t * __a, float32x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv4sf (__o, (float32x4_t) __val.val[3], 3);
-  __builtin_aarch64_st4v4sf ((__builtin_aarch64_simd_sf *) __a, __o);
+  __builtin_aarch64_st4v4sf ((__builtin_aarch64_simd_sf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_f64 (float64_t * __a, float64x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2df (__o, (float64x2_t) __val.val[3], 3);
-  __builtin_aarch64_st4v2df ((__builtin_aarch64_simd_df *) __a, __o);
+  __builtin_aarch64_st4v2df ((__builtin_aarch64_simd_df *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_p64 (poly64_t * __a, poly64x2x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv2di_ssps (__o,
-            (poly64x2_t) __val.val[3], 3);
-  __builtin_aarch64_st4v2di ((__builtin_aarch64_simd_di *) __a, __o);
+  __builtin_aarch64_st4v2di_sp ((__builtin_aarch64_simd_di *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -24413,51 +22189,28 @@ __extension__ extern __inline int8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx4_s8 (int8x8_t __r, int8x8x4_t __tab, int8x8_t __idx)
 {
-  int8x8_t __result;
   int8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_s8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_s8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = __builtin_aarch64_tbx4v8qi (__r, __o, __idx);
-  return __result;
+  return __builtin_aarch64_qtbx2v8qi (__r, __temp, __idx);
 }
 __extension__ extern __inline uint8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx4_u8 (uint8x8_t __r, uint8x8x4_t __tab, uint8x8_t __idx)
 {
-  uint8x8_t __result;
   uint8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_u8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_u8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (uint8x8_t)__builtin_aarch64_tbx4v8qi ((int8x8_t)__r, __o,
-        (int8x8_t)__idx);
-  return __result;
+  return __builtin_aarch64_qtbx2v8qi_uuuu (__r, __temp, __idx);
 }
 __extension__ extern __inline poly8x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vtbx4_p8 (poly8x8_t __r, poly8x8x4_t __tab, uint8x8_t __idx)
 {
-  poly8x8_t __result;
   poly8x16x2_t __temp;
-  __builtin_aarch64_simd_oi __o;
   __temp.val[0] = vcombine_p8 (__tab.val[0], __tab.val[1]);
   __temp.val[1] = vcombine_p8 (__tab.val[2], __tab.val[3]);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv16qi (__o,
-        (int8x16_t) __temp.val[1], 1);
-  __result = (poly8x8_t)__builtin_aarch64_tbx4v8qi ((int8x8_t)__r, __o,
-        (int8x8_t)__idx);
-  return __result;
+  return __builtin_aarch64_qtbx2v8qi_pppu (__r, __temp, __idx);
 }
 __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -26547,13 +24300,13 @@ __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndn_f16 (float16x4_t __a)
 {
-  return __builtin_aarch64_frintnv4hf (__a);
+  return __builtin_aarch64_roundevenv4hf (__a);
 }
 __extension__ extern __inline float16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vrndnq_f16 (float16x8_t __a)
 {
-  return __builtin_aarch64_frintnv8hf (__a);
+  return __builtin_aarch64_roundevenv8hf (__a);
 }
 __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -26799,13 +24552,13 @@ __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmax_f16 (float16x4_t __a, float16x4_t __b)
 {
-  return __builtin_aarch64_smax_nanv4hf (__a, __b);
+  return __builtin_aarch64_fmax_nanv4hf (__a, __b);
 }
 __extension__ extern __inline float16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmaxq_f16 (float16x8_t __a, float16x8_t __b)
 {
-  return __builtin_aarch64_smax_nanv8hf (__a, __b);
+  return __builtin_aarch64_fmax_nanv8hf (__a, __b);
 }
 __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -26823,13 +24576,13 @@ __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vmin_f16 (float16x4_t __a, float16x4_t __b)
 {
-  return __builtin_aarch64_smin_nanv4hf (__a, __b);
+  return __builtin_aarch64_fmin_nanv4hf (__a, __b);
 }
 __extension__ extern __inline float16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vminq_f16 (float16x8_t __a, float16x8_t __b)
 {
-  return __builtin_aarch64_smin_nanv8hf (__a, __b);
+  return __builtin_aarch64_fmin_nanv8hf (__a, __b);
 }
 __extension__ extern __inline float16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -27246,25 +24999,25 @@ __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vdot_u32 (uint32x2_t __r, uint8x8_t __a, uint8x8_t __b)
 {
-  return __builtin_aarch64_udotv8qi_uuuu (__r, __a, __b);
+  return __builtin_aarch64_udot_prodv8qi_uuuu (__a, __b, __r);
 }
 __extension__ extern __inline uint32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vdotq_u32 (uint32x4_t __r, uint8x16_t __a, uint8x16_t __b)
 {
-  return __builtin_aarch64_udotv16qi_uuuu (__r, __a, __b);
+  return __builtin_aarch64_udot_prodv16qi_uuuu (__a, __b, __r);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vdot_s32 (int32x2_t __r, int8x8_t __a, int8x8_t __b)
 {
-  return __builtin_aarch64_sdotv8qi (__r, __a, __b);
+  return __builtin_aarch64_sdot_prodv8qi (__a, __b, __r);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vdotq_s32 (int32x4_t __r, int8x16_t __a, int8x16_t __b)
 {
-  return __builtin_aarch64_sdotv16qi (__r, __a, __b);
+  return __builtin_aarch64_sdot_prodv16qi (__a, __b, __r);
 }
 __extension__ extern __inline uint32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -28220,7 +25973,7 @@ __extension__ extern __inline bfloat16x8_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vcombine_bf16 (bfloat16x4_t __a, bfloat16x4_t __b)
 {
-  return (bfloat16x8_t)__builtin_aarch64_combinev4bf (__a, __b);
+  return __builtin_aarch64_combinev4bf (__a, __b);
 }
 __extension__ extern __inline bfloat16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -28286,65 +26039,38 @@ __extension__ extern __inline bfloat16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_bf16_x2 (const bfloat16_t *__a)
 {
-  bfloat16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_bf16_x2 (const bfloat16_t *__a)
 {
-  bfloat16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld1x2v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld1x2v8bf (
+    (const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_bf16_x3 (const bfloat16_t *__a)
 {
-  bfloat16x4x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v4bf ((const __builtin_aarch64_simd_bf *) __a);
-  __i.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 0);
-  __i.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 1);
-  __i.val[2] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_bf16_x3 (const bfloat16_t *__a)
 {
-  bfloat16x8x3_t __i;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld1x3v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  __i.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 0);
-  __i.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 1);
-  __i.val[2] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 2);
-  return __i;
+  return __builtin_aarch64_ld1x3v8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1_bf16_x4 (const bfloat16_t *__a)
 {
-  union { bfloat16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v4bf ((const __builtin_aarch64_simd_bf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld1q_bf16_x4 (const bfloat16_t *__a)
 {
-  union { bfloat16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __au;
-  __au.__o
-    = __builtin_aarch64_ld1x4v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  return __au.__i;
+  return __builtin_aarch64_ld1x4v8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -28374,145 +26100,73 @@ __extension__ extern __inline bfloat16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v4bf (__a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v4bf (__a);
 }
 __extension__ extern __inline bfloat16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2v8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregoiv4bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld2q_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x2_t ret;
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_ld2rv8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregoiv8bf (__o, 1);
-  return ret;
+  return __builtin_aarch64_ld2rv8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 1);
-  ret.val[2] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 1);
-  ret.val[2] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3v8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 1);
-  ret.val[2] = (bfloat16x4_t) __builtin_aarch64_get_dregciv4bf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x3_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld3q_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x3_t ret;
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_ld3rv8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 1);
-  ret.val[2] = (bfloat16x8_t) __builtin_aarch64_get_qregciv8bf (__o, 2);
-  return ret;
+  return __builtin_aarch64_ld3rv8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 1);
-  ret.val[2] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 2);
-  ret.val[3] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4v8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 1);
-  ret.val[2] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 2);
-  ret.val[3] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4v8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x4x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x4x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv4bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 0);
-  ret.val[1] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 1);
-  ret.val[2] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 2);
-  ret.val[3] = (bfloat16x4_t) __builtin_aarch64_get_dregxiv4bf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv4bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline bfloat16x8x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vld4q_dup_bf16 (const bfloat16_t * __a)
 {
-  bfloat16x8x4_t ret;
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_ld4rv8bf ((const __builtin_aarch64_simd_bf *) __a);
-  ret.val[0] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 0);
-  ret.val[1] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 1);
-  ret.val[2] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 2);
-  ret.val[3] = (bfloat16x8_t) __builtin_aarch64_get_qregxiv8bf (__o, 3);
-  return ret;
+  return __builtin_aarch64_ld4rv8bf ((const __builtin_aarch64_simd_bf *) __a);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -28524,60 +26178,37 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_bf16_x2 (bfloat16_t * __a, bfloat16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  bfloat16x8x2_t __temp;
-  __temp.val[0] = vcombine_bf16 (__val.val[0], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_bf16 (__val.val[1], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __temp.val[1], 1);
-  __builtin_aarch64_st1x2v4bf (__a, __o);
+  __builtin_aarch64_st1x2v4bf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_bf16_x2 (bfloat16_t * __a, bfloat16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __val.val[1], 1);
-  __builtin_aarch64_st1x2v8bf (__a, __o);
+  __builtin_aarch64_st1x2v8bf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1_bf16_x3 (bfloat16_t * __a, bfloat16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  bfloat16x8x3_t __temp;
-  __temp.val[0] = vcombine_bf16 (__val.val[0], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_bf16 (__val.val[1], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_bf16 (__val.val[2], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st1x3v4bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st1x3v4bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst1q_bf16_x3 (bfloat16_t * __a, bfloat16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st1x3v8bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st1x3v8bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1_bf16_x4 (bfloat16_t * __a, bfloat16x4x4_t val)
+vst1_bf16_x4 (bfloat16_t * __a, bfloat16x4x4_t __val)
 {
-  union { bfloat16x4x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v4bf ((__builtin_aarch64_simd_bf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v4bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
-vst1q_bf16_x4 (bfloat16_t * __a, bfloat16x8x4_t val)
+vst1q_bf16_x4 (bfloat16_t * __a, bfloat16x8x4_t __val)
 {
-  union { bfloat16x8x4_t __i; __builtin_aarch64_simd_xi __o; } __u = { val };
-  __builtin_aarch64_st1x4v8bf ((__builtin_aarch64_simd_bf *) __a, __u.__o);
+  __builtin_aarch64_st1x4v8bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -28601,73 +26232,37 @@ __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2_bf16 (bfloat16_t * __a, bfloat16x4x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  bfloat16x8x2_t __temp;
-  __temp.val[0] = vcombine_bf16 (__val.val[0], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_bf16 (__val.val[1], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __temp.val[1], 1);
-  __builtin_aarch64_st2v4bf (__a, __o);
+  __builtin_aarch64_st2v4bf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst2q_bf16 (bfloat16_t * __a, bfloat16x8x2_t __val)
 {
-  __builtin_aarch64_simd_oi __o;
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregoiv8bf (__o, __val.val[1], 1);
-  __builtin_aarch64_st2v8bf (__a, __o);
+  __builtin_aarch64_st2v8bf (__a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3_bf16 (bfloat16_t * __a, bfloat16x4x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  bfloat16x8x3_t __temp;
-  __temp.val[0] = vcombine_bf16 (__val.val[0], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_bf16 (__val.val[1], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_bf16 (__val.val[2], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __temp.val[2], 2);
-  __builtin_aarch64_st3v4bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st3v4bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst3q_bf16 (bfloat16_t * __a, bfloat16x8x3_t __val)
 {
-  __builtin_aarch64_simd_ci __o;
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregciv8bf (__o, (bfloat16x8_t) __val.val[2], 2);
-  __builtin_aarch64_st3v8bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st3v8bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4_bf16 (bfloat16_t * __a, bfloat16x4x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  bfloat16x8x4_t __temp;
-  __temp.val[0] = vcombine_bf16 (__val.val[0], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[1] = vcombine_bf16 (__val.val[1], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[2] = vcombine_bf16 (__val.val[2], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __temp.val[3] = vcombine_bf16 (__val.val[3], vcreate_bf16 (__AARCH64_UINT64_C (0)));
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __temp.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __temp.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __temp.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __temp.val[3], 3);
-  __builtin_aarch64_st4v4bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st4v4bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline void
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vst4q_bf16 (bfloat16_t * __a, bfloat16x8x4_t __val)
 {
-  __builtin_aarch64_simd_xi __o;
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __val.val[0], 0);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __val.val[1], 1);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __val.val[2], 2);
-  __o = __builtin_aarch64_set_qregxiv8bf (__o, (bfloat16x8_t) __val.val[3], 3);
-  __builtin_aarch64_st4v8bf ((__builtin_aarch64_simd_bf *) __a, __o);
+  __builtin_aarch64_st4v8bf ((__builtin_aarch64_simd_bf *) __a, __val);
 }
 __extension__ extern __inline bfloat16x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -29183,24 +26778,90 @@ vcopyq_laneq_bf16 (bfloat16x8_t __a, const int __lane1,
   return __aarch64_vset_lane_any (__aarch64_vget_lane_any (__b, __lane2),
       __a, __lane1);
 }
-__LD2_LANE_FUNC (bfloat16x4x2_t, bfloat16x4_t, bfloat16x8x2_t, bfloat16_t, v4bf,
-   v8bf, bf, bf16, bfloat16x8_t)
-__LD2Q_LANE_FUNC (bfloat16x8x2_t, bfloat16x8_t, bfloat16_t, v8bf, bf, bf16)
-__LD3_LANE_FUNC (bfloat16x4x3_t, bfloat16x4_t, bfloat16x8x3_t, bfloat16_t, v4bf,
-   v8bf, bf, bf16, bfloat16x8_t)
-__LD3Q_LANE_FUNC (bfloat16x8x3_t, bfloat16x8_t, bfloat16_t, v8bf, bf, bf16)
-__LD4_LANE_FUNC (bfloat16x4x4_t, bfloat16x4_t, bfloat16x8x4_t, bfloat16_t, v4bf,
-   v8bf, bf, bf16, bfloat16x8_t)
-__LD4Q_LANE_FUNC (bfloat16x8x4_t, bfloat16x8_t, bfloat16_t, v8bf, bf, bf16)
-__ST2_LANE_FUNC (bfloat16x4x2_t, bfloat16x8x2_t, bfloat16_t, v4bf, v8bf, bf,
-   bf16, bfloat16x8_t)
-__ST2Q_LANE_FUNC (bfloat16x8x2_t, bfloat16_t, v8bf, bf, bf16)
-__ST3_LANE_FUNC (bfloat16x4x3_t, bfloat16x8x3_t, bfloat16_t, v4bf, v8bf, bf,
-   bf16, bfloat16x8_t)
-__ST3Q_LANE_FUNC (bfloat16x8x3_t, bfloat16_t, v8bf, bf, bf16)
-__ST4_LANE_FUNC (bfloat16x4x4_t, bfloat16x8x4_t, bfloat16_t, v4bf, v8bf, bf,
-   bf16, bfloat16x8_t)
-__ST4Q_LANE_FUNC (bfloat16x8x4_t, bfloat16_t, v8bf, bf, bf16)
+__extension__ extern __inline bfloat16x4x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2_lane_bf16 (const bfloat16_t * __ptr, bfloat16x4x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev4bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline bfloat16x8x2_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld2q_lane_bf16 (const bfloat16_t * __ptr, bfloat16x8x2_t __b, const int __c)
+{
+  return __builtin_aarch64_ld2_lanev8bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline bfloat16x4x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3_lane_bf16 (const bfloat16_t * __ptr, bfloat16x4x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev4bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline bfloat16x8x3_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld3q_lane_bf16 (const bfloat16_t * __ptr, bfloat16x8x3_t __b, const int __c)
+{
+  return __builtin_aarch64_ld3_lanev8bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline bfloat16x4x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4_lane_bf16 (const bfloat16_t * __ptr, bfloat16x4x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev4bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline bfloat16x8x4_t
+__attribute__ ((__always_inline__, __gnu_inline__,__artificial__))
+vld4q_lane_bf16 (const bfloat16_t * __ptr, bfloat16x8x4_t __b, const int __c)
+{
+  return __builtin_aarch64_ld4_lanev8bf (
+   (__builtin_aarch64_simd_bf *) __ptr, __b, __c);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2_lane_bf16 (bfloat16_t *__ptr, bfloat16x4x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev4bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst2q_lane_bf16 (bfloat16_t *__ptr, bfloat16x8x2_t __val, const int __lane)
+{
+  __builtin_aarch64_st2_lanev8bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3_lane_bf16 (bfloat16_t *__ptr, bfloat16x4x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev4bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst3q_lane_bf16 (bfloat16_t *__ptr, bfloat16x8x3_t __val, const int __lane)
+{
+  __builtin_aarch64_st3_lanev8bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4_lane_bf16 (bfloat16_t *__ptr, bfloat16x4x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev4bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
+__extension__ extern __inline void
+__attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
+vst4q_lane_bf16 (bfloat16_t *__ptr, bfloat16x8x4_t __val, const int __lane)
+{
+  __builtin_aarch64_st4_lanev8bf ((__builtin_aarch64_simd_bf *) __ptr, __val,
+      __lane);
+}
 #pragma GCC pop_options
 #pragma GCC push_options
 #pragma GCC target ("arch=armv8.2-a+i8mm")
@@ -29208,13 +26869,13 @@ __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vusdot_s32 (int32x2_t __r, uint8x8_t __a, int8x8_t __b)
 {
-  return __builtin_aarch64_usdotv8qi_ssus (__r, __a, __b);
+  return __builtin_aarch64_usdot_prodv8qi_suss (__a, __b, __r);
 }
 __extension__ extern __inline int32x4_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
 vusdotq_s32 (int32x4_t __r, uint8x16_t __a, int8x16_t __b)
 {
-  return __builtin_aarch64_usdotv16qi_ssus (__r, __a, __b);
+  return __builtin_aarch64_usdot_prodv16qi_suss (__a, __b, __r);
 }
 __extension__ extern __inline int32x2_t
 __attribute__ ((__always_inline__, __gnu_inline__, __artificial__))
@@ -29385,17 +27046,5 @@ vaddq_p128 (poly128_t __a, poly128_t __b)
 #undef __aarch64_vdupq_laneq_u16
 #undef __aarch64_vdupq_laneq_u32
 #undef __aarch64_vdupq_laneq_u64
-#undef __LD2_LANE_FUNC
-#undef __LD2Q_LANE_FUNC
-#undef __LD3_LANE_FUNC
-#undef __LD3Q_LANE_FUNC
-#undef __LD4_LANE_FUNC
-#undef __LD4Q_LANE_FUNC
-#undef __ST2_LANE_FUNC
-#undef __ST2Q_LANE_FUNC
-#undef __ST3_LANE_FUNC
-#undef __ST3Q_LANE_FUNC
-#undef __ST4_LANE_FUNC
-#undef __ST4Q_LANE_FUNC
 #endif
 #endif
diff --git a/third_party/aarch64/arm_sve.internal.h b/third_party/aarch64/arm_sve.internal.h
index a4fb4000a..c224bdba3 100644
--- a/third_party/aarch64/arm_sve.internal.h
+++ b/third_party/aarch64/arm_sve.internal.h
@@ -5,6 +5,6 @@
 typedef __fp16 float16_t;
 typedef float float32_t;
 typedef double float64_t;
-#pragma GCC aarch64 "third_party/aarch64/arm_sve.internal.h"
+#pragma GCC aarch64 "arm_sve.h"
 #endif
 #endif
diff --git a/third_party/aarch64/upgrade.sh b/third_party/aarch64/upgrade.sh
index 04376e53a..1e4e6c5b7 100755
--- a/third_party/aarch64/upgrade.sh
+++ b/third_party/aarch64/upgrade.sh
@@ -13,17 +13,16 @@
 # 3. You should fix up the `#pragma GCC aarch64` things.
 #
 
-s=/opt/cross11portcosmo/lib/gcc/aarch64-linux-musl/11.2.0/include
+s=/opt/goodies/include
 d=third_party/aarch64
 
 FILES='
+acc_prof
 arm_acle
+arm_bf16
 arm_fp16
 arm_neon
-acc_prof
-arm_bf16
 arm_sve
-acc_prof
 openacc
 '
 
diff --git a/third_party/argon2/argon2.c b/third_party/argon2/argon2.c
index 323120a69..24f04d4d8 100644
--- a/third_party/argon2/argon2.c
+++ b/third_party/argon2/argon2.c
@@ -22,11 +22,6 @@
 #include "third_party/argon2/core.h"
 #include "third_party/argon2/encoding.h"
 
-asm(".ident\t\"\\n\\n\
-argon2 (CC0 or Apache2)\\n\
-Copyright 2016 Daniel Dinu, Dmitry Khovratovich\\n\
-Copyright 2016 Jean-Philippe Aumasson, Samuel Neves\"");
-
 /**
  * Function that gives the string representation of an argon2_type.
  * @param type The argon2_type that we want the string for
diff --git a/third_party/argon2/blake2b.c b/third_party/argon2/blake2b.c
index f62c988f2..26392c322 100644
--- a/third_party/argon2/blake2b.c
+++ b/third_party/argon2/blake2b.c
@@ -21,11 +21,6 @@
 #include "third_party/argon2/blake2.h"
 #include "third_party/argon2/core.h"
 
-asm(".ident\t\"\\n\\n\
-argon2 (CC0 or Apache2)\\n\
-Copyright 2016 Daniel Dinu, Dmitry Khovratovich\\n\
-Copyright 2016 Jean-Philippe Aumasson, Samuel Neves\"");
-
 /* Ensure param structs have not been wrongly padded */
 /* Poor man's static_assert */
 enum {
diff --git a/third_party/argon2/core.c b/third_party/argon2/core.c
index 340fc22c2..72f04ee50 100644
--- a/third_party/argon2/core.c
+++ b/third_party/argon2/core.c
@@ -20,10 +20,10 @@
 #include "third_party/argon2/blake2.h"
 #include "third_party/argon2/core.h"
 
-asm(".ident\t\"\\n\\n\
-argon2 (CC0 or Apache2)\\n\
-Copyright 2016 Daniel Dinu, Dmitry Khovratovich\\n\
-Copyright 2016 Jean-Philippe Aumasson, Samuel Neves\"");
+__notice(argon2_notice, "\
+argon2 (CC0 or Apache2)\n\
+Copyright 2016 Daniel Dinu, Dmitry Khovratovich\n\
+Copyright 2016 Jean-Philippe Aumasson, Samuel Neves");
 
 int FLAG_clear_internal_memory = 1;
 
diff --git a/third_party/argon2/encoding.c b/third_party/argon2/encoding.c
index 22f4707fc..22a50dcf3 100644
--- a/third_party/argon2/encoding.c
+++ b/third_party/argon2/encoding.c
@@ -21,11 +21,6 @@
 #include "libc/str/str.h"
 #include "third_party/argon2/core.h"
 
-asm(".ident\t\"\\n\\n\
-argon2 (CC0 or Apache2)\\n\
-Copyright 2016 Daniel Dinu, Dmitry Khovratovich\\n\
-Copyright 2016 Jean-Philippe Aumasson, Samuel Neves\"");
-
 /*
  * Example code for a decoder and encoder of "hash strings", with Argon2
  * parameters.
diff --git a/third_party/argon2/ref.c b/third_party/argon2/ref.c
index 1809aca03..f21c902de 100644
--- a/third_party/argon2/ref.c
+++ b/third_party/argon2/ref.c
@@ -22,11 +22,6 @@
 #include "third_party/argon2/blamka-round-ref.h"
 #include "third_party/argon2/core.h"
 
-asm(".ident\t\"\\n\\n\
-argon2 (CC0 or Apache2)\\n\
-Copyright 2016 Daniel Dinu, Dmitry Khovratovich\\n\
-Copyright 2016 Jean-Philippe Aumasson, Samuel Neves\"");
-
 /*
  * Argon2 reference source code package - reference C implementations
  *
diff --git a/third_party/awk/main.c b/third_party/awk/main.c
index d1ec5b755..07fba0946 100644
--- a/third_party/awk/main.c
+++ b/third_party/awk/main.c
@@ -39,28 +39,29 @@
 #include "libc/sysv/consts/sig.h"
 #include "third_party/awk/awk.h"
 
-asm(".ident\t\"\\n\\n\
-Copyright (C) Lucent Technologies 1997\\n\
-All Rights Reserved\\n\
-\\n\
-Permission to use, copy, modify, and distribute this software and\\n\
-its documentation for any purpose and without fee is hereby\\n\
-granted, provided that the above copyright notice appear in all\\n\
-copies and that both that the copyright notice and this\\n\
-permission notice and warranty disclaimer appear in supporting\\n\
-documentation, and that the name Lucent Technologies or any of\\n\
-its entities not be used in advertising or publicity pertaining\\n\
-to distribution of the software without specific, written prior\\n\
-permission.\\n\
-\\n\
-LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,\\n\
-INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.\\n\
-IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY\\n\
-SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES\\n\
-WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER\\n\
-IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,\\n\
-ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF\\n\
-THIS SOFTWARE.\"");
+__notice(awk_notice, "\
+The One True Awk\n\
+Copyright (C) Lucent Technologies 1997\n\
+All Rights Reserved\n\
+\n\
+Permission to use, copy, modify, and distribute this software and\n\
+its documentation for any purpose and without fee is hereby\n\
+granted, provided that the above copyright notice appear in all\n\
+copies and that both that the copyright notice and this\n\
+permission notice and warranty disclaimer appear in supporting\n\
+documentation, and that the name Lucent Technologies or any of\n\
+its entities not be used in advertising or publicity pertaining\n\
+to distribution of the software without specific, written prior\n\
+permission.\n\
+\n\
+LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,\n\
+INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.\n\
+IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY\n\
+SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES\n\
+WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER\n\
+IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,\n\
+ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF\n\
+THIS SOFTWARE.");
 
 const char	*version = "version 20220530";
 extern	int	nfields;
diff --git a/third_party/bzip2/crctable.c b/third_party/bzip2/crctable.c
index 2fee0a90a..2c21bd6a5 100644
--- a/third_party/bzip2/crctable.c
+++ b/third_party/bzip2/crctable.c
@@ -20,6 +20,7 @@
 
 uint32_t BZ2_crc32Table[256];
 
+__attribute__((__constructor__(10)))
 static textstartup void BZ2_crc32Table_init() {
   unsigned i, j, u;
   for (i = 0; i < 256; ++i) {
@@ -38,10 +39,6 @@ static textstartup void BZ2_crc32Table_init() {
   }
 }
 
-const void *const BZ2_crc32Table_ctor[] initarray = {
-    BZ2_crc32Table_init,
-};
-
 #else
 
 const uint32_t BZ2_crc32Table[256] = {
diff --git a/third_party/chibicc/chibicc.c b/third_party/chibicc/chibicc.c
index 3273b3077..8224433a5 100644
--- a/third_party/chibicc/chibicc.c
+++ b/third_party/chibicc/chibicc.c
@@ -9,11 +9,10 @@
 #include "libc/sysv/consts/sig.h"
 #include "libc/x/xasprintf.h"
 
-asm(".ident\t\"\\n\\n\
-chibicc (MIT/ISC License)\\n\
-Copyright 2019 Rui Ueyama\\n\
-Copyright 2020 Justine Alexandra Roberts Tunney\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(chibicc_notice, "\
+chibicc (MIT/ISC License)\n\
+Copyright 2019 Rui Ueyama\n\
+Copyright 2020 Justine Alexandra Roberts Tunney");
 
 typedef enum {
   FILE_NONE,
diff --git a/third_party/compiler_rt/comprt.S b/third_party/compiler_rt/comprt.S
index 84814f711..95060b658 100644
--- a/third_party/compiler_rt/comprt.S
+++ b/third_party/compiler_rt/comprt.S
@@ -7,12 +7,8 @@ huge_compiler_rt_license:
 	.endobj	huge_compiler_rt_license,globl,hidden
 	.previous
 
-.ident "\n
-compiler_rt (Licensed MIT)
-Copyright (c) 2009-2015 by the contributors listed in:
-github.com/llvm-mirror/compiler-rt/blob/master/CREDITS.TXT"
-
-.ident "\n
+	.section .notice,"aR",@progbits
+	.asciz	"\n\n\
 compiler_rt (Licensed \"University of Illinois/NCSA Open Source License\")
 Copyright (c) 2009-2018 by the contributors listed in:
 github.com/llvm-mirror/compiler-rt/blob/master/CREDITS.TXT
diff --git a/third_party/compiler_rt/cpu_model.h b/third_party/compiler_rt/cpu_model.h
deleted file mode 100644
index f5d2ba440..000000000
--- a/third_party/compiler_rt/cpu_model.h
+++ /dev/null
@@ -1,33 +0,0 @@
-//===-- cpu_model_common.c - Utilities for cpu model detection ----*- C -*-===//
-//
-// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
-// See https://llvm.org/LICENSE.txt for license information.
-// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-//
-//===----------------------------------------------------------------------===//
-//
-//  This file implements common utilities for runtime cpu model detection.
-//
-//===----------------------------------------------------------------------===//
-
-#ifndef COMPILER_RT_LIB_BUILTINS_CPU_MODEL_COMMON_H
-#define COMPILER_RT_LIB_BUILTINS_CPU_MODEL_COMMON_H
-
-#if __has_attribute(constructor)
-#if __GNUC__ >= 9
-// Ordinarily init priorities below 101 are disallowed as they are reserved for
-// the implementation. However, we are the implementation, so silence the
-// diagnostic, since it doesn't apply to us.
-#pragma GCC diagnostic ignored "-Wprio-ctor-dtor"
-#endif
-// We're choosing init priority 90 to force our constructors to run before any
-// constructors in the end user application (starting at priority 101). This
-// value matches the libgcc choice for the same functions.
-#define CONSTRUCTOR_ATTRIBUTE __attribute__((__constructor__(90)))
-#else
-// FIXME: For MSVC, we should make a function pointer global in .CRT$X?? so that
-// this runs during initialization.
-#define CONSTRUCTOR_ATTRIBUTE
-#endif
-
-#endif
diff --git a/third_party/compiler_rt/ilogbl.c b/third_party/compiler_rt/ilogbl.c
index 73ccc6352..09e4c9630 100644
--- a/third_party/compiler_rt/ilogbl.c
+++ b/third_party/compiler_rt/ilogbl.c
@@ -30,11 +30,7 @@
 #include "libc/tinymath/internal.h"
 #include "libc/tinymath/ldshape.internal.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /**
  * Returns log₂𝑥 exponent part of double.
diff --git a/third_party/compiler_rt/logbl.c b/third_party/compiler_rt/logbl.c
index 171b9aa44..27c639045 100644
--- a/third_party/compiler_rt/logbl.c
+++ b/third_party/compiler_rt/logbl.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/math.h"
 #if !(LDBL_MANT_DIG == 53 && LDBL_MAX_EXP == 1024)
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 long double logbl(long double x)
 {
diff --git a/third_party/dlmalloc/init.inc b/third_party/dlmalloc/init.inc
index b635d84bf..684e68995 100644
--- a/third_party/dlmalloc/init.inc
+++ b/third_party/dlmalloc/init.inc
@@ -8,7 +8,7 @@ static void dlmalloc_post_fork_child(void)  { (void)INITIAL_LOCK(&(gm)->mutex);
 #endif /* LOCK_AT_FORK */
 
 /* Initialize mparams */
-__attribute__((__constructor__)) int init_mparams(void) {
+__attribute__((__constructor__(50))) int init_mparams(void) {
 #ifdef NEED_GLOBAL_LOCK_INIT
   if (malloc_global_mutex_status <= 0)
     init_malloc_global_mutex();
diff --git a/third_party/double-conversion/bignum-dtoa.cc b/third_party/double-conversion/bignum-dtoa.cc
index 5b9325158..a9a75e4f8 100644
--- a/third_party/double-conversion/bignum-dtoa.cc
+++ b/third_party/double-conversion/bignum-dtoa.cc
@@ -28,11 +28,7 @@
 #include "third_party/double-conversion/bignum.h"
 #include "third_party/double-conversion/ieee.h"
 #include "third_party/libcxx/cmath"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/bignum.cc b/third_party/double-conversion/bignum.cc
index 62853945a..cadef28b0 100644
--- a/third_party/double-conversion/bignum.cc
+++ b/third_party/double-conversion/bignum.cc
@@ -28,11 +28,7 @@
 #include "third_party/double-conversion/utils.h"
 #include "third_party/libcxx/algorithm"
 #include "third_party/libcxx/cstring"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/cached-powers.cc b/third_party/double-conversion/cached-powers.cc
index 18253cfc4..05156a099 100644
--- a/third_party/double-conversion/cached-powers.cc
+++ b/third_party/double-conversion/cached-powers.cc
@@ -29,11 +29,7 @@
 #include "third_party/libcxx/climits"
 #include "third_party/libcxx/cmath"
 #include "third_party/libcxx/cstdarg"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/double-to-string.cc b/third_party/double-conversion/double-to-string.cc
index a58a5c925..0dfe0dc41 100644
--- a/third_party/double-conversion/double-to-string.cc
+++ b/third_party/double-conversion/double-to-string.cc
@@ -33,11 +33,7 @@
 #include "third_party/libcxx/algorithm"
 #include "third_party/libcxx/climits"
 #include "third_party/libcxx/cmath"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/dubble.c b/third_party/double-conversion/dubble.c
new file mode 100644
index 000000000..a73cf09aa
--- /dev/null
+++ b/third_party/double-conversion/dubble.c
@@ -0,0 +1,3 @@
+__notice(double_conversion_notice, "\
+double-conversion (BSD-3 License)\n\
+Copyright 2006-2012 the V8 project authors");
diff --git a/third_party/double-conversion/fast-dtoa.cc b/third_party/double-conversion/fast-dtoa.cc
index 60195c26d..84fcfd36b 100644
--- a/third_party/double-conversion/fast-dtoa.cc
+++ b/third_party/double-conversion/fast-dtoa.cc
@@ -28,11 +28,7 @@
 #include "third_party/double-conversion/diy-fp.h"
 #include "third_party/double-conversion/fast-dtoa.h"
 #include "third_party/double-conversion/ieee.h"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/fixed-dtoa.cc b/third_party/double-conversion/fixed-dtoa.cc
index cca64aca7..d00093609 100644
--- a/third_party/double-conversion/fixed-dtoa.cc
+++ b/third_party/double-conversion/fixed-dtoa.cc
@@ -27,11 +27,7 @@
 #include "third_party/double-conversion/fixed-dtoa.h"
 #include "third_party/double-conversion/ieee.h"
 #include "third_party/libcxx/cmath"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/double-conversion/string-to-double.cc b/third_party/double-conversion/string-to-double.cc
index 0d88f2d60..cc29f9555 100644
--- a/third_party/double-conversion/string-to-double.cc
+++ b/third_party/double-conversion/string-to-double.cc
@@ -31,11 +31,7 @@
 #include "third_party/libcxx/climits"
 #include "third_party/libcxx/cmath"
 #include "third_party/libcxx/locale"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 #ifdef _MSC_VER
 #  if _MSC_VER >= 1900
diff --git a/third_party/double-conversion/strtod.cc b/third_party/double-conversion/strtod.cc
index 9c00ef19c..bf6dde672 100644
--- a/third_party/double-conversion/strtod.cc
+++ b/third_party/double-conversion/strtod.cc
@@ -30,11 +30,7 @@
 #include "third_party/double-conversion/strtod.h"
 #include "third_party/libcxx/climits"
 #include "third_party/libcxx/cstdarg"
-
-asm(".ident\t\"\\n\\n\
-double-conversion (BSD-3 License)\\n\
-Copyright 2006-2012 the V8 project authors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("double_conversion_notice");
 
 namespace double_conversion {
 
diff --git a/third_party/gdtoa/gdtoa.internal.h b/third_party/gdtoa/gdtoa.internal.h
index bbb4d1e2c..67f11af0a 100644
--- a/third_party/gdtoa/gdtoa.internal.h
+++ b/third_party/gdtoa/gdtoa.internal.h
@@ -4,12 +4,7 @@
 #include "libc/str/str.h"
 #include "third_party/gdtoa/gdtoa.h"
 
-asm(".ident\t\"\\n\\n\
-gdtoa (MIT License)\\n\
-The author of this software is David M. Gay\\n\
-Kudos go to Guy L. Steele, Jr. and Jon L. White\\n\
-Copyright (C) 1997, 1998, 2000 by Lucent Technologies\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("gdtoa_notice");
 
 #define IEEE_Arith          1
 #define IEEE_8087           1
diff --git a/third_party/gdtoa/misc.c b/third_party/gdtoa/misc.c
index 43b1d4729..75d3883d8 100644
--- a/third_party/gdtoa/misc.c
+++ b/third_party/gdtoa/misc.c
@@ -98,7 +98,7 @@ __gdtoa_Bclear(void)
 	__gdtoa_unlock();
 }
 
-__attribute__((__constructor__)) static void
+__attribute__((__constructor__(60))) static void
 __gdtoa_Binit(void)
 {
 	__gdtoa_initlock();
diff --git a/third_party/gdtoa/notice.c b/third_party/gdtoa/notice.c
new file mode 100644
index 000000000..1046f9235
--- /dev/null
+++ b/third_party/gdtoa/notice.c
@@ -0,0 +1,5 @@
+__notice(gdtoa_notice, "\
+gdtoa (MIT License)\n\
+The author of this software is David M. Gay\n\
+Kudos go to Guy L. Steele, Jr. and Jon L. White\n\
+Copyright (C) 1997, 1998, 2000 by Lucent Technologies");
diff --git a/third_party/getopt/getopt.c b/third_party/getopt/getopt.c
index f1d7b716e..e51509a51 100644
--- a/third_party/getopt/getopt.c
+++ b/third_party/getopt/getopt.c
@@ -37,10 +37,8 @@
 #include "libc/errno.h"
 #include "third_party/getopt/getopt.internal.h"
 
-asm(".ident\t\"\\n\
-getopt (BSD-3)\\n\
-Copyright 1987, 1993, 1994 The Regents of the University of California\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(notice_getopt, "getopt (BSD-3)\n\
+Copyright 1987, 1993, 1994 The Regents of the University of California");
 
 #define BADCH  '?'
 #define BADARG ':'
diff --git a/third_party/intel/amxbf16intrin.internal.h b/third_party/intel/amxbf16intrin.internal.h
index d7b31ecb0..8a1ed88fb 100644
--- a/third_party/intel/amxbf16intrin.internal.h
+++ b/third_party/intel/amxbf16intrin.internal.h
@@ -9,7 +9,7 @@
 #pragma GCC target("amx-bf16")
 #define __DISABLE_AMX_BF16__
 #endif
-#if defined(__x86_64__) && defined(__AMX_BF16__)
+#if defined(__x86_64__)
 #define _tile_dpbf16ps_internal(dst,src1,src2) __asm__ volatile ("{tdpbf16ps\t%%tmm"#src2", %%tmm"#src1", %%tmm"#dst"|tdpbf16ps\t%%tmm"#dst", %%tmm"#src1", %%tmm"#src2"}" ::)
 #define _tile_dpbf16ps(dst,src1,src2) _tile_dpbf16ps_internal (dst, src1, src2)
 #endif
diff --git a/third_party/intel/amxint8intrin.internal.h b/third_party/intel/amxint8intrin.internal.h
index f1d25d11f..e5c566735 100644
--- a/third_party/intel/amxint8intrin.internal.h
+++ b/third_party/intel/amxint8intrin.internal.h
@@ -9,7 +9,7 @@
 #pragma GCC target("amx-int8")
 #define __DISABLE_AMX_INT8__
 #endif
-#if defined(__x86_64__) && defined(__AMX_INT8__)
+#if defined(__x86_64__)
 #define _tile_int8_dp_internal(name,dst,src1,src2) __asm__ volatile ("{"#name"\t%%tmm"#src2", %%tmm"#src1", %%tmm"#dst"|"#name"\t%%tmm"#dst", %%tmm"#src1", %%tmm"#src2"}" ::)
 #define _tile_dpbssd(dst,src1,src2) _tile_int8_dp_internal (tdpbssd, dst, src1, src2)
 #define _tile_dpbsud(dst,src1,src2) _tile_int8_dp_internal (tdpbsud, dst, src1, src2)
diff --git a/third_party/intel/amxtileintrin.internal.h b/third_party/intel/amxtileintrin.internal.h
index 3913c900e..00e403118 100644
--- a/third_party/intel/amxtileintrin.internal.h
+++ b/third_party/intel/amxtileintrin.internal.h
@@ -9,7 +9,7 @@
 #pragma GCC target("amx-tile")
 #define __DISABLE_AMX_TILE__
 #endif
-#if defined(__x86_64__) && defined(__AMX_TILE__)
+#if defined(__x86_64__)
 extern __inline void
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _tile_loadconfig (const void *__config)
@@ -29,11 +29,11 @@ _tile_release (void)
   __asm__ volatile ("tilerelease" ::);
 }
 #define _tile_loadd(dst,base,stride) _tile_loadd_internal (dst, base, stride)
-#define _tile_loadd_internal(dst,base,stride) __asm__ volatile ("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" :: "r" ((const void*) base), "r" ((long) stride))
+#define _tile_loadd_internal(dst,base,stride) __asm__ volatile ("{tileloadd\t(%0,%1,1), %%tmm"#dst"|tileloadd\t%%tmm"#dst", [%0+%1*1]}" :: "r" ((const void*) (base)), "r" ((long) (stride)))
 #define _tile_stream_loadd(dst,base,stride) _tile_stream_loadd_internal (dst, base, stride)
-#define _tile_stream_loadd_internal(dst,base,stride) __asm__ volatile ("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" :: "r" ((const void*) base), "r" ((long) stride))
+#define _tile_stream_loadd_internal(dst,base,stride) __asm__ volatile ("{tileloaddt1\t(%0,%1,1), %%tmm"#dst"|tileloaddt1\t%%tmm"#dst", [%0+%1*1]}" :: "r" ((const void*) (base)), "r" ((long) (stride)))
 #define _tile_stored(dst,base,stride) _tile_stored_internal (dst, base, stride)
-#define _tile_stored_internal(src,base,stride) __asm__ volatile ("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" :: "r" ((void*) base), "r" ((long) stride) : "memory")
+#define _tile_stored_internal(src,base,stride) __asm__ volatile ("{tilestored\t%%tmm"#src", (%0,%1,1)|tilestored\t[%0+%1*1], %%tmm"#src"}" :: "r" ((void*) (base)), "r" ((long) (stride)) : "memory")
 #define _tile_zero(dst) _tile_zero_internal (dst)
 #define _tile_zero_internal(dst) __asm__ volatile ("tilezero\t%%tmm"#dst ::)
 #endif
diff --git a/third_party/intel/avx512bf16intrin.internal.h b/third_party/intel/avx512bf16intrin.internal.h
index 088340d6a..5bc3f8358 100644
--- a/third_party/intel/avx512bf16intrin.internal.h
+++ b/third_party/intel/avx512bf16intrin.internal.h
@@ -11,6 +11,14 @@
 #endif
 typedef short __v32bh __attribute__ ((__vector_size__ (64)));
 typedef short __m512bh __attribute__ ((__vector_size__ (64), __may_alias__));
+extern __inline float
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsbh_ss (__bfloat16 __A)
+{
+  union{ float a; unsigned int b;} __tmp;
+  __tmp.b = ((unsigned int)(__A)) << 16;
+  return __tmp.a;
+}
 extern __inline __m512bh
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_cvtne2ps_pbh (__m512 __A, __m512 __B)
@@ -65,6 +73,29 @@ _mm512_maskz_dpbf16_ps (__mmask16 __A, __m512 __B, __m512bh __C, __m512bh __D)
 {
   return (__m512)__builtin_ia32_dpbf16ps_v16sf_maskz(__B, __C, __D, __A);
 }
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtpbh_ps (__m256bh __A)
+{
+  return (__m512)_mm512_castsi512_ps ((__m512i)_mm512_slli_epi32 (
+  (__m512i)_mm512_cvtepi16_epi32 ((__m256i)__A), 16));
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtpbh_ps (__mmask16 __U, __m256bh __A)
+{
+  return (__m512)_mm512_castsi512_ps ((__m512i) _mm512_slli_epi32 (
+  (__m512i)_mm512_maskz_cvtepi16_epi32 (
+  (__mmask16)__U, (__m256i)__A), 16));
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtpbh_ps (__m512 __S, __mmask16 __U, __m256bh __A)
+{
+  return (__m512)_mm512_castsi512_ps ((__m512i)(_mm512_mask_slli_epi32 (
+  (__m512i)__S, (__mmask16)__U,
+  (__m512i)_mm512_cvtepi16_epi32 ((__m256i)__A), 16)));
+}
 #ifdef __DISABLE_AVX512BF16__
 #undef __DISABLE_AVX512BF16__
 #pragma GCC pop_options
diff --git a/third_party/intel/avx512bf16vlintrin.internal.h b/third_party/intel/avx512bf16vlintrin.internal.h
index 83019cf5f..216196fcf 100644
--- a/third_party/intel/avx512bf16vlintrin.internal.h
+++ b/third_party/intel/avx512bf16vlintrin.internal.h
@@ -13,6 +13,7 @@ typedef short __v16bh __attribute__ ((__vector_size__ (32)));
 typedef short __v8bh __attribute__ ((__vector_size__ (16)));
 typedef short __m256bh __attribute__ ((__vector_size__ (32), __may_alias__));
 typedef short __m128bh __attribute__ ((__vector_size__ (16), __may_alias__));
+typedef unsigned short __bfloat16;
 extern __inline __m256bh
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm256_cvtne2ps_pbh (__m256 __A, __m256 __B)
@@ -121,6 +122,61 @@ _mm_maskz_dpbf16_ps (__mmask8 __A, __m128 __B, __m128bh __C, __m128bh __D)
 {
   return (__m128)__builtin_ia32_dpbf16ps_v4sf_maskz(__B, __C, __D, __A);
 }
+extern __inline __bfloat16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtness_sbh (float __A)
+{
+  __v4sf __V = {__A, 0, 0, 0};
+  __v8hi __R = __builtin_ia32_cvtneps2bf16_v4sf_mask ((__v4sf)__V,
+        (__v8hi)_mm_undefined_si128 (), (__mmask8)-1);
+  return __R[0];
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpbh_ps (__m128bh __A)
+{
+  return (__m128)_mm_castsi128_ps ((__m128i)_mm_slli_epi32 (
+  (__m128i)_mm_cvtepi16_epi32 ((__m128i)__A), 16));
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtpbh_ps (__m128bh __A)
+{
+  return (__m256)_mm256_castsi256_ps ((__m256i)_mm256_slli_epi32 (
+  (__m256i)_mm256_cvtepi16_epi32 ((__m128i)__A), 16));
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtpbh_ps (__mmask8 __U, __m128bh __A)
+{
+  return (__m128)_mm_castsi128_ps ((__m128i)_mm_slli_epi32 (
+  (__m128i)_mm_maskz_cvtepi16_epi32 (
+  (__mmask8)__U, (__m128i)__A), 16));
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtpbh_ps (__mmask8 __U, __m128bh __A)
+{
+  return (__m256)_mm256_castsi256_ps ((__m256i)_mm256_slli_epi32 (
+  (__m256i)_mm256_maskz_cvtepi16_epi32 (
+  (__mmask8)__U, (__m128i)__A), 16));
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtpbh_ps (__m128 __S, __mmask8 __U, __m128bh __A)
+{
+  return (__m128)_mm_castsi128_ps ((__m128i)_mm_mask_slli_epi32 (
+  (__m128i)__S, (__mmask8)__U, (__m128i)_mm_cvtepi16_epi32 (
+  (__m128i)__A), 16));
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtpbh_ps (__m256 __S, __mmask8 __U, __m128bh __A)
+{
+  return (__m256)_mm256_castsi256_ps ((__m256i)_mm256_mask_slli_epi32 (
+  (__m256i)__S, (__mmask8)__U, (__m256i)_mm256_cvtepi16_epi32 (
+  (__m128i)__A), 16));
+}
 #ifdef __DISABLE_AVX512BF16VL__
 #undef __DISABLE_AVX512BF16VL__
 #pragma GCC pop_options
diff --git a/third_party/intel/avx512dqintrin.internal.h b/third_party/intel/avx512dqintrin.internal.h
index 38d8010f1..0cf6a7080 100644
--- a/third_party/intel/avx512dqintrin.internal.h
+++ b/third_party/intel/avx512dqintrin.internal.h
@@ -2248,9 +2248,9 @@ _mm512_fpclass_ps_mask (__m512 __A, const int __imm)
 #define _mm_mask_fpclass_ss_mask(X, C, U) ((__mmask8) __builtin_ia32_fpclassss_mask ((__v4sf) (__m128) (X), (int) (C), (__mmask8) (U)))
 #define _mm_mask_fpclass_sd_mask(X, C, U) ((__mmask8) __builtin_ia32_fpclasssd_mask ((__v2df) (__m128d) (X), (int) (C), (__mmask8) (U)))
 #define _mm512_mask_fpclass_pd_mask(u, X, C) ((__mmask8) __builtin_ia32_fpclasspd512_mask ((__v8df) (__m512d) (X), (int) (C), (__mmask8)(u)))
-#define _mm512_mask_fpclass_ps_mask(u, x, c) ((__mmask16) __builtin_ia32_fpclassps512_mask ((__v16sf) (__m512) (x), (int) (c),(__mmask8)(u)))
+#define _mm512_mask_fpclass_ps_mask(u, x, c) ((__mmask16) __builtin_ia32_fpclassps512_mask ((__v16sf) (__m512) (x), (int) (c),(__mmask16)(u)))
 #define _mm512_fpclass_pd_mask(X, C) ((__mmask8) __builtin_ia32_fpclasspd512_mask ((__v8df) (__m512d) (X), (int) (C), (__mmask8)-1))
-#define _mm512_fpclass_ps_mask(x, c) ((__mmask16) __builtin_ia32_fpclassps512_mask ((__v16sf) (__m512) (x), (int) (c),(__mmask8)-1))
+#define _mm512_fpclass_ps_mask(x, c) ((__mmask16) __builtin_ia32_fpclassps512_mask ((__v16sf) (__m512) (x), (int) (c),(__mmask16)-1))
 #define _mm_reduce_sd(A, B, C) ((__m128d) __builtin_ia32_reducesd_mask ((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), (int)(C), (__v2df) _mm_setzero_pd (), (__mmask8)-1))
 #define _mm_mask_reduce_sd(W, U, A, B, C) ((__m128d) __builtin_ia32_reducesd_mask ((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), (int)(C), (__v2df)(__m128d)(W), (__mmask8)(U)))
 #define _mm_maskz_reduce_sd(U, A, B, C) ((__m128d) __builtin_ia32_reducesd_mask ((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), (int)(C), (__v2df) _mm_setzero_pd (), (__mmask8)(U)))
diff --git a/third_party/intel/avx512fintrin.internal.h b/third_party/intel/avx512fintrin.internal.h
index 79158ec85..a2c2c788c 100644
--- a/third_party/intel/avx512fintrin.internal.h
+++ b/third_party/intel/avx512fintrin.internal.h
@@ -1351,74 +1351,92 @@ _mm_maskz_sub_round_ss (__mmask8 __U, __m128 __A, __m128 __B,
 #define _mm_mask_sub_round_ss(W, U, A, B, C) (__m128)__builtin_ia32_subss_mask_round(A, B, W, U, C)
 #define _mm_maskz_sub_round_ss(U, A, B, C) (__m128)__builtin_ia32_subss_mask_round(A, B, (__v4sf)_mm_setzero_ps(), U, C)
 #endif
+typedef enum
+{
+  _MM_TERNLOG_A = 0xF0,
+  _MM_TERNLOG_B = 0xCC,
+  _MM_TERNLOG_C = 0xAA
+} _MM_TERNLOG_ENUM;
 #ifdef __OPTIMIZE__
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_ternarylogic_epi64 (__m512i __A, __m512i __B, __m512i __C,
       const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogq512_mask ((__v8di) __A,
-           (__v8di) __B,
-           (__v8di) __C, __imm,
-           (__mmask8) -1);
+  return (__m512i)
+    __builtin_ia32_pternlogq512_mask ((__v8di) __A,
+          (__v8di) __B,
+          (__v8di) __C,
+          (unsigned char) __imm,
+          (__mmask8) -1);
 }
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_mask_ternarylogic_epi64 (__m512i __A, __mmask8 __U, __m512i __B,
     __m512i __C, const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogq512_mask ((__v8di) __A,
-           (__v8di) __B,
-           (__v8di) __C, __imm,
-           (__mmask8) __U);
+  return (__m512i)
+    __builtin_ia32_pternlogq512_mask ((__v8di) __A,
+          (__v8di) __B,
+          (__v8di) __C,
+          (unsigned char) __imm,
+          (__mmask8) __U);
 }
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_maskz_ternarylogic_epi64 (__mmask8 __U, __m512i __A, __m512i __B,
      __m512i __C, const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogq512_maskz ((__v8di) __A,
-            (__v8di) __B,
-            (__v8di) __C,
-            __imm, (__mmask8) __U);
+  return (__m512i)
+    __builtin_ia32_pternlogq512_maskz ((__v8di) __A,
+           (__v8di) __B,
+           (__v8di) __C,
+           (unsigned char) __imm,
+           (__mmask8) __U);
 }
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_ternarylogic_epi32 (__m512i __A, __m512i __B, __m512i __C,
       const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogd512_mask ((__v16si) __A,
-           (__v16si) __B,
-           (__v16si) __C,
-           __imm, (__mmask16) -1);
+  return (__m512i)
+    __builtin_ia32_pternlogd512_mask ((__v16si) __A,
+          (__v16si) __B,
+          (__v16si) __C,
+          (unsigned char) __imm,
+          (__mmask16) -1);
 }
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_mask_ternarylogic_epi32 (__m512i __A, __mmask16 __U, __m512i __B,
     __m512i __C, const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogd512_mask ((__v16si) __A,
-           (__v16si) __B,
-           (__v16si) __C,
-           __imm, (__mmask16) __U);
+  return (__m512i)
+    __builtin_ia32_pternlogd512_mask ((__v16si) __A,
+          (__v16si) __B,
+          (__v16si) __C,
+          (unsigned char) __imm,
+          (__mmask16) __U);
 }
 extern __inline __m512i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm512_maskz_ternarylogic_epi32 (__mmask16 __U, __m512i __A, __m512i __B,
      __m512i __C, const int __imm)
 {
-  return (__m512i) __builtin_ia32_pternlogd512_maskz ((__v16si) __A,
-            (__v16si) __B,
-            (__v16si) __C,
-            __imm, (__mmask16) __U);
+  return (__m512i)
+    __builtin_ia32_pternlogd512_maskz ((__v16si) __A,
+           (__v16si) __B,
+           (__v16si) __C,
+           (unsigned char) __imm,
+           (__mmask16) __U);
 }
 #else
-#define _mm512_ternarylogic_epi64(A, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_mask ((__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), (int)(I), (__mmask8)-1))
-#define _mm512_mask_ternarylogic_epi64(A, U, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_mask ((__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), (int)(I), (__mmask8)(U)))
-#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_maskz ((__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C), (int)(I), (__mmask8)(U)))
-#define _mm512_ternarylogic_epi32(A, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_mask ((__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), (int)(I), (__mmask16)-1))
-#define _mm512_mask_ternarylogic_epi32(A, U, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_mask ((__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), (int)(I), (__mmask16)(U)))
-#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_maskz ((__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C), (int)(I), (__mmask16)(U)))
+#define _mm512_ternarylogic_epi64(A, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_mask ((__v8di) (__m512i) (A), (__v8di) (__m512i) (B), (__v8di) (__m512i) (C), (unsigned char) (I), (__mmask8) -1))
+#define _mm512_mask_ternarylogic_epi64(A, U, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_mask ((__v8di) (__m512i) (A), (__v8di) (__m512i) (B), (__v8di) (__m512i) (C), (unsigned char)(I), (__mmask8) (U)))
+#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m512i) __builtin_ia32_pternlogq512_maskz ((__v8di) (__m512i) (A), (__v8di) (__m512i) (B), (__v8di) (__m512i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm512_ternarylogic_epi32(A, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_mask ((__v16si) (__m512i) (A), (__v16si) (__m512i) (B), (__v16si) (__m512i) (C), (unsigned char) (I), (__mmask16) -1))
+#define _mm512_mask_ternarylogic_epi32(A, U, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_mask ((__v16si) (__m512i) (A), (__v16si) (__m512i) (B), (__v16si) (__m512i) (C), (unsigned char) (I), (__mmask16) (U)))
+#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m512i) __builtin_ia32_pternlogd512_maskz ((__v16si) (__m512i) (A), (__v16si) (__m512i) (B), (__v16si) (__m512i) (C), (unsigned char) (I), (__mmask16) (U)))
 #endif
 extern __inline __m512d
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -2680,14 +2698,18 @@ _mm_maskz_scalef_round_ss (__mmask8 __U, __m128 __A, __m128 __B, const int __R)
             (__mmask8) __U, __R);
 }
 #else
-#define _mm512_scalef_round_pd(A, B, C) (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_undefined_pd(), -1, C)
-#define _mm512_mask_scalef_round_pd(W, U, A, B, C) (__m512d)__builtin_ia32_scalefpd512_mask(A, B, W, U, C)
-#define _mm512_maskz_scalef_round_pd(U, A, B, C) (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_setzero_pd(), U, C)
-#define _mm512_scalef_round_ps(A, B, C) (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_undefined_ps(), -1, C)
-#define _mm512_mask_scalef_round_ps(W, U, A, B, C) (__m512)__builtin_ia32_scalefps512_mask(A, B, W, U, C)
-#define _mm512_maskz_scalef_round_ps(U, A, B, C) (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_setzero_ps(), U, C)
-#define _mm_scalef_round_sd(A, B, C) (__m128d)__builtin_ia32_scalefsd_mask_round (A, B, (__v2df)_mm_setzero_pd (), -1, C)
-#define _mm_scalef_round_ss(A, B, C) (__m128)__builtin_ia32_scalefss_mask_round (A, B, (__v4sf)_mm_setzero_ps (), -1, C)
+#define _mm512_scalef_round_pd(A, B, C) ((__m512d) __builtin_ia32_scalefpd512_mask((A), (B), (__v8df) _mm512_undefined_pd(), -1, (C)))
+#define _mm512_mask_scalef_round_pd(W, U, A, B, C) ((__m512d) __builtin_ia32_scalefpd512_mask((A), (B), (W), (U), (C)))
+#define _mm512_maskz_scalef_round_pd(U, A, B, C) ((__m512d) __builtin_ia32_scalefpd512_mask((A), (B), (__v8df) _mm512_setzero_pd(), (U), (C)))
+#define _mm512_scalef_round_ps(A, B, C) ((__m512) __builtin_ia32_scalefps512_mask((A), (B), (__v16sf) _mm512_undefined_ps(), -1, (C)))
+#define _mm512_mask_scalef_round_ps(W, U, A, B, C) ((__m512) __builtin_ia32_scalefps512_mask((A), (B), (W), (U), (C)))
+#define _mm512_maskz_scalef_round_ps(U, A, B, C) ((__m512) __builtin_ia32_scalefps512_mask((A), (B), (__v16sf) _mm512_setzero_ps(), (U), (C)))
+#define _mm_scalef_round_sd(A, B, C) ((__m128d) __builtin_ia32_scalefsd_mask_round ((A), (B), (__v2df) _mm_undefined_pd (), -1, (C)))
+#define _mm_scalef_round_ss(A, B, C) ((__m128) __builtin_ia32_scalefss_mask_round ((A), (B), (__v4sf) _mm_undefined_ps (), -1, (C)))
+#define _mm_mask_scalef_round_sd(W, U, A, B, C) ((__m128d) __builtin_ia32_scalefsd_mask_round ((A), (B), (W), (U), (C)))
+#define _mm_mask_scalef_round_ss(W, U, A, B, C) ((__m128) __builtin_ia32_scalefss_mask_round ((A), (B), (W), (U), (C)))
+#define _mm_maskz_scalef_round_sd(U, A, B, C) ((__m128d) __builtin_ia32_scalefsd_mask_round ((A), (B), (__v2df) _mm_setzero_pd (), (U), (C)))
+#define _mm_maskz_scalef_round_ss(U, A, B, C) ((__m128) __builtin_ia32_scalefss_mask_round ((A), (B), (__v4sf) _mm_setzero_ps (), (U), (C)))
 #endif
 #define _mm_mask_scalef_sd(W, U, A, B) _mm_mask_scalef_round_sd ((W), (U), (A), (B), _MM_FROUND_CUR_DIRECTION)
 #define _mm_maskz_scalef_sd(U, A, B) _mm_maskz_scalef_round_sd ((U), (A), (B), _MM_FROUND_CUR_DIRECTION)
diff --git a/third_party/intel/avx512fp16intrin.internal.h b/third_party/intel/avx512fp16intrin.internal.h
new file mode 100644
index 000000000..4f249cc30
--- /dev/null
+++ b/third_party/intel/avx512fp16intrin.internal.h
@@ -0,0 +1,5452 @@
+#if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
+#ifndef _IMMINTRIN_H_INCLUDED
+#error "Never use <avx512fp16intrin.h> directly; include <immintrin.h> instead."
+#endif
+#ifndef __AVX512FP16INTRIN_H_INCLUDED
+#define __AVX512FP16INTRIN_H_INCLUDED
+#ifndef __AVX512FP16__
+#pragma GCC push_options
+#pragma GCC target("avx512fp16")
+#define __DISABLE_AVX512FP16__
+#endif
+typedef _Float16 __v8hf __attribute__ ((__vector_size__ (16)));
+typedef _Float16 __v16hf __attribute__ ((__vector_size__ (32)));
+typedef _Float16 __v32hf __attribute__ ((__vector_size__ (64)));
+typedef _Float16 __m128h __attribute__ ((__vector_size__ (16), __may_alias__));
+typedef _Float16 __m256h __attribute__ ((__vector_size__ (32), __may_alias__));
+typedef _Float16 __m512h __attribute__ ((__vector_size__ (64), __may_alias__));
+typedef _Float16 __m128h_u __attribute__ ((__vector_size__ (16), __may_alias__, __aligned__ (1)));
+typedef _Float16 __m256h_u __attribute__ ((__vector_size__ (32), __may_alias__, __aligned__ (1)));
+typedef _Float16 __m512h_u __attribute__ ((__vector_size__ (64), __may_alias__, __aligned__ (1)));
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ph (_Float16 __A7, _Float16 __A6, _Float16 __A5,
+     _Float16 __A4, _Float16 __A3, _Float16 __A2,
+     _Float16 __A1, _Float16 __A0)
+{
+  return __extension__ (__m128h)(__v8hf){ __A0, __A1, __A2, __A3,
+       __A4, __A5, __A6, __A7 };
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_set_ph (_Float16 __A15, _Float16 __A14, _Float16 __A13,
+        _Float16 __A12, _Float16 __A11, _Float16 __A10,
+        _Float16 __A9, _Float16 __A8, _Float16 __A7,
+        _Float16 __A6, _Float16 __A5, _Float16 __A4,
+        _Float16 __A3, _Float16 __A2, _Float16 __A1,
+        _Float16 __A0)
+{
+  return __extension__ (__m256h)(__v16hf){ __A0, __A1, __A2, __A3,
+        __A4, __A5, __A6, __A7,
+        __A8, __A9, __A10, __A11,
+        __A12, __A13, __A14, __A15 };
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_set_ph (_Float16 __A31, _Float16 __A30, _Float16 __A29,
+        _Float16 __A28, _Float16 __A27, _Float16 __A26,
+        _Float16 __A25, _Float16 __A24, _Float16 __A23,
+        _Float16 __A22, _Float16 __A21, _Float16 __A20,
+        _Float16 __A19, _Float16 __A18, _Float16 __A17,
+        _Float16 __A16, _Float16 __A15, _Float16 __A14,
+        _Float16 __A13, _Float16 __A12, _Float16 __A11,
+        _Float16 __A10, _Float16 __A9, _Float16 __A8,
+        _Float16 __A7, _Float16 __A6, _Float16 __A5,
+        _Float16 __A4, _Float16 __A3, _Float16 __A2,
+        _Float16 __A1, _Float16 __A0)
+{
+  return __extension__ (__m512h)(__v32hf){ __A0, __A1, __A2, __A3,
+        __A4, __A5, __A6, __A7,
+        __A8, __A9, __A10, __A11,
+        __A12, __A13, __A14, __A15,
+        __A16, __A17, __A18, __A19,
+        __A20, __A21, __A22, __A23,
+        __A24, __A25, __A26, __A27,
+        __A28, __A29, __A30, __A31 };
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2,
+      _Float16 __A3, _Float16 __A4, _Float16 __A5,
+      _Float16 __A6, _Float16 __A7)
+{
+  return _mm_set_ph (__A7, __A6, __A5, __A4, __A3, __A2, __A1, __A0);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2,
+  _Float16 __A3, _Float16 __A4, _Float16 __A5,
+  _Float16 __A6, _Float16 __A7, _Float16 __A8,
+  _Float16 __A9, _Float16 __A10, _Float16 __A11,
+  _Float16 __A12, _Float16 __A13, _Float16 __A14,
+  _Float16 __A15)
+{
+  return _mm256_set_ph (__A15, __A14, __A13, __A12, __A11, __A10, __A9,
+   __A8, __A7, __A6, __A5, __A4, __A3, __A2, __A1,
+   __A0);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_setr_ph (_Float16 __A0, _Float16 __A1, _Float16 __A2,
+  _Float16 __A3, _Float16 __A4, _Float16 __A5,
+  _Float16 __A6, _Float16 __A7, _Float16 __A8,
+  _Float16 __A9, _Float16 __A10, _Float16 __A11,
+  _Float16 __A12, _Float16 __A13, _Float16 __A14,
+  _Float16 __A15, _Float16 __A16, _Float16 __A17,
+  _Float16 __A18, _Float16 __A19, _Float16 __A20,
+  _Float16 __A21, _Float16 __A22, _Float16 __A23,
+  _Float16 __A24, _Float16 __A25, _Float16 __A26,
+  _Float16 __A27, _Float16 __A28, _Float16 __A29,
+  _Float16 __A30, _Float16 __A31)
+{
+  return _mm512_set_ph (__A31, __A30, __A29, __A28, __A27, __A26, __A25,
+   __A24, __A23, __A22, __A21, __A20, __A19, __A18,
+   __A17, __A16, __A15, __A14, __A13, __A12, __A11,
+   __A10, __A9, __A8, __A7, __A6, __A5, __A4, __A3,
+   __A2, __A1, __A0);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_ph (_Float16 __A)
+{
+  return _mm_set_ph (__A, __A, __A, __A, __A, __A, __A, __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_set1_ph (_Float16 __A)
+{
+  return _mm256_set_ph (__A, __A, __A, __A, __A, __A, __A, __A,
+   __A, __A, __A, __A, __A, __A, __A, __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_set1_ph (_Float16 __A)
+{
+  return _mm512_set_ph (__A, __A, __A, __A, __A, __A, __A, __A,
+   __A, __A, __A, __A, __A, __A, __A, __A,
+   __A, __A, __A, __A, __A, __A, __A, __A,
+   __A, __A, __A, __A, __A, __A, __A, __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setzero_ph (void)
+{
+  return _mm_set1_ph (0.0f);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_setzero_ph (void)
+{
+  return _mm256_set1_ph (0.0f);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_setzero_ph (void)
+{
+  return _mm512_set1_ph (0.0f);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_undefined_ph (void)
+{
+  __m128h __Y = __Y;
+  return __Y;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_undefined_ph (void)
+{
+  __m256h __Y = __Y;
+  return __Y;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_undefined_ph (void)
+{
+  __m512h __Y = __Y;
+  return __Y;
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_h (__m128h __A)
+{
+  return __A[0];
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtsh_h (__m256h __A)
+{
+  return __A[0];
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtsh_h (__m512h __A)
+{
+  return __A[0];
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph_ps (__m512h __a)
+{
+  return (__m512) __a;
+}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph_pd (__m512h __a)
+{
+  return (__m512d) __a;
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph_si512 (__m512h __a)
+{
+  return (__m512i) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph512_ph128 (__m512h __A)
+{
+  union
+  {
+    __m128h a[4];
+    __m512h v;
+  } u = { .v = __A };
+  return u.a[0];
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph512_ph256 (__m512h __A)
+{
+  union
+  {
+    __m256h a[2];
+    __m512h v;
+  } u = { .v = __A };
+  return u.a[0];
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph128_ph512 (__m128h __A)
+{
+  union
+  {
+    __m128h a[4];
+    __m512h v;
+  } u;
+  u.a[0] = __A;
+  return u.v;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castph256_ph512 (__m256h __A)
+{
+  union
+  {
+    __m256h a[2];
+    __m512h v;
+  } u;
+  u.a[0] = __A;
+  return u.v;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_zextph128_ph512 (__m128h __A)
+{
+  return (__m512h) _mm512_insertf32x4 (_mm512_setzero_ps (),
+           (__m128) __A, 0);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_zextph256_ph512 (__m256h __A)
+{
+  return (__m512h) _mm512_insertf64x4 (_mm512_setzero_pd (),
+           (__m256d) __A, 0);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castps_ph (__m512 __a)
+{
+  return (__m512h) __a;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castpd_ph (__m512d __a)
+{
+  return (__m512h) __a;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_castsi512_ph (__m512i __a)
+{
+  return (__m512h) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_sh (_Float16 __F)
+{
+  return _mm_set_ph (0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, __F);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_sh (void const *__P)
+{
+  return _mm_set_ph (0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
+       *(_Float16 const *) __P);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_load_ph (void const *__P)
+{
+  return *(const __m512h *) __P;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_load_ph (void const *__P)
+{
+  return *(const __m256h *) __P;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ph (void const *__P)
+{
+  return *(const __m128h *) __P;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_loadu_ph (void const *__P)
+{
+  return *(const __m512h_u *) __P;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_loadu_ph (void const *__P)
+{
+  return *(const __m256h_u *) __P;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadu_ph (void const *__P)
+{
+  return *(const __m128h_u *) __P;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_sh (void *__P, __m128h __A)
+{
+  *(_Float16 *) __P = ((__v8hf)__A)[0];
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_store_ph (void *__P, __m512h __A)
+{
+   *(__m512h *) __P = __A;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_store_ph (void *__P, __m256h __A)
+{
+   *(__m256h *) __P = __A;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ph (void *__P, __m128h __A)
+{
+   *(__m128h *) __P = __A;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_storeu_ph (void *__P, __m512h __A)
+{
+   *(__m512h_u *) __P = __A;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_storeu_ph (void *__P, __m256h __A)
+{
+   *(__m256h_u *) __P = __A;
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeu_ph (void *__P, __m128h __A)
+{
+   *(__m128h_u *) __P = __A;
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_abs_ph (__m512h __A)
+{
+  return (__m512h) _mm512_and_epi32 ( _mm512_set1_epi32 (0x7FFF7FFF),
+          (__m512i) __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_add_ph (__m512h __A, __m512h __B)
+{
+  return (__m512h) ((__v32hf) __A + (__v32hf) __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_add_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_addph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_add_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_addph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sub_ph (__m512h __A, __m512h __B)
+{
+  return (__m512h) ((__v32hf) __A - (__v32hf) __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sub_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_subph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sub_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_subph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mul_ph (__m512h __A, __m512h __B)
+{
+  return (__m512h) ((__v32hf) __A * (__v32hf) __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_mul_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_mulph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_mul_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_mulph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_div_ph (__m512h __A, __m512h __B)
+{
+  return (__m512h) ((__v32hf) __A / (__v32hf) __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_div_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_divph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_div_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_divph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_add_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_addph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_add_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_addph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_add_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_addph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sub_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_subph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sub_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_subph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sub_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_subph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mul_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_mulph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_mul_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_mulph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_mul_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_mulph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_div_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_divph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_div_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_divph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_div_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_divph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+#else
+#define _mm512_add_round_ph(A, B, C) ((__m512h)__builtin_ia32_addph512_mask_round((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_add_round_ph(A, B, C, D, E) ((__m512h)__builtin_ia32_addph512_mask_round((C), (D), (A), (B), (E)))
+#define _mm512_maskz_add_round_ph(A, B, C, D) ((__m512h)__builtin_ia32_addph512_mask_round((B), (C), _mm512_setzero_ph (), (A), (D)))
+#define _mm512_sub_round_ph(A, B, C) ((__m512h)__builtin_ia32_subph512_mask_round((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_sub_round_ph(A, B, C, D, E) ((__m512h)__builtin_ia32_subph512_mask_round((C), (D), (A), (B), (E)))
+#define _mm512_maskz_sub_round_ph(A, B, C, D) ((__m512h)__builtin_ia32_subph512_mask_round((B), (C), _mm512_setzero_ph (), (A), (D)))
+#define _mm512_mul_round_ph(A, B, C) ((__m512h)__builtin_ia32_mulph512_mask_round((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_mul_round_ph(A, B, C, D, E) ((__m512h)__builtin_ia32_mulph512_mask_round((C), (D), (A), (B), (E)))
+#define _mm512_maskz_mul_round_ph(A, B, C, D) ((__m512h)__builtin_ia32_mulph512_mask_round((B), (C), _mm512_setzero_ph (), (A), (D)))
+#define _mm512_div_round_ph(A, B, C) ((__m512h)__builtin_ia32_divph512_mask_round((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_div_round_ph(A, B, C, D, E) ((__m512h)__builtin_ia32_divph512_mask_round((C), (D), (A), (B), (E)))
+#define _mm512_maskz_div_round_ph(A, B, C, D) ((__m512h)__builtin_ia32_divph512_mask_round((B), (C), _mm512_setzero_ph (), (A), (D)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_conj_pch (__m512h __A)
+{
+  return (__m512h) _mm512_xor_epi32 ((__m512i) __A, _mm512_set1_epi32 (1<<31));
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_conj_pch (__m512h __W, __mmask16 __U, __m512h __A)
+{
+  return (__m512h)
+    __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A),
+       (__v16sf) __W,
+       (__mmask16) __U);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_conj_pch (__mmask16 __U, __m512h __A)
+{
+  return (__m512h)
+    __builtin_ia32_movaps512_mask ((__v16sf) _mm512_conj_pch (__A),
+       (__v16sf) _mm512_setzero_ps (),
+       (__mmask16) __U);
+}
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_sh (__m128h __A, __m128h __B)
+{
+  __A[0] += __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_add_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_addsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_add_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_addsh_mask (__B, __C, _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_sh (__m128h __A, __m128h __B)
+{
+  __A[0] -= __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sub_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_subsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sub_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_subsh_mask (__B, __C, _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_sh (__m128h __A, __m128h __B)
+{
+  __A[0] *= __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_mul_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_mulsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_mul_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_mulsh_mask (__B, __C, _mm_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_sh (__m128h __A, __m128h __B)
+{
+  __A[0] /= __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_div_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_divsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_div_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_divsh_mask (__B, __C, _mm_setzero_ph (),
+        __A);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_addsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_add_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_addsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_add_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_addsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_subsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sub_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_subsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sub_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_subsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_mulsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_mul_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_mulsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_mul_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_mulsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_divsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_div_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_divsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_div_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_divsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+#else
+#define _mm_add_round_sh(A, B, C) ((__m128h)__builtin_ia32_addsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_add_round_sh(A, B, C, D, E) ((__m128h)__builtin_ia32_addsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_add_round_sh(A, B, C, D) ((__m128h)__builtin_ia32_addsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#define _mm_sub_round_sh(A, B, C) ((__m128h)__builtin_ia32_subsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_sub_round_sh(A, B, C, D, E) ((__m128h)__builtin_ia32_subsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_sub_round_sh(A, B, C, D) ((__m128h)__builtin_ia32_subsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#define _mm_mul_round_sh(A, B, C) ((__m128h)__builtin_ia32_mulsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_mul_round_sh(A, B, C, D, E) ((__m128h)__builtin_ia32_mulsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_mul_round_sh(A, B, C, D) ((__m128h)__builtin_ia32_mulsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#define _mm_div_round_sh(A, B, C) ((__m128h)__builtin_ia32_divsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_div_round_sh(A, B, C, D, E) ((__m128h)__builtin_ia32_divsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_div_round_sh(A, B, C, D) ((__m128h)__builtin_ia32_divsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_max_ph (__m512h __A, __m512h __B)
+{
+  return __builtin_ia32_maxph512_mask (__A, __B,
+           _mm512_setzero_ph (),
+           (__mmask32) -1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_max_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_maxph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_max_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_maxph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_min_ph (__m512h __A, __m512h __B)
+{
+  return __builtin_ia32_minph512_mask (__A, __B,
+           _mm512_setzero_ph (),
+           (__mmask32) -1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_min_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_minph512_mask (__C, __D, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_min_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_minph512_mask (__B, __C,
+           _mm512_setzero_ph (), __A);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_max_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_maxph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_max_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_maxph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_max_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_maxph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_min_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_minph512_mask_round (__A, __B,
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_min_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+     __m512h __D, const int __E)
+{
+  return __builtin_ia32_minph512_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_min_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_minph512_mask_round (__B, __C,
+          _mm512_setzero_ph (),
+          __A, __D);
+}
+#else
+#define _mm512_max_round_ph(A, B, C) (__builtin_ia32_maxph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_max_round_ph(A, B, C, D, E) (__builtin_ia32_maxph512_mask_round ((C), (D), (A), (B), (E)))
+#define _mm512_maskz_max_round_ph(A, B, C, D) (__builtin_ia32_maxph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), (D)))
+#define _mm512_min_round_ph(A, B, C) (__builtin_ia32_minph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_min_round_ph(A, B, C, D, E) (__builtin_ia32_minph512_mask_round ((C), (D), (A), (B), (E)))
+#define _mm512_maskz_min_round_ph(A, B, C, D) (__builtin_ia32_minph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), (D)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_sh (__m128h __A, __m128h __B)
+{
+  __A[0] = __A[0] > __B[0] ? __A[0] : __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_max_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_maxsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_max_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_maxsh_mask (__B, __C, _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_sh (__m128h __A, __m128h __B)
+{
+  __A[0] = __A[0] < __B[0] ? __A[0] : __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_min_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_minsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_min_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_minsh_mask (__B, __C, _mm_setzero_ph (),
+        __A);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_maxsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_max_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_maxsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_max_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_maxsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_minsh_mask_round (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_min_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, const int __E)
+{
+  return __builtin_ia32_minsh_mask_round (__C, __D, __A, __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_min_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+   const int __D)
+{
+  return __builtin_ia32_minsh_mask_round (__B, __C,
+       _mm_setzero_ph (),
+       __A, __D);
+}
+#else
+#define _mm_max_round_sh(A, B, C) (__builtin_ia32_maxsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_max_round_sh(A, B, C, D, E) (__builtin_ia32_maxsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_max_round_sh(A, B, C, D) (__builtin_ia32_maxsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#define _mm_min_round_sh(A, B, C) (__builtin_ia32_minsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_min_round_sh(A, B, C, D, E) (__builtin_ia32_minsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_min_round_sh(A, B, C, D) (__builtin_ia32_minsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#endif
+#ifdef __OPTIMIZE
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cmp_ph_mask (__m512h __A, __m512h __B, const int __C)
+{
+  return (__mmask32) __builtin_ia32_cmpph512_mask (__A, __B, __C,
+         (__mmask32) -1);
+}
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cmp_ph_mask (__mmask32 __A, __m512h __B, __m512h __C,
+    const int __D)
+{
+  return (__mmask32) __builtin_ia32_cmpph512_mask (__B, __C, __D,
+         __A);
+}
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cmp_round_ph_mask (__m512h __A, __m512h __B, const int __C,
+     const int __D)
+{
+  return (__mmask32) __builtin_ia32_cmpph512_mask_round (__A, __B,
+        __C, (__mmask32) -1,
+        __D);
+}
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cmp_round_ph_mask (__mmask32 __A, __m512h __B, __m512h __C,
+          const int __D, const int __E)
+{
+  return (__mmask32) __builtin_ia32_cmpph512_mask_round (__B, __C,
+        __D, __A,
+        __E);
+}
+#else
+#define _mm512_cmp_ph_mask(A, B, C) (__builtin_ia32_cmpph512_mask ((A), (B), (C), (-1)))
+#define _mm512_mask_cmp_ph_mask(A, B, C, D) (__builtin_ia32_cmpph512_mask ((B), (C), (D), (A)))
+#define _mm512_cmp_round_ph_mask(A, B, C, D) (__builtin_ia32_cmpph512_mask_round ((A), (B), (C), (-1), (D)))
+#define _mm512_mask_cmp_round_ph_mask(A, B, C, D, E) (__builtin_ia32_cmpph512_mask_round ((B), (C), (D), (A), (E)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmp_sh_mask (__m128h __A, __m128h __B, const int __C)
+{
+  return (__mmask8)
+    __builtin_ia32_cmpsh_mask_round (__A, __B,
+         __C, (__mmask8) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cmp_sh_mask (__mmask8 __A, __m128h __B, __m128h __C,
+        const int __D)
+{
+  return (__mmask8)
+    __builtin_ia32_cmpsh_mask_round (__B, __C,
+         __D, __A,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmp_round_sh_mask (__m128h __A, __m128h __B, const int __C,
+         const int __D)
+{
+  return (__mmask8) __builtin_ia32_cmpsh_mask_round (__A, __B,
+           __C, (__mmask8) -1,
+           __D);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cmp_round_sh_mask (__mmask8 __A, __m128h __B, __m128h __C,
+       const int __D, const int __E)
+{
+  return (__mmask8) __builtin_ia32_cmpsh_mask_round (__B, __C,
+           __D, __A,
+           __E);
+}
+#else
+#define _mm_cmp_sh_mask(A, B, C) (__builtin_ia32_cmpsh_mask_round ((A), (B), (C), (-1), (_MM_FROUND_CUR_DIRECTION)))
+#define _mm_mask_cmp_sh_mask(A, B, C, D) (__builtin_ia32_cmpsh_mask_round ((B), (C), (D), (A), (_MM_FROUND_CUR_DIRECTION)))
+#define _mm_cmp_round_sh_mask(A, B, C, D) (__builtin_ia32_cmpsh_mask_round ((A), (B), (C), (-1), (D)))
+#define _mm_mask_cmp_round_sh_mask(A, B, C, D, E) (__builtin_ia32_cmpsh_mask_round ((B), (C), (D), (A), (E)))
+#endif
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comieq_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_EQ_OS,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comilt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_LT_OS,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comile_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_LE_OS,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comigt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_GT_OS,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comige_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_GE_OS,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comineq_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_NEQ_US,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomieq_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_EQ_OQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomilt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_LT_OQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomile_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_LE_OQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomigt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_GT_OQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomige_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_GE_OQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomineq_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, _CMP_NEQ_UQ,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comi_sh (__m128h __A, __m128h __B, const int __P)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, __P,
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comi_round_sh (__m128h __A, __m128h __B, const int __P, const int __R)
+{
+  return __builtin_ia32_cmpsh_mask_round (__A, __B, __P,
+       (__mmask8) -1,__R);
+}
+#else
+#define _mm_comi_round_sh(A, B, P, R) (__builtin_ia32_cmpsh_mask_round ((A), (B), (P), (__mmask8) (-1), (R)))
+#define _mm_comi_sh(A, B, P) (__builtin_ia32_cmpsh_mask_round ((A), (B), (P), (__mmask8) (-1), _MM_FROUND_CUR_DIRECTION))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sqrt_ph (__m512h __A)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__A,
+           _mm512_setzero_ph(),
+           (__mmask32) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sqrt_ph (__m512h __A, __mmask32 __B, __m512h __C)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sqrt_ph (__mmask32 __A, __m512h __B)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__B,
+           _mm512_setzero_ph (),
+           __A,
+           _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_sqrt_round_ph (__m512h __A, const int __B)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__A,
+           _mm512_setzero_ph(),
+           (__mmask32) -1, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_sqrt_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+      const int __D)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_sqrt_round_ph (__mmask32 __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_sqrtph512_mask_round (__B,
+           _mm512_setzero_ph (),
+           __A, __C);
+}
+#else
+#define _mm512_sqrt_round_ph(A, B) (__builtin_ia32_sqrtph512_mask_round ((A), _mm512_setzero_ph (), (__mmask32)-1, (B)))
+#define _mm512_mask_sqrt_round_ph(A, B, C, D) (__builtin_ia32_sqrtph512_mask_round ((C), (A), (B), (D)))
+#define _mm512_maskz_sqrt_round_ph(A, B, C) (__builtin_ia32_sqrtph512_mask_round ((B), _mm512_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_rsqrt_ph (__m512h __A)
+{
+  return __builtin_ia32_rsqrtph512_mask (__A, _mm512_setzero_ph (),
+      (__mmask32) -1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_rsqrt_ph (__m512h __A, __mmask32 __B, __m512h __C)
+{
+  return __builtin_ia32_rsqrtph512_mask (__C, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_rsqrt_ph (__mmask32 __A, __m512h __B)
+{
+  return __builtin_ia32_rsqrtph512_mask (__B, _mm512_setzero_ph (),
+      __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_rsqrtsh_mask (__B, __A, _mm_setzero_ph (),
+          (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_rsqrt_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_rsqrtsh_mask (__D, __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_rsqrt_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_rsqrtsh_mask (__C, __B, _mm_setzero_ph (),
+          __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__B, __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sqrt_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__D, __C, __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sqrt_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__C, __B,
+        _mm_setzero_ph (),
+        __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__B, __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sqrt_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+   __m128h __D, const int __E)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__D, __C, __A, __B,
+        __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sqrt_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+    const int __D)
+{
+  return __builtin_ia32_sqrtsh_mask_round (__C, __B,
+        _mm_setzero_ph (),
+        __A, __D);
+}
+#else
+#define _mm_sqrt_round_sh(A, B, C) (__builtin_ia32_sqrtsh_mask_round ((B), (A), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_sqrt_round_sh(A, B, C, D, E) (__builtin_ia32_sqrtsh_mask_round ((D), (C), (A), (B), (E)))
+#define _mm_maskz_sqrt_round_sh(A, B, C, D) (__builtin_ia32_sqrtsh_mask_round ((C), (B), _mm_setzero_ph (), (A), (D)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_rcp_ph (__m512h __A)
+{
+  return __builtin_ia32_rcpph512_mask (__A, _mm512_setzero_ph (),
+           (__mmask32) -1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_rcp_ph (__m512h __A, __mmask32 __B, __m512h __C)
+{
+  return __builtin_ia32_rcpph512_mask (__C, __A, __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_rcp_ph (__mmask32 __A, __m512h __B)
+{
+  return __builtin_ia32_rcpph512_mask (__B, _mm512_setzero_ph (),
+           __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_rcpsh_mask (__B, __A, _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_rcp_sh (__m128h __A, __mmask32 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_rcpsh_mask (__D, __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_rcp_sh (__mmask32 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_rcpsh_mask (__C, __B, _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_scalef_ph (__m512h __A, __m512h __B)
+{
+  return __builtin_ia32_scalefph512_mask_round (__A, __B,
+      _mm512_setzero_ph (),
+      (__mmask32) -1,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_scalef_ph (__m512h __A, __mmask32 __B, __m512h __C, __m512h __D)
+{
+  return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_scalef_ph (__mmask32 __A, __m512h __B, __m512h __C)
+{
+  return __builtin_ia32_scalefph512_mask_round (__B, __C,
+      _mm512_setzero_ph (),
+      __A,
+      _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_scalef_round_ph (__m512h __A, __m512h __B, const int __C)
+{
+  return __builtin_ia32_scalefph512_mask_round (__A, __B,
+      _mm512_setzero_ph (),
+      (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_scalef_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+        __m512h __D, const int __E)
+{
+  return __builtin_ia32_scalefph512_mask_round (__C, __D, __A, __B,
+      __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_scalef_round_ph (__mmask32 __A, __m512h __B, __m512h __C,
+         const int __D)
+{
+  return __builtin_ia32_scalefph512_mask_round (__B, __C,
+      _mm512_setzero_ph (),
+      __A, __D);
+}
+#else
+#define _mm512_scalef_round_ph(A, B, C) (__builtin_ia32_scalefph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_scalef_round_ph(A, B, C, D, E) (__builtin_ia32_scalefph512_mask_round ((C), (D), (A), (B), (E)))
+#define _mm512_maskz_scalef_round_ph(A, B, C, D) (__builtin_ia32_scalefph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), (D)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_scalef_sh (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_scalefsh_mask_round (__A, __B,
+          _mm_setzero_ph (),
+          (__mmask8) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_scalef_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_scalefsh_mask_round (__C, __D, __A, __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_scalef_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_scalefsh_mask_round (__B, __C,
+          _mm_setzero_ph (),
+          __A,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_scalef_round_sh (__m128h __A, __m128h __B, const int __C)
+{
+  return __builtin_ia32_scalefsh_mask_round (__A, __B,
+          _mm_setzero_ph (),
+          (__mmask8) -1, __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_scalef_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+     __m128h __D, const int __E)
+{
+  return __builtin_ia32_scalefsh_mask_round (__C, __D, __A, __B,
+          __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_scalef_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+      const int __D)
+{
+  return __builtin_ia32_scalefsh_mask_round (__B, __C,
+          _mm_setzero_ph (),
+          __A, __D);
+}
+#else
+#define _mm_scalef_round_sh(A, B, C) (__builtin_ia32_scalefsh_mask_round ((A), (B), _mm_setzero_ph (), (__mmask8)-1, (C)))
+#define _mm_mask_scalef_round_sh(A, B, C, D, E) (__builtin_ia32_scalefsh_mask_round ((C), (D), (A), (B), (E)))
+#define _mm_maskz_scalef_round_sh(A, B, C, D) (__builtin_ia32_scalefsh_mask_round ((B), (C), _mm_setzero_ph (), (A), (D)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_ph (__m512h __A, int __B)
+{
+  return __builtin_ia32_reduceph512_mask_round (__A, __B,
+      _mm512_setzero_ph (),
+      (__mmask32) -1,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_reduce_ph (__m512h __A, __mmask32 __B, __m512h __C, int __D)
+{
+  return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_reduce_ph (__mmask32 __A, __m512h __B, int __C)
+{
+  return __builtin_ia32_reduceph512_mask_round (__B, __C,
+      _mm512_setzero_ph (),
+      __A,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_round_ph (__m512h __A, int __B, const int __C)
+{
+  return __builtin_ia32_reduceph512_mask_round (__A, __B,
+      _mm512_setzero_ph (),
+      (__mmask32) -1, __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_reduce_round_ph (__m512h __A, __mmask32 __B, __m512h __C,
+        int __D, const int __E)
+{
+  return __builtin_ia32_reduceph512_mask_round (__C, __D, __A, __B,
+      __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_reduce_round_ph (__mmask32 __A, __m512h __B, int __C,
+         const int __D)
+{
+  return __builtin_ia32_reduceph512_mask_round (__B, __C,
+      _mm512_setzero_ph (),
+      __A, __D);
+}
+#else
+#define _mm512_reduce_ph(A, B) (__builtin_ia32_reduceph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm512_mask_reduce_ph(A, B, C, D) (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_maskz_reduce_ph(A, B, C) (__builtin_ia32_reduceph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_reduce_round_ph(A, B, C) (__builtin_ia32_reduceph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_reduce_round_ph(A, B, C, D, E) (__builtin_ia32_reduceph512_mask_round ((C), (D), (A), (B), (E)))
+#define _mm512_maskz_reduce_round_ph(A, B, C, D) (__builtin_ia32_reduceph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), (D)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_sh (__m128h __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_reducesh_mask_round (__A, __B, __C,
+          _mm_setzero_ph (),
+          (__mmask8) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_reduce_sh (__m128h __A, __mmask8 __B, __m128h __C,
+      __m128h __D, int __E)
+{
+  return __builtin_ia32_reducesh_mask_round (__C, __D, __E, __A, __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_reduce_sh (__mmask8 __A, __m128h __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_reducesh_mask_round (__B, __C, __D,
+          _mm_setzero_ph (), __A,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_round_sh (__m128h __A, __m128h __B, int __C, const int __D)
+{
+  return __builtin_ia32_reducesh_mask_round (__A, __B, __C,
+          _mm_setzero_ph (),
+          (__mmask8) -1, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_reduce_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+     __m128h __D, int __E, const int __F)
+{
+  return __builtin_ia32_reducesh_mask_round (__C, __D, __E, __A,
+          __B, __F);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_reduce_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+      int __D, const int __E)
+{
+  return __builtin_ia32_reducesh_mask_round (__B, __C, __D,
+          _mm_setzero_ph (),
+          __A, __E);
+}
+#else
+#define _mm_reduce_sh(A, B, C) (__builtin_ia32_reducesh_mask_round ((A), (B), (C), _mm_setzero_ph (), (__mmask8)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm_mask_reduce_sh(A, B, C, D, E) (__builtin_ia32_reducesh_mask_round ((C), (D), (E), (A), (B), _MM_FROUND_CUR_DIRECTION))
+#define _mm_maskz_reduce_sh(A, B, C, D) (__builtin_ia32_reducesh_mask_round ((B), (C), (D), _mm_setzero_ph (), (A), _MM_FROUND_CUR_DIRECTION))
+#define _mm_reduce_round_sh(A, B, C, D) (__builtin_ia32_reducesh_mask_round ((A), (B), (C), _mm_setzero_ph (), (__mmask8)-1, (D)))
+#define _mm_mask_reduce_round_sh(A, B, C, D, E, F) (__builtin_ia32_reducesh_mask_round ((C), (D), (E), (A), (B), (F)))
+#define _mm_maskz_reduce_round_sh(A, B, C, D, E) (__builtin_ia32_reducesh_mask_round ((B), (C), (D), _mm_setzero_ph (), (A), (E)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_roundscale_ph (__m512h __A, int __B)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__A, __B,
+        _mm512_setzero_ph (),
+        (__mmask32) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_roundscale_ph (__m512h __A, __mmask32 __B,
+      __m512h __C, int __D)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_roundscale_ph (__mmask32 __A, __m512h __B, int __C)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__B, __C,
+        _mm512_setzero_ph (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_roundscale_round_ph (__m512h __A, int __B, const int __C)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__A, __B,
+        _mm512_setzero_ph (),
+        (__mmask32) -1,
+        __C);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_roundscale_round_ph (__m512h __A, __mmask32 __B,
+     __m512h __C, int __D, const int __E)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__C, __D, __A,
+        __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_roundscale_round_ph (__mmask32 __A, __m512h __B, int __C,
+      const int __D)
+{
+  return __builtin_ia32_rndscaleph512_mask_round (__B, __C,
+        _mm512_setzero_ph (),
+        __A, __D);
+}
+#else
+#define _mm512_roundscale_ph(A, B) (__builtin_ia32_rndscaleph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm512_mask_roundscale_ph(A, B, C, D) (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_maskz_roundscale_ph(A, B, C) (__builtin_ia32_rndscaleph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_roundscale_round_ph(A, B, C) (__builtin_ia32_rndscaleph512_mask_round ((A), (B), _mm512_setzero_ph (), (__mmask32)-1, (C)))
+#define _mm512_mask_roundscale_round_ph(A, B, C, D, E) (__builtin_ia32_rndscaleph512_mask_round ((C), (D), (A), (B), (E)))
+#define _mm512_maskz_roundscale_round_ph(A, B, C, D) (__builtin_ia32_rndscaleph512_mask_round ((B), (C), _mm512_setzero_ph (), (A), (D)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_roundscale_sh (__m128h __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__A, __B, __C,
+            _mm_setzero_ph (),
+            (__mmask8) -1,
+            _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_roundscale_sh (__m128h __A, __mmask8 __B, __m128h __C,
+   __m128h __D, int __E)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__C, __D, __E, __A, __B,
+            _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_roundscale_sh (__mmask8 __A, __m128h __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__B, __C, __D,
+            _mm_setzero_ph (), __A,
+            _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_roundscale_round_sh (__m128h __A, __m128h __B, int __C, const int __D)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__A, __B, __C,
+            _mm_setzero_ph (),
+            (__mmask8) -1,
+            __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_roundscale_round_sh (__m128h __A, __mmask8 __B, __m128h __C,
+         __m128h __D, int __E, const int __F)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__C, __D, __E,
+            __A, __B, __F);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_roundscale_round_sh (__mmask8 __A, __m128h __B, __m128h __C,
+          int __D, const int __E)
+{
+  return __builtin_ia32_rndscalesh_mask_round (__B, __C, __D,
+            _mm_setzero_ph (),
+            __A, __E);
+}
+#else
+#define _mm_roundscale_sh(A, B, C) (__builtin_ia32_rndscalesh_mask_round ((A), (B), (C), _mm_setzero_ph (), (__mmask8)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm_mask_roundscale_sh(A, B, C, D, E) (__builtin_ia32_rndscalesh_mask_round ((C), (D), (E), (A), (B), _MM_FROUND_CUR_DIRECTION))
+#define _mm_maskz_roundscale_sh(A, B, C, D) (__builtin_ia32_rndscalesh_mask_round ((B), (C), (D), _mm_setzero_ph (), (A), _MM_FROUND_CUR_DIRECTION))
+#define _mm_roundscale_round_sh(A, B, C, D) (__builtin_ia32_rndscalesh_mask_round ((A), (B), (C), _mm_setzero_ph (), (__mmask8)-1, (D)))
+#define _mm_mask_roundscale_round_sh(A, B, C, D, E, F) (__builtin_ia32_rndscalesh_mask_round ((C), (D), (E), (A), (B), (F)))
+#define _mm_maskz_roundscale_round_sh(A, B, C, D, E) (__builtin_ia32_rndscalesh_mask_round ((B), (C), (D), _mm_setzero_ph (), (A), (E)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __mmask8
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fpclass_sh_mask (__m128h __A, const int __imm)
+{
+  return (__mmask8) __builtin_ia32_fpclasssh_mask ((__v8hf) __A, __imm,
+         (__mmask8) -1);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fpclass_sh_mask (__mmask8 __U, __m128h __A, const int __imm)
+{
+  return (__mmask8) __builtin_ia32_fpclasssh_mask ((__v8hf) __A, __imm, __U);
+}
+#else
+#define _mm_fpclass_sh_mask(X, C) ((__mmask8) __builtin_ia32_fpclasssh_mask ((__v8hf) (__m128h) (X), (int) (C), (__mmask8) (-1)))
+#define _mm_mask_fpclass_sh_mask(U, X, C) ((__mmask8) __builtin_ia32_fpclasssh_mask ((__v8hf) (__m128h) (X), (int) (C), (__mmask8) (U)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fpclass_ph_mask (__mmask32 __U, __m512h __A,
+        const int __imm)
+{
+  return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A,
+             __imm, __U);
+}
+extern __inline __mmask32
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fpclass_ph_mask (__m512h __A, const int __imm)
+{
+  return (__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) __A,
+             __imm,
+             (__mmask32) -1);
+}
+#else
+#define _mm512_mask_fpclass_ph_mask(u, x, c) ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), (int) (c),(__mmask8)(u)))
+#define _mm512_fpclass_ph_mask(x, c) ((__mmask32) __builtin_ia32_fpclassph512_mask ((__v32hf) (__m512h) (x), (int) (c),(__mmask8)-1))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getexp_sh (__m128h __A, __m128h __B)
+{
+  return (__m128h)
+    __builtin_ia32_getexpsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+     (__v8hf) _mm_setzero_ph (),
+     (__mmask8) -1,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getexp_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h)
+    __builtin_ia32_getexpsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+     (__v8hf) __W, (__mmask8) __U,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getexp_sh (__mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h)
+    __builtin_ia32_getexpsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+     (__v8hf) _mm_setzero_ph (),
+     (__mmask8) __U,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getexp_ph (__m512h __A)
+{
+  return (__m512h)
+    __builtin_ia32_getexpph512_mask ((__v32hf) __A,
+         (__v32hf) _mm512_setzero_ph (),
+         (__mmask32) -1, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getexp_ph (__m512h __W, __mmask32 __U, __m512h __A)
+{
+  return (__m512h)
+    __builtin_ia32_getexpph512_mask ((__v32hf) __A, (__v32hf) __W,
+         (__mmask32) __U, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getexp_ph (__mmask32 __U, __m512h __A)
+{
+  return (__m512h)
+    __builtin_ia32_getexpph512_mask ((__v32hf) __A,
+         (__v32hf) _mm512_setzero_ph (),
+         (__mmask32) __U, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getexp_round_sh (__m128h __A, __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_getexpsh_mask_round ((__v8hf) __A,
+             (__v8hf) __B,
+             _mm_setzero_ph (),
+             (__mmask8) -1,
+             __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getexp_round_sh (__m128h __W, __mmask8 __U, __m128h __A,
+     __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_getexpsh_mask_round ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __W,
+             (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getexp_round_sh (__mmask8 __U, __m128h __A, __m128h __B,
+      const int __R)
+{
+  return (__m128h) __builtin_ia32_getexpsh_mask_round ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf)
+             _mm_setzero_ph (),
+             (__mmask8) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getexp_round_ph (__m512h __A, const int __R)
+{
+  return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A,
+          (__v32hf)
+          _mm512_setzero_ph (),
+          (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getexp_round_ph (__m512h __W, __mmask32 __U, __m512h __A,
+        const int __R)
+{
+  return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A,
+          (__v32hf) __W,
+          (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getexp_round_ph (__mmask32 __U, __m512h __A, const int __R)
+{
+  return (__m512h) __builtin_ia32_getexpph512_mask ((__v32hf) __A,
+          (__v32hf)
+          _mm512_setzero_ph (),
+          (__mmask32) __U, __R);
+}
+#else
+#define _mm_getexp_round_sh(A, B, R) ((__m128h)__builtin_ia32_getexpsh_mask_round((__v8hf)(__m128h)(A), (__v8hf)(__m128h)(B), (__v8hf)_mm_setzero_ph(), (__mmask8)-1, R))
+#define _mm_mask_getexp_round_sh(W, U, A, B, C) (__m128h)__builtin_ia32_getexpsh_mask_round(A, B, W, U, C)
+#define _mm_maskz_getexp_round_sh(U, A, B, C) (__m128h)__builtin_ia32_getexpsh_mask_round(A, B, (__v8hf)_mm_setzero_ph(), U, C)
+#define _mm512_getexp_round_ph(A, R) ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), (__v32hf)_mm512_setzero_ph(), (__mmask32)-1, R))
+#define _mm512_mask_getexp_round_ph(W, U, A, R) ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), (__v32hf)(__m512h)(W), (__mmask32)(U), R))
+#define _mm512_maskz_getexp_round_ph(U, A, R) ((__m512h)__builtin_ia32_getexpph512_mask((__v32hf)(__m512h)(A), (__v32hf)_mm512_setzero_ph(), (__mmask32)(U), R))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getmant_sh (__m128h __A, __m128h __B,
+  _MM_MANTISSA_NORM_ENUM __C,
+  _MM_MANTISSA_SIGN_ENUM __D)
+{
+  return (__m128h)
+    __builtin_ia32_getmantsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+      (__D << 2) | __C, _mm_setzero_ph (),
+      (__mmask8) -1,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getmant_sh (__m128h __W, __mmask8 __U, __m128h __A,
+       __m128h __B, _MM_MANTISSA_NORM_ENUM __C,
+       _MM_MANTISSA_SIGN_ENUM __D)
+{
+  return (__m128h)
+    __builtin_ia32_getmantsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+      (__D << 2) | __C, (__v8hf) __W,
+      __U, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getmant_sh (__mmask8 __U, __m128h __A, __m128h __B,
+        _MM_MANTISSA_NORM_ENUM __C,
+        _MM_MANTISSA_SIGN_ENUM __D)
+{
+  return (__m128h)
+    __builtin_ia32_getmantsh_mask_round ((__v8hf) __A, (__v8hf) __B,
+      (__D << 2) | __C,
+      (__v8hf) _mm_setzero_ph(),
+      __U, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getmant_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B,
+     _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           _mm512_setzero_ph (),
+           (__mmask32) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getmant_ph (__m512h __W, __mmask32 __U, __m512h __A,
+   _MM_MANTISSA_NORM_ENUM __B,
+   _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           (__v32hf) __W, __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getmant_ph (__mmask32 __U, __m512h __A,
+    _MM_MANTISSA_NORM_ENUM __B,
+    _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           (__v32hf)
+           _mm512_setzero_ph (),
+           __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getmant_round_sh (__m128h __A, __m128h __B,
+        _MM_MANTISSA_NORM_ENUM __C,
+        _MM_MANTISSA_SIGN_ENUM __D, const int __R)
+{
+  return (__m128h) __builtin_ia32_getmantsh_mask_round ((__v8hf) __A,
+       (__v8hf) __B,
+       (__D << 2) | __C,
+       _mm_setzero_ph (),
+       (__mmask8) -1,
+       __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getmant_round_sh (__m128h __W, __mmask8 __U, __m128h __A,
+      __m128h __B, _MM_MANTISSA_NORM_ENUM __C,
+      _MM_MANTISSA_SIGN_ENUM __D, const int __R)
+{
+  return (__m128h) __builtin_ia32_getmantsh_mask_round ((__v8hf) __A,
+       (__v8hf) __B,
+       (__D << 2) | __C,
+       (__v8hf) __W,
+       __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getmant_round_sh (__mmask8 __U, __m128h __A, __m128h __B,
+       _MM_MANTISSA_NORM_ENUM __C,
+       _MM_MANTISSA_SIGN_ENUM __D, const int __R)
+{
+  return (__m128h) __builtin_ia32_getmantsh_mask_round ((__v8hf) __A,
+       (__v8hf) __B,
+       (__D << 2) | __C,
+       (__v8hf)
+       _mm_setzero_ph(),
+       __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_getmant_round_ph (__m512h __A, _MM_MANTISSA_NORM_ENUM __B,
+    _MM_MANTISSA_SIGN_ENUM __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           _mm512_setzero_ph (),
+           (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_getmant_round_ph (__m512h __W, __mmask32 __U, __m512h __A,
+         _MM_MANTISSA_NORM_ENUM __B,
+         _MM_MANTISSA_SIGN_ENUM __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           (__v32hf) __W, __U,
+           __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_getmant_round_ph (__mmask32 __U, __m512h __A,
+          _MM_MANTISSA_NORM_ENUM __B,
+          _MM_MANTISSA_SIGN_ENUM __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_getmantph512_mask ((__v32hf) __A,
+           (__C << 2) | __B,
+           (__v32hf)
+           _mm512_setzero_ph (),
+           __U, __R);
+}
+#else
+#define _mm512_getmant_ph(X, B, C) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h) _mm512_setzero_ph(), (__mmask32)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm512_mask_getmant_ph(W, U, X, B, C) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h)(W), (__mmask32)(U), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_maskz_getmant_ph(U, X, B, C) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h) _mm512_setzero_ph(), (__mmask32)(U), _MM_FROUND_CUR_DIRECTION))
+#define _mm_getmant_sh(X, Y, C, D) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h) _mm_setzero_ph (), (__mmask8)-1, _MM_FROUND_CUR_DIRECTION))
+#define _mm_mask_getmant_sh(W, U, X, Y, C, D) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h)(W), (__mmask8)(U), _MM_FROUND_CUR_DIRECTION))
+#define _mm_maskz_getmant_sh(U, X, Y, C, D) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h) _mm_setzero_ph(), (__mmask8)(U), _MM_FROUND_CUR_DIRECTION))
+#define _mm512_getmant_round_ph(X, B, C, R) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h) _mm512_setzero_ph(), (__mmask32)-1, (R)))
+#define _mm512_mask_getmant_round_ph(W, U, X, B, C, R) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h)(W), (__mmask32)(U), (R)))
+#define _mm512_maskz_getmant_round_ph(U, X, B, C, R) ((__m512h)__builtin_ia32_getmantph512_mask ((__v32hf)(__m512h)(X), (int)(((C)<<2) | (B)), (__v32hf)(__m512h) _mm512_setzero_ph(), (__mmask32)(U), (R)))
+#define _mm_getmant_round_sh(X, Y, C, D, R) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h) _mm_setzero_ph (), (__mmask8)-1, (R)))
+#define _mm_mask_getmant_round_sh(W, U, X, Y, C, D, R) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h)(W), (__mmask8)(U), (R)))
+#define _mm_maskz_getmant_round_sh(U, X, Y, C, D, R) ((__m128h)__builtin_ia32_getmantsh_mask_round ((__v8hf)(__m128h)(X), (__v8hf)(__m128h)(Y), (int)(((D)<<2) | (C)), (__v8hf)(__m128h) _mm_setzero_ph(), (__mmask8)(U), (R)))
+#endif
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi16_si128 (short __A)
+{
+  return _mm_set_epi16 (0, 0, 0, 0, 0, 0, 0, __A);
+}
+extern __inline short
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi128_si16 (__m128i __A)
+{
+  return __builtin_ia32_vec_ext_v8hi ((__v8hi)__A, 0);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_load_sh (__m128h __A, __mmask8 __B, _Float16 const* __C)
+{
+  return __builtin_ia32_loadsh_mask (__C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_load_sh (__mmask8 __A, _Float16 const* __B)
+{
+  return __builtin_ia32_loadsh_mask (__B, _mm_setzero_ph (), __A);
+}
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_store_sh (_Float16 const* __A, __mmask8 __B, __m128h __C)
+{
+  __builtin_ia32_storesh_mask (__A, __C, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_move_sh (__m128h __A, __m128h __B)
+{
+  __A[0] = __B[0];
+  return __A;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_move_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_vmovsh_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_move_sh (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_vmovsh_mask (__B, __C, _mm_setzero_ph (), __A);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epi32 (__m256h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__A,
+         (__v16si)
+         _mm512_setzero_si512 (),
+         (__mmask16) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epi32 (__m512i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__C,
+         (__v16si) __A,
+         __B,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epi32 (__mmask16 __A, __m256h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__B,
+         (__v16si)
+         _mm512_setzero_si512 (),
+         __A,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epi32 (__m256h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__A,
+         (__v16si)
+         _mm512_setzero_si512 (),
+         (__mmask16) -1,
+         __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epi32 (__m512i __A, __mmask16 __B, __m256h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__C,
+         (__v16si) __A,
+         __B,
+         __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epi32 (__mmask16 __A, __m256h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2dq512_mask_round (__B,
+         (__v16si)
+         _mm512_setzero_si512 (),
+         __A,
+         __C);
+}
+#else
+#define _mm512_cvt_roundph_epi32(A, B) ((__m512i) __builtin_ia32_vcvtph2dq512_mask_round ((A), (__v16si) _mm512_setzero_si512 (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvt_roundph_epi32(A, B, C, D) ((__m512i) __builtin_ia32_vcvtph2dq512_mask_round ((C), (__v16si)(A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epi32(A, B, C) ((__m512i) __builtin_ia32_vcvtph2dq512_mask_round ((B), (__v16si) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epu32 (__m256h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__A,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          (__mmask16) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epu32 (__m512i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__C,
+          (__v16si) __A,
+          __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epu32 (__mmask16 __A, __m256h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__B,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          __A,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epu32 (__m256h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__A,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          (__mmask16) -1,
+          __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epu32 (__m512i __A, __mmask16 __B, __m256h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__C,
+          (__v16si) __A,
+          __B,
+          __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epu32 (__mmask16 __A, __m256h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2udq512_mask_round (__B,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          __A,
+          __C);
+}
+#else
+#define _mm512_cvt_roundph_epu32(A, B) ((__m512i) __builtin_ia32_vcvtph2udq512_mask_round ((A), (__v16si) _mm512_setzero_si512 (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvt_roundph_epu32(A, B, C, D) ((__m512i) __builtin_ia32_vcvtph2udq512_mask_round ((C), (__v16si)(A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epu32(A, B, C) ((__m512i) __builtin_ia32_vcvtph2udq512_mask_round ((B), (__v16si) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epi32 (__m256h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__A,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          (__mmask16) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epi32 (__m512i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__C,
+          (__v16si) __A,
+          __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epi32 (__mmask16 __A, __m256h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__B,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          __A,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epi32 (__m256h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__A,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          (__mmask16) -1,
+          __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epi32 (__m512i __A, __mmask16 __B,
+    __m256h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__C,
+          (__v16si) __A,
+          __B,
+          __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epi32 (__mmask16 __A, __m256h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2dq512_mask_round (__B,
+          (__v16si)
+          _mm512_setzero_si512 (),
+          __A,
+          __C);
+}
+#else
+#define _mm512_cvtt_roundph_epi32(A, B) ((__m512i) __builtin_ia32_vcvttph2dq512_mask_round ((A), (__v16si) (_mm512_setzero_si512 ()), (__mmask16)(-1), (B)))
+#define _mm512_mask_cvtt_roundph_epi32(A, B, C, D) ((__m512i) __builtin_ia32_vcvttph2dq512_mask_round ((C), (__v16si)(A), (B), (D)))
+#define _mm512_maskz_cvtt_roundph_epi32(A, B, C) ((__m512i) __builtin_ia32_vcvttph2dq512_mask_round ((B), (__v16si) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epu32 (__m256h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__A,
+           (__v16si)
+           _mm512_setzero_si512 (),
+           (__mmask16) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epu32 (__m512i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__C,
+           (__v16si) __A,
+           __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epu32 (__mmask16 __A, __m256h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__B,
+           (__v16si)
+           _mm512_setzero_si512 (),
+           __A,
+           _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epu32 (__m256h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__A,
+           (__v16si)
+           _mm512_setzero_si512 (),
+           (__mmask16) -1,
+           __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epu32 (__m512i __A, __mmask16 __B,
+    __m256h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__C,
+           (__v16si) __A,
+           __B,
+           __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epu32 (__mmask16 __A, __m256h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2udq512_mask_round (__B,
+           (__v16si)
+           _mm512_setzero_si512 (),
+           __A,
+           __C);
+}
+#else
+#define _mm512_cvtt_roundph_epu32(A, B) ((__m512i) __builtin_ia32_vcvttph2udq512_mask_round ((A), (__v16si) _mm512_setzero_si512 (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvtt_roundph_epu32(A, B, C, D) ((__m512i) __builtin_ia32_vcvttph2udq512_mask_round ((C), (__v16si)(A), (B), (D)))
+#define _mm512_maskz_cvtt_roundph_epu32(A, B, C) ((__m512i) __builtin_ia32_vcvttph2udq512_mask_round ((B), (__v16si) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtepi32_ph (__m512i __A)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __A,
+       _mm256_setzero_ph (),
+       (__mmask16) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepi32_ph (__m256h __A, __mmask16 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __C,
+       __A,
+       __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepi32_ph (__mmask16 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __B,
+       _mm256_setzero_ph (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepi32_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __A,
+       _mm256_setzero_ph (),
+       (__mmask16) -1,
+       __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepi32_ph (__m256h __A, __mmask16 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __C,
+       __A,
+       __B,
+       __D);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepi32_ph (__mmask16 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtdq2ph512_mask_round ((__v16si) __B,
+       _mm256_setzero_ph (),
+       __A,
+       __C);
+}
+#else
+#define _mm512_cvt_roundepi32_ph(A, B) (__builtin_ia32_vcvtdq2ph512_mask_round ((__v16si)(A), _mm256_setzero_ph (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvt_roundepi32_ph(A, B, C, D) (__builtin_ia32_vcvtdq2ph512_mask_round ((__v16si)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundepi32_ph(A, B, C) (__builtin_ia32_vcvtdq2ph512_mask_round ((__v16si)(B), _mm256_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtepu32_ph (__m512i __A)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __A,
+        _mm256_setzero_ph (),
+        (__mmask16) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepu32_ph (__m256h __A, __mmask16 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __C,
+        __A,
+        __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepu32_ph (__mmask16 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __B,
+        _mm256_setzero_ph (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepu32_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __A,
+        _mm256_setzero_ph (),
+        (__mmask16) -1,
+        __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepu32_ph (__m256h __A, __mmask16 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __C,
+        __A,
+        __B,
+        __D);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepu32_ph (__mmask16 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtudq2ph512_mask_round ((__v16si) __B,
+        _mm256_setzero_ph (),
+        __A,
+        __C);
+}
+#else
+#define _mm512_cvt_roundepu32_ph(A, B) (__builtin_ia32_vcvtudq2ph512_mask_round ((__v16si)(A), _mm256_setzero_ph (), (__mmask16)-1, B))
+#define _mm512_mask_cvt_roundepu32_ph(A, B, C, D) (__builtin_ia32_vcvtudq2ph512_mask_round ((__v16si)C, A, B, D))
+#define _mm512_maskz_cvt_roundepu32_ph(A, B, C) (__builtin_ia32_vcvtudq2ph512_mask_round ((__v16si)B, _mm256_setzero_ph (), A, C))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epi64 (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__A,
+       _mm512_setzero_si512 (),
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epi64 (__m512i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__C, __A, __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__B,
+       _mm512_setzero_si512 (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epi64 (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__A,
+       _mm512_setzero_si512 (),
+       (__mmask8) -1,
+       __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epi64 (__m512i __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epi64 (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_vcvtph2qq512_mask_round (__B,
+       _mm512_setzero_si512 (),
+       __A,
+       __C);
+}
+#else
+#define _mm512_cvt_roundph_epi64(A, B) (__builtin_ia32_vcvtph2qq512_mask_round ((A), _mm512_setzero_si512 (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundph_epi64(A, B, C, D) (__builtin_ia32_vcvtph2qq512_mask_round ((C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epi64(A, B, C) (__builtin_ia32_vcvtph2qq512_mask_round ((B), _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__A,
+        _mm512_setzero_si512 (),
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epu64 (__m512i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__C, __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__B,
+        _mm512_setzero_si512 (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epu64 (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__A,
+        _mm512_setzero_si512 (),
+        (__mmask8) -1,
+        __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epu64 (__m512i __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epu64 (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_vcvtph2uqq512_mask_round (__B,
+        _mm512_setzero_si512 (),
+        __A,
+        __C);
+}
+#else
+#define _mm512_cvt_roundph_epu64(A, B) (__builtin_ia32_vcvtph2uqq512_mask_round ((A), _mm512_setzero_si512 (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundph_epu64(A, B, C, D) (__builtin_ia32_vcvtph2uqq512_mask_round ((C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epu64(A, B, C) (__builtin_ia32_vcvtph2uqq512_mask_round ((B), _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epi64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__A,
+        _mm512_setzero_si512 (),
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epi64 (__m512i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__C, __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__B,
+        _mm512_setzero_si512 (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epi64 (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__A,
+        _mm512_setzero_si512 (),
+        (__mmask8) -1,
+        __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epi64 (__m512i __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epi64 (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_vcvttph2qq512_mask_round (__B,
+        _mm512_setzero_si512 (),
+        __A,
+        __C);
+}
+#else
+#define _mm512_cvtt_roundph_epi64(A, B) (__builtin_ia32_vcvttph2qq512_mask_round ((A), _mm512_setzero_si512 (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvtt_roundph_epi64(A, B, C, D) __builtin_ia32_vcvttph2qq512_mask_round ((C), (A), (B), (D))
+#define _mm512_maskz_cvtt_roundph_epi64(A, B, C) (__builtin_ia32_vcvttph2qq512_mask_round ((B), _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__A,
+         _mm512_setzero_si512 (),
+         (__mmask8) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epu64 (__m512i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__C, __A, __B,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__B,
+         _mm512_setzero_si512 (),
+         __A,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epu64 (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__A,
+         _mm512_setzero_si512 (),
+         (__mmask8) -1,
+         __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epu64 (__m512i __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epu64 (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_vcvttph2uqq512_mask_round (__B,
+         _mm512_setzero_si512 (),
+         __A,
+         __C);
+}
+#else
+#define _mm512_cvtt_roundph_epu64(A, B) (__builtin_ia32_vcvttph2uqq512_mask_round ((A), _mm512_setzero_si512 (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvtt_roundph_epu64(A, B, C, D) __builtin_ia32_vcvttph2uqq512_mask_round ((C), (A), (B), (D))
+#define _mm512_maskz_cvtt_roundph_epu64(A, B, C) (__builtin_ia32_vcvttph2uqq512_mask_round ((B), _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtepi64_ph (__m512i __A)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __A,
+       _mm_setzero_ph (),
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepi64_ph (__m128h __A, __mmask8 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __C,
+       __A,
+       __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepi64_ph (__mmask8 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __B,
+       _mm_setzero_ph (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepi64_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __A,
+       _mm_setzero_ph (),
+       (__mmask8) -1,
+       __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepi64_ph (__m128h __A, __mmask8 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __C,
+       __A,
+       __B,
+       __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepi64_ph (__mmask8 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtqq2ph512_mask_round ((__v8di) __B,
+       _mm_setzero_ph (),
+       __A,
+       __C);
+}
+#else
+#define _mm512_cvt_roundepi64_ph(A, B) (__builtin_ia32_vcvtqq2ph512_mask_round ((__v8di)(A), _mm_setzero_ph (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundepi64_ph(A, B, C, D) (__builtin_ia32_vcvtqq2ph512_mask_round ((__v8di)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundepi64_ph(A, B, C) (__builtin_ia32_vcvtqq2ph512_mask_round ((__v8di)(B), _mm_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtepu64_ph (__m512i __A)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepu64_ph (__m128h __A, __mmask8 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __C,
+        __A,
+        __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepu64_ph (__mmask8 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __B,
+        _mm_setzero_ph (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepu64_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1,
+        __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepu64_ph (__m128h __A, __mmask8 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __C,
+        __A,
+        __B,
+        __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepu64_ph (__mmask8 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di) __B,
+        _mm_setzero_ph (),
+        __A,
+        __C);
+}
+#else
+#define _mm512_cvt_roundepu64_ph(A, B) (__builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di)(A), _mm_setzero_ph (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundepu64_ph(A, B, C, D) (__builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundepu64_ph(A, B, C) (__builtin_ia32_vcvtuqq2ph512_mask_round ((__v8di)(B), _mm_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epi16 (__m512h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__A,
+           (__v32hi)
+           _mm512_setzero_si512 (),
+           (__mmask32) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epi16 (__m512i __A, __mmask32 __B, __m512h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__C,
+           (__v32hi) __A,
+           __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epi16 (__mmask32 __A, __m512h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__B,
+           (__v32hi)
+           _mm512_setzero_si512 (),
+           __A,
+           _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epi16 (__m512h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__A,
+           (__v32hi)
+           _mm512_setzero_si512 (),
+           (__mmask32) -1,
+           __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epi16 (__m512i __A, __mmask32 __B, __m512h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__C,
+           (__v32hi) __A,
+           __B,
+           __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epi16 (__mmask32 __A, __m512h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2w512_mask_round (__B,
+           (__v32hi)
+           _mm512_setzero_si512 (),
+           __A,
+           __C);
+}
+#else
+#define _mm512_cvt_roundph_epi16(A, B) ((__m512i)__builtin_ia32_vcvtph2w512_mask_round ((A), (__v32hi) _mm512_setzero_si512 (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvt_roundph_epi16(A, B, C, D) ((__m512i)__builtin_ia32_vcvtph2w512_mask_round ((C), (__v32hi)(A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epi16(A, B, C) ((__m512i)__builtin_ia32_vcvtph2w512_mask_round ((B), (__v32hi) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_epu16 (__m512h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__A,
+            (__v32hi)
+            _mm512_setzero_si512 (),
+            (__mmask32) -1,
+            _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_epu16 (__m512i __A, __mmask32 __B, __m512h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__C, (__v32hi) __A, __B,
+            _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_epu16 (__mmask32 __A, __m512h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__B,
+            (__v32hi)
+            _mm512_setzero_si512 (),
+            __A,
+            _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_epu16 (__m512h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__A,
+            (__v32hi)
+            _mm512_setzero_si512 (),
+            (__mmask32) -1,
+            __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_epu16 (__m512i __A, __mmask32 __B, __m512h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__C, (__v32hi) __A, __B, __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_epu16 (__mmask32 __A, __m512h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvtph2uw512_mask_round (__B,
+            (__v32hi)
+            _mm512_setzero_si512 (),
+            __A,
+            __C);
+}
+#else
+#define _mm512_cvt_roundph_epu16(A, B) ((__m512i) __builtin_ia32_vcvtph2uw512_mask_round ((A), (__v32hi) _mm512_setzero_si512 (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvt_roundph_epu16(A, B, C, D) ((__m512i) __builtin_ia32_vcvtph2uw512_mask_round ((C), (__v32hi)(A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_epu16(A, B, C) ((__m512i) __builtin_ia32_vcvtph2uw512_mask_round ((B), (__v32hi) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epi16 (__m512h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__A,
+         (__v32hi)
+         _mm512_setzero_si512 (),
+         (__mmask32) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epi16 (__m512i __A, __mmask32 __B, __m512h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__C,
+         (__v32hi) __A,
+         __B,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epi16 (__mmask32 __A, __m512h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__B,
+         (__v32hi)
+         _mm512_setzero_si512 (),
+         __A,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epi16 (__m512h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__A,
+         (__v32hi)
+         _mm512_setzero_si512 (),
+         (__mmask32) -1,
+         __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epi16 (__m512i __A, __mmask32 __B,
+    __m512h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__C,
+         (__v32hi) __A,
+         __B,
+         __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epi16 (__mmask32 __A, __m512h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2w512_mask_round (__B,
+         (__v32hi)
+         _mm512_setzero_si512 (),
+         __A,
+         __C);
+}
+#else
+#define _mm512_cvtt_roundph_epi16(A, B) ((__m512i) __builtin_ia32_vcvttph2w512_mask_round ((A), (__v32hi) _mm512_setzero_si512 (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvtt_roundph_epi16(A, B, C, D) ((__m512i) __builtin_ia32_vcvttph2w512_mask_round ((C), (__v32hi)(A), (B), (D)))
+#define _mm512_maskz_cvtt_roundph_epi16(A, B, C) ((__m512i) __builtin_ia32_vcvttph2w512_mask_round ((B), (__v32hi) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvttph_epu16 (__m512h __A)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__A,
+          (__v32hi)
+          _mm512_setzero_si512 (),
+          (__mmask32) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvttph_epu16 (__m512i __A, __mmask32 __B, __m512h __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__C,
+          (__v32hi) __A,
+          __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvttph_epu16 (__mmask32 __A, __m512h __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__B,
+          (__v32hi)
+          _mm512_setzero_si512 (),
+          __A,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtt_roundph_epu16 (__m512h __A, int __B)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__A,
+          (__v32hi)
+          _mm512_setzero_si512 (),
+          (__mmask32) -1,
+          __B);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtt_roundph_epu16 (__m512i __A, __mmask32 __B,
+    __m512h __C, int __D)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__C,
+          (__v32hi) __A,
+          __B,
+          __D);
+}
+extern __inline __m512i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtt_roundph_epu16 (__mmask32 __A, __m512h __B, int __C)
+{
+  return (__m512i)
+    __builtin_ia32_vcvttph2uw512_mask_round (__B,
+          (__v32hi)
+          _mm512_setzero_si512 (),
+          __A,
+          __C);
+}
+#else
+#define _mm512_cvtt_roundph_epu16(A, B) ((__m512i) __builtin_ia32_vcvttph2uw512_mask_round ((A), (__v32hi) _mm512_setzero_si512 (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvtt_roundph_epu16(A, B, C, D) ((__m512i) __builtin_ia32_vcvttph2uw512_mask_round ((C), (__v32hi)(A), (B), (D)))
+#define _mm512_maskz_cvtt_roundph_epu16(A, B, C) ((__m512i) __builtin_ia32_vcvttph2uw512_mask_round ((B), (__v32hi) _mm512_setzero_si512 (), (A), (C)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtepi16_ph (__m512i __A)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __A,
+      _mm512_setzero_ph (),
+      (__mmask32) -1,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepi16_ph (__m512h __A, __mmask32 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __C,
+      __A,
+      __B,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepi16_ph (__mmask32 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __B,
+      _mm512_setzero_ph (),
+      __A,
+      _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepi16_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __A,
+      _mm512_setzero_ph (),
+      (__mmask32) -1,
+      __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepi16_ph (__m512h __A, __mmask32 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __C,
+      __A,
+      __B,
+      __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepi16_ph (__mmask32 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtw2ph512_mask_round ((__v32hi) __B,
+      _mm512_setzero_ph (),
+      __A,
+      __C);
+}
+#else
+#define _mm512_cvt_roundepi16_ph(A, B) (__builtin_ia32_vcvtw2ph512_mask_round ((__v32hi)(A), _mm512_setzero_ph (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvt_roundepi16_ph(A, B, C, D) (__builtin_ia32_vcvtw2ph512_mask_round ((__v32hi)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundepi16_ph(A, B, C) (__builtin_ia32_vcvtw2ph512_mask_round ((__v32hi)(B), _mm512_setzero_ph (), (A), (C)))
+#endif
+  extern __inline __m512h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+  _mm512_cvtepu16_ph (__m512i __A)
+  {
+    return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __A,
+         _mm512_setzero_ph (),
+         (__mmask32) -1,
+         _MM_FROUND_CUR_DIRECTION);
+  }
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtepu16_ph (__m512h __A, __mmask32 __B, __m512i __C)
+{
+  return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __C,
+       __A,
+       __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtepu16_ph (__mmask32 __A, __m512i __B)
+{
+  return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __B,
+       _mm512_setzero_ph (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundepu16_ph (__m512i __A, int __B)
+{
+  return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __A,
+       _mm512_setzero_ph (),
+       (__mmask32) -1,
+       __B);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundepu16_ph (__m512h __A, __mmask32 __B, __m512i __C, int __D)
+{
+  return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __C,
+       __A,
+       __B,
+       __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundepu16_ph (__mmask32 __A, __m512i __B, int __C)
+{
+  return __builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi) __B,
+       _mm512_setzero_ph (),
+       __A,
+       __C);
+}
+#else
+#define _mm512_cvt_roundepu16_ph(A, B) (__builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi)(A), _mm512_setzero_ph (), (__mmask32)-1, (B)))
+#define _mm512_mask_cvt_roundepu16_ph(A, B, C, D) (__builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundepu16_ph(A, B, C) (__builtin_ia32_vcvtuw2ph512_mask_round ((__v32hi)(B), _mm512_setzero_ph (), (A), (C)))
+#endif
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_i32 (__m128h __A)
+{
+  return (int) __builtin_ia32_vcvtsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline unsigned
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_u32 (__m128h __A)
+{
+  return (int) __builtin_ia32_vcvtsh2usi32_round (__A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_i32 (__m128h __A, const int __R)
+{
+  return (int) __builtin_ia32_vcvtsh2si32_round (__A, __R);
+}
+extern __inline unsigned
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_u32 (__m128h __A, const int __R)
+{
+  return (int) __builtin_ia32_vcvtsh2usi32_round (__A, __R);
+}
+#else
+#define _mm_cvt_roundsh_i32(A, B) ((int)__builtin_ia32_vcvtsh2si32_round ((A), (B)))
+#define _mm_cvt_roundsh_u32(A, B) ((int)__builtin_ia32_vcvtsh2usi32_round ((A), (B)))
+#endif
+#ifdef __x86_64__
+extern __inline long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_i64 (__m128h __A)
+{
+  return (long long)
+    __builtin_ia32_vcvtsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline unsigned long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_u64 (__m128h __A)
+{
+  return (long long)
+    __builtin_ia32_vcvtsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_i64 (__m128h __A, const int __R)
+{
+  return (long long) __builtin_ia32_vcvtsh2si64_round (__A, __R);
+}
+extern __inline unsigned long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_u64 (__m128h __A, const int __R)
+{
+  return (long long) __builtin_ia32_vcvtsh2usi64_round (__A, __R);
+}
+#else
+#define _mm_cvt_roundsh_i64(A, B) ((long long)__builtin_ia32_vcvtsh2si64_round ((A), (B)))
+#define _mm_cvt_roundsh_u64(A, B) ((long long)__builtin_ia32_vcvtsh2usi64_round ((A), (B)))
+#endif
+#endif
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsh_i32 (__m128h __A)
+{
+  return (int)
+    __builtin_ia32_vcvttsh2si32_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline unsigned
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsh_u32 (__m128h __A)
+{
+  return (int)
+    __builtin_ia32_vcvttsh2usi32_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_roundsh_i32 (__m128h __A, const int __R)
+{
+  return (int) __builtin_ia32_vcvttsh2si32_round (__A, __R);
+}
+extern __inline unsigned
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_roundsh_u32 (__m128h __A, const int __R)
+{
+  return (int) __builtin_ia32_vcvttsh2usi32_round (__A, __R);
+}
+#else
+#define _mm_cvtt_roundsh_i32(A, B) ((int)__builtin_ia32_vcvttsh2si32_round ((A), (B)))
+#define _mm_cvtt_roundsh_u32(A, B) ((int)__builtin_ia32_vcvttsh2usi32_round ((A), (B)))
+#endif
+#ifdef __x86_64__
+extern __inline long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsh_i64 (__m128h __A)
+{
+  return (long long)
+    __builtin_ia32_vcvttsh2si64_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline unsigned long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttsh_u64 (__m128h __A)
+{
+  return (long long)
+    __builtin_ia32_vcvttsh2usi64_round (__A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_roundsh_i64 (__m128h __A, const int __R)
+{
+  return (long long) __builtin_ia32_vcvttsh2si64_round (__A, __R);
+}
+extern __inline unsigned long long
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_roundsh_u64 (__m128h __A, const int __R)
+{
+  return (long long) __builtin_ia32_vcvttsh2usi64_round (__A, __R);
+}
+#else
+#define _mm_cvtt_roundsh_i64(A, B) ((long long)__builtin_ia32_vcvttsh2si64_round ((A), (B)))
+#define _mm_cvtt_roundsh_u64(A, B) ((long long)__builtin_ia32_vcvttsh2usi64_round ((A), (B)))
+#endif
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvti32_sh (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvtsi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtu32_sh (__m128h __A, unsigned int __B)
+{
+  return __builtin_ia32_vcvtusi2sh32_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundi32_sh (__m128h __A, int __B, const int __R)
+{
+  return __builtin_ia32_vcvtsi2sh32_round (__A, __B, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundu32_sh (__m128h __A, unsigned int __B, const int __R)
+{
+  return __builtin_ia32_vcvtusi2sh32_round (__A, __B, __R);
+}
+#else
+#define _mm_cvt_roundi32_sh(A, B, C) (__builtin_ia32_vcvtsi2sh32_round ((A), (B), (C)))
+#define _mm_cvt_roundu32_sh(A, B, C) (__builtin_ia32_vcvtusi2sh32_round ((A), (B), (C)))
+#endif
+#ifdef __x86_64__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvti64_sh (__m128h __A, long long __B)
+{
+  return __builtin_ia32_vcvtsi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtu64_sh (__m128h __A, unsigned long long __B)
+{
+  return __builtin_ia32_vcvtusi2sh64_round (__A, __B, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundi64_sh (__m128h __A, long long __B, const int __R)
+{
+  return __builtin_ia32_vcvtsi2sh64_round (__A, __B, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundu64_sh (__m128h __A, unsigned long long __B, const int __R)
+{
+  return __builtin_ia32_vcvtusi2sh64_round (__A, __B, __R);
+}
+#else
+#define _mm_cvt_roundi64_sh(A, B, C) (__builtin_ia32_vcvtsi2sh64_round ((A), (B), (C)))
+#define _mm_cvt_roundu64_sh(A, B, C) (__builtin_ia32_vcvtusi2sh64_round ((A), (B), (C)))
+#endif
+#endif
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtph_pd (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__A,
+       _mm512_setzero_pd (),
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtph_pd (__m512d __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__C, __A, __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtph_pd (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__B,
+       _mm512_setzero_pd (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundph_pd (__m128h __A, int __B)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__A,
+       _mm512_setzero_pd (),
+       (__mmask8) -1,
+       __B);
+}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundph_pd (__m512d __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundph_pd (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_vcvtph2pd512_mask_round (__B,
+       _mm512_setzero_pd (),
+       __A,
+       __C);
+}
+#else
+#define _mm512_cvt_roundph_pd(A, B) (__builtin_ia32_vcvtph2pd512_mask_round ((A), _mm512_setzero_pd (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundph_pd(A, B, C, D) (__builtin_ia32_vcvtph2pd512_mask_round ((C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundph_pd(A, B, C) (__builtin_ia32_vcvtph2pd512_mask_round ((B), _mm512_setzero_pd (), (A), (C)))
+#endif
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtxph_ps (__m256h __A)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__A,
+        _mm512_setzero_ps (),
+        (__mmask16) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtxph_ps (__m512 __A, __mmask16 __B, __m256h __C)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__C, __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtxph_ps (__mmask16 __A, __m256h __B)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__B,
+        _mm512_setzero_ps (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtx_roundph_ps (__m256h __A, int __B)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__A,
+        _mm512_setzero_ps (),
+        (__mmask16) -1,
+        __B);
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtx_roundph_ps (__m512 __A, __mmask16 __B, __m256h __C, int __D)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__C, __A, __B, __D);
+}
+extern __inline __m512
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtx_roundph_ps (__mmask16 __A, __m256h __B, int __C)
+{
+  return __builtin_ia32_vcvtph2psx512_mask_round (__B,
+        _mm512_setzero_ps (),
+        __A,
+        __C);
+}
+#else
+#define _mm512_cvtx_roundph_ps(A, B) (__builtin_ia32_vcvtph2psx512_mask_round ((A), _mm512_setzero_ps (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvtx_roundph_ps(A, B, C, D) (__builtin_ia32_vcvtph2psx512_mask_round ((C), (A), (B), (D)))
+#define _mm512_maskz_cvtx_roundph_ps(A, B, C) (__builtin_ia32_vcvtph2psx512_mask_round ((B), _mm512_setzero_ps (), (A), (C)))
+#endif
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtxps_ph (__m512 __A)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __A,
+        _mm256_setzero_ph (),
+        (__mmask16) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtxps_ph (__m256h __A, __mmask16 __B, __m512 __C)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __C,
+        __A, __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtxps_ph (__mmask16 __A, __m512 __B)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __B,
+        _mm256_setzero_ph (),
+        __A,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtx_roundps_ph (__m512 __A, int __B)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __A,
+        _mm256_setzero_ph (),
+        (__mmask16) -1,
+        __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtx_roundps_ph (__m256h __A, __mmask16 __B, __m512 __C, int __D)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __C,
+        __A, __B, __D);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtx_roundps_ph (__mmask16 __A, __m512 __B, int __C)
+{
+  return __builtin_ia32_vcvtps2phx512_mask_round ((__v16sf) __B,
+        _mm256_setzero_ph (),
+        __A, __C);
+}
+#else
+#define _mm512_cvtx_roundps_ph(A, B) (__builtin_ia32_vcvtps2phx512_mask_round ((__v16sf)(A), _mm256_setzero_ph (), (__mmask16)-1, (B)))
+#define _mm512_mask_cvtx_roundps_ph(A, B, C, D) (__builtin_ia32_vcvtps2phx512_mask_round ((__v16sf)(C), (A), (B), (D)))
+#define _mm512_maskz_cvtx_roundps_ph(A, B, C) (__builtin_ia32_vcvtps2phx512_mask_round ((__v16sf)(B), _mm256_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvtpd_ph (__m512d __A)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __A,
+       _mm_setzero_ph (),
+       (__mmask8) -1,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvtpd_ph (__m128h __A, __mmask8 __B, __m512d __C)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __C,
+       __A, __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvtpd_ph (__mmask8 __A, __m512d __B)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __B,
+       _mm_setzero_ph (),
+       __A,
+       _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_cvt_roundpd_ph (__m512d __A, int __B)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __A,
+       _mm_setzero_ph (),
+       (__mmask8) -1,
+       __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_cvt_roundpd_ph (__m128h __A, __mmask8 __B, __m512d __C, int __D)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __C,
+       __A, __B, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_cvt_roundpd_ph (__mmask8 __A, __m512d __B, int __C)
+{
+  return __builtin_ia32_vcvtpd2ph512_mask_round ((__v8df) __B,
+       _mm_setzero_ph (),
+       __A, __C);
+}
+#else
+#define _mm512_cvt_roundpd_ph(A, B) (__builtin_ia32_vcvtpd2ph512_mask_round ((__v8df)(A), _mm_setzero_ph (), (__mmask8)-1, (B)))
+#define _mm512_mask_cvt_roundpd_ph(A, B, C, D) (__builtin_ia32_vcvtpd2ph512_mask_round ((__v8df)(C), (A), (B), (D)))
+#define _mm512_maskz_cvt_roundpd_ph(A, B, C) (__builtin_ia32_vcvtpd2ph512_mask_round ((__v8df)(B), _mm_setzero_ph (), (A), (C)))
+#endif
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_ss (__m128 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A,
+           _mm_setzero_ps (),
+           (__mmask8) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtsh_ss (__m128 __A, __mmask8 __B, __m128 __C,
+    __m128h __D)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtsh_ss (__mmask8 __A, __m128 __B,
+     __m128h __C)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B,
+           _mm_setzero_ps (),
+           __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsh_sd (__m128d __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A,
+           _mm_setzero_pd (),
+           (__mmask8) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtsh_sd (__m128d __A, __mmask8 __B, __m128d __C,
+    __m128h __D)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtsh_sd (__mmask8 __A, __m128d __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B,
+           _mm_setzero_pd (),
+           __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_ss (__m128 __A, __m128h __B, const int __R)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__B, __A,
+           _mm_setzero_ps (),
+           (__mmask8) -1, __R);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvt_roundsh_ss (__m128 __A, __mmask8 __B, __m128 __C,
+    __m128h __D, const int __R)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__D, __C, __A, __B, __R);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvt_roundsh_ss (__mmask8 __A, __m128 __B,
+     __m128h __C, const int __R)
+{
+  return __builtin_ia32_vcvtsh2ss_mask_round (__C, __B,
+           _mm_setzero_ps (),
+           __A, __R);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsh_sd (__m128d __A, __m128h __B, const int __R)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__B, __A,
+           _mm_setzero_pd (),
+           (__mmask8) -1, __R);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvt_roundsh_sd (__m128d __A, __mmask8 __B, __m128d __C,
+    __m128h __D, const int __R)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__D, __C, __A, __B, __R);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvt_roundsh_sd (__mmask8 __A, __m128d __B, __m128h __C, const int __R)
+{
+  return __builtin_ia32_vcvtsh2sd_mask_round (__C, __B,
+           _mm_setzero_pd (),
+           __A, __R);
+}
+#else
+#define _mm_cvt_roundsh_ss(A, B, R) (__builtin_ia32_vcvtsh2ss_mask_round ((B), (A), _mm_setzero_ps (), (__mmask8) -1, (R)))
+#define _mm_mask_cvt_roundsh_ss(A, B, C, D, R) (__builtin_ia32_vcvtsh2ss_mask_round ((D), (C), (A), (B), (R)))
+#define _mm_maskz_cvt_roundsh_ss(A, B, C, R) (__builtin_ia32_vcvtsh2ss_mask_round ((C), (B), _mm_setzero_ps (), (A), (R)))
+#define _mm_cvt_roundsh_sd(A, B, R) (__builtin_ia32_vcvtsh2sd_mask_round ((B), (A), _mm_setzero_pd (), (__mmask8) -1, (R)))
+#define _mm_mask_cvt_roundsh_sd(A, B, C, D, R) (__builtin_ia32_vcvtsh2sd_mask_round ((D), (C), (A), (B), (R)))
+#define _mm_maskz_cvt_roundsh_sd(A, B, C, R) (__builtin_ia32_vcvtsh2sd_mask_round ((C), (B), _mm_setzero_pd (), (A), (R)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_sh (__m128h __A, __m128 __B)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__B, __A,
+           _mm_setzero_ph (),
+           (__mmask8) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtss_sh (__mmask8 __A, __m128h __B, __m128 __C)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__C, __B,
+           _mm_setzero_ph (),
+           __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsd_sh (__m128h __A, __m128d __B)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A,
+           _mm_setzero_ph (),
+           (__mmask8) -1,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtsd_sh (__mmask8 __A, __m128h __B, __m128d __C)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B,
+           _mm_setzero_ph (),
+           __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundss_sh (__m128h __A, __m128 __B, const int __R)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__B, __A,
+           _mm_setzero_ph (),
+           (__mmask8) -1, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvt_roundss_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128 __D,
+    const int __R)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__D, __C, __A, __B, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvt_roundss_sh (__mmask8 __A, __m128h __B, __m128 __C,
+     const int __R)
+{
+  return __builtin_ia32_vcvtss2sh_mask_round (__C, __B,
+           _mm_setzero_ph (),
+           __A, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_roundsd_sh (__m128h __A, __m128d __B, const int __R)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__B, __A,
+           _mm_setzero_ph (),
+           (__mmask8) -1, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvt_roundsd_sh (__m128h __A, __mmask8 __B, __m128h __C, __m128d __D,
+    const int __R)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__D, __C, __A, __B, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvt_roundsd_sh (__mmask8 __A, __m128h __B, __m128d __C,
+     const int __R)
+{
+  return __builtin_ia32_vcvtsd2sh_mask_round (__C, __B,
+           _mm_setzero_ph (),
+           __A, __R);
+}
+#else
+#define _mm_cvt_roundss_sh(A, B, R) (__builtin_ia32_vcvtss2sh_mask_round ((B), (A), _mm_setzero_ph (), (__mmask8) -1, R))
+#define _mm_mask_cvt_roundss_sh(A, B, C, D, R) (__builtin_ia32_vcvtss2sh_mask_round ((D), (C), (A), (B), (R)))
+#define _mm_maskz_cvt_roundss_sh(A, B, C, R) (__builtin_ia32_vcvtss2sh_mask_round ((C), (B), _mm_setzero_ph (), A, R))
+#define _mm_cvt_roundsd_sh(A, B, R) (__builtin_ia32_vcvtsd2sh_mask_round ((B), (A), _mm_setzero_ph (), (__mmask8) -1, R))
+#define _mm_mask_cvt_roundsd_sh(A, B, C, D, R) (__builtin_ia32_vcvtsd2sh_mask_round ((D), (C), (A), (B), (R)))
+#define _mm_maskz_cvt_roundsd_sh(A, B, C, R) (__builtin_ia32_vcvtsd2sh_mask_round ((C), (B), _mm_setzero_ph (), (A), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmaddsub_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) -1,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmaddsub_ph (__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) __U,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmaddsub_ph (__m512h __A, __m512h __B, __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask3 ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmaddsub_ph (__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_maskz ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmaddsub_round_ph (__m512h __A, __m512h __B, __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmaddsub_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmaddsub_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_mask3 ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmaddsub_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddsubph512_maskz ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fmaddsub_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfmaddsubph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fmaddsub_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfmaddsubph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fmaddsub_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfmaddsubph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fmaddsub_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfmaddsubph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+  _mm512_fmsubadd_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) -1,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmsubadd_ph (__m512h __A, __mmask32 __U,
+    __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) __U,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmsubadd_ph (__m512h __A, __m512h __B,
+     __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask3 ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmsubadd_ph (__mmask32 __U, __m512h __A,
+     __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_maskz ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmsubadd_round_ph (__m512h __A, __m512h __B,
+     __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmsubadd_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmsubadd_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_mask3 ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmsubadd_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubaddph512_maskz ((__v32hf) __A,
+      (__v32hf) __B,
+      (__v32hf) __C,
+      (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fmsubadd_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfmsubaddph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fmsubadd_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfmsubaddph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fmsubadd_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfmsubaddph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fmsubadd_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfmsubaddph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+  _mm512_fmadd_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddph512_mask ((__v32hf) __A,
+         (__v32hf) __B,
+         (__v32hf) __C,
+         (__mmask32) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmadd_ph (__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddph512_mask ((__v32hf) __A,
+         (__v32hf) __B,
+         (__v32hf) __C,
+         (__mmask32) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmadd_ph (__m512h __A, __m512h __B, __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddph512_mask3 ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmadd_ph (__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddph512_maskz ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmadd_round_ph (__m512h __A, __m512h __B, __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmaddph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmadd_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmaddph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmadd_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmaddph512_mask3 ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmadd_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmaddph512_maskz ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fmadd_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfmaddph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fmadd_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfmaddph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fmadd_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfmaddph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fmadd_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfmaddph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmadd_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmaddph512_mask ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmadd_ph (__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmaddph512_mask ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmadd_ph (__m512h __A, __m512h __B, __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmaddph512_mask3 ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           (__mmask32) __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmadd_ph (__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmaddph512_maskz ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           (__mmask32) __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmadd_round_ph (__m512h __A, __m512h __B, __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmaddph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmadd_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmaddph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmadd_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmaddph512_mask3 ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmadd_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmaddph512_maskz ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fnmadd_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfnmaddph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fnmadd_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfnmaddph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fnmadd_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfnmaddph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fnmadd_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfnmaddph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmsub_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubph512_mask ((__v32hf) __A,
+         (__v32hf) __B,
+         (__v32hf) __C,
+         (__mmask32) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmsub_ph (__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubph512_mask ((__v32hf) __A,
+         (__v32hf) __B,
+         (__v32hf) __C,
+         (__mmask32) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmsub_ph (__m512h __A, __m512h __B, __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubph512_mask3 ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmsub_ph (__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmsubph512_maskz ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmsub_round_ph (__m512h __A, __m512h __B, __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmsubph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmsub_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmsubph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmsub_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmsubph512_mask3 ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmsub_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfmsubph512_maskz ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fmsub_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfmsubph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fmsub_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfmsubph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fmsub_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfmsubph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fmsub_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfmsubph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmsub_ph (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmsubph512_mask ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) -1,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmsub_ph (__m512h __A, __mmask32 __U, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmsubph512_mask ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          (__mmask32) __U,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmsub_ph (__m512h __A, __m512h __B, __m512h __C, __mmask32 __U)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmsubph512_mask3 ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           (__mmask32) __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmsub_ph (__mmask32 __U, __m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfnmsubph512_maskz ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           (__mmask32) __U,
+           _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fnmsub_round_ph (__m512h __A, __m512h __B, __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmsubph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) -1, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fnmsub_round_ph (__m512h __A, __mmask32 __U, __m512h __B,
+          __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmsubph512_mask ((__v32hf) __A,
+             (__v32hf) __B,
+             (__v32hf) __C,
+             (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fnmsub_round_ph (__m512h __A, __m512h __B, __m512h __C,
+    __mmask32 __U, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmsubph512_mask3 ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fnmsub_round_ph (__mmask32 __U, __m512h __A, __m512h __B,
+    __m512h __C, const int __R)
+{
+  return (__m512h) __builtin_ia32_vfnmsubph512_maskz ((__v32hf) __A,
+       (__v32hf) __B,
+       (__v32hf) __C,
+       (__mmask32) __U, __R);
+}
+#else
+#define _mm512_fnmsub_round_ph(A, B, C, R) ((__m512h)__builtin_ia32_vfnmsubph512_mask ((A), (B), (C), -1, (R)))
+#define _mm512_mask_fnmsub_round_ph(A, U, B, C, R) ((__m512h)__builtin_ia32_vfnmsubph512_mask ((A), (B), (C), (U), (R)))
+#define _mm512_mask3_fnmsub_round_ph(A, B, C, U, R) ((__m512h)__builtin_ia32_vfnmsubph512_mask3 ((A), (B), (C), (U), (R)))
+#define _mm512_maskz_fnmsub_round_ph(U, A, B, C, R) ((__m512h)__builtin_ia32_vfnmsubph512_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_sh (__m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) __U,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) -1,
+        __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B,
+    const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U,
+     const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A,
+     __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+#else
+#define _mm_fmadd_round_sh(A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (-1), (R)))
+#define _mm_mask_fmadd_round_sh(A, U, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), (C), (U), (R)))
+#define _mm_mask3_fmadd_round_sh(A, B, C, U, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask3 ((A), (B), (C), (U), (R)))
+#define _mm_maskz_fmadd_round_sh(U, A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) -1,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmadd_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) __U,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmadd_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmadd_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) -1,
+         __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmadd_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B,
+    const int __R)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        (__v8hf) __B,
+        (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmadd_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U,
+     const int __R)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmadd_round_sh (__mmask8 __U, __m128h __W, __m128h __A,
+     __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfnmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+#else
+#define _mm_fnmadd_round_sh(A, B, C, R) ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (-1), (R)))
+#define _mm_mask_fnmadd_round_sh(A, U, B, C, R) ((__m128h) __builtin_ia32_vfnmaddsh3_mask ((A), (B), (C), (U), (R)))
+#define _mm_mask3_fnmadd_round_sh(A, B, C, U, R) ((__m128h) __builtin_ia32_vfnmaddsh3_mask3 ((A), (B), (C), (U), (R)))
+#define _mm_maskz_fnmadd_round_sh(U, A, B, C, R) ((__m128h) __builtin_ia32_vfnmaddsh3_maskz ((A), (B), (C), (U), (R)))
+#endif
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmsub_sh (__m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) __U,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         -(__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) -1,
+        __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B,
+    const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        (__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U,
+     const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W,
+         (__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A,
+     __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         (__v8hf) __A,
+         -(__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+#else
+#define _mm_fmsub_round_sh(A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (-1), (R)))
+#define _mm_mask_fmsub_round_sh(A, U, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), (B), -(C), (U), (R)))
+#define _mm_mask3_fmsub_round_sh(A, B, C, U, R) ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), (B), (C), (U), (R)))
+#define _mm_maskz_fmsub_round_sh(U, A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), (B), -(C), (U), (R)))
+#endif
+extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        -(__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) -1,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmsub_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        -(__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) __U,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmsub_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W,
+         -(__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmsub_sh (__mmask8 __U, __m128h __W, __m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         -(__v8hf) __A,
+         -(__v8hf) __B,
+         (__mmask8) __U,
+         _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        -(__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) -1,
+        __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmsub_round_sh (__m128h __W, __mmask8 __U, __m128h __A, __m128h __B,
+    const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_mask ((__v8hf) __W,
+        -(__v8hf) __A,
+        -(__v8hf) __B,
+        (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmsub_round_sh (__m128h __W, __m128h __A, __m128h __B, __mmask8 __U,
+     const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmsubsh3_mask3 ((__v8hf) __W,
+         -(__v8hf) __A,
+         (__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmsub_round_sh (__mmask8 __U, __m128h __W, __m128h __A,
+     __m128h __B, const int __R)
+{
+  return (__m128h) __builtin_ia32_vfmaddsh3_maskz ((__v8hf) __W,
+         -(__v8hf) __A,
+         -(__v8hf) __B,
+         (__mmask8) __U, __R);
+}
+#else
+#define _mm_fnmsub_round_sh(A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (-1), (R)))
+#define _mm_mask_fnmsub_round_sh(A, U, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_mask ((A), -(B), -(C), (U), (R)))
+#define _mm_mask3_fnmsub_round_sh(A, B, C, U, R) ((__m128h) __builtin_ia32_vfmsubsh3_mask3 ((A), -(B), (C), (U), (R)))
+#define _mm_maskz_fnmsub_round_sh(U, A, B, C, R) ((__m128h) __builtin_ia32_vfmaddsh3_maskz ((A), -(B), -(C), (U), (R)))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fcmadd_pch (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_round ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fcmadd_pch (__m512h __A, __mmask16 __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_mask_round ((__v32hf) __A,
+          (__v32hf) __C,
+          (__v32hf) __D, __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fcmadd_pch (__m512h __A, __m512h __B, __m512h __C, __mmask16 __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_mask3_round ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           __D, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fcmadd_pch (__mmask16 __A, __m512h __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_maskz_round ((__v32hf) __B,
+           (__v32hf) __C,
+           (__v32hf) __D,
+           __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmadd_pch (__m512h __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_round ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmadd_pch (__m512h __A, __mmask16 __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_mask_round ((__v32hf) __A,
+         (__v32hf) __C,
+         (__v32hf) __D, __B,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmadd_pch (__m512h __A, __m512h __B, __m512h __C, __mmask16 __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_mask3_round ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          __D, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmadd_pch (__mmask16 __A, __m512h __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_maskz_round ((__v32hf) __B,
+          (__v32hf) __C,
+          (__v32hf) __D,
+          __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fcmadd_round_pch (__m512h __A, __m512h __B, __m512h __C, const int __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_round ((__v32hf) __A,
+     (__v32hf) __B,
+     (__v32hf) __C,
+     __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fcmadd_round_pch (__m512h __A, __mmask16 __B, __m512h __C,
+         __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_mask_round ((__v32hf) __A,
+          (__v32hf) __C,
+          (__v32hf) __D, __B,
+          __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fcmadd_round_pch (__m512h __A, __m512h __B, __m512h __C,
+          __mmask16 __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_mask3_round ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           __D, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fcmadd_round_pch (__mmask16 __A, __m512h __B, __m512h __C,
+          __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmaddcph512_maskz_round ((__v32hf) __B,
+           (__v32hf) __C,
+           (__v32hf) __D,
+           __A, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmadd_round_pch (__m512h __A, __m512h __B, __m512h __C, const int __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_round ((__v32hf) __A,
+           (__v32hf) __B,
+           (__v32hf) __C,
+           __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmadd_round_pch (__m512h __A, __mmask16 __B, __m512h __C,
+        __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_mask_round ((__v32hf) __A,
+         (__v32hf) __C,
+         (__v32hf) __D, __B,
+         __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask3_fmadd_round_pch (__m512h __A, __m512h __B, __m512h __C,
+         __mmask16 __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_mask3_round ((__v32hf) __A,
+          (__v32hf) __B,
+          (__v32hf) __C,
+          __D, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmadd_round_pch (__mmask16 __A, __m512h __B, __m512h __C,
+         __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfmaddcph512_maskz_round ((__v32hf) __B,
+          (__v32hf) __C,
+          (__v32hf) __D,
+          __A, __E);
+}
+#else
+#define _mm512_fcmadd_round_pch(A, B, C, D) (__m512h) __builtin_ia32_vfcmaddcph512_round ((A), (B), (C), (D))
+#define _mm512_mask_fcmadd_round_pch(A, B, C, D, E) ((__m512h) __builtin_ia32_vfcmaddcph512_mask_round ((__v32hf) (A), (__v32hf) (C), (__v32hf) (D), (B), (E)))
+#define _mm512_mask3_fcmadd_round_pch(A, B, C, D, E) ((__m512h) __builtin_ia32_vfcmaddcph512_mask3_round ((A), (B), (C), (D), (E)))
+#define _mm512_maskz_fcmadd_round_pch(A, B, C, D, E) (__m512h) __builtin_ia32_vfcmaddcph512_maskz_round ((B), (C), (D), (A), (E))
+#define _mm512_fmadd_round_pch(A, B, C, D) (__m512h) __builtin_ia32_vfmaddcph512_round ((A), (B), (C), (D))
+#define _mm512_mask_fmadd_round_pch(A, B, C, D, E) ((__m512h) __builtin_ia32_vfmaddcph512_mask_round ((__v32hf) (A), (__v32hf) (C), (__v32hf) (D), (B), (E)))
+#define _mm512_mask3_fmadd_round_pch(A, B, C, D, E) (__m512h) __builtin_ia32_vfmaddcph512_mask3_round ((A), (B), (C), (D), (E))
+#define _mm512_maskz_fmadd_round_pch(A, B, C, D, E) (__m512h) __builtin_ia32_vfmaddcph512_maskz_round ((B), (C), (D), (A), (E))
+#endif
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fcmul_pch (__m512h __A, __m512h __B)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_round ((__v32hf) __A,
+           (__v32hf) __B,
+           _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fcmul_pch (__m512h __A, __mmask16 __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_mask_round ((__v32hf) __C,
+         (__v32hf) __D,
+         (__v32hf) __A,
+         __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fcmul_pch (__mmask16 __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_mask_round ((__v32hf) __B,
+         (__v32hf) __C,
+         _mm512_setzero_ph (),
+         __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmul_pch (__m512h __A, __m512h __B)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_round ((__v32hf) __A,
+          (__v32hf) __B,
+          _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmul_pch (__m512h __A, __mmask16 __B, __m512h __C, __m512h __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_mask_round ((__v32hf) __C,
+        (__v32hf) __D,
+        (__v32hf) __A,
+        __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmul_pch (__mmask16 __A, __m512h __B, __m512h __C)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_mask_round ((__v32hf) __B,
+        (__v32hf) __C,
+        _mm512_setzero_ph (),
+        __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fcmul_round_pch (__m512h __A, __m512h __B, const int __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_round ((__v32hf) __A,
+           (__v32hf) __B, __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fcmul_round_pch (__m512h __A, __mmask16 __B, __m512h __C,
+        __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_mask_round ((__v32hf) __C,
+         (__v32hf) __D,
+         (__v32hf) __A,
+         __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fcmul_round_pch (__mmask16 __A, __m512h __B,
+         __m512h __C, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfcmulcph512_mask_round ((__v32hf) __B,
+         (__v32hf) __C,
+         _mm512_setzero_ph (),
+         __A, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_fmul_round_pch (__m512h __A, __m512h __B, const int __D)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_round ((__v32hf) __A,
+          (__v32hf) __B,
+          __D);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_fmul_round_pch (__m512h __A, __mmask16 __B, __m512h __C,
+       __m512h __D, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_mask_round ((__v32hf) __C,
+        (__v32hf) __D,
+        (__v32hf) __A,
+        __B, __E);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_fmul_round_pch (__mmask16 __A, __m512h __B,
+        __m512h __C, const int __E)
+{
+  return (__m512h)
+    __builtin_ia32_vfmulcph512_mask_round ((__v32hf) __B,
+        (__v32hf) __C,
+        _mm512_setzero_ph (),
+        __A, __E);
+}
+#else
+#define _mm512_fcmul_round_pch(A, B, D) (__m512h) __builtin_ia32_vfcmulcph512_round ((A), (B), (D))
+#define _mm512_mask_fcmul_round_pch(A, B, C, D, E) (__m512h) __builtin_ia32_vfcmulcph512_mask_round ((C), (D), (A), (B), (E))
+#define _mm512_maskz_fcmul_round_pch(A, B, C, E) (__m512h) __builtin_ia32_vfcmulcph512_mask_round ((B), (C), (__v32hf) _mm512_setzero_ph (), (A), (E))
+#define _mm512_fmul_round_pch(A, B, D) (__m512h) __builtin_ia32_vfmulcph512_round ((A), (B), (D))
+#define _mm512_mask_fmul_round_pch(A, B, C, D, E) (__m512h) __builtin_ia32_vfmulcph512_mask_round ((C), (D), (A), (B), (E))
+#define _mm512_maskz_fmul_round_pch(A, B, C, E) (__m512h) __builtin_ia32_vfmulcph512_mask_round ((B), (C), (__v32hf) _mm512_setzero_ph (), (A), (E))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A,
+       (__v8hf) __C,
+       (__v8hf) __D, __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A,
+        (__v8hf) __B,
+        (__v8hf) __C, __D,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B,
+        (__v8hf) __C,
+        (__v8hf) __D,
+        __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmadd_sch (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A,
+         (__v8hf) __B,
+         (__v8hf) __C,
+         _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A,
+      (__v8hf) __C,
+      (__v8hf) __D, __B,
+      _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_sch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C, __D,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_sch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B,
+       (__v8hf) __C,
+       (__v8hf) __D,
+       __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_sch (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_round ((__v8hf) __A,
+        (__v8hf) __B,
+        (__v8hf) __C,
+        _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C,
+      __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) __A,
+       (__v8hf) __C,
+       (__v8hf) __D,
+       __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C,
+       __mmask8 __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) __A,
+        (__v8hf) __B,
+        (__v8hf) __C,
+        __D, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C,
+       __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_maskz_round ((__v8hf) __B,
+        (__v8hf) __C,
+        (__v8hf) __D,
+        __A, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcsh_round ((__v8hf) __A,
+         (__v8hf) __B,
+         (__v8hf) __C,
+         __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_round_sch (__m128h __A, __mmask8 __B, __m128h __C,
+     __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) __A,
+      (__v8hf) __C,
+      (__v8hf) __D,
+      __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C,
+      __mmask8 __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       __D, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_round_sch (__mmask8 __A, __m128h __B, __m128h __C,
+      __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_maskz_round ((__v8hf) __B,
+       (__v8hf) __C,
+       (__v8hf) __D,
+       __A, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_round_sch (__m128h __A, __m128h __B, __m128h __C, const int __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcsh_round ((__v8hf) __A,
+        (__v8hf) __B,
+        (__v8hf) __C,
+        __D);
+}
+#else
+#define _mm_mask_fcmadd_round_sch(A, B, C, D, E) ((__m128h) __builtin_ia32_vfcmaddcsh_mask_round ((__v8hf) (A), (__v8hf) (C), (__v8hf) (D), (B), (E)))
+#define _mm_mask3_fcmadd_round_sch(A, B, C, D, E) ((__m128h) __builtin_ia32_vfcmaddcsh_mask3_round ((__v8hf) (A), (__v8hf) (B), (__v8hf) (C), (D), (E)))
+#define _mm_maskz_fcmadd_round_sch(A, B, C, D, E) __builtin_ia32_vfcmaddcsh_maskz_round ((B), (C), (D), (A), (E))
+#define _mm_fcmadd_round_sch(A, B, C, D) __builtin_ia32_vfcmaddcsh_round ((A), (B), (C), (D))
+#define _mm_mask_fmadd_round_sch(A, B, C, D, E) ((__m128h) __builtin_ia32_vfmaddcsh_mask_round ((__v8hf) (A), (__v8hf) (C), (__v8hf) (D), (B), (E)))
+#define _mm_mask3_fmadd_round_sch(A, B, C, D, E) ((__m128h) __builtin_ia32_vfmaddcsh_mask3_round ((__v8hf) (A), (__v8hf) (B), (__v8hf) (C), (D), (E)))
+#define _mm_maskz_fmadd_round_sch(A, B, C, D, E) __builtin_ia32_vfmaddcsh_maskz_round ((B), (C), (D), (A), (E))
+#define _mm_fmadd_round_sch(A, B, C, D) __builtin_ia32_vfmaddcsh_round ((A), (B), (C), (D))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmul_sch (__m128h __A, __m128h __B)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_round ((__v8hf) __A,
+        (__v8hf) __B,
+        _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C,
+      (__v8hf) __D,
+      (__v8hf) __A,
+      __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmul_sch (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B,
+      (__v8hf) __C,
+      _mm_setzero_ph (),
+      __A, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmul_sch (__m128h __A, __m128h __B)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_round ((__v8hf) __A,
+       (__v8hf) __B,
+       _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmul_sch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C,
+     (__v8hf) __D,
+     (__v8hf) __A,
+     __B, _MM_FROUND_CUR_DIRECTION);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmul_sch (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B,
+     (__v8hf) __C,
+     _mm_setzero_ph (),
+     __A, _MM_FROUND_CUR_DIRECTION);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmul_round_sch (__m128h __A, __m128h __B, const int __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_round ((__v8hf) __A,
+        (__v8hf) __B,
+        __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C,
+     __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C,
+      (__v8hf) __D,
+      (__v8hf) __A,
+      __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C,
+      const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B,
+      (__v8hf) __C,
+      _mm_setzero_ph (),
+      __A, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmul_round_sch (__m128h __A, __m128h __B, const int __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_round ((__v8hf) __A,
+       (__v8hf) __B, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmul_round_sch (__m128h __A, __mmask8 __B, __m128h __C,
+    __m128h __D, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C,
+     (__v8hf) __D,
+     (__v8hf) __A,
+     __B, __E);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmul_round_sch (__mmask8 __A, __m128h __B, __m128h __C, const int __E)
+{
+  return (__m128h)
+    __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B,
+     (__v8hf) __C,
+     _mm_setzero_ph (),
+     __A, __E);
+}
+#else
+#define _mm_fcmul_round_sch(__A, __B, __D) (__m128h) __builtin_ia32_vfcmulcsh_round ((__v8hf) __A, (__v8hf) __B, __D)
+#define _mm_mask_fcmul_round_sch(__A, __B, __C, __D, __E) (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __C, (__v8hf) __D, (__v8hf) __A, __B, __E)
+#define _mm_maskz_fcmul_round_sch(__A, __B, __C, __E) (__m128h) __builtin_ia32_vfcmulcsh_mask_round ((__v8hf) __B, (__v8hf) __C, _mm_setzero_ph (), __A, __E)
+#define _mm_fmul_round_sch(__A, __B, __D) (__m128h) __builtin_ia32_vfmulcsh_round ((__v8hf) __A, (__v8hf) __B, __D)
+#define _mm_mask_fmul_round_sch(__A, __B, __C, __D, __E) (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __C, (__v8hf) __D, (__v8hf) __A, __B, __E)
+#define _mm_maskz_fmul_round_sch(__A, __B, __C, __E) (__m128h) __builtin_ia32_vfmulcsh_mask_round ((__v8hf) __B, (__v8hf) __C, _mm_setzero_ph (), __A, __E)
+#endif
+#define _MM512_REDUCE_OP(op) __m256h __T1 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 0); __m256h __T2 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 1); __m256h __T3 = (__T1 op __T2); __m128h __T4 = (__m128h) _mm256_extractf128_pd ((__m256d) __T3, 0); __m128h __T5 = (__m128h) _mm256_extractf128_pd ((__m256d) __T3, 1); __m128h __T6 = (__T4 op __T5); __m128h __T7 = (__m128h) __builtin_shuffle ((__m128h)__T6, (__v8hi) { 4, 5, 6, 7, 0, 1, 2, 3 }); __m128h __T8 = (__T6 op __T7); __m128h __T9 = (__m128h) __builtin_shuffle ((__m128h)__T8, (__v8hi) { 2, 3, 0, 1, 4, 5, 6, 7 }); __m128h __T10 = __T8 op __T9; return __T10[0] op __T10[1]
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_add_ph (__m512h __A)
+{
+   _MM512_REDUCE_OP (+);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_mul_ph (__m512h __A)
+{
+   _MM512_REDUCE_OP (*);
+}
+#undef _MM512_REDUCE_OP
+#ifdef __AVX512VL__
+#define _MM512_REDUCE_OP(op) __m256h __T1 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 0); __m256h __T2 = (__m256h) _mm512_extractf64x4_pd ((__m512d) __A, 1); __m256h __T3 = __builtin_ia32_##op##ph256_mask (__T1, __T2, _mm256_setzero_ph (), (__mmask16) -1); __m128h __T4 = (__m128h) _mm256_extractf128_pd ((__m256d) __T3, 0); __m128h __T5 = (__m128h) _mm256_extractf128_pd ((__m256d) __T3, 1); __m128h __T6 = __builtin_ia32_##op##ph128_mask (__T4, __T5, _mm_setzero_ph (),(__mmask8) -1); __m128h __T7 = (__m128h) __builtin_shuffle ((__m128h)__T6, (__v8hi) { 2, 3, 0, 1, 6, 7, 4, 5 }); __m128h __T8 = (__m128h) __builtin_ia32_##op##ph128_mask (__T6, __T7, _mm_setzero_ph (),(__mmask8) -1); __m128h __T9 = (__m128h) __builtin_shuffle ((__m128h)__T8, (__v8hi) { 4, 5 }); __m128h __T10 = __builtin_ia32_##op##ph128_mask (__T8, __T9, _mm_setzero_ph (),(__mmask8) -1); __m128h __T11 = (__m128h) __builtin_shuffle (__T10, (__v8hi) { 1, 0 }); __m128h __T12 = __builtin_ia32_##op##ph128_mask (__T10, __T11, _mm_setzero_ph (),(__mmask8) -1); return __T12[0]
+#else
+#define _MM512_REDUCE_OP(op) __m512h __T1 = (__m512h) __builtin_shuffle ((__m512d) __A, (__v8di) { 4, 5, 6, 7, 0, 0, 0, 0 }); __m512h __T2 = _mm512_##op##_ph (__A, __T1); __m512h __T3 = (__m512h) __builtin_shuffle ((__m512d) __T2, (__v8di) { 2, 3, 0, 0, 0, 0, 0, 0 }); __m512h __T4 = _mm512_##op##_ph (__T2, __T3); __m512h __T5 = (__m512h) __builtin_shuffle ((__m512d) __T4, (__v8di) { 1, 0, 0, 0, 0, 0, 0, 0 }); __m512h __T6 = _mm512_##op##_ph (__T4, __T5); __m512h __T7 = (__m512h) __builtin_shuffle ((__m512) __T6, (__v16si) { 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }); __m512h __T8 = _mm512_##op##_ph (__T6, __T7); __m512h __T9 = (__m512h) __builtin_shuffle (__T8, (__v32hi) { 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }); __m512h __T10 = _mm512_##op##_ph (__T8, __T9); return __T10[0]
+#endif
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_min_ph (__m512h __A)
+{
+  _MM512_REDUCE_OP (min);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_reduce_max_ph (__m512h __A)
+{
+  _MM512_REDUCE_OP (max);
+}
+#undef _MM512_REDUCE_OP
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_blend_ph (__mmask32 __U, __m512h __A, __m512h __W)
+{
+  return (__m512h) __builtin_ia32_movdquhi512_mask ((__v32hi) __W,
+          (__v32hi) __A,
+          (__mmask32) __U);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_permutex2var_ph (__m512h __A, __m512i __I, __m512h __B)
+{
+  return (__m512h) __builtin_ia32_vpermi2varhi512_mask ((__v32hi) __A,
+             (__v32hi) __I,
+             (__v32hi) __B,
+             (__mmask32)-1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_permutexvar_ph (__m512i __A, __m512h __B)
+{
+  return (__m512h) __builtin_ia32_permvarhi512_mask ((__v32hi) __B,
+           (__v32hi) __A,
+           (__v32hi)
+           (_mm512_setzero_ph ()),
+           (__mmask32)-1);
+}
+extern __inline __m512h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_set1_pch (_Float16 _Complex __A)
+{
+  union
+  {
+    _Float16 _Complex a;
+    float b;
+  } u = { .a = __A};
+  return (__m512h) _mm512_set1_ps (u.b);
+}
+#define _mm512_mul_pch(A, B) _mm512_fmul_pch ((A), (B))
+#define _mm512_mask_mul_pch(W, U, A, B) _mm512_mask_fmul_pch ((W), (U), (A), (B))
+#define _mm512_maskz_mul_pch(U, A, B) _mm512_maskz_fmul_pch ((U), (A), (B))
+#define _mm512_mul_round_pch(A, B, R) _mm512_fmul_round_pch ((A), (B), (R))
+#define _mm512_mask_mul_round_pch(W, U, A, B, R) _mm512_mask_fmul_round_pch ((W), (U), (A), (B), (R))
+#define _mm512_maskz_mul_round_pch(U, A, B, R) _mm512_maskz_fmul_round_pch ((U), (A), (B), (R))
+#define _mm512_cmul_pch(A, B) _mm512_fcmul_pch ((A), (B))
+#define _mm512_mask_cmul_pch(W, U, A, B) _mm512_mask_fcmul_pch ((W), (U), (A), (B))
+#define _mm512_maskz_cmul_pch(U, A, B) _mm512_maskz_fcmul_pch ((U), (A), (B))
+#define _mm512_cmul_round_pch(A, B, R) _mm512_fcmul_round_pch ((A), (B), (R))
+#define _mm512_mask_cmul_round_pch(W, U, A, B, R) _mm512_mask_fcmul_round_pch ((W), (U), (A), (B), (R))
+#define _mm512_maskz_cmul_round_pch(U, A, B, R) _mm512_maskz_fcmul_round_pch ((U), (A), (B), (R))
+#define _mm_mul_sch(A, B) _mm_fmul_sch ((A), (B))
+#define _mm_mask_mul_sch(W, U, A, B) _mm_mask_fmul_sch ((W), (U), (A), (B))
+#define _mm_maskz_mul_sch(U, A, B) _mm_maskz_fmul_sch ((U), (A), (B))
+#define _mm_mul_round_sch(A, B, R) _mm_fmul_round_sch ((A), (B), (R))
+#define _mm_mask_mul_round_sch(W, U, A, B, R) _mm_mask_fmul_round_sch ((W), (U), (A), (B), (R))
+#define _mm_maskz_mul_round_sch(U, A, B, R) _mm_maskz_fmul_round_sch ((U), (A), (B), (R))
+#define _mm_cmul_sch(A, B) _mm_fcmul_sch ((A), (B))
+#define _mm_mask_cmul_sch(W, U, A, B) _mm_mask_fcmul_sch ((W), (U), (A), (B))
+#define _mm_maskz_cmul_sch(U, A, B) _mm_maskz_fcmul_sch ((U), (A), (B))
+#define _mm_cmul_round_sch(A, B, R) _mm_fcmul_round_sch ((A), (B), (R))
+#define _mm_mask_cmul_round_sch(W, U, A, B, R) _mm_mask_fcmul_round_sch ((W), (U), (A), (B), (R))
+#define _mm_maskz_cmul_round_sch(U, A, B, R) _mm_maskz_fcmul_round_sch ((U), (A), (B), (R))
+#ifdef __DISABLE_AVX512FP16__
+#undef __DISABLE_AVX512FP16__
+#pragma GCC pop_options
+#endif
+#endif
+#endif
diff --git a/third_party/intel/avx512fp16vlintrin.internal.h b/third_party/intel/avx512fp16vlintrin.internal.h
new file mode 100644
index 000000000..d42a041a9
--- /dev/null
+++ b/third_party/intel/avx512fp16vlintrin.internal.h
@@ -0,0 +1,2819 @@
+#if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
+#ifndef _IMMINTRIN_H_INCLUDED
+#error "Never use <avx512fp16vlintrin.h> directly; include <immintrin.h> instead."
+#endif
+#ifndef __AVX512FP16VLINTRIN_H_INCLUDED
+#define __AVX512FP16VLINTRIN_H_INCLUDED
+#if !defined(__AVX512VL__) || !defined(__AVX512FP16__)
+#pragma GCC push_options
+#pragma GCC target("avx512fp16,avx512vl")
+#define __DISABLE_AVX512FP16VL__
+#endif
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castph_ps (__m128h __a)
+{
+  return (__m128) __a;
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castph_ps (__m256h __a)
+{
+  return (__m256) __a;
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castph_pd (__m128h __a)
+{
+  return (__m128d) __a;
+}
+extern __inline __m256d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castph_pd (__m256h __a)
+{
+  return (__m256d) __a;
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castph_si128 (__m128h __a)
+{
+  return (__m128i) __a;
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castph_si256 (__m256h __a)
+{
+  return (__m256i) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castps_ph (__m128 __a)
+{
+  return (__m128h) __a;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castps_ph (__m256 __a)
+{
+  return (__m256h) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castpd_ph (__m128d __a)
+{
+  return (__m128h) __a;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castpd_ph (__m256d __a)
+{
+  return (__m256h) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_castsi128_ph (__m128i __a)
+{
+  return (__m128h) __a;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castsi256_ph (__m256i __a)
+{
+  return (__m256h) __a;
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castph256_ph128 (__m256h __A)
+{
+  union
+  {
+    __m128h a[2];
+    __m256h v;
+  } u = { .v = __A };
+  return u.a[0];
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_castph128_ph256 (__m128h __A)
+{
+  union
+  {
+    __m128h a[2];
+    __m256h v;
+  } u;
+  u.a[0] = __A;
+  return u.v;
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_zextph128_ph256 (__m128h __A)
+{
+  return (__m256h) _mm256_insertf128_ps (_mm256_setzero_ps (),
+      (__m128) __A, 0);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_conj_pch (__m256h __A)
+{
+  return (__m256h) _mm256_xor_epi32 ((__m256i) __A, _mm256_set1_epi32 (1<<31));
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_conj_pch (__m256h __W, __mmask8 __U, __m256h __A)
+{
+  return (__m256h) __builtin_ia32_movaps256_mask ((__v8sf)
+         _mm256_conj_pch (__A),
+        (__v8sf) __W,
+        (__mmask8) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_conj_pch (__mmask8 __U, __m256h __A)
+{
+  return (__m256h) __builtin_ia32_movaps256_mask ((__v8sf)
+         _mm256_conj_pch (__A),
+        (__v8sf)
+         _mm256_setzero_ps (),
+        (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_conj_pch (__m128h __A)
+{
+  return (__m128h) _mm_xor_epi32 ((__m128i) __A, _mm_set1_epi32 (1<<31));
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_conj_pch (__m128h __W, __mmask8 __U, __m128h __A)
+{
+  return (__m128h) __builtin_ia32_movaps128_mask ((__v4sf) _mm_conj_pch (__A),
+        (__v4sf) __W,
+        (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_conj_pch (__mmask8 __U, __m128h __A)
+{
+  return (__m128h) __builtin_ia32_movaps128_mask ((__v4sf) _mm_conj_pch (__A),
+        (__v4sf) _mm_setzero_ps (),
+        (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_ph (__m128h __A, __m128h __B)
+{
+  return (__m128h) ((__v8hf) __A + (__v8hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_add_ph (__m256h __A, __m256h __B)
+{
+  return (__m256h) ((__v16hf) __A + (__v16hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_add_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_addph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_add_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_addph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_add_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_addph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_add_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_addph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_ph (__m128h __A, __m128h __B)
+{
+  return (__m128h) ((__v8hf) __A - (__v8hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sub_ph (__m256h __A, __m256h __B)
+{
+  return (__m256h) ((__v16hf) __A - (__v16hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sub_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_subph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_sub_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_subph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sub_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_subph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_sub_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_subph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_ph (__m128h __A, __m128h __B)
+{
+  return (__m128h) ((__v8hf) __A * (__v8hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mul_ph (__m256h __A, __m256h __B)
+{
+  return (__m256h) ((__v16hf) __A * (__v16hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_mul_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_mulph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_mul_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_mulph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_mul_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_mulph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_mul_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_mulph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_ph (__m128h __A, __m128h __B)
+{
+  return (__m128h) ((__v8hf) __A / (__v8hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_div_ph (__m256h __A, __m256h __B)
+{
+  return (__m256h) ((__v16hf) __A / (__v16hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_div_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_divph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_div_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_divph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_div_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_divph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_div_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_divph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_ph (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_maxph128_mask (__A, __B,
+           _mm_setzero_ph (),
+           (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_max_ph (__m256h __A, __m256h __B)
+{
+  return __builtin_ia32_maxph256_mask (__A, __B,
+           _mm256_setzero_ph (),
+           (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_max_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_maxph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_max_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_maxph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_max_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_maxph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_max_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_maxph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_ph (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_minph128_mask (__A, __B,
+           _mm_setzero_ph (),
+           (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_min_ph (__m256h __A, __m256h __B)
+{
+  return __builtin_ia32_minph256_mask (__A, __B,
+           _mm256_setzero_ph (),
+           (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_min_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_minph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_min_ph (__m256h __A, __mmask16 __B, __m256h __C, __m256h __D)
+{
+  return __builtin_ia32_minph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_min_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_minph128_mask (__B, __C, _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_min_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_minph256_mask (__B, __C,
+           _mm256_setzero_ph (), __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_abs_ph (__m128h __A)
+{
+  return (__m128h) _mm_and_si128 ( _mm_set1_epi32 (0x7FFF7FFF),
+       (__m128i) __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_abs_ph (__m256h __A)
+{
+  return (__m256h) _mm256_and_si256 ( _mm256_set1_epi32 (0x7FFF7FFF),
+          (__m256i) __A);
+}
+#ifdef __OPTIMIZE
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmp_ph_mask (__m128h __A, __m128h __B, const int __C)
+{
+  return (__mmask8) __builtin_ia32_cmpph128_mask (__A, __B, __C,
+        (__mmask8) -1);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cmp_ph_mask (__mmask8 __A, __m128h __B, __m128h __C,
+        const int __D)
+{
+  return (__mmask8) __builtin_ia32_cmpph128_mask (__B, __C, __D, __A);
+}
+extern __inline __mmask16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmp_ph_mask (__m256h __A, __m256h __B, const int __C)
+{
+  return (__mmask16) __builtin_ia32_cmpph256_mask (__A, __B, __C,
+         (__mmask16) -1);
+}
+extern __inline __mmask16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cmp_ph_mask (__mmask16 __A, __m256h __B, __m256h __C,
+        const int __D)
+{
+  return (__mmask16) __builtin_ia32_cmpph256_mask (__B, __C, __D,
+         __A);
+}
+#else
+#define _mm_cmp_ph_mask(A, B, C) (__builtin_ia32_cmpph128_mask ((A), (B), (C), (-1)))
+#define _mm_mask_cmp_ph_mask(A, B, C, D) (__builtin_ia32_cmpph128_mask ((B), (C), (D), (A)))
+#define _mm256_cmp_ph_mask(A, B, C) (__builtin_ia32_cmpph256_mask ((A), (B), (C), (-1)))
+#define _mm256_mask_cmp_ph_mask(A, B, C, D) (__builtin_ia32_cmpph256_mask ((B), (C), (D), (A)))
+#endif
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_ph (__m128h __A)
+{
+  return __builtin_ia32_sqrtph128_mask (__A, _mm_setzero_ph (),
+     (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sqrt_ph (__m256h __A)
+{
+  return __builtin_ia32_sqrtph256_mask (__A, _mm256_setzero_ph (),
+     (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sqrt_ph (__m128h __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_sqrtph128_mask (__C, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_sqrt_ph (__m256h __A, __mmask16 __B, __m256h __C)
+{
+  return __builtin_ia32_sqrtph256_mask (__C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sqrt_ph (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_sqrtph128_mask (__B, _mm_setzero_ph (),
+     __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_sqrt_ph (__mmask16 __A, __m256h __B)
+{
+  return __builtin_ia32_sqrtph256_mask (__B, _mm256_setzero_ph (),
+     __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_ph (__m128h __A)
+{
+  return __builtin_ia32_rsqrtph128_mask (__A, _mm_setzero_ph (),
+      (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_rsqrt_ph (__m256h __A)
+{
+  return __builtin_ia32_rsqrtph256_mask (__A, _mm256_setzero_ph (),
+      (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_rsqrt_ph (__m128h __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_rsqrtph128_mask (__C, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_rsqrt_ph (__m256h __A, __mmask16 __B, __m256h __C)
+{
+  return __builtin_ia32_rsqrtph256_mask (__C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_rsqrt_ph (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_rsqrtph128_mask (__B, _mm_setzero_ph (), __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_rsqrt_ph (__mmask16 __A, __m256h __B)
+{
+  return __builtin_ia32_rsqrtph256_mask (__B, _mm256_setzero_ph (),
+      __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_ph (__m128h __A)
+{
+  return __builtin_ia32_rcpph128_mask (__A, _mm_setzero_ph (),
+           (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_rcp_ph (__m256h __A)
+{
+  return __builtin_ia32_rcpph256_mask (__A, _mm256_setzero_ph (),
+           (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_rcp_ph (__m128h __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_rcpph128_mask (__C, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_rcp_ph (__m256h __A, __mmask16 __B, __m256h __C)
+{
+  return __builtin_ia32_rcpph256_mask (__C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_rcp_ph (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_rcpph128_mask (__B, _mm_setzero_ph (), __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_rcp_ph (__mmask16 __A, __m256h __B)
+{
+  return __builtin_ia32_rcpph256_mask (__B, _mm256_setzero_ph (),
+           __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_scalef_ph (__m128h __A, __m128h __B)
+{
+  return __builtin_ia32_scalefph128_mask (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_scalef_ph (__m256h __A, __m256h __B)
+{
+  return __builtin_ia32_scalefph256_mask (__A, __B,
+       _mm256_setzero_ph (),
+       (__mmask16) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_scalef_ph (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return __builtin_ia32_scalefph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_scalef_ph (__m256h __A, __mmask16 __B, __m256h __C,
+         __m256h __D)
+{
+  return __builtin_ia32_scalefph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_scalef_ph (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return __builtin_ia32_scalefph128_mask (__B, __C,
+       _mm_setzero_ph (), __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_scalef_ph (__mmask16 __A, __m256h __B, __m256h __C)
+{
+  return __builtin_ia32_scalefph256_mask (__B, __C,
+       _mm256_setzero_ph (),
+       __A);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_ph (__m128h __A, int __B)
+{
+  return __builtin_ia32_reduceph128_mask (__A, __B,
+       _mm_setzero_ph (),
+       (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_reduce_ph (__m128h __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_reduceph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_reduce_ph (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_reduceph128_mask (__B, __C,
+       _mm_setzero_ph (), __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_reduce_ph (__m256h __A, int __B)
+{
+  return __builtin_ia32_reduceph256_mask (__A, __B,
+       _mm256_setzero_ph (),
+       (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_reduce_ph (__m256h __A, __mmask16 __B, __m256h __C, int __D)
+{
+  return __builtin_ia32_reduceph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_reduce_ph (__mmask16 __A, __m256h __B, int __C)
+{
+  return __builtin_ia32_reduceph256_mask (__B, __C,
+       _mm256_setzero_ph (),
+       __A);
+}
+#else
+#define _mm_reduce_ph(A, B) (__builtin_ia32_reduceph128_mask ((A), (B), _mm_setzero_ph (), ((__mmask8)-1)))
+#define _mm_mask_reduce_ph(A, B, C, D) (__builtin_ia32_reduceph128_mask ((C), (D), (A), (B)))
+#define _mm_maskz_reduce_ph(A, B, C) (__builtin_ia32_reduceph128_mask ((B), (C), _mm_setzero_ph (), (A)))
+#define _mm256_reduce_ph(A, B) (__builtin_ia32_reduceph256_mask ((A), (B), _mm256_setzero_ph (), ((__mmask16)-1)))
+#define _mm256_mask_reduce_ph(A, B, C, D) (__builtin_ia32_reduceph256_mask ((C), (D), (A), (B)))
+#define _mm256_maskz_reduce_ph(A, B, C) (__builtin_ia32_reduceph256_mask ((B), (C), _mm256_setzero_ph (), (A)))
+#endif
+#ifdef __OPTIMIZE__
+  extern __inline __m128h
+  __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+  _mm_roundscale_ph (__m128h __A, int __B)
+  {
+    return __builtin_ia32_rndscaleph128_mask (__A, __B,
+           _mm_setzero_ph (),
+           (__mmask8) -1);
+  }
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_roundscale_ph (__m128h __A, __mmask8 __B, __m128h __C, int __D)
+{
+  return __builtin_ia32_rndscaleph128_mask (__C, __D, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_roundscale_ph (__mmask8 __A, __m128h __B, int __C)
+{
+  return __builtin_ia32_rndscaleph128_mask (__B, __C,
+         _mm_setzero_ph (), __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_roundscale_ph (__m256h __A, int __B)
+{
+  return __builtin_ia32_rndscaleph256_mask (__A, __B,
+         _mm256_setzero_ph (),
+         (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_roundscale_ph (__m256h __A, __mmask16 __B, __m256h __C,
+      int __D)
+{
+  return __builtin_ia32_rndscaleph256_mask (__C, __D, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_roundscale_ph (__mmask16 __A, __m256h __B, int __C)
+{
+  return __builtin_ia32_rndscaleph256_mask (__B, __C,
+         _mm256_setzero_ph (),
+         __A);
+}
+#else
+#define _mm_roundscale_ph(A, B) (__builtin_ia32_rndscaleph128_mask ((A), (B), _mm_setzero_ph (), ((__mmask8)-1)))
+#define _mm_mask_roundscale_ph(A, B, C, D) (__builtin_ia32_rndscaleph128_mask ((C), (D), (A), (B)))
+#define _mm_maskz_roundscale_ph(A, B, C) (__builtin_ia32_rndscaleph128_mask ((B), (C), _mm_setzero_ph (), (A)))
+#define _mm256_roundscale_ph(A, B) (__builtin_ia32_rndscaleph256_mask ((A), (B), _mm256_setzero_ph(), ((__mmask16)-1)))
+#define _mm256_mask_roundscale_ph(A, B, C, D) (__builtin_ia32_rndscaleph256_mask ((C), (D), (A), (B)))
+#define _mm256_maskz_roundscale_ph(A, B, C) (__builtin_ia32_rndscaleph256_mask ((B), (C), _mm256_setzero_ph (), (A)))
+#endif
+#ifdef __OPTIMIZE__
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+  _mm_mask_fpclass_ph_mask (__mmask8 __U, __m128h __A, const int __imm)
+{
+  return (__mmask8) __builtin_ia32_fpclassph128_mask ((__v8hf) __A,
+            __imm, __U);
+}
+extern __inline __mmask8
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fpclass_ph_mask (__m128h __A, const int __imm)
+{
+  return (__mmask8) __builtin_ia32_fpclassph128_mask ((__v8hf) __A,
+            __imm,
+            (__mmask8) -1);
+}
+extern __inline __mmask16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fpclass_ph_mask (__mmask16 __U, __m256h __A, const int __imm)
+{
+  return (__mmask16) __builtin_ia32_fpclassph256_mask ((__v16hf) __A,
+             __imm, __U);
+}
+extern __inline __mmask16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fpclass_ph_mask (__m256h __A, const int __imm)
+{
+  return (__mmask16) __builtin_ia32_fpclassph256_mask ((__v16hf) __A,
+             __imm,
+             (__mmask16) -1);
+}
+#else
+#define _mm_fpclass_ph_mask(X, C) ((__mmask8) __builtin_ia32_fpclassph128_mask ((__v8hf) (__m128h) (X), (int) (C),(__mmask8)-1))
+#define _mm_mask_fpclass_ph_mask(u, X, C) ((__mmask8) __builtin_ia32_fpclassph128_mask ((__v8hf) (__m128h) (X), (int) (C),(__mmask8)(u)))
+#define _mm256_fpclass_ph_mask(X, C) ((__mmask16) __builtin_ia32_fpclassph256_mask ((__v16hf) (__m256h) (X), (int) (C),(__mmask16)-1))
+#define _mm256_mask_fpclass_ph_mask(u, X, C) ((__mmask16) __builtin_ia32_fpclassph256_mask ((__v16hf) (__m256h) (X), (int) (C),(__mmask16)(u)))
+#endif
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_getexp_ph (__m256h __A)
+{
+  return (__m256h) __builtin_ia32_getexpph256_mask ((__v16hf) __A,
+          (__v16hf)
+          _mm256_setzero_ph (),
+          (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_getexp_ph (__m256h __W, __mmask16 __U, __m256h __A)
+{
+  return (__m256h) __builtin_ia32_getexpph256_mask ((__v16hf) __A,
+          (__v16hf) __W,
+          (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_getexp_ph (__mmask16 __U, __m256h __A)
+{
+  return (__m256h) __builtin_ia32_getexpph256_mask ((__v16hf) __A,
+          (__v16hf)
+          _mm256_setzero_ph (),
+          (__mmask16) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getexp_ph (__m128h __A)
+{
+  return (__m128h) __builtin_ia32_getexpph128_mask ((__v8hf) __A,
+          (__v8hf)
+          _mm_setzero_ph (),
+          (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getexp_ph (__m128h __W, __mmask8 __U, __m128h __A)
+{
+  return (__m128h) __builtin_ia32_getexpph128_mask ((__v8hf) __A,
+          (__v8hf) __W,
+          (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getexp_ph (__mmask8 __U, __m128h __A)
+{
+  return (__m128h) __builtin_ia32_getexpph128_mask ((__v8hf) __A,
+          (__v8hf)
+          _mm_setzero_ph (),
+          (__mmask8) __U);
+}
+#ifdef __OPTIMIZE__
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_getmant_ph (__m256h __A, _MM_MANTISSA_NORM_ENUM __B,
+     _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m256h) __builtin_ia32_getmantph256_mask ((__v16hf) __A,
+           (__C << 2) | __B,
+           (__v16hf)
+           _mm256_setzero_ph (),
+           (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_getmant_ph (__m256h __W, __mmask16 __U, __m256h __A,
+   _MM_MANTISSA_NORM_ENUM __B,
+   _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m256h) __builtin_ia32_getmantph256_mask ((__v16hf) __A,
+           (__C << 2) | __B,
+           (__v16hf) __W,
+           (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_getmant_ph (__mmask16 __U, __m256h __A,
+    _MM_MANTISSA_NORM_ENUM __B,
+    _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m256h) __builtin_ia32_getmantph256_mask ((__v16hf) __A,
+           (__C << 2) | __B,
+           (__v16hf)
+           _mm256_setzero_ph (),
+           (__mmask16) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getmant_ph (__m128h __A, _MM_MANTISSA_NORM_ENUM __B,
+  _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m128h) __builtin_ia32_getmantph128_mask ((__v8hf) __A,
+           (__C << 2) | __B,
+           (__v8hf)
+           _mm_setzero_ph (),
+           (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_getmant_ph (__m128h __W, __mmask8 __U, __m128h __A,
+       _MM_MANTISSA_NORM_ENUM __B,
+       _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m128h) __builtin_ia32_getmantph128_mask ((__v8hf) __A,
+           (__C << 2) | __B,
+           (__v8hf) __W,
+           (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_getmant_ph (__mmask8 __U, __m128h __A,
+        _MM_MANTISSA_NORM_ENUM __B,
+        _MM_MANTISSA_SIGN_ENUM __C)
+{
+  return (__m128h) __builtin_ia32_getmantph128_mask ((__v8hf) __A,
+           (__C << 2) | __B,
+           (__v8hf)
+           _mm_setzero_ph (),
+           (__mmask8) __U);
+}
+#else
+#define _mm256_getmant_ph(X, B, C) ((__m256h) __builtin_ia32_getmantph256_mask ((__v16hf)(__m256h) (X), (int)(((C)<<2) | (B)), (__v16hf)(__m256h)_mm256_setzero_ph (), (__mmask16)-1))
+#define _mm256_mask_getmant_ph(W, U, X, B, C) ((__m256h) __builtin_ia32_getmantph256_mask ((__v16hf)(__m256h) (X), (int)(((C)<<2) | (B)), (__v16hf)(__m256h)(W), (__mmask16)(U)))
+#define _mm256_maskz_getmant_ph(U, X, B, C) ((__m256h) __builtin_ia32_getmantph256_mask ((__v16hf)(__m256h) (X), (int)(((C)<<2) | (B)), (__v16hf)(__m256h)_mm256_setzero_ph (), (__mmask16)(U)))
+#define _mm_getmant_ph(X, B, C) ((__m128h) __builtin_ia32_getmantph128_mask ((__v8hf)(__m128h) (X), (int)(((C)<<2) | (B)), (__v8hf)(__m128h)_mm_setzero_ph (), (__mmask8)-1))
+#define _mm_mask_getmant_ph(W, U, X, B, C) ((__m128h) __builtin_ia32_getmantph128_mask ((__v8hf)(__m128h) (X), (int)(((C)<<2) | (B)), (__v8hf)(__m128h)(W), (__mmask8)(U)))
+#define _mm_maskz_getmant_ph(U, X, B, C) ((__m128h) __builtin_ia32_getmantph128_mask ((__v8hf)(__m128h) (X), (int)(((C)<<2) | (B)), (__v8hf)(__m128h)_mm_setzero_ph (), (__mmask8)(U)))
+#endif
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epi32 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2dq128_mask (__A,
+          (__v4si)
+          _mm_setzero_si128 (),
+          (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epi32 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2dq128_mask (__C, ( __v4si) __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epi32 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2dq128_mask (__B,
+          (__v4si) _mm_setzero_si128 (),
+          __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epi32 (__m128h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2dq256_mask (__A,
+          (__v8si)
+          _mm256_setzero_si256 (),
+          (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epi32 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2dq256_mask (__C, ( __v8si) __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epi32 (__mmask8 __A, __m128h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2dq256_mask (__B,
+          (__v8si)
+          _mm256_setzero_si256 (),
+          __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epu32 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2udq128_mask (__A,
+           (__v4si)
+           _mm_setzero_si128 (),
+           (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epu32 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2udq128_mask (__C, ( __v4si) __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epu32 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2udq128_mask (__B,
+           (__v4si)
+           _mm_setzero_si128 (),
+           __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epu32 (__m128h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2udq256_mask (__A,
+           (__v8si)
+           _mm256_setzero_si256 (),
+           (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epu32 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2udq256_mask (__C, ( __v8si) __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epu32 (__mmask8 __A, __m128h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2udq256_mask (__B,
+           (__v8si) _mm256_setzero_si256 (),
+           __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epi32 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2dq128_mask (__A,
+           (__v4si) _mm_setzero_si128 (),
+           (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epi32 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)__builtin_ia32_vcvttph2dq128_mask (__C,
+           ( __v4si) __A,
+           __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epi32 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2dq128_mask (__B,
+           (__v4si) _mm_setzero_si128 (),
+           __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epi32 (__m128h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2dq256_mask (__A,
+           (__v8si)
+           _mm256_setzero_si256 (),
+           (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epi32 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2dq256_mask (__C,
+           ( __v8si) __A,
+           __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epi32 (__mmask8 __A, __m128h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2dq256_mask (__B,
+           (__v8si)
+           _mm256_setzero_si256 (),
+           __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epu32 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2udq128_mask (__A,
+     (__v4si)
+     _mm_setzero_si128 (),
+     (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epu32 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2udq128_mask (__C,
+     ( __v4si) __A,
+     __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epu32 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2udq128_mask (__B,
+     (__v4si)
+     _mm_setzero_si128 (),
+     __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epu32 (__m128h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2udq256_mask (__A,
+     (__v8si)
+     _mm256_setzero_si256 (), (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epu32 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2udq256_mask (__C,
+     ( __v8si) __A,
+     __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epu32 (__mmask8 __A, __m128h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2udq256_mask (__B,
+     (__v8si)
+     _mm256_setzero_si256 (),
+     __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepi32_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtdq2ph128_mask ((__v4si) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepi32_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtdq2ph128_mask ((__v4si) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepi32_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtdq2ph128_mask ((__v4si) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepi32_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtdq2ph256_mask ((__v8si) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepi32_ph (__m128h __A, __mmask8 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtdq2ph256_mask ((__v8si) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepi32_ph (__mmask8 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtdq2ph256_mask ((__v8si) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepu32_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtudq2ph128_mask ((__v4si) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepu32_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtudq2ph128_mask ((__v4si) __C,
+         __A,
+         __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepu32_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtudq2ph128_mask ((__v4si) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepu32_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtudq2ph256_mask ((__v8si) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepu32_ph (__m128h __A, __mmask8 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtudq2ph256_mask ((__v8si) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepu32_ph (__mmask8 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtudq2ph256_mask ((__v8si) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epi64 (__m128h __A)
+{
+  return
+    __builtin_ia32_vcvtph2qq128_mask (__A,
+          _mm_setzero_si128 (),
+          (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epi64 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2qq128_mask (__C, __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2qq128_mask (__B,
+        _mm_setzero_si128 (),
+        __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epi64 (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2qq256_mask (__A,
+        _mm256_setzero_si256 (),
+        (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epi64 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2qq256_mask (__C, __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2qq256_mask (__B,
+        _mm256_setzero_si256 (),
+        __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2uqq128_mask (__A,
+         _mm_setzero_si128 (),
+         (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epu64 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2uqq128_mask (__C, __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2uqq128_mask (__B,
+         _mm_setzero_si128 (),
+         __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2uqq256_mask (__A,
+         _mm256_setzero_si256 (),
+         (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epu64 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2uqq256_mask (__C, __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2uqq256_mask (__B,
+         _mm256_setzero_si256 (),
+         __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epi64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2qq128_mask (__A,
+         _mm_setzero_si128 (),
+         (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epi64 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2qq128_mask (__C,
+         __A,
+         __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2qq128_mask (__B,
+         _mm_setzero_si128 (),
+         __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epi64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2qq256_mask (__A,
+         _mm256_setzero_si256 (),
+         (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epi64 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2qq256_mask (__C,
+         __A,
+         __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epi64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2qq256_mask (__B,
+         _mm256_setzero_si256 (),
+         __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2uqq128_mask (__A,
+          _mm_setzero_si128 (),
+          (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epu64 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2uqq128_mask (__C,
+          __A,
+          __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2uqq128_mask (__B,
+          _mm_setzero_si128 (),
+          __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epu64 (__m128h __A)
+{
+  return __builtin_ia32_vcvttph2uqq256_mask (__A,
+          _mm256_setzero_si256 (),
+          (__mmask8) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epu64 (__m256i __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvttph2uqq256_mask (__C,
+          __A,
+          __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epu64 (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvttph2uqq256_mask (__B,
+          _mm256_setzero_si256 (),
+          __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepi64_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtqq2ph128_mask ((__v2di) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepi64_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtqq2ph128_mask ((__v2di) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepi64_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtqq2ph128_mask ((__v2di) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepi64_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtqq2ph256_mask ((__v4di) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepi64_ph (__m128h __A, __mmask8 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtqq2ph256_mask ((__v4di) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepi64_ph (__mmask8 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtqq2ph256_mask ((__v4di) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepu64_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtuqq2ph128_mask ((__v2di) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepu64_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtuqq2ph128_mask ((__v2di) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepu64_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtuqq2ph128_mask ((__v2di) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepu64_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtuqq2ph256_mask ((__v4di) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepu64_ph (__m128h __A, __mmask8 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtuqq2ph256_mask ((__v4di) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepu64_ph (__mmask8 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtuqq2ph256_mask ((__v4di) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epi16 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2w128_mask (__A,
+         (__v8hi)
+         _mm_setzero_si128 (),
+         (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epi16 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2w128_mask (__C, ( __v8hi) __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epi16 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2w128_mask (__B,
+         (__v8hi)
+         _mm_setzero_si128 (),
+         __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epi16 (__m256h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2w256_mask (__A,
+         (__v16hi)
+         _mm256_setzero_si256 (),
+         (__mmask16) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epi16 (__m256i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2w256_mask (__C, ( __v16hi) __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epi16 (__mmask16 __A, __m256h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2w256_mask (__B,
+         (__v16hi)
+         _mm256_setzero_si256 (),
+         __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_epu16 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2uw128_mask (__A,
+          (__v8hi)
+          _mm_setzero_si128 (),
+          (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_epu16 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2uw128_mask (__C, ( __v8hi) __A, __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_epu16 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvtph2uw128_mask (__B,
+          (__v8hi)
+          _mm_setzero_si128 (),
+          __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_epu16 (__m256h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2uw256_mask (__A,
+          (__v16hi)
+          _mm256_setzero_si256 (),
+          (__mmask16) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_epu16 (__m256i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2uw256_mask (__C, ( __v16hi) __A, __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_epu16 (__mmask16 __A, __m256h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvtph2uw256_mask (__B,
+          (__v16hi)
+          _mm256_setzero_si256 (),
+          __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epi16 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2w128_mask (__A,
+          (__v8hi)
+          _mm_setzero_si128 (),
+          (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epi16 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2w128_mask (__C,
+          ( __v8hi) __A,
+          __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epi16 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2w128_mask (__B,
+          (__v8hi)
+          _mm_setzero_si128 (),
+          __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epi16 (__m256h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2w256_mask (__A,
+          (__v16hi)
+          _mm256_setzero_si256 (),
+          (__mmask16) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epi16 (__m256i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2w256_mask (__C,
+          ( __v16hi) __A,
+          __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epi16 (__mmask16 __A, __m256h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2w256_mask (__B,
+          (__v16hi)
+          _mm256_setzero_si256 (),
+          __A);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttph_epu16 (__m128h __A)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2uw128_mask (__A,
+           (__v8hi)
+           _mm_setzero_si128 (),
+           (__mmask8) -1);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvttph_epu16 (__m128i __A, __mmask8 __B, __m128h __C)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2uw128_mask (__C,
+           ( __v8hi) __A,
+           __B);
+}
+extern __inline __m128i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvttph_epu16 (__mmask8 __A, __m128h __B)
+{
+  return (__m128i)
+    __builtin_ia32_vcvttph2uw128_mask (__B,
+           (__v8hi)
+           _mm_setzero_si128 (),
+           __A);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvttph_epu16 (__m256h __A)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2uw256_mask (__A,
+           (__v16hi)
+           _mm256_setzero_si256 (),
+           (__mmask16) -1);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvttph_epu16 (__m256i __A, __mmask16 __B, __m256h __C)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2uw256_mask (__C,
+           ( __v16hi) __A,
+           __B);
+}
+extern __inline __m256i
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvttph_epu16 (__mmask16 __A, __m256h __B)
+{
+  return (__m256i)
+    __builtin_ia32_vcvttph2uw256_mask (__B,
+           (__v16hi) _mm256_setzero_si256 (),
+           __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepi16_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtw2ph128_mask ((__v8hi) __A,
+       _mm_setzero_ph (),
+       (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepi16_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtw2ph128_mask ((__v8hi) __C,
+       __A,
+       __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepi16_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtw2ph128_mask ((__v8hi) __B,
+       _mm_setzero_ph (),
+       __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepi16_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtw2ph256_mask ((__v16hi) __A,
+       _mm256_setzero_ph (),
+       (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepi16_ph (__m256h __A, __mmask16 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtw2ph256_mask ((__v16hi) __C,
+       __A,
+       __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepi16_ph (__mmask16 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtw2ph256_mask ((__v16hi) __B,
+       _mm256_setzero_ph (),
+       __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtepu16_ph (__m128i __A)
+{
+  return __builtin_ia32_vcvtuw2ph128_mask ((__v8hi) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtepu16_ph (__m128h __A, __mmask8 __B, __m128i __C)
+{
+  return __builtin_ia32_vcvtuw2ph128_mask ((__v8hi) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtepu16_ph (__mmask8 __A, __m128i __B)
+{
+  return __builtin_ia32_vcvtuw2ph128_mask ((__v8hi) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtepu16_ph (__m256i __A)
+{
+  return __builtin_ia32_vcvtuw2ph256_mask ((__v16hi) __A,
+        _mm256_setzero_ph (),
+        (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtepu16_ph (__m256h __A, __mmask16 __B, __m256i __C)
+{
+  return __builtin_ia32_vcvtuw2ph256_mask ((__v16hi) __C, __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtepu16_ph (__mmask16 __A, __m256i __B)
+{
+  return __builtin_ia32_vcvtuw2ph256_mask ((__v16hi) __B,
+        _mm256_setzero_ph (),
+        __A);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtph_pd (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2pd128_mask (__A,
+        _mm_setzero_pd (),
+        (__mmask8) -1);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtph_pd (__m128d __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2pd128_mask (__C, __A, __B);
+}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtph_pd (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2pd128_mask (__B, _mm_setzero_pd (), __A);
+}
+extern __inline __m256d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtph_pd (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2pd256_mask (__A,
+        _mm256_setzero_pd (),
+        (__mmask8) -1);
+}
+extern __inline __m256d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtph_pd (__m256d __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2pd256_mask (__C, __A, __B);
+}
+extern __inline __m256d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtph_pd (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2pd256_mask (__B,
+        _mm256_setzero_pd (),
+        __A);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtxph_ps (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2psx128_mask (__A,
+        _mm_setzero_ps (),
+        (__mmask8) -1);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtxph_ps (__m128 __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2psx128_mask (__C, __A, __B);
+}
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtxph_ps (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2psx128_mask (__B, _mm_setzero_ps (), __A);
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtxph_ps (__m128h __A)
+{
+  return __builtin_ia32_vcvtph2psx256_mask (__A,
+         _mm256_setzero_ps (),
+         (__mmask8) -1);
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtxph_ps (__m256 __A, __mmask8 __B, __m128h __C)
+{
+  return __builtin_ia32_vcvtph2psx256_mask (__C, __A, __B);
+}
+extern __inline __m256
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtxph_ps (__mmask8 __A, __m128h __B)
+{
+  return __builtin_ia32_vcvtph2psx256_mask (__B,
+         _mm256_setzero_ps (),
+         __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtxps_ph (__m128 __A)
+{
+  return __builtin_ia32_vcvtps2phx128_mask ((__v4sf) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtxps_ph (__m128h __A, __mmask8 __B, __m128 __C)
+{
+  return __builtin_ia32_vcvtps2phx128_mask ((__v4sf) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtxps_ph (__mmask8 __A, __m128 __B)
+{
+  return __builtin_ia32_vcvtps2phx128_mask ((__v4sf) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtxps_ph (__m256 __A)
+{
+  return __builtin_ia32_vcvtps2phx256_mask ((__v8sf) __A,
+         _mm_setzero_ph (),
+         (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtxps_ph (__m128h __A, __mmask8 __B, __m256 __C)
+{
+  return __builtin_ia32_vcvtps2phx256_mask ((__v8sf) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtxps_ph (__mmask8 __A, __m256 __B)
+{
+  return __builtin_ia32_vcvtps2phx256_mask ((__v8sf) __B,
+         _mm_setzero_ph (),
+         __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpd_ph (__m128d __A)
+{
+  return __builtin_ia32_vcvtpd2ph128_mask ((__v2df) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_cvtpd_ph (__m128h __A, __mmask8 __B, __m128d __C)
+{
+  return __builtin_ia32_vcvtpd2ph128_mask ((__v2df) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_cvtpd_ph (__mmask8 __A, __m128d __B)
+{
+  return __builtin_ia32_vcvtpd2ph128_mask ((__v2df) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_cvtpd_ph (__m256d __A)
+{
+  return __builtin_ia32_vcvtpd2ph256_mask ((__v4df) __A,
+        _mm_setzero_ph (),
+        (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_cvtpd_ph (__m128h __A, __mmask8 __B, __m256d __C)
+{
+  return __builtin_ia32_vcvtpd2ph256_mask ((__v4df) __C, __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_cvtpd_ph (__mmask8 __A, __m256d __B)
+{
+  return __builtin_ia32_vcvtpd2ph256_mask ((__v4df) __B,
+        _mm_setzero_ph (),
+        __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmaddsub_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h)__builtin_ia32_vfmaddsubph256_mask ((__v16hf)__A,
+            (__v16hf)__B,
+            (__v16hf)__C,
+            (__mmask16)-1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmaddsub_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddsubph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fmaddsub_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfmaddsubph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmaddsub_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddsubph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmaddsub_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h)__builtin_ia32_vfmaddsubph128_mask ((__v8hf)__A,
+            (__v8hf)__B,
+            (__v8hf)__C,
+            (__mmask8)-1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmaddsub_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddsubph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmaddsub_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmaddsubph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmaddsub_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddsubph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmsubadd_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmsubadd_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fmsubadd_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfmsubaddph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmsubadd_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubaddph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmsubadd_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmsubadd_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmsubadd_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmsubaddph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmsubadd_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubaddph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmadd_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmadd_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fmadd_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfmaddph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmadd_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmaddph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fnmadd_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fnmadd_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmaddph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fnmadd_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfnmaddph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fnmadd_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmaddph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmadd_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmadd_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmaddph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmadd_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfnmaddph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmadd_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmaddph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmsub_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmsub_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fmsub_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfmsubph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmsub_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmsubph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmsub_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmsub_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmsub_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfmsubph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmsub_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmsubph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fnmsub_ph (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmsubph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) -1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fnmsub_ph (__m256h __A, __mmask16 __U, __m256h __B,
+    __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmsubph256_mask ((__v16hf) __A,
+             (__v16hf) __B,
+             (__v16hf) __C,
+             (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fnmsub_ph (__m256h __A, __m256h __B, __m256h __C,
+     __mmask16 __U)
+{
+  return (__m256h) __builtin_ia32_vfnmsubph256_mask3 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fnmsub_ph (__mmask16 __U, __m256h __A, __m256h __B,
+     __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfnmsubph256_maskz ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C,
+       (__mmask16)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fnmsub_ph (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmsubph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) -1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fnmsub_ph (__m128h __A, __mmask8 __U, __m128h __B,
+        __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmsubph128_mask ((__v8hf) __A,
+             (__v8hf) __B,
+             (__v8hf) __C,
+             (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fnmsub_ph (__m128h __A, __m128h __B, __m128h __C,
+         __mmask8 __U)
+{
+  return (__m128h) __builtin_ia32_vfnmsubph128_mask3 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fnmsub_ph (__mmask8 __U, __m128h __A, __m128h __B,
+         __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfnmsubph128_maskz ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C,
+       (__mmask8)
+       __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmadd_pch (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmaddcph128 ((__v8hf) __A,
+      (__v8hf) __B,
+      (__v8hf) __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmadd_pch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcph128_mask ((__v8hf) __A,
+          (__v8hf) __C,
+          (__v8hf) __D, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fmadd_pch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfmaddcph128_mask3 ((__v8hf) __A,
+           (__v8hf) __B,
+           (__v8hf) __C, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmadd_pch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D)
+{
+  return (__m128h) __builtin_ia32_vfmaddcph128_maskz ((__v8hf) __B,
+            (__v8hf) __C,
+            (__v8hf) __D, __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmadd_pch (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmaddcph256 ((__v16hf) __A,
+      (__v16hf) __B,
+      (__v16hf) __C);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmadd_pch (__m256h __A, __mmask8 __B, __m256h __C, __m256h __D)
+{
+  return (__m256h)
+     __builtin_ia32_vfmaddcph256_mask ((__v16hf) __A,
+           (__v16hf) __C,
+           (__v16hf) __D, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fmadd_pch (__m256h __A, __m256h __B, __m256h __C, __mmask8 __D)
+{
+  return (__m256h)
+    __builtin_ia32_vfmaddcph256_mask3 ((__v16hf) __A,
+           (__v16hf) __B,
+           (__v16hf) __C, __D);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmadd_pch (__mmask8 __A, __m256h __B, __m256h __C, __m256h __D)
+{
+  return (__m256h)__builtin_ia32_vfmaddcph256_maskz ((__v16hf) __B,
+           (__v16hf) __C,
+           (__v16hf) __D, __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmadd_pch (__m128h __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfcmaddcph128 ((__v8hf) __A,
+       (__v8hf) __B,
+       (__v8hf) __C);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmadd_pch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)
+     __builtin_ia32_vfcmaddcph128_mask ((__v8hf) __A,
+     (__v8hf) __C,
+     (__v8hf) __D, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask3_fcmadd_pch (__m128h __A, __m128h __B, __m128h __C, __mmask8 __D)
+{
+  return (__m128h)
+    __builtin_ia32_vfcmaddcph128_mask3 ((__v8hf) __A,
+     (__v8hf) __B,
+     (__v8hf) __C, __D);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmadd_pch (__mmask8 __A, __m128h __B, __m128h __C, __m128h __D)
+{
+  return (__m128h)__builtin_ia32_vfcmaddcph128_maskz ((__v8hf) __B,
+            (__v8hf) __C,
+            (__v8hf) __D, __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fcmadd_pch (__m256h __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfcmaddcph256 ((__v16hf) __A,
+       (__v16hf) __B,
+       (__v16hf) __C);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fcmadd_pch (__m256h __A, __mmask8 __B, __m256h __C, __m256h __D)
+{
+  return (__m256h)
+     __builtin_ia32_vfcmaddcph256_mask ((__v16hf) __A,
+     (__v16hf) __C,
+     (__v16hf) __D, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask3_fcmadd_pch (__m256h __A, __m256h __B, __m256h __C, __mmask8 __D)
+{
+  return (__m256h)
+    __builtin_ia32_vfcmaddcph256_mask3 ((__v16hf) __A,
+     (__v16hf) __B,
+     (__v16hf) __C, __D);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fcmadd_pch (__mmask8 __A, __m256h __B, __m256h __C, __m256h __D)
+{
+  return (__m256h) __builtin_ia32_vfcmaddcph256_maskz ((__v16hf) __B,
+             (__v16hf) __C,
+             (__v16hf) __D, __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fmul_pch (__m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfmulcph128 ((__v8hf) __A, (__v8hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fmul_pch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h) __builtin_ia32_vfmulcph128_mask ((__v8hf) __C,
+          (__v8hf) __D,
+          (__v8hf) __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fmul_pch (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfmulcph128_mask ((__v8hf) __B,
+          (__v8hf) __C,
+          _mm_setzero_ph (),
+          __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fmul_pch (__m256h __A, __m256h __B)
+{
+  return (__m256h) __builtin_ia32_vfmulcph256 ((__v16hf) __A,
+            (__v16hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fmul_pch (__m256h __A, __mmask8 __B, __m256h __C, __m256h __D)
+{
+  return (__m256h) __builtin_ia32_vfmulcph256_mask ((__v16hf) __C,
+          (__v16hf) __D,
+          (__v16hf) __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fmul_pch (__mmask8 __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfmulcph256_mask ((__v16hf) __B,
+          (__v16hf) __C,
+          _mm256_setzero_ph (),
+          __A);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_fcmul_pch (__m128h __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vfcmulcph128 ((__v8hf) __A,
+      (__v8hf) __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_fcmul_pch (__m128h __A, __mmask8 __B, __m128h __C, __m128h __D)
+{
+  return (__m128h) __builtin_ia32_vfcmulcph128_mask ((__v8hf) __C,
+           (__v8hf) __D,
+           (__v8hf) __A, __B);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_fcmul_pch (__mmask8 __A, __m128h __B, __m128h __C)
+{
+  return (__m128h) __builtin_ia32_vfcmulcph128_mask ((__v8hf) __B,
+           (__v8hf) __C,
+           _mm_setzero_ph (),
+           __A);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_fcmul_pch (__m256h __A, __m256h __B)
+{
+  return (__m256h) __builtin_ia32_vfcmulcph256 ((__v16hf) __A, (__v16hf) __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_fcmul_pch (__m256h __A, __mmask8 __B, __m256h __C, __m256h __D)
+{
+  return (__m256h) __builtin_ia32_vfcmulcph256_mask ((__v16hf) __C,
+           (__v16hf) __D,
+           (__v16hf) __A, __B);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_fcmul_pch (__mmask8 __A, __m256h __B, __m256h __C)
+{
+  return (__m256h) __builtin_ia32_vfcmulcph256_mask ((__v16hf) __B,
+           (__v16hf) __C,
+           _mm256_setzero_ph (),
+           __A);
+}
+#define _MM256_REDUCE_OP(op) __m128h __T1 = (__m128h) _mm256_extractf128_pd ((__m256d) __A, 0); __m128h __T2 = (__m128h) _mm256_extractf128_pd ((__m256d) __A, 1); __m128h __T3 = (__T1 op __T2); __m128h __T4 = (__m128h) __builtin_shuffle (__T3, (__v8hi) { 4, 5, 6, 7, 0, 1, 2, 3 }); __m128h __T5 = (__T3) op (__T4); __m128h __T6 = (__m128h) __builtin_shuffle (__T5, (__v8hi) { 2, 3, 0, 1, 4, 5, 6, 7 }); __m128h __T7 = __T5 op __T6; return __T7[0] op __T7[1]
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_reduce_add_ph (__m256h __A)
+{
+  _MM256_REDUCE_OP (+);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_reduce_mul_ph (__m256h __A)
+{
+  _MM256_REDUCE_OP (*);
+}
+#undef _MM256_REDUCE_OP
+#define _MM256_REDUCE_OP(op) __m128h __T1 = (__m128h) _mm256_extractf128_pd ((__m256d) __A, 0); __m128h __T2 = (__m128h) _mm256_extractf128_pd ((__m256d) __A, 1); __m128h __T3 = _mm_##op (__T1, __T2); __m128h __T4 = (__m128h) __builtin_shuffle (__T3, (__v8hi) { 2, 3, 0, 1, 6, 7, 4, 5 }); __m128h __T5 = _mm_##op (__T3, __T4); __m128h __T6 = (__m128h) __builtin_shuffle (__T5, (__v8hi) { 4, 5 }); __m128h __T7 = _mm_##op (__T5, __T6); __m128h __T8 = (__m128h) __builtin_shuffle (__T7, (__v8hi) { 1, 0 }); __m128h __T9 = _mm_##op (__T7, __T8); return __T9[0]
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_reduce_min_ph (__m256h __A)
+{
+  _MM256_REDUCE_OP (min_ph);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_reduce_max_ph (__m256h __A)
+{
+  _MM256_REDUCE_OP (max_ph);
+}
+#define _MM_REDUCE_OP(op) __m128h __T1 = (__m128h) __builtin_shuffle (__A, (__v8hi) { 4, 5, 6, 7, 0, 1, 2, 3 }); __m128h __T2 = (__A) op (__T1); __m128h __T3 = (__m128h) __builtin_shuffle (__T2, (__v8hi){ 2, 3, 0, 1, 4, 5, 6, 7 }); __m128h __T4 = __T2 op __T3; return __T4[0] op __T4[1]
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_add_ph (__m128h __A)
+{
+  _MM_REDUCE_OP (+);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_mul_ph (__m128h __A)
+{
+  _MM_REDUCE_OP (*);
+}
+#undef _MM_REDUCE_OP
+#define _MM_REDUCE_OP(op) __m128h __T1 = (__m128h) __builtin_shuffle (__A, (__v8hi) { 2, 3, 0, 1, 6, 7, 4, 5 }); __m128h __T2 = _mm_##op (__A, __T1); __m128h __T3 = (__m128h) __builtin_shuffle (__T2, (__v8hi){ 4, 5 }); __m128h __T4 = _mm_##op (__T2, __T3); __m128h __T5 = (__m128h) __builtin_shuffle (__T4, (__v8hi){ 1, 0 }); __m128h __T6 = _mm_##op (__T4, __T5); return __T6[0]
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_min_ph (__m128h __A)
+{
+  _MM_REDUCE_OP (min_ph);
+}
+extern __inline _Float16
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_reduce_max_ph (__m128h __A)
+{
+  _MM_REDUCE_OP (max_ph);
+}
+#undef _MM256_REDUCE_OP
+#undef _MM_REDUCE_OP
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_blend_ph (__mmask16 __U, __m256h __A, __m256h __W)
+{
+  return (__m256h) __builtin_ia32_movdquhi256_mask ((__v16hi) __W,
+          (__v16hi) __A,
+          (__mmask16) __U);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_permutex2var_ph (__m256h __A, __m256i __I, __m256h __B)
+{
+  return (__m256h) __builtin_ia32_vpermi2varhi256_mask ((__v16hi) __A,
+             (__v16hi) __I,
+             (__v16hi) __B,
+             (__mmask16)-1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_permutexvar_ph (__m256i __A, __m256h __B)
+{
+  return (__m256h) __builtin_ia32_permvarhi256_mask ((__v16hi) __B,
+           (__v16hi) __A,
+           (__v16hi)
+           (_mm256_setzero_ph ()),
+           (__mmask16)-1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_blend_ph (__mmask8 __U, __m128h __A, __m128h __W)
+{
+  return (__m128h) __builtin_ia32_movdquhi128_mask ((__v8hi) __W,
+          (__v8hi) __A,
+          (__mmask8) __U);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_permutex2var_ph (__m128h __A, __m128i __I, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_vpermi2varhi128_mask ((__v8hi) __A,
+             (__v8hi) __I,
+             (__v8hi) __B,
+             (__mmask8)-1);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_permutexvar_ph (__m128i __A, __m128h __B)
+{
+  return (__m128h) __builtin_ia32_permvarhi128_mask ((__v8hi) __B,
+           (__v8hi) __A,
+           (__v8hi)
+           (_mm_setzero_ph ()),
+           (__mmask8)-1);
+}
+extern __inline __m256h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_set1_pch (_Float16 _Complex __A)
+{
+  union
+  {
+    _Float16 _Complex a;
+    float b;
+  } u = { .a = __A };
+  return (__m256h) _mm256_set1_ps (u.b);
+}
+extern __inline __m128h
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_pch (_Float16 _Complex __A)
+{
+  union
+  {
+    _Float16 _Complex a;
+    float b;
+  } u = { .a = __A };
+  return (__m128h) _mm_set1_ps (u.b);
+}
+#define _mm_mul_pch(A, B) _mm_fmul_pch ((A), (B))
+#define _mm_mask_mul_pch(W, U, A, B) _mm_mask_fmul_pch ((W), (U), (A), (B))
+#define _mm_maskz_mul_pch(U, A, B) _mm_maskz_fmul_pch ((U), (A), (B))
+#define _mm256_mul_pch(A, B) _mm256_fmul_pch ((A), (B))
+#define _mm256_mask_mul_pch(W, U, A, B) _mm256_mask_fmul_pch ((W), (U), (A), (B))
+#define _mm256_maskz_mul_pch(U, A, B) _mm256_maskz_fmul_pch ((U), (A), (B))
+#define _mm_cmul_pch(A, B) _mm_fcmul_pch ((A), (B))
+#define _mm_mask_cmul_pch(W, U, A, B) _mm_mask_fcmul_pch ((W), (U), (A), (B))
+#define _mm_maskz_cmul_pch(U, A, B) _mm_maskz_fcmul_pch ((U), (A), (B))
+#define _mm256_cmul_pch(A, B) _mm256_fcmul_pch ((A), (B))
+#define _mm256_mask_cmul_pch(W, U, A, B) _mm256_mask_fcmul_pch ((W), (U), (A), (B))
+#define _mm256_maskz_cmul_pch(U, A, B) _mm256_maskz_fcmul_pch((U), (A), (B))
+#ifdef __DISABLE_AVX512FP16VL__
+#undef __DISABLE_AVX512FP16VL__
+#pragma GCC pop_options
+#endif
+#endif
+#endif
diff --git a/third_party/intel/avx512vlintrin.internal.h b/third_party/intel/avx512vlintrin.internal.h
index 73d7e3054..493f80abb 100644
--- a/third_party/intel/avx512vlintrin.internal.h
+++ b/third_party/intel/avx512vlintrin.internal.h
@@ -9486,10 +9486,12 @@ __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm256_ternarylogic_epi64 (__m256i __A, __m256i __B, __m256i __C,
       const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogq256_mask ((__v4di) __A,
-           (__v4di) __B,
-           (__v4di) __C, __imm,
-           (__mmask8) -1);
+  return (__m256i)
+    __builtin_ia32_pternlogq256_mask ((__v4di) __A,
+          (__v4di) __B,
+          (__v4di) __C,
+          (unsigned char) __imm,
+          (__mmask8) -1);
 }
 extern __inline __m256i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -9497,10 +9499,12 @@ _mm256_mask_ternarylogic_epi64 (__m256i __A, __mmask8 __U,
     __m256i __B, __m256i __C,
     const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogq256_mask ((__v4di) __A,
-           (__v4di) __B,
-           (__v4di) __C, __imm,
-           (__mmask8) __U);
+  return (__m256i)
+    __builtin_ia32_pternlogq256_mask ((__v4di) __A,
+          (__v4di) __B,
+          (__v4di) __C,
+          (unsigned char) __imm,
+          (__mmask8) __U);
 }
 extern __inline __m256i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -9508,21 +9512,24 @@ _mm256_maskz_ternarylogic_epi64 (__mmask8 __U, __m256i __A,
      __m256i __B, __m256i __C,
      const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogq256_maskz ((__v4di) __A,
-            (__v4di) __B,
-            (__v4di) __C,
-            __imm,
-            (__mmask8) __U);
+  return (__m256i)
+    __builtin_ia32_pternlogq256_maskz ((__v4di) __A,
+           (__v4di) __B,
+           (__v4di) __C,
+           (unsigned char) __imm,
+           (__mmask8) __U);
 }
 extern __inline __m256i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm256_ternarylogic_epi32 (__m256i __A, __m256i __B, __m256i __C,
       const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogd256_mask ((__v8si) __A,
-           (__v8si) __B,
-           (__v8si) __C, __imm,
-           (__mmask8) -1);
+  return (__m256i)
+    __builtin_ia32_pternlogd256_mask ((__v8si) __A,
+          (__v8si) __B,
+          (__v8si) __C,
+          (unsigned char) __imm,
+          (__mmask8) -1);
 }
 extern __inline __m256i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -9530,10 +9537,12 @@ _mm256_mask_ternarylogic_epi32 (__m256i __A, __mmask8 __U,
     __m256i __B, __m256i __C,
     const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogd256_mask ((__v8si) __A,
-           (__v8si) __B,
-           (__v8si) __C, __imm,
-           (__mmask8) __U);
+  return (__m256i)
+    __builtin_ia32_pternlogd256_mask ((__v8si) __A,
+          (__v8si) __B,
+          (__v8si) __C,
+          (unsigned char) __imm,
+          (__mmask8) __U);
 }
 extern __inline __m256i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -9541,73 +9550,88 @@ _mm256_maskz_ternarylogic_epi32 (__mmask8 __U, __m256i __A,
      __m256i __B, __m256i __C,
      const int __imm)
 {
-  return (__m256i) __builtin_ia32_pternlogd256_maskz ((__v8si) __A,
-            (__v8si) __B,
-            (__v8si) __C,
-            __imm,
-            (__mmask8) __U);
+  return (__m256i)
+    __builtin_ia32_pternlogd256_maskz ((__v8si) __A,
+           (__v8si) __B,
+           (__v8si) __C,
+           (unsigned char) __imm,
+           (__mmask8) __U);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_ternarylogic_epi64 (__m128i __A, __m128i __B, __m128i __C,
    const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogq128_mask ((__v2di) __A,
-           (__v2di) __B,
-           (__v2di) __C, __imm,
-           (__mmask8) -1);
+  return (__m128i)
+    __builtin_ia32_pternlogq128_mask ((__v2di) __A,
+          (__v2di) __B,
+          (__v2di) __C,
+          (unsigned char) __imm,
+          (__mmask8) -1);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_mask_ternarylogic_epi64 (__m128i __A, __mmask8 __U,
-        __m128i __B, __m128i __C, const int __imm)
+        __m128i __B, __m128i __C,
+        const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogq128_mask ((__v2di) __A,
-           (__v2di) __B,
-           (__v2di) __C, __imm,
-           (__mmask8) __U);
+  return (__m128i)
+    __builtin_ia32_pternlogq128_mask ((__v2di) __A,
+          (__v2di) __B,
+          (__v2di) __C,
+          (unsigned char) __imm,
+          (__mmask8) __U);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_maskz_ternarylogic_epi64 (__mmask8 __U, __m128i __A,
-         __m128i __B, __m128i __C, const int __imm)
+         __m128i __B, __m128i __C,
+         const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogq128_maskz ((__v2di) __A,
-            (__v2di) __B,
-            (__v2di) __C,
-            __imm,
-            (__mmask8) __U);
+  return (__m128i)
+    __builtin_ia32_pternlogq128_maskz ((__v2di) __A,
+           (__v2di) __B,
+           (__v2di) __C,
+           (unsigned char) __imm,
+           (__mmask8) __U);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_ternarylogic_epi32 (__m128i __A, __m128i __B, __m128i __C,
    const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogd128_mask ((__v4si) __A,
-           (__v4si) __B,
-           (__v4si) __C, __imm,
-           (__mmask8) -1);
+  return (__m128i)
+    __builtin_ia32_pternlogd128_mask ((__v4si) __A,
+          (__v4si) __B,
+          (__v4si) __C,
+          (unsigned char) __imm,
+          (__mmask8) -1);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_mask_ternarylogic_epi32 (__m128i __A, __mmask8 __U,
-        __m128i __B, __m128i __C, const int __imm)
+        __m128i __B, __m128i __C,
+        const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogd128_mask ((__v4si) __A,
-           (__v4si) __B,
-           (__v4si) __C, __imm,
-           (__mmask8) __U);
+  return (__m128i)
+    __builtin_ia32_pternlogd128_mask ((__v4si) __A,
+          (__v4si) __B,
+          (__v4si) __C,
+          (unsigned char) __imm,
+          (__mmask8) __U);
 }
 extern __inline __m128i
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
 _mm_maskz_ternarylogic_epi32 (__mmask8 __U, __m128i __A,
-         __m128i __B, __m128i __C, const int __imm)
+         __m128i __B, __m128i __C,
+         const int __imm)
 {
-  return (__m128i) __builtin_ia32_pternlogd128_maskz ((__v4si) __A,
-            (__v4si) __B,
-            (__v4si) __C,
-            __imm,
-            (__mmask8) __U);
+  return (__m128i)
+    __builtin_ia32_pternlogd128_maskz ((__v4si) __A,
+           (__v4si) __B,
+           (__v4si) __C,
+           (unsigned char) __imm,
+           (__mmask8) __U);
 }
 extern __inline __m256
 __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
@@ -11335,18 +11359,18 @@ _mm256_permutex_pd (__m256d __X, const int __M)
 #define _mm_maskz_slli_epi32(U, X, C) ((__m128i)__builtin_ia32_pslldi128_mask ((__v4si)(__m128i)(X), (int)(C), (__v4si)(__m128i)_mm_setzero_si128 (), (__mmask8)(U)))
 #define _mm_mask_slli_epi64(W, U, X, C) ((__m128i)__builtin_ia32_psllqi128_mask ((__v2di)(__m128i)(X), (int)(C), (__v2di)(__m128i)(W), (__mmask8)(U)))
 #define _mm_maskz_slli_epi64(U, X, C) ((__m128i)__builtin_ia32_psllqi128_mask ((__v2di)(__m128i)(X), (int)(C), (__v2di)(__m128i)_mm_setzero_si128 (), (__mmask8)(U)))
-#define _mm256_ternarylogic_epi64(A, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_mask ((__v4di)(__m256i)(A), (__v4di)(__m256i)(B), (__v4di)(__m256i)(C), (int)(I), (__mmask8)-1))
-#define _mm256_mask_ternarylogic_epi64(A, U, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_mask ((__v4di)(__m256i)(A), (__v4di)(__m256i)(B), (__v4di)(__m256i)(C), (int)(I), (__mmask8)(U)))
-#define _mm256_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_maskz ((__v4di)(__m256i)(A), (__v4di)(__m256i)(B), (__v4di)(__m256i)(C), (int)(I), (__mmask8)(U)))
-#define _mm256_ternarylogic_epi32(A, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_mask ((__v8si)(__m256i)(A), (__v8si)(__m256i)(B), (__v8si)(__m256i)(C), (int)(I), (__mmask8)-1))
-#define _mm256_mask_ternarylogic_epi32(A, U, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_mask ((__v8si)(__m256i)(A), (__v8si)(__m256i)(B), (__v8si)(__m256i)(C), (int)(I), (__mmask8)(U)))
-#define _mm256_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_maskz ((__v8si)(__m256i)(A), (__v8si)(__m256i)(B), (__v8si)(__m256i)(C), (int)(I), (__mmask8)(U)))
-#define _mm_ternarylogic_epi64(A, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_mask ((__v2di)(__m128i)(A), (__v2di)(__m128i)(B), (__v2di)(__m128i)(C), (int)(I), (__mmask8)-1))
-#define _mm_mask_ternarylogic_epi64(A, U, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_mask ((__v2di)(__m128i)(A), (__v2di)(__m128i)(B), (__v2di)(__m128i)(C), (int)(I), (__mmask8)(U)))
-#define _mm_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_maskz ((__v2di)(__m128i)(A), (__v2di)(__m128i)(B), (__v2di)(__m128i)(C), (int)(I), (__mmask8)(U)))
-#define _mm_ternarylogic_epi32(A, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_mask ((__v4si)(__m128i)(A), (__v4si)(__m128i)(B), (__v4si)(__m128i)(C), (int)(I), (__mmask8)-1))
-#define _mm_mask_ternarylogic_epi32(A, U, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_mask ((__v4si)(__m128i)(A), (__v4si)(__m128i)(B), (__v4si)(__m128i)(C), (int)(I), (__mmask8)(U)))
-#define _mm_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_maskz ((__v4si)(__m128i)(A), (__v4si)(__m128i)(B), (__v4si)(__m128i)(C), (int)(I), (__mmask8)(U)))
+#define _mm256_ternarylogic_epi64(A, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_mask ((__v4di) (__m256i) (A), (__v4di) (__m256i) (B), (__v4di) (__m256i) (C), (unsigned char) (I), (__mmask8) -1))
+#define _mm256_mask_ternarylogic_epi64(A, U, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_mask ((__v4di) (__m256i) (A), (__v4di) (__m256i) (B), (__v4di) (__m256i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm256_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m256i) __builtin_ia32_pternlogq256_maskz ((__v4di) (__m256i) (A), (__v4di) (__m256i) (B), (__v4di) (__m256i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm256_ternarylogic_epi32(A, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_mask ((__v8si) (__m256i) (A), (__v8si) (__m256i) (B), (__v8si) (__m256i) (C), (unsigned char) (I), (__mmask8) -1))
+#define _mm256_mask_ternarylogic_epi32(A, U, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_mask ((__v8si) (__m256i) (A), (__v8si) (__m256i) (B), (__v8si) (__m256i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm256_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m256i) __builtin_ia32_pternlogd256_maskz ((__v8si) (__m256i) (A), (__v8si) (__m256i) (B), (__v8si) (__m256i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm_ternarylogic_epi64(A, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_mask ((__v2di) (__m128i) (A), (__v2di) (__m128i) (B), (__v2di) (__m128i) (C), (unsigned char) (I), (__mmask8) -1))
+#define _mm_mask_ternarylogic_epi64(A, U, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_mask ((__v2di) (__m128i) (A), (__v2di) (__m128i) (B), (__v2di) (__m128i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm_maskz_ternarylogic_epi64(U, A, B, C, I) ((__m128i) __builtin_ia32_pternlogq128_maskz ((__v2di) (__m128i) (A), (__v2di) (__m128i) (B), (__v2di) (__m128i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm_ternarylogic_epi32(A, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_mask ((__v4si) (__m128i) (A), (__v4si) (__m128i) (B), (__v4si) (__m128i) (C), (unsigned char) (I), (__mmask8) -1))
+#define _mm_mask_ternarylogic_epi32(A, U, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_mask ((__v4si) (__m128i) (A), (__v4si) (__m128i) (B), (__v4si) (__m128i) (C), (unsigned char) (I), (__mmask8) (U)))
+#define _mm_maskz_ternarylogic_epi32(U, A, B, C, I) ((__m128i) __builtin_ia32_pternlogd128_maskz ((__v4si) (__m128i) (A), (__v4si) (__m128i) (B), (__v4si) (__m128i) (C), (unsigned char) (I), (__mmask8) (U)))
 #define _mm256_roundscale_ps(A, B) ((__m256) __builtin_ia32_rndscaleps_256_mask ((__v8sf)(__m256)(A), (int)(B), (__v8sf)(__m256)_mm256_setzero_ps (), (__mmask8)-1))
 #define _mm256_mask_roundscale_ps(W, U, A, B) ((__m256) __builtin_ia32_rndscaleps_256_mask ((__v8sf)(__m256)(A), (int)(B), (__v8sf)(__m256)(W), (__mmask8)(U)))
 #define _mm256_maskz_roundscale_ps(U, A, B) ((__m256) __builtin_ia32_rndscaleps_256_mask ((__v8sf)(__m256)(A), (int)(B), (__v8sf)(__m256)_mm256_setzero_ps (), (__mmask8)(U)))
diff --git a/third_party/intel/bmiintrin.internal.h b/third_party/intel/bmiintrin.internal.h
index cf2042f1b..0c17f0bd2 100644
--- a/third_party/intel/bmiintrin.internal.h
+++ b/third_party/intel/bmiintrin.internal.h
@@ -14,12 +14,22 @@ __tzcnt_u16 (unsigned short __X)
 {
   return __builtin_ia32_tzcnt_u16 (__X);
 }
+extern __inline unsigned short __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_tzcnt_u16 (unsigned short __X)
+{
+  return __builtin_ia32_tzcnt_u16 (__X);
+}
 extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 __andn_u32 (unsigned int __X, unsigned int __Y)
 {
   return ~__X & __Y;
 }
 extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_andn_u32 (unsigned int __X, unsigned int __Y)
+{
+  return __andn_u32 (__X, __Y);
+}
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 __bextr_u32 (unsigned int __X, unsigned int __Y)
 {
   return __builtin_ia32_bextr_u32 (__X, __Y);
@@ -76,6 +86,11 @@ __andn_u64 (unsigned long long __X, unsigned long long __Y)
   return ~__X & __Y;
 }
 extern __inline unsigned long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_andn_u64 (unsigned long long __X, unsigned long long __Y)
+{
+  return __andn_u64 (__X, __Y);
+}
+extern __inline unsigned long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 __bextr_u64 (unsigned long long __X, unsigned long long __Y)
 {
   return __builtin_ia32_bextr_u64 (__X, __Y);
diff --git a/third_party/intel/cpuid.internal.h b/third_party/intel/cpuid.internal.h
index 081886d9b..3f082193c 100644
--- a/third_party/intel/cpuid.internal.h
+++ b/third_party/intel/cpuid.internal.h
@@ -48,7 +48,6 @@
 #define bit_AVX2 (1 << 5)
 #define bit_BMI2 (1 << 8)
 #define bit_RTM (1 << 11)
-#define bit_MPX (1 << 14)
 #define bit_AVX512F (1 << 16)
 #define bit_AVX512DQ (1 << 17)
 #define bit_RDSEED (1 << 18)
@@ -84,6 +83,7 @@
 #define bit_AVX5124VNNIW (1 << 2)
 #define bit_AVX5124FMAPS (1 << 3)
 #define bit_AVX512VP2INTERSECT (1 << 8)
+#define bit_AVX512FP16 (1 << 23)
 #define bit_IBT (1 << 20)
 #define bit_UINTR (1 << 5)
 #define bit_PCONFIG (1 << 18)
@@ -92,8 +92,6 @@
 #define bit_AMX_BF16 (1 << 22)
 #define bit_AMX_TILE (1 << 24)
 #define bit_AMX_INT8 (1 << 25)
-#define bit_BNDREGS (1 << 3)
-#define bit_BNDCSR (1 << 4)
 #define bit_XSAVEOPT (1 << 0)
 #define bit_XSAVEC (1 << 1)
 #define bit_XSAVES (1 << 3)
diff --git a/third_party/intel/emmintrin.internal.h b/third_party/intel/emmintrin.internal.h
index d54c75287..c0c3bf59d 100644
--- a/third_party/intel/emmintrin.internal.h
+++ b/third_party/intel/emmintrin.internal.h
@@ -550,13 +550,12 @@ _mm_loadu_si64 (void const *__P)
 extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm_loadu_si32 (void const *__P)
 {
-  return _mm_set_epi32 (*(int *)__P, (int)0, (int)0, (int)0);
+  return _mm_set_epi32 (0, 0, 0, (*(__m32_u *)__P)[0]);
 }
 extern __inline __m128i __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm_loadu_si16 (void const *__P)
 {
-  return _mm_set_epi16 (*(short *)__P, (short)0, (short)0, (short)0,
-   (short)0, (short)0, (short)0, (short)0);
+  return _mm_set_epi16 (0, 0, 0, 0, 0, 0, 0, (*(__m16_u *)__P)[0]);
 }
 extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm_store_si128 (__m128i *__P, __m128i __B)
diff --git a/third_party/intel/ia32intrin.internal.h b/third_party/intel/ia32intrin.internal.h
index faeaff648..f6a5172c7 100644
--- a/third_party/intel/ia32intrin.internal.h
+++ b/third_party/intel/ia32intrin.internal.h
@@ -21,10 +21,10 @@ __bswapd (int __X)
   return __builtin_bswap32 (__X);
 }
 #ifndef __iamcu__
-#ifndef __SSE4_2__
+#ifndef __CRC32__
 #pragma GCC push_options
-#pragma GCC target("sse4.2")
-#define __DISABLE_SSE4_2__
+#pragma GCC target("crc32")
+#define __DISABLE_CRC32__
 #endif
 extern __inline unsigned int
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
@@ -44,8 +44,8 @@ __crc32d (unsigned int __C, unsigned int __V)
 {
   return __builtin_ia32_crc32si (__C, __V);
 }
-#ifdef __DISABLE_SSE4_2__
-#undef __DISABLE_SSE4_2__
+#ifdef __DISABLE_CRC32__
+#undef __DISABLE_CRC32__
 #pragma GCC pop_options
 #endif
 #endif
@@ -63,9 +63,19 @@ __rdpmc (int __S)
   return __builtin_ia32_rdpmc (__S);
 }
 #endif
-#define __rdtsc() __builtin_ia32_rdtsc ()
+extern __inline unsigned long long
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+__rdtsc (void)
+{
+  return __builtin_ia32_rdtsc ();
+}
 #ifndef __iamcu__
-#define __rdtscp(a) __builtin_ia32_rdtscp (a)
+extern __inline unsigned long long
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+__rdtscp (unsigned int *__A)
+{
+  return __builtin_ia32_rdtscp (__A);
+}
 #endif
 extern __inline unsigned char
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
@@ -130,10 +140,10 @@ __bswapq (long long __X)
 {
   return __builtin_bswap64 (__X);
 }
-#ifndef __SSE4_2__
+#ifndef __CRC32__
 #pragma GCC push_options
-#pragma GCC target("sse4.2")
-#define __DISABLE_SSE4_2__
+#pragma GCC target("crc32")
+#define __DISABLE_CRC32__
 #endif
 extern __inline unsigned long long
 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
@@ -141,8 +151,8 @@ __crc32q (unsigned long long __C, unsigned long long __V)
 {
   return __builtin_ia32_crc32di (__C, __V);
 }
-#ifdef __DISABLE_SSE4_2__
-#undef __DISABLE_SSE4_2__
+#ifdef __DISABLE_CRC32__
+#undef __DISABLE_CRC32__
 #pragma GCC pop_options
 #endif
 extern __inline long long
diff --git a/third_party/intel/immintrin.internal.h b/third_party/intel/immintrin.internal.h
index 60b7ec6d9..a932d9755 100644
--- a/third_party/intel/immintrin.internal.h
+++ b/third_party/intel/immintrin.internal.h
@@ -36,6 +36,10 @@
 #include "third_party/intel/avx512bitalgintrin.internal.h"
 #include "third_party/intel/avx512vp2intersectintrin.internal.h"
 #include "third_party/intel/avx512vp2intersectvlintrin.internal.h"
+#ifdef __SSE2__
+#include "third_party/intel/avx512fp16intrin.internal.h"
+#include "third_party/intel/avx512fp16vlintrin.internal.h"
+#endif
 #include "third_party/intel/shaintrin.internal.h"
 #include "third_party/intel/fmaintrin.internal.h"
 #include "third_party/intel/f16cintrin.internal.h"
diff --git a/third_party/intel/mwaitintrin.internal.h b/third_party/intel/mwaitintrin.internal.h
new file mode 100644
index 000000000..6405a2c9e
--- /dev/null
+++ b/third_party/intel/mwaitintrin.internal.h
@@ -0,0 +1,26 @@
+#if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
+#ifndef _MWAITINTRIN_H_INCLUDED
+#define _MWAITINTRIN_H_INCLUDED
+#ifndef __MWAIT__
+#pragma GCC push_options
+#pragma GCC target("mwait")
+#define __DISABLE_MWAIT__
+#endif
+extern __inline void
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_monitor (void const * __P, unsigned int __E, unsigned int __H)
+{
+  __builtin_ia32_monitor (__P, __E, __H);
+}
+extern __inline void
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mwait (unsigned int __E, unsigned int __H)
+{
+  __builtin_ia32_mwait (__E, __H);
+}
+#ifdef __DISABLE_MWAIT__
+#undef __DISABLE_MWAIT__
+#pragma GCC pop_options
+#endif
+#endif
+#endif
diff --git a/third_party/intel/pmmintrin.internal.h b/third_party/intel/pmmintrin.internal.h
index abac40c73..b66b8412b 100644
--- a/third_party/intel/pmmintrin.internal.h
+++ b/third_party/intel/pmmintrin.internal.h
@@ -2,6 +2,7 @@
 #ifndef _PMMINTRIN_H_INCLUDED
 #define _PMMINTRIN_H_INCLUDED
 #include "third_party/intel/emmintrin.internal.h"
+#include "third_party/intel/mwaitintrin.internal.h"
 #ifndef __SSE3__
 #pragma GCC push_options
 #pragma GCC target("sse3")
@@ -67,16 +68,6 @@ _mm_lddqu_si128 (__m128i const *__P)
 {
   return (__m128i) __builtin_ia32_lddqu ((char const *)__P);
 }
-extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm_monitor (void const * __P, unsigned int __E, unsigned int __H)
-{
-  __builtin_ia32_monitor (__P, __E, __H);
-}
-extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-_mm_mwait (unsigned int __E, unsigned int __H)
-{
-  __builtin_ia32_mwait (__E, __H);
-}
 #ifdef __DISABLE_SSE3__
 #undef __DISABLE_SSE3__
 #pragma GCC pop_options
diff --git a/third_party/intel/serializeintrin.internal.h b/third_party/intel/serializeintrin.internal.h
index a68abbf43..22fa1c97b 100644
--- a/third_party/intel/serializeintrin.internal.h
+++ b/third_party/intel/serializeintrin.internal.h
@@ -9,7 +9,12 @@
 #pragma GCC target("serialize")
 #define __DISABLE_SERIALIZE__
 #endif
-#define _serialize() __builtin_ia32_serialize ()
+extern __inline void
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_serialize (void)
+{
+  __builtin_ia32_serialize ();
+}
 #ifdef __DISABLE_SERIALIZE__
 #undef __DISABLE_SERIALIZE__
 #pragma GCC pop_options
diff --git a/third_party/intel/smmintrin.internal.h b/third_party/intel/smmintrin.internal.h
index 2fae19a10..5179c6e2e 100644
--- a/third_party/intel/smmintrin.internal.h
+++ b/third_party/intel/smmintrin.internal.h
@@ -524,15 +524,10 @@ _mm_cmpgt_epi64 (__m128i __X, __m128i __Y)
 #pragma GCC pop_options
 #endif
 #include "third_party/intel/popcntintrin.internal.h"
-#ifndef __SSE4_1__
+#ifndef __CRC32__
 #pragma GCC push_options
-#pragma GCC target("sse4.1")
-#define __DISABLE_SSE4_1__
-#endif
-#ifndef __SSE4_2__
-#pragma GCC push_options
-#pragma GCC target("sse4.2")
-#define __DISABLE_SSE4_2__
+#pragma GCC target("crc32")
+#define __DISABLE_CRC32__
 #endif
 extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
 _mm_crc32_u8 (unsigned int __C, unsigned char __V)
@@ -556,12 +551,8 @@ _mm_crc32_u64 (unsigned long long __C, unsigned long long __V)
   return __builtin_ia32_crc32di (__C, __V);
 }
 #endif
-#ifdef __DISABLE_SSE4_2__
-#undef __DISABLE_SSE4_2__
-#pragma GCC pop_options
-#endif
-#ifdef __DISABLE_SSE4_1__
-#undef __DISABLE_SSE4_1__
+#ifdef __DISABLE_CRC32__
+#undef __DISABLE_CRC32__
 #pragma GCC pop_options
 #endif
 #endif
diff --git a/third_party/intel/upgrade.sh b/third_party/intel/upgrade.sh
index 02458cbe1..f5f32ddae 100755
--- a/third_party/intel/upgrade.sh
+++ b/third_party/intel/upgrade.sh
@@ -1,32 +1,27 @@
 #!/bin/sh
 
-s=/opt/cross11portcosmo/lib/gcc/x86_64-linux-musl/11.2.0/include
+s=/opt/include
 d=third_party/intel
 
 FILES='
+adxintrin
+ammintrin
 amxbf16intrin
 amxint8intrin
 amxtileintrin
-avx512bf16intrin
-avx512bf16vlintrin
-avx512vp2intersectintrin
-avx512vp2intersectvlintrin
-avxvnniintrin
-enqcmdintrin
-hresetintrin
-keylockerintrin
-serializeintrin
-tsxldtrkintrin
-uintrintrin
-x86gprintrin
+avx2intrin
 avx5124fmapsintrin
 avx5124vnniwintrin
+avx512bf16intrin
+avx512bf16vlintrin
 avx512bitalgintrin
 avx512bwintrin
 avx512cdintrin
 avx512dqintrin
 avx512erintrin
 avx512fintrin
+avx512fp16intrin
+avx512fp16vlintrin
 avx512ifmaintrin
 avx512ifmavlintrin
 avx512pfintrin
@@ -39,12 +34,12 @@ avx512vldqintrin
 avx512vlintrin
 avx512vnniintrin
 avx512vnnivlintrin
+avx512vp2intersectintrin
+avx512vp2intersectvlintrin
 avx512vpopcntdqintrin
 avx512vpopcntdqvlintrin
-adxintrin
-ammintrin
-avx2intrin
 avxintrin
+avxvnniintrin
 bmi2intrin
 bmiintrin
 cetintrin
@@ -54,19 +49,23 @@ clwbintrin
 clzerointrin
 cpuid
 emmintrin
+enqcmdintrin
 f16cintrin
 fma4intrin
 fmaintrin
 fxsrintrin
 gfniintrin
+hresetintrin
 ia32intrin
 immintrin
+keylockerintrin
 lwpintrin
 lzcntintrin
 mm3dnow
 mm_malloc
 mmintrin
 movdirintrin
+mwaitintrin
 mwaitxintrin
 nmmintrin
 pconfigintrin
@@ -76,16 +75,20 @@ popcntintrin
 prfchwintrin
 rdseedintrin
 rtmintrin
+serializeintrin
 sgxintrin
 shaintrin
 smmintrin
 tbmintrin
 tmmintrin
+tsxldtrkintrin
+uintrintrin
 vaesintrin
 vpclmulqdqintrin
 waitpkgintrin
 wbnoinvdintrin
 wmmintrin
+x86gprintrin
 x86intrin
 xmmintrin
 xopintrin
diff --git a/third_party/intel/vaesintrin.internal.h b/third_party/intel/vaesintrin.internal.h
index e0b577112..6a55221af 100644
--- a/third_party/intel/vaesintrin.internal.h
+++ b/third_party/intel/vaesintrin.internal.h
@@ -1,7 +1,4 @@
 #if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
-#ifndef _IMMINTRIN_H_INCLUDED
-# error "Never use <vaesintrin.h> directly; include <immintrin.h> instead."
-#endif
 #ifndef __VAESINTRIN_H_INCLUDED
 #define __VAESINTRIN_H_INCLUDED
 #if !defined(__VAES__) || !defined(__AVX__)
diff --git a/third_party/intel/x86gprintrin.internal.h b/third_party/intel/x86gprintrin.internal.h
index 3f8aedf78..875718588 100644
--- a/third_party/intel/x86gprintrin.internal.h
+++ b/third_party/intel/x86gprintrin.internal.h
@@ -1,6 +1,11 @@
 #if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
 #ifndef _X86GPRINTRIN_H_INCLUDED
 #define _X86GPRINTRIN_H_INCLUDED
+#if !defined _SOFT_FLOAT || defined __MMX__ || defined __SSE__
+#pragma GCC push_options
+#pragma GCC target("general-regs-only")
+#define __DISABLE_GENERAL_REGS_ONLY__
+#endif
 #include "third_party/intel/ia32intrin.internal.h"
 #ifndef __iamcu__
 #include "third_party/intel/adxintrin.internal.h"
@@ -16,6 +21,7 @@
 #include "third_party/intel/lzcntintrin.internal.h"
 #include "third_party/intel/lwpintrin.internal.h"
 #include "third_party/intel/movdirintrin.internal.h"
+#include "third_party/intel/mwaitintrin.internal.h"
 #include "third_party/intel/mwaitxintrin.internal.h"
 #include "third_party/intel/pconfigintrin.internal.h"
 #include "third_party/intel/popcntintrin.internal.h"
@@ -175,5 +181,9 @@ _ptwrite32 (unsigned __B)
 #pragma GCC pop_options
 #endif
 #endif
+#ifdef __DISABLE_GENERAL_REGS_ONLY__
+#undef __DISABLE_GENERAL_REGS_ONLY__
+#pragma GCC pop_options
+#endif
 #endif
 #endif
diff --git a/third_party/intel/xmmintrin.internal.h b/third_party/intel/xmmintrin.internal.h
index 7c744991b..3a97b9fa1 100644
--- a/third_party/intel/xmmintrin.internal.h
+++ b/third_party/intel/xmmintrin.internal.h
@@ -1,9 +1,10 @@
 #if defined(__x86_64__) && !(__ASSEMBLER__ + __LINKER__ + 0)
 #ifndef _XMMINTRIN_H_INCLUDED
 #define _XMMINTRIN_H_INCLUDED
-#include "third_party/intel/mm_malloc.internal.h"
 #include "third_party/intel/mmintrin.internal.h"
-enum _mm_hint {
+#include "third_party/intel/mm_malloc.internal.h"
+enum _mm_hint
+{
   _MM_HINT_ET0 = 7,
   _MM_HINT_ET1 = 6,
   _MM_HINT_T0 = 3,
@@ -12,953 +13,950 @@ enum _mm_hint {
   _MM_HINT_NTA = 0
 };
 #ifdef __OPTIMIZE__
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_prefetch(const void *__P, enum _mm_hint __I) {
-  __builtin_prefetch(__P, (__I & 0x4) >> 2, __I & 0x3);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_prefetch (const void *__P, enum _mm_hint __I)
+{
+  __builtin_prefetch (__P, (__I & 0x4) >> 2, __I & 0x3);
 }
 #else
-#define _mm_prefetch(P, I) __builtin_prefetch((P), ((I & 0x4) >> 2), (I & 0x3))
+#define _mm_prefetch(P, I) __builtin_prefetch ((P), ((I & 0x4) >> 2), (I & 0x3))
 #endif
 #ifndef __SSE__
 #pragma GCC push_options
 #pragma GCC target("sse")
 #define __DISABLE_SSE__
 #endif
-typedef float __m128 __attribute__((__vector_size__(16), __may_alias__));
-typedef float __m128_u
-    __attribute__((__vector_size__(16), __may_alias__, __aligned__(1)));
-typedef float __v4sf __attribute__((__vector_size__(16)));
-#define _MM_SHUFFLE(fp3, fp2, fp1, fp0) \
-  (((fp3) << 6) | ((fp2) << 4) | ((fp1) << 2) | (fp0))
-#define _MM_EXCEPT_MASK       0x003f
-#define _MM_EXCEPT_INVALID    0x0001
-#define _MM_EXCEPT_DENORM     0x0002
-#define _MM_EXCEPT_DIV_ZERO   0x0004
-#define _MM_EXCEPT_OVERFLOW   0x0008
-#define _MM_EXCEPT_UNDERFLOW  0x0010
-#define _MM_EXCEPT_INEXACT    0x0020
-#define _MM_MASK_MASK         0x1f80
-#define _MM_MASK_INVALID      0x0080
-#define _MM_MASK_DENORM       0x0100
-#define _MM_MASK_DIV_ZERO     0x0200
-#define _MM_MASK_OVERFLOW     0x0400
-#define _MM_MASK_UNDERFLOW    0x0800
-#define _MM_MASK_INEXACT      0x1000
-#define _MM_ROUND_MASK        0x6000
-#define _MM_ROUND_NEAREST     0x0000
-#define _MM_ROUND_DOWN        0x2000
-#define _MM_ROUND_UP          0x4000
+typedef float __m128 __attribute__ ((__vector_size__ (16), __may_alias__));
+typedef float __m128_u __attribute__ ((__vector_size__ (16), __may_alias__, __aligned__ (1)));
+typedef float __v4sf __attribute__ ((__vector_size__ (16)));
+#define _MM_SHUFFLE(fp3,fp2,fp1,fp0) (((fp3) << 6) | ((fp2) << 4) | ((fp1) << 2) | (fp0))
+#define _MM_EXCEPT_MASK 0x003f
+#define _MM_EXCEPT_INVALID 0x0001
+#define _MM_EXCEPT_DENORM 0x0002
+#define _MM_EXCEPT_DIV_ZERO 0x0004
+#define _MM_EXCEPT_OVERFLOW 0x0008
+#define _MM_EXCEPT_UNDERFLOW 0x0010
+#define _MM_EXCEPT_INEXACT 0x0020
+#define _MM_MASK_MASK 0x1f80
+#define _MM_MASK_INVALID 0x0080
+#define _MM_MASK_DENORM 0x0100
+#define _MM_MASK_DIV_ZERO 0x0200
+#define _MM_MASK_OVERFLOW 0x0400
+#define _MM_MASK_UNDERFLOW 0x0800
+#define _MM_MASK_INEXACT 0x1000
+#define _MM_ROUND_MASK 0x6000
+#define _MM_ROUND_NEAREST 0x0000
+#define _MM_ROUND_DOWN 0x2000
+#define _MM_ROUND_UP 0x4000
 #define _MM_ROUND_TOWARD_ZERO 0x6000
-#define _MM_FLUSH_ZERO_MASK   0x8000
-#define _MM_FLUSH_ZERO_ON     0x8000
-#define _MM_FLUSH_ZERO_OFF    0x0000
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_undefined_ps(void) {
+#define _MM_FLUSH_ZERO_MASK 0x8000
+#define _MM_FLUSH_ZERO_ON 0x8000
+#define _MM_FLUSH_ZERO_OFF 0x0000
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_undefined_ps (void)
+{
   __m128 __Y = __Y;
   return __Y;
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_setzero_ps(void) {
-  return __extension__(__m128){0.0f, 0.0f, 0.0f, 0.0f};
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setzero_ps (void)
+{
+  return __extension__ (__m128){ 0.0f, 0.0f, 0.0f, 0.0f };
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_add_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_addss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_addss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sub_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_subss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_subss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_mul_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_mulss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_mulss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_div_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_divss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_divss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sqrt_ss(__m128 __A) {
-  return (__m128)__builtin_ia32_sqrtss((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_ss (__m128 __A)
+{
+  return (__m128) __builtin_ia32_sqrtss ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_rcp_ss(__m128 __A) {
-  return (__m128)__builtin_ia32_rcpss((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_ss (__m128 __A)
+{
+  return (__m128) __builtin_ia32_rcpss ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_rsqrt_ss(__m128 __A) {
-  return (__m128)__builtin_ia32_rsqrtss((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_ss (__m128 __A)
+{
+  return (__m128) __builtin_ia32_rsqrtss ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_min_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_minss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_minss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_max_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_maxss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_maxss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_add_ps(__m128 __A, __m128 __B) {
-  return (__m128)((__v4sf)__A + (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_add_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) ((__v4sf)__A + (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sub_ps(__m128 __A, __m128 __B) {
-  return (__m128)((__v4sf)__A - (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sub_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) ((__v4sf)__A - (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_mul_ps(__m128 __A, __m128 __B) {
-  return (__m128)((__v4sf)__A * (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mul_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) ((__v4sf)__A * (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_div_ps(__m128 __A, __m128 __B) {
-  return (__m128)((__v4sf)__A / (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_div_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) ((__v4sf)__A / (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sqrt_ps(__m128 __A) {
-  return (__m128)__builtin_ia32_sqrtps((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sqrt_ps (__m128 __A)
+{
+  return (__m128) __builtin_ia32_sqrtps ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_rcp_ps(__m128 __A) {
-  return (__m128)__builtin_ia32_rcpps((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rcp_ps (__m128 __A)
+{
+  return (__m128) __builtin_ia32_rcpps ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_rsqrt_ps(__m128 __A) {
-  return (__m128)__builtin_ia32_rsqrtps((__v4sf)__A);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_rsqrt_ps (__m128 __A)
+{
+  return (__m128) __builtin_ia32_rsqrtps ((__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_min_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_minps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_minps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_max_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_maxps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_maxps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_and_ps(__m128 __A, __m128 __B) {
-  return __builtin_ia32_andps(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_and_ps (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_andps (__A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_andnot_ps(__m128 __A, __m128 __B) {
-  return __builtin_ia32_andnps(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_andnot_ps (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_andnps (__A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_or_ps(__m128 __A, __m128 __B) {
-  return __builtin_ia32_orps(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_or_ps (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_orps (__A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_xor_ps(__m128 __A, __m128 __B) {
-  return __builtin_ia32_xorps(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_xor_ps (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_xorps (__A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpeq_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpeqss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpeqss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmplt_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpltss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpltss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmple_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpless((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpless ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpgt_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movss(
-      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpltss((__v4sf)__B, (__v4sf)__A));
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movss ((__v4sf) __A,
+     (__v4sf)
+     __builtin_ia32_cmpltss ((__v4sf) __B,
+        (__v4sf)
+        __A));
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpge_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movss(
-      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpless((__v4sf)__B, (__v4sf)__A));
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movss ((__v4sf) __A,
+     (__v4sf)
+     __builtin_ia32_cmpless ((__v4sf) __B,
+        (__v4sf)
+        __A));
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpneq_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpneqss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpneqss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnlt_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpnltss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpnltss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnle_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpnless((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpnless ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpngt_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movss(
-      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpnltss((__v4sf)__B, (__v4sf)__A));
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movss ((__v4sf) __A,
+     (__v4sf)
+     __builtin_ia32_cmpnltss ((__v4sf) __B,
+         (__v4sf)
+         __A));
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnge_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movss(
-      (__v4sf)__A, (__v4sf)__builtin_ia32_cmpnless((__v4sf)__B, (__v4sf)__A));
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movss ((__v4sf) __A,
+     (__v4sf)
+     __builtin_ia32_cmpnless ((__v4sf) __B,
+         (__v4sf)
+         __A));
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpord_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpordss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpordss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpunord_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpunordss((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpunordss ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpeq_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpeqps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpeq_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpeqps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmplt_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpltps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmplt_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpltps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmple_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpleps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmple_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpleps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpgt_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpgtps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpgt_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpgtps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpge_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpgeps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpge_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpgeps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpneq_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpneqps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpneq_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpneqps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnlt_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpnltps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnlt_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpnltps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnle_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpnleps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnle_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpnleps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpngt_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpngtps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpngt_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpngtps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpnge_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpngeps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpnge_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpngeps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpord_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpordps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpord_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpordps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cmpunord_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_cmpunordps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cmpunord_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_cmpunordps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comieq_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comieq((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comieq_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comieq ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comilt_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comilt((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comilt_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comilt ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comile_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comile((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comile_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comile ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comigt_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comigt((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comigt_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comigt ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comige_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comige((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comige_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comige ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_comineq_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_comineq((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_comineq_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_comineq ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomieq_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomieq((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomieq_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomieq ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomilt_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomilt((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomilt_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomilt ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomile_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomile((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomile_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomile ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomigt_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomigt((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomigt_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomigt ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomige_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomige((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomige_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomige ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_ucomineq_ss(__m128 __A, __m128 __B) {
-  return __builtin_ia32_ucomineq((__v4sf)__A, (__v4sf)__B);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_ucomineq_ss (__m128 __A, __m128 __B)
+{
+  return __builtin_ia32_ucomineq ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtss_si32(__m128 __A) {
-  return __builtin_ia32_cvtss2si((__v4sf)__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si32 (__m128 __A)
+{
+  return __builtin_ia32_cvtss2si ((__v4sf) __A);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvt_ss2si(__m128 __A) {
-  return _mm_cvtss_si32(__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_ss2si (__m128 __A)
+{
+  return _mm_cvtss_si32 (__A);
 }
 #ifdef __x86_64__
-extern __inline long long
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtss_si64(__m128 __A) {
-  return __builtin_ia32_cvtss2si64((__v4sf)__A);
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si64 (__m128 __A)
+{
+  return __builtin_ia32_cvtss2si64 ((__v4sf) __A);
 }
-extern __inline long long
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtss_si64x(__m128 __A) {
-  return __builtin_ia32_cvtss2si64((__v4sf)__A);
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_si64x (__m128 __A)
+{
+  return __builtin_ia32_cvtss2si64 ((__v4sf) __A);
 }
 #endif
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtps_pi32(__m128 __A) {
-  return (__m64)__builtin_ia32_cvtps2pi((__v4sf)__A);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi32 (__m128 __A)
+{
+  return (__m64) __builtin_ia32_cvtps2pi ((__v4sf) __A);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvt_ps2pi(__m128 __A) {
-  return _mm_cvtps_pi32(__A);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_ps2pi (__m128 __A)
+{
+  return _mm_cvtps_pi32 (__A);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvttss_si32(__m128 __A) {
-  return __builtin_ia32_cvttss2si((__v4sf)__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si32 (__m128 __A)
+{
+  return __builtin_ia32_cvttss2si ((__v4sf) __A);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtt_ss2si(__m128 __A) {
-  return _mm_cvttss_si32(__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_ss2si (__m128 __A)
+{
+  return _mm_cvttss_si32 (__A);
 }
 #ifdef __x86_64__
-extern __inline long long
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvttss_si64(__m128 __A) {
-  return __builtin_ia32_cvttss2si64((__v4sf)__A);
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si64 (__m128 __A)
+{
+  return __builtin_ia32_cvttss2si64 ((__v4sf) __A);
 }
-extern __inline long long
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvttss_si64x(__m128 __A) {
-  return __builtin_ia32_cvttss2si64((__v4sf)__A);
+extern __inline long long __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttss_si64x (__m128 __A)
+{
+  return __builtin_ia32_cvttss2si64 ((__v4sf) __A);
 }
 #endif
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvttps_pi32(__m128 __A) {
-  return (__m64)__builtin_ia32_cvttps2pi((__v4sf)__A);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvttps_pi32 (__m128 __A)
+{
+  return (__m64) __builtin_ia32_cvttps2pi ((__v4sf) __A);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtt_ps2pi(__m128 __A) {
-  return _mm_cvttps_pi32(__A);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtt_ps2pi (__m128 __A)
+{
+  return _mm_cvttps_pi32 (__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtsi32_ss(__m128 __A, int __B) {
-  return (__m128)__builtin_ia32_cvtsi2ss((__v4sf)__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi32_ss (__m128 __A, int __B)
+{
+  return (__m128) __builtin_ia32_cvtsi2ss ((__v4sf) __A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvt_si2ss(__m128 __A, int __B) {
-  return _mm_cvtsi32_ss(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_si2ss (__m128 __A, int __B)
+{
+  return _mm_cvtsi32_ss (__A, __B);
 }
 #ifdef __x86_64__
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtsi64_ss(__m128 __A, long long __B) {
-  return (__m128)__builtin_ia32_cvtsi642ss((__v4sf)__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64_ss (__m128 __A, long long __B)
+{
+  return (__m128) __builtin_ia32_cvtsi642ss ((__v4sf) __A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtsi64x_ss(__m128 __A, long long __B) {
-  return (__m128)__builtin_ia32_cvtsi642ss((__v4sf)__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtsi64x_ss (__m128 __A, long long __B)
+{
+  return (__m128) __builtin_ia32_cvtsi642ss ((__v4sf) __A, __B);
 }
 #endif
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpi32_ps(__m128 __A, __m64 __B) {
-  return (__m128)__builtin_ia32_cvtpi2ps((__v4sf)__A, (__v2si)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi32_ps (__m128 __A, __m64 __B)
+{
+  return (__m128) __builtin_ia32_cvtpi2ps ((__v4sf) __A, (__v2si)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvt_pi2ps(__m128 __A, __m64 __B) {
-  return _mm_cvtpi32_ps(__A, __B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvt_pi2ps (__m128 __A, __m64 __B)
+{
+  return _mm_cvtpi32_ps (__A, __B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpi16_ps(__m64 __A) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi16_ps (__m64 __A)
+{
   __v4hi __sign;
   __v2si __hisi, __losi;
   __v4sf __zero, __ra, __rb;
-  __sign = __builtin_ia32_pcmpgtw((__v4hi)0LL, (__v4hi)__A);
-  __losi = (__v2si)__builtin_ia32_punpcklwd((__v4hi)__A, __sign);
-  __hisi = (__v2si)__builtin_ia32_punpckhwd((__v4hi)__A, __sign);
-  __zero = (__v4sf)_mm_setzero_ps();
-  __ra = __builtin_ia32_cvtpi2ps(__zero, __losi);
-  __rb = __builtin_ia32_cvtpi2ps(__ra, __hisi);
-  return (__m128)__builtin_ia32_movlhps(__ra, __rb);
+  __sign = __builtin_ia32_pcmpgtw ((__v4hi)0LL, (__v4hi)__A);
+  __losi = (__v2si) __builtin_ia32_punpcklwd ((__v4hi)__A, __sign);
+  __hisi = (__v2si) __builtin_ia32_punpckhwd ((__v4hi)__A, __sign);
+  __zero = (__v4sf) _mm_setzero_ps ();
+  __ra = __builtin_ia32_cvtpi2ps (__zero, __losi);
+  __rb = __builtin_ia32_cvtpi2ps (__ra, __hisi);
+  return (__m128) __builtin_ia32_movlhps (__ra, __rb);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpu16_ps(__m64 __A) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpu16_ps (__m64 __A)
+{
   __v2si __hisi, __losi;
   __v4sf __zero, __ra, __rb;
-  __losi = (__v2si)__builtin_ia32_punpcklwd((__v4hi)__A, (__v4hi)0LL);
-  __hisi = (__v2si)__builtin_ia32_punpckhwd((__v4hi)__A, (__v4hi)0LL);
-  __zero = (__v4sf)_mm_setzero_ps();
-  __ra = __builtin_ia32_cvtpi2ps(__zero, __losi);
-  __rb = __builtin_ia32_cvtpi2ps(__ra, __hisi);
-  return (__m128)__builtin_ia32_movlhps(__ra, __rb);
+  __losi = (__v2si) __builtin_ia32_punpcklwd ((__v4hi)__A, (__v4hi)0LL);
+  __hisi = (__v2si) __builtin_ia32_punpckhwd ((__v4hi)__A, (__v4hi)0LL);
+  __zero = (__v4sf) _mm_setzero_ps ();
+  __ra = __builtin_ia32_cvtpi2ps (__zero, __losi);
+  __rb = __builtin_ia32_cvtpi2ps (__ra, __hisi);
+  return (__m128) __builtin_ia32_movlhps (__ra, __rb);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpi8_ps(__m64 __A) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi8_ps (__m64 __A)
+{
   __v8qi __sign;
-  __sign = __builtin_ia32_pcmpgtb((__v8qi)0LL, (__v8qi)__A);
-  __A = (__m64)__builtin_ia32_punpcklbw((__v8qi)__A, __sign);
+  __sign = __builtin_ia32_pcmpgtb ((__v8qi)0LL, (__v8qi)__A);
+  __A = (__m64) __builtin_ia32_punpcklbw ((__v8qi)__A, __sign);
   return _mm_cvtpi16_ps(__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpu8_ps(__m64 __A) {
-  __A = (__m64)__builtin_ia32_punpcklbw((__v8qi)__A, (__v8qi)0LL);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpu8_ps(__m64 __A)
+{
+  __A = (__m64) __builtin_ia32_punpcklbw ((__v8qi)__A, (__v8qi)0LL);
   return _mm_cvtpu16_ps(__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtpi32x2_ps(__m64 __A, __m64 __B) {
-  __v4sf __zero = (__v4sf)_mm_setzero_ps();
-  __v4sf __sfa = __builtin_ia32_cvtpi2ps(__zero, (__v2si)__A);
-  __v4sf __sfb = __builtin_ia32_cvtpi2ps(__sfa, (__v2si)__B);
-  return (__m128)__builtin_ia32_movlhps(__sfa, __sfb);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtpi32x2_ps(__m64 __A, __m64 __B)
+{
+  __v4sf __zero = (__v4sf) _mm_setzero_ps ();
+  __v4sf __sfa = __builtin_ia32_cvtpi2ps (__zero, (__v2si)__A);
+  __v4sf __sfb = __builtin_ia32_cvtpi2ps (__sfa, (__v2si)__B);
+  return (__m128) __builtin_ia32_movlhps (__sfa, __sfb);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtps_pi16(__m128 __A) {
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi16(__m128 __A)
+{
   __v4sf __hisf = (__v4sf)__A;
-  __v4sf __losf = __builtin_ia32_movhlps(__hisf, __hisf);
-  __v2si __hisi = __builtin_ia32_cvtps2pi(__hisf);
-  __v2si __losi = __builtin_ia32_cvtps2pi(__losf);
-  return (__m64)__builtin_ia32_packssdw(__hisi, __losi);
+  __v4sf __losf = __builtin_ia32_movhlps (__hisf, __hisf);
+  __v2si __hisi = __builtin_ia32_cvtps2pi (__hisf);
+  __v2si __losi = __builtin_ia32_cvtps2pi (__losf);
+  return (__m64) __builtin_ia32_packssdw (__hisi, __losi);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtps_pi8(__m128 __A) {
-  __v4hi __tmp = (__v4hi)_mm_cvtps_pi16(__A);
-  return (__m64)__builtin_ia32_packsswb(__tmp, (__v4hi)0LL);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtps_pi8(__m128 __A)
+{
+  __v4hi __tmp = (__v4hi) _mm_cvtps_pi16 (__A);
+  return (__m64) __builtin_ia32_packsswb (__tmp, (__v4hi)0LL);
 }
 #ifdef __OPTIMIZE__
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_shuffle_ps(__m128 __A, __m128 __B, int const __mask) {
-  return (__m128)__builtin_ia32_shufps((__v4sf)__A, (__v4sf)__B, __mask);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_ps (__m128 __A, __m128 __B, int const __mask)
+{
+  return (__m128) __builtin_ia32_shufps ((__v4sf)__A, (__v4sf)__B, __mask);
 }
 #else
-#define _mm_shuffle_ps(A, B, MASK)                                         \
-  ((__m128)__builtin_ia32_shufps((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), \
-                                 (int)(MASK)))
+#define _mm_shuffle_ps(A, B, MASK) ((__m128) __builtin_ia32_shufps ((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), (int)(MASK)))
 #endif
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_unpackhi_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_unpckhps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpackhi_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_unpckhps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_unpacklo_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_unpcklps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_unpacklo_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_unpcklps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_loadh_pi(__m128 __A, __m64 const *__P) {
-  return (__m128)__builtin_ia32_loadhps((__v4sf)__A, (const __v2sf *)__P);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadh_pi (__m128 __A, __m64 const *__P)
+{
+  return (__m128) __builtin_ia32_loadhps ((__v4sf)__A, (const __v2sf *)__P);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_storeh_pi(__m64 *__P, __m128 __A) {
-  __builtin_ia32_storehps((__v2sf *)__P, (__v4sf)__A);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeh_pi (__m64 *__P, __m128 __A)
+{
+  __builtin_ia32_storehps ((__v2sf *)__P, (__v4sf)__A);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_movehl_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movhlps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movehl_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movhlps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_movelh_ps(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_ia32_movlhps((__v4sf)__A, (__v4sf)__B);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movelh_ps (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_ia32_movlhps ((__v4sf)__A, (__v4sf)__B);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_loadl_pi(__m128 __A, __m64 const *__P) {
-  return (__m128)__builtin_ia32_loadlps((__v4sf)__A, (const __v2sf *)__P);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadl_pi (__m128 __A, __m64 const *__P)
+{
+  return (__m128) __builtin_ia32_loadlps ((__v4sf)__A, (const __v2sf *)__P);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_storel_pi(__m64 *__P, __m128 __A) {
-  __builtin_ia32_storelps((__v2sf *)__P, (__v4sf)__A);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storel_pi (__m64 *__P, __m128 __A)
+{
+  __builtin_ia32_storelps ((__v2sf *)__P, (__v4sf)__A);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_movemask_ps(__m128 __A) {
-  return __builtin_ia32_movmskps((__v4sf)__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_ps (__m128 __A)
+{
+  return __builtin_ia32_movmskps ((__v4sf)__A);
 }
-extern __inline unsigned int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_getcsr(void) {
-  return __builtin_ia32_stmxcsr();
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_getcsr (void)
+{
+  return __builtin_ia32_stmxcsr ();
 }
-extern __inline unsigned int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_GET_EXCEPTION_STATE(void) {
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_GET_EXCEPTION_STATE (void)
+{
   return _mm_getcsr() & _MM_EXCEPT_MASK;
 }
-extern __inline unsigned int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_GET_EXCEPTION_MASK(void) {
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_GET_EXCEPTION_MASK (void)
+{
   return _mm_getcsr() & _MM_MASK_MASK;
 }
-extern __inline unsigned int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_GET_ROUNDING_MODE(void) {
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_GET_ROUNDING_MODE (void)
+{
   return _mm_getcsr() & _MM_ROUND_MASK;
 }
-extern __inline unsigned int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_GET_FLUSH_ZERO_MODE(void) {
+extern __inline unsigned int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_GET_FLUSH_ZERO_MODE (void)
+{
   return _mm_getcsr() & _MM_FLUSH_ZERO_MASK;
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_setcsr(unsigned int __I) {
-  __builtin_ia32_ldmxcsr(__I);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setcsr (unsigned int __I)
+{
+  __builtin_ia32_ldmxcsr (__I);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_SET_EXCEPTION_STATE(unsigned int __mask) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_SET_EXCEPTION_STATE(unsigned int __mask)
+{
   _mm_setcsr((_mm_getcsr() & ~_MM_EXCEPT_MASK) | __mask);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_SET_EXCEPTION_MASK(unsigned int __mask) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_SET_EXCEPTION_MASK (unsigned int __mask)
+{
   _mm_setcsr((_mm_getcsr() & ~_MM_MASK_MASK) | __mask);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_SET_ROUNDING_MODE(unsigned int __mode) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_SET_ROUNDING_MODE (unsigned int __mode)
+{
   _mm_setcsr((_mm_getcsr() & ~_MM_ROUND_MASK) | __mode);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _MM_SET_FLUSH_ZERO_MODE(unsigned int __mode) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_MM_SET_FLUSH_ZERO_MODE (unsigned int __mode)
+{
   _mm_setcsr((_mm_getcsr() & ~_MM_FLUSH_ZERO_MASK) | __mode);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_set_ss(float __F) {
-  return __extension__(__m128)(__v4sf){__F, 0.0f, 0.0f, 0.0f};
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ss (float __F)
+{
+  return __extension__ (__m128)(__v4sf){ __F, 0.0f, 0.0f, 0.0f };
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_set1_ps(float __F) {
-  return __extension__(__m128)(__v4sf){__F, __F, __F, __F};
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set1_ps (float __F)
+{
+  return __extension__ (__m128)(__v4sf){ __F, __F, __F, __F };
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_set_ps1(float __F) {
-  return _mm_set1_ps(__F);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ps1 (float __F)
+{
+  return _mm_set1_ps (__F);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_load_ss(float const *__P) {
-  return _mm_set_ss(*__P);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ss (float const *__P)
+{
+  return _mm_set_ss (*__P);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_load1_ps(float const *__P) {
-  return _mm_set1_ps(*__P);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load1_ps (float const *__P)
+{
+  return _mm_set1_ps (*__P);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_load_ps1(float const *__P) {
-  return _mm_load1_ps(__P);
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ps1 (float const *__P)
+{
+  return _mm_load1_ps (__P);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_load_ps(float const *__P) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_load_ps (float const *__P)
+{
   return *(__m128 *)__P;
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_loadu_ps(float const *__P) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadu_ps (float const *__P)
+{
   return *(__m128_u *)__P;
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_loadr_ps(float const *__P) {
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_loadr_ps (float const *__P)
+{
   __v4sf __tmp = *(__v4sf *)__P;
-  return (__m128)__builtin_ia32_shufps(__tmp, __tmp, _MM_SHUFFLE(0, 1, 2, 3));
+  return (__m128) __builtin_ia32_shufps (__tmp, __tmp, _MM_SHUFFLE (0,1,2,3));
 }
-extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__,
-                                      __artificial__))
-_mm_set_ps(const float __Z, const float __Y, const float __X, const float __W) {
-  return __extension__(__m128)(__v4sf){__W, __X, __Y, __Z};
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_set_ps (const float __Z, const float __Y, const float __X, const float __W)
+{
+  return __extension__ (__m128)(__v4sf){ __W, __X, __Y, __Z };
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_setr_ps(float __Z, float __Y, float __X, float __W) {
-  return __extension__(__m128)(__v4sf){__Z, __Y, __X, __W};
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_setr_ps (float __Z, float __Y, float __X, float __W)
+{
+  return __extension__ (__m128)(__v4sf){ __Z, __Y, __X, __W };
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_store_ss(float *__P, __m128 __A) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ss (float *__P, __m128 __A)
+{
   *__P = ((__v4sf)__A)[0];
 }
-extern __inline float
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_cvtss_f32(__m128 __A) {
+extern __inline float __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_cvtss_f32 (__m128 __A)
+{
   return ((__v4sf)__A)[0];
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_store_ps(float *__P, __m128 __A) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ps (float *__P, __m128 __A)
+{
   *(__m128 *)__P = __A;
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_storeu_ps(float *__P, __m128 __A) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storeu_ps (float *__P, __m128 __A)
+{
   *(__m128_u *)__P = __A;
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_store1_ps(float *__P, __m128 __A) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store1_ps (float *__P, __m128 __A)
+{
   __v4sf __va = (__v4sf)__A;
-  __v4sf __tmp = __builtin_ia32_shufps(__va, __va, _MM_SHUFFLE(0, 0, 0, 0));
-  _mm_storeu_ps(__P, __tmp);
+  __v4sf __tmp = __builtin_ia32_shufps (__va, __va, _MM_SHUFFLE (0,0,0,0));
+  _mm_storeu_ps (__P, __tmp);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_store_ps1(float *__P, __m128 __A) {
-  _mm_store1_ps(__P, __A);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_store_ps1 (float *__P, __m128 __A)
+{
+  _mm_store1_ps (__P, __A);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_storer_ps(float *__P, __m128 __A) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_storer_ps (float *__P, __m128 __A)
+{
   __v4sf __va = (__v4sf)__A;
-  __v4sf __tmp = __builtin_ia32_shufps(__va, __va, _MM_SHUFFLE(0, 1, 2, 3));
-  _mm_store_ps(__P, __tmp);
+  __v4sf __tmp = __builtin_ia32_shufps (__va, __va, _MM_SHUFFLE (0,1,2,3));
+  _mm_store_ps (__P, __tmp);
 }
-extern __inline __m128
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_move_ss(__m128 __A, __m128 __B) {
-  return (__m128)__builtin_shuffle(
-      (__v4sf)__A, (__v4sf)__B,
-      __extension__(__attribute__((__vector_size__(16))) int){4, 1, 2, 3});
+extern __inline __m128 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_move_ss (__m128 __A, __m128 __B)
+{
+  return (__m128) __builtin_shuffle ((__v4sf)__A, (__v4sf)__B,
+                                     __extension__
+                                     (__attribute__((__vector_size__ (16))) int)
+                                     {4,1,2,3});
 }
 #ifdef __OPTIMIZE__
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_extract_pi16(__m64 const __A, int const __N) {
-  return (unsigned short)__builtin_ia32_vec_ext_v4hi((__v4hi)__A, __N);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_extract_pi16 (__m64 const __A, int const __N)
+{
+  return (unsigned short) __builtin_ia32_vec_ext_v4hi ((__v4hi)__A, __N);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pextrw(__m64 const __A, int const __N) {
-  return _mm_extract_pi16(__A, __N);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pextrw (__m64 const __A, int const __N)
+{
+  return _mm_extract_pi16 (__A, __N);
 }
 #else
-#define _mm_extract_pi16(A, N)                                          \
-  ((int)(unsigned short)__builtin_ia32_vec_ext_v4hi((__v4hi)(__m64)(A), \
-                                                    (int)(N)))
+#define _mm_extract_pi16(A, N) ((int) (unsigned short) __builtin_ia32_vec_ext_v4hi ((__v4hi)(__m64)(A), (int)(N)))
 #define _m_pextrw(A, N) _mm_extract_pi16(A, N)
 #endif
 #ifdef __OPTIMIZE__
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_insert_pi16(__m64 const __A, int const __D, int const __N) {
-  return (__m64)__builtin_ia32_vec_set_v4hi((__v4hi)__A, __D, __N);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_insert_pi16 (__m64 const __A, int const __D, int const __N)
+{
+  return (__m64) __builtin_ia32_vec_set_v4hi ((__v4hi)__A, __D, __N);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pinsrw(__m64 const __A, int const __D, int const __N) {
-  return _mm_insert_pi16(__A, __D, __N);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pinsrw (__m64 const __A, int const __D, int const __N)
+{
+  return _mm_insert_pi16 (__A, __D, __N);
 }
 #else
-#define _mm_insert_pi16(A, D, N) \
-  ((__m64)__builtin_ia32_vec_set_v4hi((__v4hi)(__m64)(A), (int)(D), (int)(N)))
+#define _mm_insert_pi16(A, D, N) ((__m64) __builtin_ia32_vec_set_v4hi ((__v4hi)(__m64)(A), (int)(D), (int)(N)))
 #define _m_pinsrw(A, D, N) _mm_insert_pi16(A, D, N)
 #endif
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_max_pi16(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pmaxsw((__v4hi)__A, (__v4hi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_pi16 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pmaxsw ((__v4hi)__A, (__v4hi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pmaxsw(__m64 __A, __m64 __B) {
-  return _mm_max_pi16(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmaxsw (__m64 __A, __m64 __B)
+{
+  return _mm_max_pi16 (__A, __B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_max_pu8(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pmaxub((__v8qi)__A, (__v8qi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_max_pu8 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pmaxub ((__v8qi)__A, (__v8qi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pmaxub(__m64 __A, __m64 __B) {
-  return _mm_max_pu8(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmaxub (__m64 __A, __m64 __B)
+{
+  return _mm_max_pu8 (__A, __B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_min_pi16(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pminsw((__v4hi)__A, (__v4hi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_pi16 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pminsw ((__v4hi)__A, (__v4hi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pminsw(__m64 __A, __m64 __B) {
-  return _mm_min_pi16(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pminsw (__m64 __A, __m64 __B)
+{
+  return _mm_min_pi16 (__A, __B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_min_pu8(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pminub((__v8qi)__A, (__v8qi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_min_pu8 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pminub ((__v8qi)__A, (__v8qi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pminub(__m64 __A, __m64 __B) {
-  return _mm_min_pu8(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pminub (__m64 __A, __m64 __B)
+{
+  return _mm_min_pu8 (__A, __B);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_movemask_pi8(__m64 __A) {
-  return __builtin_ia32_pmovmskb((__v8qi)__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_movemask_pi8 (__m64 __A)
+{
+  return __builtin_ia32_pmovmskb ((__v8qi)__A);
 }
-extern __inline int
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pmovmskb(__m64 __A) {
-  return _mm_movemask_pi8(__A);
+extern __inline int __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmovmskb (__m64 __A)
+{
+  return _mm_movemask_pi8 (__A);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_mulhi_pu16(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pmulhuw((__v4hi)__A, (__v4hi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mulhi_pu16 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pmulhuw ((__v4hi)__A, (__v4hi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pmulhuw(__m64 __A, __m64 __B) {
-  return _mm_mulhi_pu16(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pmulhuw (__m64 __A, __m64 __B)
+{
+  return _mm_mulhi_pu16 (__A, __B);
 }
 #ifdef __OPTIMIZE__
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_shuffle_pi16(__m64 __A, int const __N) {
-  return (__m64)__builtin_ia32_pshufw((__v4hi)__A, __N);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_shuffle_pi16 (__m64 __A, int const __N)
+{
+  return (__m64) __builtin_ia32_pshufw ((__v4hi)__A, __N);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pshufw(__m64 __A, int const __N) {
-  return _mm_shuffle_pi16(__A, __N);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pshufw (__m64 __A, int const __N)
+{
+  return _mm_shuffle_pi16 (__A, __N);
 }
 #else
-#define _mm_shuffle_pi16(A, N) \
-  ((__m64)__builtin_ia32_pshufw((__v4hi)(__m64)(A), (int)(N)))
-#define _m_pshufw(A, N) _mm_shuffle_pi16(A, N)
+#define _mm_shuffle_pi16(A, N) ((__m64) __builtin_ia32_pshufw ((__v4hi)(__m64)(A), (int)(N)))
+#define _m_pshufw(A, N) _mm_shuffle_pi16 (A, N)
 #endif
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_maskmove_si64(__m64 __A, __m64 __N, char *__P) {
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskmove_si64 (__m64 __A, __m64 __N, char *__P)
+{
 #ifdef __MMX_WITH_SSE__
-  typedef long long __v2di __attribute__((__vector_size__(16)));
-  typedef char __v16qi __attribute__((__vector_size__(16)));
-  __v2di __A128 = __extension__(__v2di){((__v1di)__A)[0], 0};
-  __v2di __N128 = __extension__(__v2di){((__v1di)__N)[0], 0};
-  __SIZE_TYPE__ offset = ((__SIZE_TYPE__)__P) & 0xf;
-  if (offset) {
-    if (offset > 8) offset = 8;
-    __P = (char *)(((__SIZE_TYPE__)__P) - offset);
-    switch (offset) {
-      case 1:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 8);
-        break;
-      case 2:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 2 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 2 * 8);
-        break;
-      case 3:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 3 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 3 * 8);
-        break;
-      case 4:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 4 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 4 * 8);
-        break;
-      case 5:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 5 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 5 * 8);
-        break;
-      case 6:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 6 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 6 * 8);
-        break;
-      case 7:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 7 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 7 * 8);
-        break;
-      case 8:
-        __A128 = __builtin_ia32_pslldqi128(__A128, 8 * 8);
-        __N128 = __builtin_ia32_pslldqi128(__N128, 8 * 8);
-        break;
-      default:
-        break;
+  typedef long long __v2di __attribute__ ((__vector_size__ (16)));
+  typedef char __v16qi __attribute__ ((__vector_size__ (16)));
+  __v2di __A128 = __extension__ (__v2di) { ((__v1di) __A)[0], 0 };
+  __v2di __N128 = __extension__ (__v2di) { ((__v1di) __N)[0], 0 };
+  __SIZE_TYPE__ offset = ((__SIZE_TYPE__) __P) & 0xf;
+  if (offset)
+    {
+      if (offset > 8)
+ offset = 8;
+      __P = (char *) (((__SIZE_TYPE__) __P) - offset);
+      switch (offset)
+ {
+ case 1:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 8);
+   break;
+ case 2:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 2 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 2 * 8);
+   break;
+ case 3:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 3 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 3 * 8);
+   break;
+ case 4:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 4 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 4 * 8);
+   break;
+ case 5:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 5 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 5 * 8);
+   break;
+ case 6:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 6 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 6 * 8);
+   break;
+ case 7:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 7 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 7 * 8);
+   break;
+ case 8:
+   __A128 = __builtin_ia32_pslldqi128 (__A128, 8 * 8);
+   __N128 = __builtin_ia32_pslldqi128 (__N128, 8 * 8);
+   break;
+ default:
+   break;
+ }
     }
-  }
-  __builtin_ia32_maskmovdqu((__v16qi)__A128, (__v16qi)__N128, __P);
+  __builtin_ia32_maskmovdqu ((__v16qi)__A128, (__v16qi)__N128, __P);
 #else
-  __builtin_ia32_maskmovq((__v8qi)__A, (__v8qi)__N, __P);
+  __builtin_ia32_maskmovq ((__v8qi)__A, (__v8qi)__N, __P);
 #endif
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_maskmovq(__m64 __A, __m64 __N, char *__P) {
-  _mm_maskmove_si64(__A, __N, __P);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_maskmovq (__m64 __A, __m64 __N, char *__P)
+{
+  _mm_maskmove_si64 (__A, __N, __P);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_avg_pu8(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pavgb((__v8qi)__A, (__v8qi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_pu8 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pavgb ((__v8qi)__A, (__v8qi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pavgb(__m64 __A, __m64 __B) {
-  return _mm_avg_pu8(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pavgb (__m64 __A, __m64 __B)
+{
+  return _mm_avg_pu8 (__A, __B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_avg_pu16(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_pavgw((__v4hi)__A, (__v4hi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_avg_pu16 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_pavgw ((__v4hi)__A, (__v4hi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_pavgw(__m64 __A, __m64 __B) {
-  return _mm_avg_pu16(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_pavgw (__m64 __A, __m64 __B)
+{
+  return _mm_avg_pu16 (__A, __B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sad_pu8(__m64 __A, __m64 __B) {
-  return (__m64)__builtin_ia32_psadbw((__v8qi)__A, (__v8qi)__B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sad_pu8 (__m64 __A, __m64 __B)
+{
+  return (__m64) __builtin_ia32_psadbw ((__v8qi)__A, (__v8qi)__B);
 }
-extern __inline __m64
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _m_psadbw(__m64 __A, __m64 __B) {
-  return _mm_sad_pu8(__A, __B);
+extern __inline __m64 __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_m_psadbw (__m64 __A, __m64 __B)
+{
+  return _mm_sad_pu8 (__A, __B);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_stream_pi(__m64 *__P, __m64 __A) {
-  __builtin_ia32_movntq((unsigned long long *)__P, (unsigned long long)__A);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_pi (__m64 *__P, __m64 __A)
+{
+  __builtin_ia32_movntq ((unsigned long long *)__P, (unsigned long long)__A);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_stream_ps(float *__P, __m128 __A) {
-  __builtin_ia32_movntps(__P, (__v4sf)__A);
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_stream_ps (float *__P, __m128 __A)
+{
+  __builtin_ia32_movntps (__P, (__v4sf)__A);
 }
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_sfence(void) {
-  __builtin_ia32_sfence();
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sfence (void)
+{
+  __builtin_ia32_sfence ();
 }
-#define _MM_TRANSPOSE4_PS(row0, row1, row2, row3)                      \
-  do {                                                                 \
-    __v4sf __r0 = (row0), __r1 = (row1), __r2 = (row2), __r3 = (row3); \
-    __v4sf __t0 = __builtin_ia32_unpcklps(__r0, __r1);                 \
-    __v4sf __t1 = __builtin_ia32_unpcklps(__r2, __r3);                 \
-    __v4sf __t2 = __builtin_ia32_unpckhps(__r0, __r1);                 \
-    __v4sf __t3 = __builtin_ia32_unpckhps(__r2, __r3);                 \
-    (row0) = __builtin_ia32_movlhps(__t0, __t1);                       \
-    (row1) = __builtin_ia32_movhlps(__t1, __t0);                       \
-    (row2) = __builtin_ia32_movlhps(__t2, __t3);                       \
-    (row3) = __builtin_ia32_movhlps(__t3, __t2);                       \
-  } while (0)
+#define _MM_TRANSPOSE4_PS(row0, row1, row2, row3) do { __v4sf __r0 = (row0), __r1 = (row1), __r2 = (row2), __r3 = (row3); __v4sf __t0 = __builtin_ia32_unpcklps (__r0, __r1); __v4sf __t1 = __builtin_ia32_unpcklps (__r2, __r3); __v4sf __t2 = __builtin_ia32_unpckhps (__r0, __r1); __v4sf __t3 = __builtin_ia32_unpckhps (__r2, __r3); (row0) = __builtin_ia32_movlhps (__t0, __t1); (row1) = __builtin_ia32_movhlps (__t1, __t0); (row2) = __builtin_ia32_movlhps (__t2, __t3); (row3) = __builtin_ia32_movhlps (__t3, __t2); } while (0)
 #include "third_party/intel/emmintrin.internal.h"
 #ifdef __DISABLE_SSE__
 #undef __DISABLE_SSE__
 #pragma GCC pop_options
 #endif
-extern __inline void
-    __attribute__((__gnu_inline__, __always_inline__, __artificial__))
-    _mm_pause(void) {
-  __builtin_ia32_pause();
+extern __inline void __attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_pause (void)
+{
+  __builtin_ia32_pause ();
 }
 #endif
 #endif
diff --git a/third_party/less/main.c b/third_party/less/main.c
index 21ed227ca..529fcb457 100644
--- a/third_party/less/main.c
+++ b/third_party/less/main.c
@@ -1,27 +1,27 @@
-asm(".ident\t\"\\n\\n\
-Less\\n\
-Copyright (C) 1984-2023  Mark Nudelman\\n\
-\\n\
-Redistribution and use in source and binary forms, with or without\\n\
-modification, are permitted provided that the following conditions\\n\
-are met:\\n\
-1. Redistributions of source code must retain the above copyright\\n\
-   notice, this list of conditions and the following disclaimer.\\n\
-2. Redistributions in binary form must reproduce the above copyright\\n\
-   notice in the documentation and/or other materials provided with\\n\
-   the distribution.\\n\
-\\n\
-THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY\\n\
-EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE\\n\
-IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR\\n\
-PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR BE LIABLE\\n\
-FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR\\n\
-CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT\\n\
-OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR\\n\
-BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,\\n\
-WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE\\n\
-OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN\\n\
-IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.\"");
+__notice(less_notice, "\
+Less\n\
+Copyright (C) 1984-2023  Mark Nudelman\n\
+\n\
+Redistribution and use in source and binary forms, with or without\n\
+modification, are permitted provided that the following conditions\n\
+are met:\n\
+1. Redistributions of source code must retain the above copyright\n\
+   notice, this list of conditions and the following disclaimer.\n\
+2. Redistributions in binary form must reproduce the above copyright\n\
+   notice in the documentation and/or other materials provided with\n\
+   the distribution.\n\
+\n\
+THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY\n\
+EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE\n\
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR\n\
+PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR BE LIABLE\n\
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR\n\
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT\n\
+OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR\n\
+BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,\n\
+WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE\n\
+OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN\n\
+IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.");
 
 /*
  * Copyright (C) 1984-2023  Mark Nudelman
diff --git a/third_party/libcxx/vector b/third_party/libcxx/vector
index 014c8a51d..91735b728 100644
--- a/third_party/libcxx/vector
+++ b/third_party/libcxx/vector
@@ -1456,6 +1456,8 @@ vector<_Tp, _Allocator>::assign(_ForwardIterator __first, _ForwardIterator __las
     else
     {
         __vdeallocate();
+        // [jart] make compiler not complain about maximum object size
+        __asm__("" : "+r"(__new_size));
         __vallocate(__recommend(__new_size));
         __construct_at_end(__first, __last, __new_size);
     }
diff --git a/third_party/linenoise/linenoise.c b/third_party/linenoise/linenoise.c
index c6386b2ba..cf673ba1f 100644
--- a/third_party/linenoise/linenoise.c
+++ b/third_party/linenoise/linenoise.c
@@ -172,11 +172,11 @@
 #include "net/http/escape.h"
 #include "tool/build/lib/case.h"
 
-asm(".ident\t\"\\n\\n\
-Cosmopolitan Linenoise (BSD-2)\\n\
-Copyright 2018-2020 Justine Tunney <jtunney@gmail.com>\\n\
-Copyright 2010-2016 Salvatore Sanfilippo <antirez@gmail.com>\\n\
-Copyright 2010-2013 Pieter Noordhuis <pcnoordhuis@gmail.com>\"");
+__notice(linenoise_notice, "\
+Cosmopolitan Linenoise (BSD-2)\n\
+Copyright 2018-2020 Justine Tunney <jtunney@gmail.com>\n\
+Copyright 2010-2016 Salvatore Sanfilippo <antirez@gmail.com>\n\
+Copyright 2010-2013 Pieter Noordhuis <pcnoordhuis@gmail.com>");
 
 #define LINENOISE_POLL_MS 50
 
@@ -2644,10 +2644,7 @@ static void linenoiseAtExit(void) {
   linenoiseRingFree();
 }
 
+__attribute__((__constructor__(99)))
 static textstartup void linenoiseInit() {
   atexit(linenoiseAtExit);
 }
-
-const void *const linenoiseCtor[] initarray = {
-    linenoiseInit,
-};
diff --git a/third_party/lua/BUILD.mk b/third_party/lua/BUILD.mk
index 60a73e960..bd62923b2 100644
--- a/third_party/lua/BUILD.mk
+++ b/third_party/lua/BUILD.mk
@@ -85,6 +85,7 @@ THIRD_PARTY_LUA_A_SRCS =						\
 	third_party/lua/llock.c						\
 	third_party/lua/lmathlib.c					\
 	third_party/lua/lmem.c						\
+	third_party/lua/lnotice.c					\
 	third_party/lua/loadlib.c					\
 	third_party/lua/lobject.c					\
 	third_party/lua/lopcodes.c					\
diff --git a/third_party/lua/lapi.c b/third_party/lua/lapi.c
index 15230810a..ae8988f2c 100644
--- a/third_party/lua/lapi.c
+++ b/third_party/lua/lapi.c
@@ -42,11 +42,7 @@
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lundump.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 const char lua_ident[] =
diff --git a/third_party/lua/lauxlib.c b/third_party/lua/lauxlib.c
index a1555ba18..1fa74d1f2 100644
--- a/third_party/lua/lauxlib.c
+++ b/third_party/lua/lauxlib.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lauxlib.h"
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /**
diff --git a/third_party/lua/lbaselib.c b/third_party/lua/lbaselib.c
index f5ab58b16..5e936c406 100644
--- a/third_party/lua/lbaselib.c
+++ b/third_party/lua/lbaselib.c
@@ -32,11 +32,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 static int luaB_print (lua_State *L) {
diff --git a/third_party/lua/lcode.c b/third_party/lua/lcode.c
index 6468d2243..c6c9c0a2b 100644
--- a/third_party/lua/lcode.c
+++ b/third_party/lua/lcode.c
@@ -42,11 +42,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /* Maximum number of registers in a Lua function (must fit in 8 bits) */
diff --git a/third_party/lua/lcorolib.c b/third_party/lua/lcorolib.c
index e15b3e749..be6f75f5d 100644
--- a/third_party/lua/lcorolib.c
+++ b/third_party/lua/lcorolib.c
@@ -31,11 +31,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 static lua_State *getco (lua_State *L) {
diff --git a/third_party/lua/ldblib.c b/third_party/lua/ldblib.c
index 8918ffcbe..74f6ed32b 100644
--- a/third_party/lua/ldblib.c
+++ b/third_party/lua/ldblib.c
@@ -32,11 +32,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/ldebug.c b/third_party/lua/ldebug.c
index 16197115f..8e103095f 100644
--- a/third_party/lua/ldebug.c
+++ b/third_party/lua/ldebug.c
@@ -42,11 +42,7 @@
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 #define noLuaClosure(f)		((f) == NULL || (f)->c.tt == LUA_VCCL)
diff --git a/third_party/lua/ldo.c b/third_party/lua/ldo.c
index 1978b1ff4..abfa70719 100644
--- a/third_party/lua/ldo.c
+++ b/third_party/lua/ldo.c
@@ -51,11 +51,7 @@
 #include "third_party/lua/lundump.h"
 #include "third_party/lua/lvm.h"
 #include "third_party/lua/lzio.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 #define errorstatus(s)	((s) > LUA_YIELD)
diff --git a/third_party/lua/ldump.c b/third_party/lua/ldump.c
index 8e4d6b7c0..a1e9666b1 100644
--- a/third_party/lua/ldump.c
+++ b/third_party/lua/ldump.c
@@ -32,11 +32,7 @@
 #include "third_party/lua/lstate.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lundump.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 typedef struct {
diff --git a/third_party/lua/lfunc.c b/third_party/lua/lfunc.c
index b3e69eef3..7ff7cafd0 100644
--- a/third_party/lua/lfunc.c
+++ b/third_party/lua/lfunc.c
@@ -37,11 +37,7 @@
 #include "third_party/lua/lstate.h"
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 CClosure *luaF_newCclosure (lua_State *L, int nupvals) {
diff --git a/third_party/lua/lgc.c b/third_party/lua/lgc.c
index 57027bb3b..a52142624 100644
--- a/third_party/lua/lgc.c
+++ b/third_party/lua/lgc.c
@@ -40,11 +40,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/linit.c b/third_party/lua/linit.c
index e625cc728..2ebb7fbd0 100644
--- a/third_party/lua/linit.c
+++ b/third_party/lua/linit.c
@@ -46,11 +46,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/liolib.c b/third_party/lua/liolib.c
index 30cc43815..c8bc69076 100644
--- a/third_party/lua/liolib.c
+++ b/third_party/lua/liolib.c
@@ -37,11 +37,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/llex.c b/third_party/lua/llex.c
index 3e1446b7f..688d2a3b0 100644
--- a/third_party/lua/llex.c
+++ b/third_party/lua/llex.c
@@ -40,11 +40,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lzio.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 #define next(ls)	(ls->current = zgetc(ls->z))
diff --git a/third_party/lua/lmathlib.c b/third_party/lua/lmathlib.c
index b0968ed16..8b1a291f4 100644
--- a/third_party/lua/lmathlib.c
+++ b/third_party/lua/lmathlib.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 #undef PI
diff --git a/third_party/lua/lmem.c b/third_party/lua/lmem.c
index 7d0ee1648..780e01bf8 100644
--- a/third_party/lua/lmem.c
+++ b/third_party/lua/lmem.c
@@ -36,11 +36,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lstate.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 #if defined(EMERGENCYGCTESTS)
diff --git a/third_party/lua/lnotice.c b/third_party/lua/lnotice.c
new file mode 100644
index 000000000..9eb4e4e17
--- /dev/null
+++ b/third_party/lua/lnotice.c
@@ -0,0 +1,3 @@
+__notice(lua_notice, "\
+Lua 5.4.3 (MIT License)\n\
+Copyright 1994–2021 Lua.org, PUC-Rio.");
diff --git a/third_party/lua/loadlib.c b/third_party/lua/loadlib.c
index 0feae4d7f..43bf0168d 100644
--- a/third_party/lua/loadlib.c
+++ b/third_party/lua/loadlib.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 /*
 ** This module contains an implementation of loadlib for Unix systems
diff --git a/third_party/lua/lobject.c b/third_party/lua/lobject.c
index 24e03a9d6..d855ac166 100644
--- a/third_party/lua/lobject.c
+++ b/third_party/lua/lobject.c
@@ -38,11 +38,7 @@
 #include "third_party/lua/lstring.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 static lua_Integer intarith (lua_State *L, int op, lua_Integer v1,
diff --git a/third_party/lua/lopcodes.c b/third_party/lua/lopcodes.c
index 75558af3a..90d726553 100644
--- a/third_party/lua/lopcodes.c
+++ b/third_party/lua/lopcodes.c
@@ -29,11 +29,7 @@
 #define LUA_CORE
 #include "third_party/lua/lopcodes.h"
 #include "third_party/lua/lprefix.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /* ORDER OP */
diff --git a/third_party/lua/loslib.c b/third_party/lua/loslib.c
index 647e473b9..4f979c59f 100644
--- a/third_party/lua/loslib.c
+++ b/third_party/lua/loslib.c
@@ -41,11 +41,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/lparser.c b/third_party/lua/lparser.c
index 96f89293e..79e598fc1 100644
--- a/third_party/lua/lparser.c
+++ b/third_party/lua/lparser.c
@@ -42,11 +42,7 @@
 #include "third_party/lua/lstring.h"
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /* maximum number of local variables per function (must be smaller
diff --git a/third_party/lua/lrepl.c b/third_party/lua/lrepl.c
index 9a4236ac5..ec65fd069 100644
--- a/third_party/lua/lrepl.c
+++ b/third_party/lua/lrepl.c
@@ -48,11 +48,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 static const char *const kKeywordHints[] = {
diff --git a/third_party/lua/lstate.c b/third_party/lua/lstate.c
index f2479b38c..e44a5cf31 100644
--- a/third_party/lua/lstate.c
+++ b/third_party/lua/lstate.c
@@ -42,11 +42,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/lstring.c b/third_party/lua/lstring.c
index 9aa128242..68efe07a0 100644
--- a/third_party/lua/lstring.c
+++ b/third_party/lua/lstring.c
@@ -36,11 +36,7 @@
 #include "third_party/lua/lstate.h"
 #include "third_party/lua/lstring.h"
 #include "third_party/lua/lua.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/lstrlib.c b/third_party/lua/lstrlib.c
index 45aecbdbc..60db68c4a 100644
--- a/third_party/lua/lstrlib.c
+++ b/third_party/lua/lstrlib.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/ltable.c b/third_party/lua/ltable.c
index a00128613..63f2867f8 100644
--- a/third_party/lua/ltable.c
+++ b/third_party/lua/ltable.c
@@ -38,11 +38,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/ltablib.c b/third_party/lua/ltablib.c
index b7a524445..fe8a1fe62 100644
--- a/third_party/lua/ltablib.c
+++ b/third_party/lua/ltablib.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lprefix.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/ltests.c b/third_party/lua/ltests.c
index b96967efb..10f18d0dc 100644
--- a/third_party/lua/ltests.c
+++ b/third_party/lua/ltests.c
@@ -43,11 +43,7 @@
 #include "third_party/lua/ltable.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/ltm.c b/third_party/lua/ltm.c
index 59b9b8dd4..b7a0d3914 100644
--- a/third_party/lua/ltm.c
+++ b/third_party/lua/ltm.c
@@ -38,11 +38,7 @@
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 static const char udatatypename[] = "userdata";
diff --git a/third_party/lua/lua.main.c b/third_party/lua/lua.main.c
index f7155e185..a49b7ae58 100644
--- a/third_party/lua/lua.main.c
+++ b/third_party/lua/lua.main.c
@@ -50,11 +50,7 @@
 #include "third_party/lua/lualib.h"
 #include "third_party/lua/lunix.h"
 #include "tool/args/args.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 STATIC_STACK_ALIGN(GetStackSize());
 
diff --git a/third_party/lua/luac.main.c b/third_party/lua/luac.main.c
index c5765d888..8b22aa237 100644
--- a/third_party/lua/luac.main.c
+++ b/third_party/lua/luac.main.c
@@ -43,11 +43,7 @@
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
 #include "third_party/lua/lundump.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 static void PrintFunction(const Proto* f, int full);
 #define luaU_print	PrintFunction
diff --git a/third_party/lua/lvm.c b/third_party/lua/lvm.c
index feaaf9cb4..36414ef32 100644
--- a/third_party/lua/lvm.c
+++ b/third_party/lua/lvm.c
@@ -41,11 +41,7 @@
 #include "third_party/lua/ltm.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lvm.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 /*
diff --git a/third_party/lua/lzio.c b/third_party/lua/lzio.c
index da9e2f6c9..3c674495f 100644
--- a/third_party/lua/lzio.c
+++ b/third_party/lua/lzio.c
@@ -34,11 +34,7 @@
 #include "third_party/lua/lstate.h"
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lzio.h"
-
-asm(".ident\t\"\\n\\n\
-Lua 5.4.3 (MIT License)\\n\
-Copyright 1994–2021 Lua.org, PUC-Rio.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("lua_notice");
 
 
 int luaZ_fill (ZIO *z) {
diff --git a/third_party/lz4cli/BUILD.mk b/third_party/lz4cli/BUILD.mk
index b1e6daf70..2440066db 100644
--- a/third_party/lz4cli/BUILD.mk
+++ b/third_party/lz4cli/BUILD.mk
@@ -58,9 +58,6 @@ o/$(MODE)/third_party/lz4cli/lz4cli.com.dbg:		\
 		$(APE_NO_MODIFY_SELF)
 	@$(APELINK)
 
-o/$(MODE)/third_party/lz4cli/lz4cli.o:			\
-		third_party/lz4cli/COPYING
-
 $(THIRD_PARTY_LZ4CLI_OBJS):				\
 		$(BUILD_FILES)				\
 		third_party/lz4cli/BUILD.mk
diff --git a/third_party/lz4cli/COPYING b/third_party/lz4cli/COPYING
index e5e0fa31d..c86990a2b 100644
--- a/third_party/lz4cli/COPYING
+++ b/third_party/lz4cli/COPYING
@@ -1,4 +1,3 @@
-.ident	"\n
 LZ4cli - LZ4 Command Line Interface (GPL v2)
 Copyright (C) Yann Collet 2011-2016
 
@@ -18,4 +17,4 @@ with this program; if not, write to the Free Software Foundation, Inc.,
 
 You can contact the author at :
 - LZ4 source repository : https://github.com/lz4/lz4
-- LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c"
+- LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c
diff --git a/third_party/lz4cli/lz4cli.c b/third_party/lz4cli/lz4cli.c
index a2e5ecc57..55e327faa 100644
--- a/third_party/lz4cli/lz4cli.c
+++ b/third_party/lz4cli/lz4cli.c
@@ -30,7 +30,27 @@
   The license of this compression CLI program is GPLv2.
 */
 
-asm(".include \"third_party/lz4cli/COPYING\"");
+__notice(lz4cli_notice, "\
+LZ4cli - LZ4 Command Line Interface (GPL v2)\n\
+Copyright (C) Yann Collet 2011-2016\n\
+\n\
+This program is free software; you can redistribute it and/or modify\n\
+it under the terms of the GNU General Public License as published by\n\
+the Free Software Foundation; either version 2 of the License, or\n\
+(at your option) any later version.\n\
+\n\
+This program is distributed in the hope that it will be useful,\n\
+but WITHOUT ANY WARRANTY; without even the implied warranty of\n\
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the\n\
+GNU General Public License for more details.\n\
+\n\
+You should have received a copy of the GNU General Public License along\n\
+with this program; if not, write to the Free Software Foundation, Inc.,\n\
+51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.\n\
+\n\
+You can contact the author at :\n\
+- LZ4 source repository : https://github.com/lz4/lz4\n\
+- LZ4 public forum : https://groups.google.com/forum/#!forum/lz4c");
 
 /****************************
 *  Includes
diff --git a/third_party/maxmind/maxminddb.c b/third_party/maxmind/maxminddb.c
index a3d176253..a51ecc495 100644
--- a/third_party/maxmind/maxminddb.c
+++ b/third_party/maxmind/maxminddb.c
@@ -37,10 +37,9 @@
 #include "libc/sysv/consts/sock.h"
 #include "tool/build/lib/case.h"
 
-asm(".ident\t\"\\n\\n\
-libmaxminddb (Apache 2.0)\\n\
-Copyright 2013-2021 MaxMind Incorporated\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(libmaxminddb_notice, "\
+libmaxminddb (Apache 2.0)\n\
+Copyright 2013-2021 MaxMind Incorporated");
 
 #define METADATA_MARKER              "\xab\xcd\xefMaxMind.com"
 #define METADATA_BLOCK_MAX_SIZE      131072 /* This is 128kb */
diff --git a/third_party/mbedtls/aes.c b/third_party/mbedtls/aes.c
index a36aa52e5..6af04dd7c 100644
--- a/third_party/mbedtls/aes.c
+++ b/third_party/mbedtls/aes.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview FIPS-197 compliant AES implementation
diff --git a/third_party/mbedtls/aesce.c b/third_party/mbedtls/aesce.c
index 2866769f5..8ede1f1b8 100644
--- a/third_party/mbedtls/aesce.c
+++ b/third_party/mbedtls/aesce.c
@@ -18,12 +18,7 @@
 #include "third_party/mbedtls/aesce.h"
 #include "libc/str/str.h"
 #include "third_party/aarch64/arm_neon.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Armv8-A Cryptographic Extension support functions for Aarch64
diff --git a/third_party/mbedtls/aesni.c b/third_party/mbedtls/aesni.c
index 13a913771..1babe33a7 100644
--- a/third_party/mbedtls/aesni.c
+++ b/third_party/mbedtls/aesni.c
@@ -19,12 +19,7 @@
 #include "libc/serialize.h"
 #include "libc/str/str.h"
 #include "third_party/mbedtls/common.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  * [AES-WP] http://software.intel.com/en-us/articles/intel-advanced-encryption-standard-aes-instructions-set
diff --git a/third_party/mbedtls/asn1parse.c b/third_party/mbedtls/asn1parse.c
index 8e6458fd1..aa419b6f4 100644
--- a/third_party/mbedtls/asn1parse.c
+++ b/third_party/mbedtls/asn1parse.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 #if defined(MBEDTLS_ASN1_PARSE_C)
 
diff --git a/third_party/mbedtls/asn1write.c b/third_party/mbedtls/asn1write.c
index 5196bd333..1e968fea3 100644
--- a/third_party/mbedtls/asn1write.c
+++ b/third_party/mbedtls/asn1write.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview ASN.1 buffer writing functionality
diff --git a/third_party/mbedtls/base64.c b/third_party/mbedtls/base64.c
index 56f5182c5..c46fcc1de 100644
--- a/third_party/mbedtls/base64.c
+++ b/third_party/mbedtls/base64.c
@@ -19,12 +19,7 @@
 #include "libc/str/str.h"
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 #define ENC "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
 
diff --git a/third_party/mbedtls/bigmul4.c b/third_party/mbedtls/bigmul4.c
index 686828503..89fbd362d 100644
--- a/third_party/mbedtls/bigmul4.c
+++ b/third_party/mbedtls/bigmul4.c
@@ -30,13 +30,11 @@
  */
 void (*Mul4x4)(uint64_t C[16], const uint64_t A[8], const uint64_t B[8]);
 
-static textstartup void Mul4x4Init()
+__attribute__((__constructor__(10))) static textstartup void Mul4x4Init()
 {
     Mul4x4 = X86_HAVE(ADX) && X86_HAVE(BMI2) ? Mul4x4Adx : Mul4x4Pure;
 }
 
-const void *const Mul4x4Ctor[] initarray = {Mul4x4Init};
-
 void Mul4x4Pure(uint64_t C[16], const uint64_t A[8], const uint64_t B[8])
 {
     uint128_t t;
diff --git a/third_party/mbedtls/bignum.c b/third_party/mbedtls/bignum.c
index a5695e4fa..96f0eb1a1 100644
--- a/third_party/mbedtls/bignum.c
+++ b/third_party/mbedtls/bignum.c
@@ -33,12 +33,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/profile.h"
 #include "third_party/mbedtls/select.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Big Numbers.
diff --git a/third_party/mbedtls/ccm.c b/third_party/mbedtls/ccm.c
index bbe5fe76c..747c313f1 100644
--- a/third_party/mbedtls/ccm.c
+++ b/third_party/mbedtls/ccm.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview NIST SP800-38C compliant CCM implementation
diff --git a/third_party/mbedtls/certs.c b/third_party/mbedtls/certs.c
index 3170df134..a946ad5a2 100644
--- a/third_party/mbedtls/certs.c
+++ b/third_party/mbedtls/certs.c
@@ -17,12 +17,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/mbedtls/certs.h"
 #include "third_party/mbedtls/common.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 #if defined(MBEDTLS_CERTS_C)
 
diff --git a/third_party/mbedtls/chacha20.c b/third_party/mbedtls/chacha20.c
index 12e3b052e..c7600c24e 100644
--- a/third_party/mbedtls/chacha20.c
+++ b/third_party/mbedtls/chacha20.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /* Parameter validation macros */
 #define CHACHA20_VALIDATE_RET( cond )                                       \
diff --git a/third_party/mbedtls/chachapoly.c b/third_party/mbedtls/chachapoly.c
index 84dc8fcd8..090b7ab4e 100644
--- a/third_party/mbedtls/chachapoly.c
+++ b/third_party/mbedtls/chachapoly.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview ChaCha20-Poly1305 AEAD construction based on RFC 7539.
diff --git a/third_party/mbedtls/cipher.c b/third_party/mbedtls/cipher.c
index a031b054f..2390f1983 100644
--- a/third_party/mbedtls/cipher.c
+++ b/third_party/mbedtls/cipher.c
@@ -26,12 +26,7 @@
 #include "third_party/mbedtls/gcm.h"
 #include "third_party/mbedtls/nist_kw.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 #if defined(MBEDTLS_CIPHER_C)
 
diff --git a/third_party/mbedtls/cipher_wrap.c b/third_party/mbedtls/cipher_wrap.c
index 176cca76f..c7926079f 100644
--- a/third_party/mbedtls/cipher_wrap.c
+++ b/third_party/mbedtls/cipher_wrap.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/gcm.h"
 #include "third_party/mbedtls/nist_kw.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 #if defined(MBEDTLS_CIPHER_C)
 
diff --git a/third_party/mbedtls/ctr_drbg.c b/third_party/mbedtls/ctr_drbg.c
index 052c741e1..01ab3b2c9 100644
--- a/third_party/mbedtls/ctr_drbg.c
+++ b/third_party/mbedtls/ctr_drbg.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/ctr_drbg.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview CTR_DRBG implementation based on AES-256 (NIST SP 800-90)
diff --git a/third_party/mbedtls/debug.c b/third_party/mbedtls/debug.c
index 462d986a4..27b395b10 100644
--- a/third_party/mbedtls/debug.c
+++ b/third_party/mbedtls/debug.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/debug.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 char mbedtls_debug_threshold;
 
diff --git a/third_party/mbedtls/des.c b/third_party/mbedtls/des.c
index 1e412fdfc..33dc7f5bb 100644
--- a/third_party/mbedtls/des.c
+++ b/third_party/mbedtls/des.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/endian.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview FIPS-46-3 compliant Triple-DES implementation
diff --git a/third_party/mbedtls/dhm.c b/third_party/mbedtls/dhm.c
index 4b9652b0c..222b2f1aa 100644
--- a/third_party/mbedtls/dhm.c
+++ b/third_party/mbedtls/dhm.c
@@ -24,12 +24,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Diffie-Hellman-Merkle key exchange
diff --git a/third_party/mbedtls/ecdh.c b/third_party/mbedtls/ecdh.c
index a0d60e41c..76114dae9 100644
--- a/third_party/mbedtls/ecdh.c
+++ b/third_party/mbedtls/ecdh.c
@@ -19,12 +19,7 @@
 #include "third_party/mbedtls/ecdh.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Elliptic curve Diffie-Hellman
diff --git a/third_party/mbedtls/ecdh_everest.c b/third_party/mbedtls/ecdh_everest.c
index 29034ce02..f0f865121 100644
--- a/third_party/mbedtls/ecdh_everest.c
+++ b/third_party/mbedtls/ecdh_everest.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/everest.h"
 #if defined(MBEDTLS_ECDH_C) && defined(MBEDTLS_ECDH_VARIANT_EVEREST_ENABLED)
 #define KEYSIZE 32
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * \brief           This function sets up the ECDH context with the information
diff --git a/third_party/mbedtls/ecdsa.c b/third_party/mbedtls/ecdsa.c
index 0317da903..0d2528f25 100644
--- a/third_party/mbedtls/ecdsa.c
+++ b/third_party/mbedtls/ecdsa.c
@@ -23,12 +23,7 @@
 #include "third_party/mbedtls/hmac_drbg.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/profile.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Elliptic curve Digital Signature Algorithm
diff --git a/third_party/mbedtls/ecp.c b/third_party/mbedtls/ecp.c
index 3ac9211f7..4b696ef83 100644
--- a/third_party/mbedtls/ecp.c
+++ b/third_party/mbedtls/ecp.c
@@ -32,12 +32,7 @@
 #include "third_party/mbedtls/hmac_drbg.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/profile.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Elliptic curves over GF(p): generic functions
diff --git a/third_party/mbedtls/ecp_curves.c b/third_party/mbedtls/ecp_curves.c
index 06bef587a..7ab258e49 100644
--- a/third_party/mbedtls/ecp_curves.c
+++ b/third_party/mbedtls/ecp_curves.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/ecp.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Elliptic curves over GF(p): curve-specific data and functions
diff --git a/third_party/mbedtls/entropy.c b/third_party/mbedtls/entropy.c
index 32f8658d7..5edc6f7af 100644
--- a/third_party/mbedtls/entropy.c
+++ b/third_party/mbedtls/entropy.c
@@ -23,12 +23,7 @@
 #include "third_party/mbedtls/entropy_poll.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Entropy accumulator implementation
diff --git a/third_party/mbedtls/error.c b/third_party/mbedtls/error.c
index b53082847..b41282740 100644
--- a/third_party/mbedtls/error.c
+++ b/third_party/mbedtls/error.c
@@ -48,12 +48,7 @@
 #include "third_party/mbedtls/sha512.h"
 #include "third_party/mbedtls/ssl.h"
 #include "third_party/mbedtls/x509.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Error message information
diff --git a/third_party/mbedtls/everest.c b/third_party/mbedtls/everest.c
index 75208b8d1..a4af39253 100644
--- a/third_party/mbedtls/everest.c
+++ b/third_party/mbedtls/everest.c
@@ -18,10 +18,11 @@
 #include "libc/serialize.h"
 #include "third_party/mbedtls/endian.h"
 
-asm(".ident\t\"\\n\\n\
-Everest (Apache 2.0)\\n\
-Copyright 2016-2018 INRIA and Microsoft Corporation\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(cosmo_everest_notice, "\
+Cosmopolitan Everest (Apache 2.0)\n\
+Copyright 2024 Justine Alexndra Roberts Tunney\n\
+Copyright 2016-2018 INRIA and Microsoft Corporation\n\
+Changes: Made C code look nice and not have pointers");
 
 #define DW(x)     (uint128_t)(x)
 #define EQ(x, y)  ((((x ^ y) | (~(x ^ y) + 1)) >> 63) - 1)
diff --git a/third_party/mbedtls/gcm.c b/third_party/mbedtls/gcm.c
index 61df3d1cd..5f330eaca 100644
--- a/third_party/mbedtls/gcm.c
+++ b/third_party/mbedtls/gcm.c
@@ -29,12 +29,7 @@
 #include "third_party/mbedtls/endian.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  NIST SP800-38D compliant GCM implementation
diff --git a/third_party/mbedtls/hkdf.c b/third_party/mbedtls/hkdf.c
index f0369c159..d5b76ea40 100644
--- a/third_party/mbedtls/hkdf.c
+++ b/third_party/mbedtls/hkdf.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/hkdf.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview HKDF implementation (RFC 5869)
diff --git a/third_party/mbedtls/hmac_drbg.c b/third_party/mbedtls/hmac_drbg.c
index a2b3c531e..fa5284d1d 100644
--- a/third_party/mbedtls/hmac_drbg.c
+++ b/third_party/mbedtls/hmac_drbg.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/hmac_drbg.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  HMAC_DRBG implementation (NIST SP 800-90)
diff --git a/third_party/mbedtls/md.c b/third_party/mbedtls/md.c
index 8942051a2..82f371cea 100644
--- a/third_party/mbedtls/md.c
+++ b/third_party/mbedtls/md.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/sha1.h"
 #include "third_party/mbedtls/sha256.h"
 #include "third_party/mbedtls/sha512.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * \file md.c
diff --git a/third_party/mbedtls/md5.c b/third_party/mbedtls/md5.c
index 10deb3367..71674bbe8 100644
--- a/third_party/mbedtls/md5.c
+++ b/third_party/mbedtls/md5.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/md.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  RFC 1321 compliant MD5 implementation
diff --git a/third_party/mbedtls/md5t.c b/third_party/mbedtls/md5t.c
index 4d050ddf9..1955eac68 100644
--- a/third_party/mbedtls/md5t.c
+++ b/third_party/mbedtls/md5t.c
@@ -18,12 +18,7 @@
 #include "libc/str/str.h"
 #include "third_party/mbedtls/md5.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  * RFC 1321 test vectors
diff --git a/third_party/mbedtls/memory_buffer_alloc.c b/third_party/mbedtls/memory_buffer_alloc.c
index b3cc7d128..283b92d0a 100644
--- a/third_party/mbedtls/memory_buffer_alloc.c
+++ b/third_party/mbedtls/memory_buffer_alloc.c
@@ -18,12 +18,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/memory_buffer_alloc.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Buffer-based memory allocator
diff --git a/third_party/mbedtls/nist_kw.c b/third_party/mbedtls/nist_kw.c
index 4523b6192..a95956301 100644
--- a/third_party/mbedtls/nist_kw.c
+++ b/third_party/mbedtls/nist_kw.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/nist_kw.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Implementation of NIST SP 800-38F key wrapping, supporting KW and KWP modes
diff --git a/third_party/mbedtls/notice.c b/third_party/mbedtls/notice.c
new file mode 100644
index 000000000..efae1a05e
--- /dev/null
+++ b/third_party/mbedtls/notice.c
@@ -0,0 +1,4 @@
+__notice(mbedtls_notice, "\
+Mbed TLS (Apache 2.0)\n\
+Copyright ARM Limited\n\
+Copyright The Mbed TLS Contributors");
diff --git a/third_party/mbedtls/oid.c b/third_party/mbedtls/oid.c
index 88083c7ec..837e909b9 100644
--- a/third_party/mbedtls/oid.c
+++ b/third_party/mbedtls/oid.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/oid.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/rsa.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Object Identifier (OID) database
diff --git a/third_party/mbedtls/pem.c b/third_party/mbedtls/pem.c
index 8b8e12d7b..6bc84f716 100644
--- a/third_party/mbedtls/pem.c
+++ b/third_party/mbedtls/pem.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/md5.h"
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Privacy Enhanced Mail (PEM) decoding
diff --git a/third_party/mbedtls/pk.c b/third_party/mbedtls/pk.c
index 736089ab0..92b3d3556 100644
--- a/third_party/mbedtls/pk.c
+++ b/third_party/mbedtls/pk.c
@@ -23,12 +23,7 @@
 #include "third_party/mbedtls/pk_internal.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/rsa.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Public Key abstraction layer
diff --git a/third_party/mbedtls/pk_wrap.c b/third_party/mbedtls/pk_wrap.c
index 9aaba9947..a649523a1 100644
--- a/third_party/mbedtls/pk_wrap.c
+++ b/third_party/mbedtls/pk_wrap.c
@@ -25,12 +25,7 @@
 #include "third_party/mbedtls/pk_internal.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/rsa.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview Public Key abstraction layer: wrapper functions
diff --git a/third_party/mbedtls/pkcs5.c b/third_party/mbedtls/pkcs5.c
index 10cc06f42..00910985b 100644
--- a/third_party/mbedtls/pkcs5.c
+++ b/third_party/mbedtls/pkcs5.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/oid.h"
 #include "third_party/mbedtls/pkcs5.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview PKCS#5 functions, e.g. PBKDF2
diff --git a/third_party/mbedtls/pkparse.c b/third_party/mbedtls/pkparse.c
index d657fd801..a5b16e2f3 100644
--- a/third_party/mbedtls/pkparse.c
+++ b/third_party/mbedtls/pkparse.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/pkcs5.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/rsa.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Public Key layer for parsing key files and structures
diff --git a/third_party/mbedtls/pkwrite.c b/third_party/mbedtls/pkwrite.c
index 91c58252c..23d52818c 100644
--- a/third_party/mbedtls/pkwrite.c
+++ b/third_party/mbedtls/pkwrite.c
@@ -26,12 +26,7 @@
 #include "third_party/mbedtls/pk.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/rsa.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Public Key layer for writing key files and structures
diff --git a/third_party/mbedtls/poly1305.c b/third_party/mbedtls/poly1305.c
index c3f543d5c..20f3f838c 100644
--- a/third_party/mbedtls/poly1305.c
+++ b/third_party/mbedtls/poly1305.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * \file poly1305.c
diff --git a/third_party/mbedtls/rsa.c b/third_party/mbedtls/rsa.c
index 158c526ba..edc6ae58f 100644
--- a/third_party/mbedtls/rsa.c
+++ b/third_party/mbedtls/rsa.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/profile.h"
 #include "third_party/mbedtls/rsa_internal.h"
 #include "third_party/mbedtls/sha1.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview The RSA public-key cryptosystem
diff --git a/third_party/mbedtls/rsa_internal.c b/third_party/mbedtls/rsa_internal.c
index aa9d03fb2..5359d5d80 100644
--- a/third_party/mbedtls/rsa_internal.c
+++ b/third_party/mbedtls/rsa_internal.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/profile.h"
 #include "third_party/mbedtls/rsa.h"
 #include "third_party/mbedtls/rsa_internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Helper functions for the RSA module
diff --git a/third_party/mbedtls/sha1.c b/third_party/mbedtls/sha1.c
index 26806ab94..be933c939 100644
--- a/third_party/mbedtls/sha1.c
+++ b/third_party/mbedtls/sha1.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/md.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview FIPS-180-1 compliant SHA-1 implementation
diff --git a/third_party/mbedtls/sha1t.c b/third_party/mbedtls/sha1t.c
index 641061b9c..fc553f34d 100644
--- a/third_party/mbedtls/sha1t.c
+++ b/third_party/mbedtls/sha1t.c
@@ -18,12 +18,7 @@
 #include "libc/str/str.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/sha1.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  * FIPS-180-1 test vectors
diff --git a/third_party/mbedtls/sha256.c b/third_party/mbedtls/sha256.c
index 3261608f1..88e621b53 100644
--- a/third_party/mbedtls/sha256.c
+++ b/third_party/mbedtls/sha256.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/endian.h"
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/md.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview FIPS-180-2 compliant SHA-256 implementation
diff --git a/third_party/mbedtls/sha256t.c b/third_party/mbedtls/sha256t.c
index 996c30019..f3b822447 100644
--- a/third_party/mbedtls/sha256t.c
+++ b/third_party/mbedtls/sha256t.c
@@ -18,12 +18,7 @@
 #include "libc/str/str.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/sha256.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  * FIPS-180-2 test vectors
diff --git a/third_party/mbedtls/sha512.c b/third_party/mbedtls/sha512.c
index a5e197e18..8f4494787 100644
--- a/third_party/mbedtls/sha512.c
+++ b/third_party/mbedtls/sha512.c
@@ -28,12 +28,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/md.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview FIPS-180-2 compliant SHA-384/512 implementation
diff --git a/third_party/mbedtls/sha512t.c b/third_party/mbedtls/sha512t.c
index d8349047f..5e4730831 100644
--- a/third_party/mbedtls/sha512t.c
+++ b/third_party/mbedtls/sha512t.c
@@ -19,12 +19,7 @@
 #include "libc/str/str.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/sha512.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  * FIPS-180-2 test vectors
diff --git a/third_party/mbedtls/shiftright.c b/third_party/mbedtls/shiftright.c
index e3be94ccd..3162eae01 100644
--- a/third_party/mbedtls/shiftright.c
+++ b/third_party/mbedtls/shiftright.c
@@ -21,8 +21,7 @@
 
 void (*ShiftRight)(uint64_t *, size_t, unsigned char);
 
+__attribute__((__constructor__(10)))
 static textstartup void ShiftRightInit(void) {
   ShiftRight = 0 && X86_HAVE(AVX) ? ShiftRightAvx : ShiftRightPure;
 }
-
-const void *const ShiftRightCtor[] initarray = {ShiftRightInit};
diff --git a/third_party/mbedtls/ssl_cache.c b/third_party/mbedtls/ssl_cache.c
index 453326702..c4d435ad9 100644
--- a/third_party/mbedtls/ssl_cache.c
+++ b/third_party/mbedtls/ssl_cache.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/ssl_cache.h"
 #include "third_party/mbedtls/ssl_internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  SSL session cache implementation
diff --git a/third_party/mbedtls/ssl_ciphersuites.c b/third_party/mbedtls/ssl_ciphersuites.c
index a7a22a1ad..812fa4cce 100644
--- a/third_party/mbedtls/ssl_ciphersuites.c
+++ b/third_party/mbedtls/ssl_ciphersuites.c
@@ -25,12 +25,7 @@
 #include "third_party/mbedtls/common.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/ssl.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  CRYPTOGRAPHY 101
diff --git a/third_party/mbedtls/ssl_cli.c b/third_party/mbedtls/ssl_cli.c
index b8f2e79b7..002c4aba9 100644
--- a/third_party/mbedtls/ssl_cli.c
+++ b/third_party/mbedtls/ssl_cli.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/ssl.h"
 #include "third_party/mbedtls/ssl_internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  SSLv3/TLSv1 client-side functions
diff --git a/third_party/mbedtls/ssl_cookie.c b/third_party/mbedtls/ssl_cookie.c
index 19eec8754..7ca95fcd4 100644
--- a/third_party/mbedtls/ssl_cookie.c
+++ b/third_party/mbedtls/ssl_cookie.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/ssl_cookie.h"
 #include "third_party/mbedtls/ssl_internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  DTLS cookie callbacks implementation
diff --git a/third_party/mbedtls/ssl_msg.c b/third_party/mbedtls/ssl_msg.c
index 68ad70f49..41fb201d3 100644
--- a/third_party/mbedtls/ssl_msg.c
+++ b/third_party/mbedtls/ssl_msg.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/ssl.h"
 #include "third_party/mbedtls/ssl_internal.h"
 #include "third_party/mbedtls/ssl_invasive.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  Generic SSL/TLS messaging layer functions
diff --git a/third_party/mbedtls/ssl_srv.c b/third_party/mbedtls/ssl_srv.c
index a51e54b6e..ef4e5c964 100644
--- a/third_party/mbedtls/ssl_srv.c
+++ b/third_party/mbedtls/ssl_srv.c
@@ -27,12 +27,7 @@
 #include "third_party/mbedtls/profile.h"
 #include "third_party/mbedtls/ssl.h"
 #include "third_party/mbedtls/ssl_internal.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  SSLv3/TLSv1 server-side functions
diff --git a/third_party/mbedtls/ssl_ticket.c b/third_party/mbedtls/ssl_ticket.c
index 01fee1941..70e4a4fa9 100644
--- a/third_party/mbedtls/ssl_ticket.c
+++ b/third_party/mbedtls/ssl_ticket.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/ssl_internal.h"
 #include "third_party/mbedtls/ssl_ticket.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  TLS server tickets callbacks implementation
diff --git a/third_party/mbedtls/ssl_tls.c b/third_party/mbedtls/ssl_tls.c
index df248f2f7..5460a4d74 100644
--- a/third_party/mbedtls/ssl_tls.c
+++ b/third_party/mbedtls/ssl_tls.c
@@ -28,12 +28,7 @@
 #include "third_party/mbedtls/ssl_ciphersuites.h"
 #include "third_party/mbedtls/ssl_internal.h"
 #include "third_party/mbedtls/version.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview SSLv3/TLSv1 shared functions
diff --git a/third_party/mbedtls/ssl_tls13_keys.c b/third_party/mbedtls/ssl_tls13_keys.c
index ae521c0c2..523959f03 100644
--- a/third_party/mbedtls/ssl_tls13_keys.c
+++ b/third_party/mbedtls/ssl_tls13_keys.c
@@ -19,12 +19,7 @@
 #include "third_party/mbedtls/hkdf.h"
 #include "third_party/mbedtls/ssl_internal.h"
 #include "third_party/mbedtls/ssl_tls13_keys.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  TLS 1.3 key schedule
diff --git a/third_party/mbedtls/test/everest_unravaged.c b/third_party/mbedtls/test/everest_unravaged.c
index cc4737be1..9c24f1f42 100644
--- a/third_party/mbedtls/test/everest_unravaged.c
+++ b/third_party/mbedtls/test/everest_unravaged.c
@@ -7,10 +7,8 @@
 #include "libc/serialize.h"
 #include "third_party/mbedtls/platform.h"
 
-asm(".ident\t\"\\n\\n\
-Everest (Apache 2.0)\\n\
-Copyright 2016-2018 INRIA and Microsoft Corporation\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(everest_notice, "Everest (Apache 2.0)\n\
+Copyright 2016-2018 INRIA and Microsoft Corporation");
 
 /*
  *  ECDH with curve-optimized implementation multiplexing
diff --git a/third_party/mbedtls/test/lib.c b/third_party/mbedtls/test/lib.c
index e284a67b4..5e72cd6b6 100644
--- a/third_party/mbedtls/test/lib.c
+++ b/third_party/mbedtls/test/lib.c
@@ -50,12 +50,7 @@
 #include "third_party/mbedtls/error.h"
 #include "libc/serialize.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 __static_yoink("zipos");
 
diff --git a/third_party/mbedtls/test/test.inc b/third_party/mbedtls/test/test.inc
index 9412aac12..f4a992d4e 100644
--- a/third_party/mbedtls/test/test.inc
+++ b/third_party/mbedtls/test/test.inc
@@ -6,11 +6,6 @@
 #include "libc/sysv/consts/exit.h"
 #include "third_party/mbedtls/config.h"
 #include "third_party/mbedtls/test/lib.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 __static_yoink("zipos");
diff --git a/third_party/mbedtls/x509.c b/third_party/mbedtls/x509.c
index dab92a7dd..62949a890 100644
--- a/third_party/mbedtls/x509.c
+++ b/third_party/mbedtls/x509.c
@@ -28,12 +28,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/x509.h"
 #include "third_party/mbedtls/x509_crt.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  X.509 common functions for parsing and verification
diff --git a/third_party/mbedtls/x509_create.c b/third_party/mbedtls/x509_create.c
index 2e2dbd8a0..ef28ec1ba 100644
--- a/third_party/mbedtls/x509_create.c
+++ b/third_party/mbedtls/x509_create.c
@@ -20,12 +20,7 @@
 #include "third_party/mbedtls/error.h"
 #include "third_party/mbedtls/oid.h"
 #include "third_party/mbedtls/x509.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  X.509 base functions for creating certificates / CSRs
diff --git a/third_party/mbedtls/x509_crl.c b/third_party/mbedtls/x509_crl.c
index 807e63873..f9460fccb 100644
--- a/third_party/mbedtls/x509_crl.c
+++ b/third_party/mbedtls/x509_crl.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/x509_crl.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview X.509 Certidicate Revocation List (CRL) parsing
diff --git a/third_party/mbedtls/x509_crt.c b/third_party/mbedtls/x509_crt.c
index 47a4c103a..da53f29d2 100644
--- a/third_party/mbedtls/x509_crt.c
+++ b/third_party/mbedtls/x509_crt.c
@@ -32,12 +32,7 @@
 #include "third_party/mbedtls/oid.h"
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  X.509 certificate parsing and verification
diff --git a/third_party/mbedtls/x509_csr.c b/third_party/mbedtls/x509_csr.c
index ee8b4941c..4ebb8d39f 100644
--- a/third_party/mbedtls/x509_csr.c
+++ b/third_party/mbedtls/x509_csr.c
@@ -21,12 +21,7 @@
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/x509_csr.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /**
  * @fileoverview X.509 Certificate Signing Request (CSR) parsing
diff --git a/third_party/mbedtls/x509write_crt.c b/third_party/mbedtls/x509write_crt.c
index 987ba9df8..7558b38ec 100644
--- a/third_party/mbedtls/x509write_crt.c
+++ b/third_party/mbedtls/x509write_crt.c
@@ -23,12 +23,7 @@
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/sha1.h"
 #include "third_party/mbedtls/x509_crt.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  X.509 certificate writing
diff --git a/third_party/mbedtls/x509write_csr.c b/third_party/mbedtls/x509write_csr.c
index 3d22704ad..ffda801af 100644
--- a/third_party/mbedtls/x509write_csr.c
+++ b/third_party/mbedtls/x509write_csr.c
@@ -22,12 +22,7 @@
 #include "third_party/mbedtls/pem.h"
 #include "third_party/mbedtls/platform.h"
 #include "third_party/mbedtls/x509_csr.h"
-
-asm(".ident\t\"\\n\\n\
-Mbed TLS (Apache 2.0)\\n\
-Copyright ARM Limited\\n\
-Copyright Mbed TLS Contributors\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("mbedtls_notice");
 
 /*
  *  X.509 Certificate Signing Request writing
diff --git a/third_party/musl/crypt.c b/third_party/musl/crypt.c
index 2bf39082e..a6a81de20 100644
--- a/third_party/musl/crypt.c
+++ b/third_party/musl/crypt.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/crypt.h"
 #include "third_party/musl/crypt.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /**
  * Encrypts password the old fashioned way.
diff --git a/third_party/musl/crypt_blowfish.c b/third_party/musl/crypt_blowfish.c
index 9dfec80c0..5586c53c6 100644
--- a/third_party/musl/crypt_blowfish.c
+++ b/third_party/musl/crypt_blowfish.c
@@ -32,11 +32,7 @@
 #include "libc/mem/gc.h"
 #include "libc/str/str.h"
 #include "third_party/musl/crypt.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /* Modified by Rich Felker in for inclusion in musl libc, based on
  * Solar Designer's second size-optimized version sent to the musl
diff --git a/third_party/musl/crypt_des.c b/third_party/musl/crypt_des.c
index f301f8b22..564059ac7 100644
--- a/third_party/musl/crypt_des.c
+++ b/third_party/musl/crypt_des.c
@@ -31,14 +31,10 @@
 #include "libc/str/str.h"
 #include "third_party/musl/crypt.internal.h"
 #include "third_party/musl/crypt_des.internal.h"
+__static_yoink("musl_libc_notice");
 
 #pragma GCC diagnostic ignored "-Wstringop-overflow"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /*
  * This version has been further modified by Rich Felker, primary author
  * and maintainer of musl libc, to remove table generation code and
diff --git a/third_party/musl/crypt_md5.c b/third_party/musl/crypt_md5.c
index 1c00345b0..5bf75c1eb 100644
--- a/third_party/musl/crypt_md5.c
+++ b/third_party/musl/crypt_md5.c
@@ -30,14 +30,10 @@
 #include "libc/literal.h"
 #include "libc/str/str.h"
 #include "third_party/musl/crypt.internal.h"
+__static_yoink("musl_libc_notice");
 
 #pragma GCC diagnostic ignored "-Wmissing-braces"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /*
  * md5 crypt implementation
  *
diff --git a/third_party/musl/crypt_r.c b/third_party/musl/crypt_r.c
index e90411ff5..f0155b8c6 100644
--- a/third_party/musl/crypt_r.c
+++ b/third_party/musl/crypt_r.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/crypt.h"
 #include "third_party/musl/crypt.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /**
  * Encrypts password the old fashioned way.
diff --git a/third_party/musl/crypt_sha256.c b/third_party/musl/crypt_sha256.c
index ec699090d..d5e2f8059 100644
--- a/third_party/musl/crypt_sha256.c
+++ b/third_party/musl/crypt_sha256.c
@@ -40,14 +40,10 @@
 #include "libc/sysv/consts/exit.h"
 #include "third_party/gdtoa/gdtoa.h"
 #include "third_party/musl/crypt.internal.h"
+__static_yoink("musl_libc_notice");
 
 #pragma GCC diagnostic ignored "-Wmissing-braces"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /*
  * public domain sha256 crypt implementation
  *
diff --git a/third_party/musl/crypt_sha512.c b/third_party/musl/crypt_sha512.c
index 122a64cff..7d3b4414b 100644
--- a/third_party/musl/crypt_sha512.c
+++ b/third_party/musl/crypt_sha512.c
@@ -40,14 +40,10 @@
 #include "libc/sysv/consts/exit.h"
 #include "third_party/gdtoa/gdtoa.h"
 #include "third_party/musl/crypt.internal.h"
+__static_yoink("musl_libc_notice");
 
 #pragma GCC diagnostic ignored "-Wmissing-braces"
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /*
  * public domain sha512 crypt implementation
  *
diff --git a/third_party/musl/encrypt.c b/third_party/musl/encrypt.c
index 6474c6a84..722ea2650 100644
--- a/third_party/musl/encrypt.c
+++ b/third_party/musl/encrypt.c
@@ -41,11 +41,7 @@
 #include "libc/sysv/consts/ok.h"
 #include "third_party/gdtoa/gdtoa.h"
 #include "third_party/musl/crypt_des.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static struct expanded_key __encrypt_key;
 
diff --git a/third_party/musl/fnmatch.c b/third_party/musl/fnmatch.c
index ed04f5c21..d2f32e259 100644
--- a/third_party/musl/fnmatch.c
+++ b/third_party/musl/fnmatch.c
@@ -28,6 +28,7 @@
 #include "libc/limits.h"
 #include "libc/str/str.h"
 #include "third_party/musl/fnmatch.h"
+__static_yoink("musl_libc_notice");
 
 /*
  * An implementation of what I call the "Sea of Stars" algorithm for
@@ -44,11 +45,6 @@
  * - Rich Felker, April 2012
  */
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #define END         0
 #define UNMATCHABLE -2
 #define BRACKET     -3
diff --git a/third_party/musl/forkpty.c b/third_party/musl/forkpty.c
index f53984a51..ea881b0fb 100644
--- a/third_party/musl/forkpty.c
+++ b/third_party/musl/forkpty.c
@@ -35,11 +35,7 @@
 #include "libc/runtime/runtime.h"
 #include "libc/errno.h"
 #include "libc/calls/termios.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int forkpty(int *pm, char *name, const struct termios *tio, const struct winsize *ws)
 {
diff --git a/third_party/musl/freeaddrinfo.c b/third_party/musl/freeaddrinfo.c
index 9fda19caa..e9bcfbd7b 100644
--- a/third_party/musl/freeaddrinfo.c
+++ b/third_party/musl/freeaddrinfo.c
@@ -29,11 +29,7 @@
 #include "libc/mem/mem.h"
 #include "third_party/musl/lookup.internal.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 void freeaddrinfo(struct addrinfo *p)
 {
diff --git a/third_party/musl/gai_strerror.c b/third_party/musl/gai_strerror.c
index 2b6e85850..7e9a83eed 100644
--- a/third_party/musl/gai_strerror.c
+++ b/third_party/musl/gai_strerror.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static const char msgs[] =
 	"Invalid flags\0"
diff --git a/third_party/musl/getaddrinfo.c b/third_party/musl/getaddrinfo.c
index 7c418d518..9c8d0a926 100644
--- a/third_party/musl/getaddrinfo.c
+++ b/third_party/musl/getaddrinfo.c
@@ -40,11 +40,7 @@
 #include "third_party/musl/lookup.internal.h"
 #include "libc/intrin/atomic.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int getaddrinfo(const char *host,
                 const char *serv,
diff --git a/third_party/musl/gethostbyaddr_r.c b/third_party/musl/gethostbyaddr_r.c
index e4934b94c..51572bb1d 100644
--- a/third_party/musl/gethostbyaddr_r.c
+++ b/third_party/musl/gethostbyaddr_r.c
@@ -32,11 +32,7 @@
 #include "libc/str/str.h"
 #include "libc/sysv/consts/af.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 errno_t gethostbyaddr_r(const void *a, socklen_t l, int af,
 	struct hostent *h, char *buf, size_t buflen,
diff --git a/third_party/musl/gethostbyname2_r.c b/third_party/musl/gethostbyname2_r.c
index 9ae8c0841..e95bf9535 100644
--- a/third_party/musl/gethostbyname2_r.c
+++ b/third_party/musl/gethostbyname2_r.c
@@ -30,11 +30,7 @@
 #include "libc/sysv/consts/af.h"
 #include "third_party/musl/lookup.internal.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 errno_t gethostbyname2_r(const char *name, int af,
 	struct hostent *h, char *buf, size_t buflen,
diff --git a/third_party/musl/getnameinfo.c b/third_party/musl/getnameinfo.c
index 62f20e009..ffced1dd3 100644
--- a/third_party/musl/getnameinfo.c
+++ b/third_party/musl/getnameinfo.c
@@ -37,11 +37,7 @@
 #include "third_party/musl/lookup.internal.h"
 #include "third_party/musl/netdb.h"
 #include "third_party/musl/resolv.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 #define PTR_MAX (64 + sizeof ".in-addr.arpa")
 #define RR_PTR 12
diff --git a/third_party/musl/getservbyname_r.c b/third_party/musl/getservbyname_r.c
index cd088ac29..d6e238fba 100644
--- a/third_party/musl/getservbyname_r.c
+++ b/third_party/musl/getservbyname_r.c
@@ -32,11 +32,7 @@
 #include "libc/sysv/consts/ipproto.h"
 #include "third_party/musl/lookup.internal.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 #define ALIGN (sizeof(struct { char a; char *b; }) - sizeof(char *))
 
diff --git a/third_party/musl/getservbyport_r.c b/third_party/musl/getservbyport_r.c
index 3749ef393..0e98a8eaf 100644
--- a/third_party/musl/getservbyport_r.c
+++ b/third_party/musl/getservbyport_r.c
@@ -31,11 +31,7 @@
 #include "libc/str/str.h"
 #include "libc/sysv/consts/af.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 errno_t getservbyport_r(int port, const char *prots,
 	struct servent *se, char *buf, size_t buflen, struct servent **res)
diff --git a/third_party/musl/getspnam_r.c b/third_party/musl/getspnam_r.c
index 3b08b1642..dae46748f 100644
--- a/third_party/musl/getspnam_r.c
+++ b/third_party/musl/getspnam_r.c
@@ -31,11 +31,7 @@
 #include <ctype.h>
 #include <pthread.h>
 #include "pwf.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /* This implementation support Openwall-style TCB passwords in place of
  * traditional shadow, if the appropriate directories and files exist.
diff --git a/third_party/musl/glob.c b/third_party/musl/glob.c
index c7156ccb3..655532d9a 100644
--- a/third_party/musl/glob.c
+++ b/third_party/musl/glob.c
@@ -36,14 +36,10 @@
 #include "libc/sysv/consts/dt.h"
 #include "libc/sysv/consts/s.h"
 #include "third_party/musl/fnmatch.h"
+__static_yoink("musl_libc_notice");
 
 #define MAXPATH 1024
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 struct GlobList {
   struct GlobList *next;
   char name[];
diff --git a/third_party/musl/grp.c b/third_party/musl/grp.c
index 7a788c7f0..17d777ef7 100644
--- a/third_party/musl/grp.c
+++ b/third_party/musl/grp.c
@@ -34,11 +34,7 @@
 #include "libc/sysv/consts/limits.h"
 #include "libc/thread/thread.h"
 #include "third_party/musl/passwd.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static unsigned atou(char **s) {
   unsigned x;
diff --git a/third_party/musl/hsearch.c b/third_party/musl/hsearch.c
index 883fec34f..e4140f404 100644
--- a/third_party/musl/hsearch.c
+++ b/third_party/musl/hsearch.c
@@ -28,11 +28,7 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 #include "third_party/musl/search.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /*
 open addressing hash table with 2^n table size
diff --git a/third_party/musl/insque.c b/third_party/musl/insque.c
index 63c00539b..cec746ec2 100644
--- a/third_party/musl/insque.c
+++ b/third_party/musl/insque.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/search.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 struct node {
 	struct node *next;
diff --git a/third_party/musl/lockf.c b/third_party/musl/lockf.c
index 33c57e3d6..a0ec7e670 100644
--- a/third_party/musl/lockf.c
+++ b/third_party/musl/lockf.c
@@ -32,11 +32,7 @@
 #include "libc/errno.h"
 #include "libc/sysv/consts/f.h"
 #include "libc/sysv/errfuns.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /**
  * Locks file.
diff --git a/third_party/musl/lookup_ipliteral.c b/third_party/musl/lookup_ipliteral.c
index 1ba9a2d5e..a02b0239b 100644
--- a/third_party/musl/lookup_ipliteral.c
+++ b/third_party/musl/lookup_ipliteral.c
@@ -32,11 +32,7 @@
 #include "libc/limits.h"
 #include "libc/sock/sock.h"
 #include "third_party/musl/lookup.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int __lookup_ipliteral(struct address buf[static 1], const char *name, int family)
 {
diff --git a/third_party/musl/lookup_name.c b/third_party/musl/lookup_name.c
index 37214b72a..291178213 100644
--- a/third_party/musl/lookup_name.c
+++ b/third_party/musl/lookup_name.c
@@ -40,11 +40,7 @@
 #include "third_party/musl/lookup.internal.h"
 #include "third_party/musl/netdb.h"
 #include "third_party/musl/resolv.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static int is_valid_hostname(const char *host)
 {
diff --git a/third_party/musl/lookup_serv.c b/third_party/musl/lookup_serv.c
index 8eb45abc7..cc7fbffb2 100644
--- a/third_party/musl/lookup_serv.c
+++ b/third_party/musl/lookup_serv.c
@@ -33,11 +33,7 @@
 #include "libc/errno.h"
 #include "libc/calls/sysdir.internal.h"
 #include "third_party/musl/lookup.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int __lookup_serv(struct service buf[static MAXSERVS],
 		  const char *name,
diff --git a/third_party/musl/lsearch.c b/third_party/musl/lsearch.c
index 1d1ee6a5c..c4b871206 100644
--- a/third_party/musl/lsearch.c
+++ b/third_party/musl/lsearch.c
@@ -28,11 +28,7 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 #include "third_party/musl/search.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 
 void *lsearch(const void *key, void *base, size_t *nelp, size_t width,
diff --git a/third_party/musl/mntent.c b/third_party/musl/mntent.c
index cc010090a..275ba586d 100644
--- a/third_party/musl/mntent.c
+++ b/third_party/musl/mntent.c
@@ -31,11 +31,7 @@
 #include "libc/stdio/stdio.h"
 #include "libc/str/str.h"
 #include "third_party/musl/mntent.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static char *internal_buf;
 static size_t internal_bufsize;
diff --git a/third_party/musl/ns_parse.c b/third_party/musl/ns_parse.c
index e3964e2fb..dad9dd71c 100644
--- a/third_party/musl/ns_parse.c
+++ b/third_party/musl/ns_parse.c
@@ -28,11 +28,7 @@
 #include "libc/errno.h"
 #include "third_party/musl/resolv.h"
 #include "third_party/musl/nameser.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 const struct _ns_flagdata _ns_flagdata[16] = {
 	{ 0x8000, 15 },
diff --git a/third_party/musl/proto.c b/third_party/musl/proto.c
index 86edea22b..6e616dbf0 100644
--- a/third_party/musl/proto.c
+++ b/third_party/musl/proto.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/str/str.h"
 #include "third_party/musl/netdb.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 /* do we really need all these?? */
 
diff --git a/third_party/musl/pwd.c b/third_party/musl/pwd.c
index a1dedd141..fc54b77cf 100644
--- a/third_party/musl/pwd.c
+++ b/third_party/musl/pwd.c
@@ -38,11 +38,7 @@
 #include "libc/str/str.h"
 #include "libc/thread/thread.h"
 #include "third_party/musl/passwd.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 #ifdef FTRACE
 // if the default mode debugging tools are enabled, and we're linking
diff --git a/third_party/musl/rand48.c b/third_party/musl/rand48.c
index fa0a94db3..5c46adf7d 100644
--- a/third_party/musl/rand48.c
+++ b/third_party/musl/rand48.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/rand48.h"
 #include "libc/str/str.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static unsigned short __seed48[7] = { 0, 0, 0, 0xe66d, 0xdeec, 0x5, 0xb };
 
diff --git a/third_party/musl/res_mkquery.c b/third_party/musl/res_mkquery.c
index f10f2148d..f955d6cfb 100644
--- a/third_party/musl/res_mkquery.c
+++ b/third_party/musl/res_mkquery.c
@@ -29,11 +29,7 @@
 #include "libc/sysv/consts/clock.h"
 #include "libc/stdio/rand.h"
 #include "third_party/musl/resolv.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int __res_mkquery(int op, const char *dname, int class, int type,
 	const unsigned char *data, int datalen,
diff --git a/third_party/musl/res_msend.c b/third_party/musl/res_msend.c
index 705a0e5b4..1a65bee68 100644
--- a/third_party/musl/res_msend.c
+++ b/third_party/musl/res_msend.c
@@ -45,11 +45,7 @@
 #include "libc/sysv/consts/tcp.h"
 #include "libc/thread/thread.h"
 #include "lookup.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static void cleanup(void *p)
 {
diff --git a/third_party/musl/res_query.c b/third_party/musl/res_query.c
index f324e58bb..04d7427f7 100644
--- a/third_party/musl/res_query.c
+++ b/third_party/musl/res_query.c
@@ -28,11 +28,7 @@
 #include "third_party/musl/resolv.h"
 #include "third_party/musl/netdb.h"
 #include "third_party/musl/resolv.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int res_query(const char *name, int class, int type, unsigned char *dest, int len)
 {
diff --git a/third_party/musl/res_querydomain.c b/third_party/musl/res_querydomain.c
index 809bc6877..e91149944 100644
--- a/third_party/musl/res_querydomain.c
+++ b/third_party/musl/res_querydomain.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/str/str.h"
 #include "third_party/musl/resolv.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int res_querydomain(const char *name, const char *domain, int class, int type, unsigned char *dest, int len)
 {
diff --git a/third_party/musl/res_send.c b/third_party/musl/res_send.c
index 1ef881370..cd28d0706 100644
--- a/third_party/musl/res_send.c
+++ b/third_party/musl/res_send.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/str/str.h"
 #include "third_party/musl/resolv.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 int __res_send(const unsigned char *msg, int msglen, unsigned char *answer, int anslen)
 {
diff --git a/third_party/musl/resolvconf.c b/third_party/musl/resolvconf.c
index bdcfbc719..ddc44b363 100644
--- a/third_party/musl/resolvconf.c
+++ b/third_party/musl/resolvconf.c
@@ -38,11 +38,7 @@
 #include "libc/sysv/consts/af.h"
 #include "libc/sock/sock.h"
 #include "third_party/musl/lookup.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 // TODO(jart): ipv6 nameservers on windows
 // TODO(jart): can we polyfill `search` on windows
diff --git a/third_party/musl/strfmon.c b/third_party/musl/strfmon.c
index 82bd8c0a9..95e590612 100644
--- a/third_party/musl/strfmon.c
+++ b/third_party/musl/strfmon.c
@@ -30,11 +30,7 @@
 #include "libc/str/locale.h"
 #include "libc/str/str.h"
 #include "libc/thread/tls.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static ssize_t vstrfmon_l(char *s, size_t n, locale_t loc, const char *fmt, va_list ap)
 {
diff --git a/third_party/musl/tdelete.c b/third_party/musl/tdelete.c
index 6023062fc..52e871341 100644
--- a/third_party/musl/tdelete.c
+++ b/third_party/musl/tdelete.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/mem/mem.h"
 #include "third_party/musl/tsearch.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 
 void *tdelete(const void *restrict key, void **restrict rootp,
diff --git a/third_party/musl/tdestroy.c b/third_party/musl/tdestroy.c
index 51a681669..17904479c 100644
--- a/third_party/musl/tdestroy.c
+++ b/third_party/musl/tdestroy.c
@@ -27,11 +27,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/mem/mem.h"
 #include "third_party/musl/tsearch.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 void tdestroy(void *root, void (*freekey)(void *))
 {
diff --git a/third_party/musl/tempnam.c b/third_party/musl/tempnam.c
index c72761ddd..93d8a6856 100644
--- a/third_party/musl/tempnam.c
+++ b/third_party/musl/tempnam.c
@@ -37,14 +37,10 @@
 #include "libc/sysv/consts/at.h"
 #include "libc/sysv/consts/clock.h"
 #include "libc/time/time.h"
+__static_yoink("musl_libc_notice");
 
 #define MAXTRIES 100
 
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 static char *
 __randname(char *template)
 {
diff --git a/third_party/musl/tfind.c b/third_party/musl/tfind.c
index 0735306b9..b4a668bcd 100644
--- a/third_party/musl/tfind.c
+++ b/third_party/musl/tfind.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/tsearch.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 void *tfind(const void *key, void *const *rootp,
 	int(*cmp)(const void *, const void *))
diff --git a/third_party/musl/tsearch.c b/third_party/musl/tsearch.c
index 1c9ceecd8..fdfd0c75e 100644
--- a/third_party/musl/tsearch.c
+++ b/third_party/musl/tsearch.c
@@ -28,11 +28,7 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 #include "third_party/musl/tsearch.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 static inline int height(struct node *n) { return n ? n->h : 0; }
 
diff --git a/third_party/musl/twalk.c b/third_party/musl/twalk.c
index 48ec2ab19..ed5bdbffc 100644
--- a/third_party/musl/twalk.c
+++ b/third_party/musl/twalk.c
@@ -26,11 +26,7 @@
 │                                                                              │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/musl/tsearch.internal.h"
-
-asm(".ident\t\"\\n\\n\
-Musl libc (MIT License)\\n\
-Copyright 2005-2020 Rich Felker, et. al.\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("musl_libc_notice");
 
 
 static void walk(const struct node *r, void (*action)(const void *, VISIT, int), int d)
diff --git a/third_party/nsync/common.c b/third_party/nsync/common.c
index 93a9cd4db..a50b6f041 100644
--- a/third_party/nsync/common.c
+++ b/third_party/nsync/common.c
@@ -34,11 +34,7 @@
 #include "third_party/nsync/mu_semaphore.h"
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* This package provides a mutex nsync_mu and a Mesa-style condition
  * variable nsync_cv. */
diff --git a/third_party/nsync/mem/nsync_counter.c b/third_party/nsync/mem/nsync_counter.c
index d974977d8..c508797fc 100644
--- a/third_party/nsync/mem/nsync_counter.c
+++ b/third_party/nsync/mem/nsync_counter.c
@@ -26,11 +26,7 @@
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
 #include "third_party/nsync/waiter.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Internal details of nsync_counter. */
 struct nsync_counter_s_ {
diff --git a/third_party/nsync/mem/nsync_cv.c b/third_party/nsync/mem/nsync_cv.c
index fb081094a..926379340 100644
--- a/third_party/nsync/mem/nsync_cv.c
+++ b/third_party/nsync/mem/nsync_cv.c
@@ -26,16 +26,12 @@
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
 #include "third_party/nsync/waiter.h"
+__static_yoink("nsync_notice");
 
 // once we're paying the cost of nsync we might as well get the benefit
 // of a better pthread_once(), since no other component pulls it in now
 __static_yoink("nsync_run_once");
 
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
-
 /* Initialize *cv. */
 void nsync_cv_init (nsync_cv *cv) {
         bzero ((void *) cv, sizeof (*cv));
diff --git a/third_party/nsync/mem/nsync_debug.c b/third_party/nsync/mem/nsync_debug.c
index e677c7ea0..1b72c1175 100644
--- a/third_party/nsync/mem/nsync_debug.c
+++ b/third_party/nsync/mem/nsync_debug.c
@@ -21,11 +21,7 @@
 #include "third_party/nsync/mu_semaphore.h"
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Routines for debugging. */
 
diff --git a/third_party/nsync/mem/nsync_mu_wait.c b/third_party/nsync/mem/nsync_mu_wait.c
index c7839950a..3c48ea599 100644
--- a/third_party/nsync/mem/nsync_mu_wait.c
+++ b/third_party/nsync/mem/nsync_mu_wait.c
@@ -22,11 +22,7 @@
 #include "third_party/nsync/mu_semaphore.h"
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Attempt to remove waiter *w from *mu's
    waiter queue.  If successful, leave the lock held in mode *l_type, and
diff --git a/third_party/nsync/mem/nsync_note.c b/third_party/nsync/mem/nsync_note.c
index 8e7276112..bdf8e9ad0 100644
--- a/third_party/nsync/mem/nsync_note.c
+++ b/third_party/nsync/mem/nsync_note.c
@@ -25,11 +25,7 @@
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
 #include "third_party/nsync/waiter.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Locking discipline for the nsync_note implementation:
 
diff --git a/third_party/nsync/mem/nsync_once.c b/third_party/nsync/mem/nsync_once.c
index 8320a6266..163923359 100644
--- a/third_party/nsync/mem/nsync_once.c
+++ b/third_party/nsync/mem/nsync_once.c
@@ -22,11 +22,7 @@
 #include "third_party/nsync/once.h"
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* An once_sync_s struct contains a lock, and a condition variable on which
    threads may wait for an nsync_once to be initialized by another thread.
diff --git a/third_party/nsync/mem/nsync_sem_wait.c b/third_party/nsync/mem/nsync_sem_wait.c
index c32a2842e..62507d686 100644
--- a/third_party/nsync/mem/nsync_sem_wait.c
+++ b/third_party/nsync/mem/nsync_sem_wait.c
@@ -23,11 +23,7 @@
 #include "third_party/nsync/common.internal.h"
 #include "third_party/nsync/mu_semaphore.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Wait until one of:
      w->sem is non-zero----decrement it and return 0.
diff --git a/third_party/nsync/mem/nsync_wait.c b/third_party/nsync/mem/nsync_wait.c
index f669730ac..9d8e95b7d 100644
--- a/third_party/nsync/mem/nsync_wait.c
+++ b/third_party/nsync/mem/nsync_wait.c
@@ -25,11 +25,7 @@
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
 #include "third_party/nsync/waiter.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 int nsync_wait_n (void *mu, void (*lock) (void *), void (*unlock) (void *),
 		  nsync_time abs_deadline,
diff --git a/third_party/nsync/mu.c b/third_party/nsync/mu.c
index 354c38621..4cae68328 100644
--- a/third_party/nsync/mu.c
+++ b/third_party/nsync/mu.c
@@ -23,11 +23,7 @@
 #include "third_party/nsync/mu_semaphore.h"
 #include "third_party/nsync/races.internal.h"
 #include "third_party/nsync/wait_s.internal.h"
-
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
+__static_yoink("nsync_notice");
 
 /* Initialize *mu. */
 void nsync_mu_init (nsync_mu *mu) {
diff --git a/third_party/nsync/mu_semaphore.c b/third_party/nsync/mu_semaphore.c
index 43f2b5ccc..493efa2c4 100644
--- a/third_party/nsync/mu_semaphore.c
+++ b/third_party/nsync/mu_semaphore.c
@@ -19,17 +19,13 @@
 #include "libc/calls/cp.internal.h"
 #include "libc/dce.h"
 #include "third_party/nsync/mu_semaphore.internal.h"
+__static_yoink("nsync_notice");
 
 /* Apple's ulock (part by Cosmo futexes) is an internal API, but:
    1. Unlike GCD it's cancellable, i.e. can be EINTR'd by signals
    2. We currently always use ulock anyway for joining threads */
 #define PREFER_GCD_OVER_ULOCK 1
 
-asm(".ident\t\"\\n\\n\
-*NSYNC (Apache 2.0)\\n\
-Copyright 2016 Google, Inc.\\n\
-https://github.com/google/nsync\"");
-
 /* Initialize *s; the initial value is 0. */
 void nsync_mu_semaphore_init (nsync_semaphore *s) {
 	if (PREFER_GCD_OVER_ULOCK && IsXnuSilicon ()) {
diff --git a/third_party/nsync/notice.c b/third_party/nsync/notice.c
new file mode 100644
index 000000000..614e3f62d
--- /dev/null
+++ b/third_party/nsync/notice.c
@@ -0,0 +1,4 @@
+__notice(nsync_notice, "\
+*NSYNC (Apache 2.0)\n\
+Copyright 2016 Google, Inc.\n\
+https://github.com/google/nsync");
diff --git a/third_party/pcre/pcre2_compile.c b/third_party/pcre/pcre2_compile.c
index fb910ed8e..4fc63cd57 100644
--- a/third_party/pcre/pcre2_compile.c
+++ b/third_party/pcre/pcre2_compile.c
@@ -1,9 +1,5 @@
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-PCRE2 (PCRE2 License)\\n\
-Copyright (c) 1997-2022 University of Cambridge\"");
-
 /*************************************************
 *      Perl-Compatible Regular Expressions       *
 *************************************************/
diff --git a/third_party/pcre/pcre2_match.c b/third_party/pcre/pcre2_match.c
index 606a850c3..897c88b29 100644
--- a/third_party/pcre/pcre2_match.c
+++ b/third_party/pcre/pcre2_match.c
@@ -1,8 +1,8 @@
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-PCRE2 (PCRE2 License)\\n\
-Copyright (c) 1997-2022 University of Cambridge\"");
+__notice(pcre2_notice, "\
+PCRE2 (PCRE2 License)\n\
+Copyright (c) 1997-2022 University of Cambridge");
 
 /*************************************************
 *      Perl-Compatible Regular Expressions       *
diff --git a/third_party/puff/puff.c b/third_party/puff/puff.c
index 94593db28..21bc92ec4 100644
--- a/third_party/puff/puff.c
+++ b/third_party/puff/puff.c
@@ -25,10 +25,9 @@
 #include "third_party/puff/puff.h"
 #include "libc/runtime/runtime.h"
 
-asm(".ident\t\"\\n\\n\
-puff (zlib License)\\n\
-Copyright 2002-203 Mark Adler\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(puff_notice, "\
+puff (zlib License)\n\
+Copyright 2002-2013 Mark Adler");
 
 // Origin: git@github.com:madler/zlib.git
 // Commit: 03614c56ad299f9b238c75aa1e66f0c08fc4fc8b
diff --git a/third_party/python/BUILD.mk b/third_party/python/BUILD.mk
index 33ef97a47..81868c829 100644
--- a/third_party/python/BUILD.mk
+++ b/third_party/python/BUILD.mk
@@ -534,6 +534,7 @@ THIRD_PARTY_PYTHON_STAGE2_A_SRCS =					\
 	third_party/python/Modules/_csv.c				\
 	third_party/python/Modules/_datetimemodule.c			\
 	third_party/python/Modules/_decimal/_decimal.c			\
+	third_party/python/Modules/_decimal/libmpdec/notice.c		\
 	third_party/python/Modules/_decimal/libmpdec/basearith.c	\
 	third_party/python/Modules/_decimal/libmpdec/constants.c	\
 	third_party/python/Modules/_decimal/libmpdec/context.c		\
diff --git a/third_party/python/Modules/_decimal/_decimal.c b/third_party/python/Modules/_decimal/_decimal.c
index 0c0903a70..3940663c9 100644
--- a/third_party/python/Modules/_decimal/_decimal.c
+++ b/third_party/python/Modules/_decimal/_decimal.c
@@ -96,11 +96,6 @@ PYTHON_PROVIDE("_decimal.setcontext");
 PYTHON_YOINK("numbers");
 PYTHON_YOINK("collections");
 
-asm(".ident\t\"\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #if !defined(MPD_VERSION_HEX) || MPD_VERSION_HEX < 0x02040100
   #error "libmpdec version >= 2.4.1 required"
 #endif
diff --git a/third_party/python/Modules/_decimal/libmpdec/basearith.c b/third_party/python/Modules/_decimal/libmpdec/basearith.c
index 893268e06..94c5835f6 100644
--- a/third_party/python/Modules/_decimal/libmpdec/basearith.c
+++ b/third_party/python/Modules/_decimal/libmpdec/basearith.c
@@ -31,11 +31,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/constants.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/typearith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /*********************************************************************/
 /*                   Calculations in base MPD_RADIX                  */
diff --git a/third_party/python/Modules/_decimal/libmpdec/constants.c b/third_party/python/Modules/_decimal/libmpdec/constants.c
index 587482f19..7f85e4f2a 100644
--- a/third_party/python/Modules/_decimal/libmpdec/constants.c
+++ b/third_party/python/Modules/_decimal/libmpdec/constants.c
@@ -29,11 +29,7 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/python/Modules/_decimal/libmpdec/constants.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* number-theory.c */
 const mpd_uint_t mpd_moduli[3] = {
diff --git a/third_party/python/Modules/_decimal/libmpdec/context.c b/third_party/python/Modules/_decimal/libmpdec/context.c
index fff1b2dae..d57742f03 100644
--- a/third_party/python/Modules/_decimal/libmpdec/context.c
+++ b/third_party/python/Modules/_decimal/libmpdec/context.c
@@ -30,11 +30,7 @@
 #include "libc/calls/calls.h"
 #include "libc/sysv/consts/sig.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 void
 mpd_dflt_traphandler(mpd_context_t *ctx)
diff --git a/third_party/python/Modules/_decimal/libmpdec/convolute.c b/third_party/python/Modules/_decimal/libmpdec/convolute.c
index 25b85d128..89cd5db13 100644
--- a/third_party/python/Modules/_decimal/libmpdec/convolute.c
+++ b/third_party/python/Modules/_decimal/libmpdec/convolute.c
@@ -36,11 +36,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/numbertheory.h"
 #include "third_party/python/Modules/_decimal/libmpdec/sixstep.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 
 /* Bignum: Fast convolution using the Number Theoretic Transform.
diff --git a/third_party/python/Modules/_decimal/libmpdec/crt.c b/third_party/python/Modules/_decimal/libmpdec/crt.c
index e20acadc5..70db6f24f 100644
--- a/third_party/python/Modules/_decimal/libmpdec/crt.c
+++ b/third_party/python/Modules/_decimal/libmpdec/crt.c
@@ -31,11 +31,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/numbertheory.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* Bignum: Chinese Remainder Theorem, extends the maximum transform length. */
 
diff --git a/third_party/python/Modules/_decimal/libmpdec/difradix2.c b/third_party/python/Modules/_decimal/libmpdec/difradix2.c
index ac59f54a6..c5f7b4fae 100644
--- a/third_party/python/Modules/_decimal/libmpdec/difradix2.c
+++ b/third_party/python/Modules/_decimal/libmpdec/difradix2.c
@@ -32,11 +32,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/numbertheory.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* Bignum: The actual transform routine (decimation in frequency). */
 
diff --git a/third_party/python/Modules/_decimal/libmpdec/fnt.c b/third_party/python/Modules/_decimal/libmpdec/fnt.c
index 4c9b6147e..cad1f4abf 100644
--- a/third_party/python/Modules/_decimal/libmpdec/fnt.c
+++ b/third_party/python/Modules/_decimal/libmpdec/fnt.c
@@ -33,11 +33,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/fnt.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/numbertheory.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* Bignum: Fast transform for medium-sized coefficients. */
 
diff --git a/third_party/python/Modules/_decimal/libmpdec/fourstep.c b/third_party/python/Modules/_decimal/libmpdec/fourstep.c
index 60912059c..36d3bb815 100644
--- a/third_party/python/Modules/_decimal/libmpdec/fourstep.c
+++ b/third_party/python/Modules/_decimal/libmpdec/fourstep.c
@@ -33,11 +33,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/sixstep.h"
 #include "third_party/python/Modules/_decimal/libmpdec/transpose.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /*
                  Cache Efficient Matrix Fourier Transform
diff --git a/third_party/python/Modules/_decimal/libmpdec/io.c b/third_party/python/Modules/_decimal/libmpdec/io.c
index bb8a92fc6..00a7af879 100644
--- a/third_party/python/Modules/_decimal/libmpdec/io.c
+++ b/third_party/python/Modules/_decimal/libmpdec/io.c
@@ -35,16 +35,12 @@
 #include "third_party/python/Modules/_decimal/libmpdec/io.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/typearith.h"
+__static_yoink("libmpdec_notice");
 
 #if __GNUC__ >= 11
 #pragma GCC diagnostic ignored "-Wmisleading-indentation"
 #endif
 
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 
 /* This file contains functions for decimal <-> string conversions, including
    PEP-3101 formatting for numeric types. */
diff --git a/third_party/python/Modules/_decimal/libmpdec/memory.c b/third_party/python/Modules/_decimal/libmpdec/memory.c
index e194159fb..98914f293 100644
--- a/third_party/python/Modules/_decimal/libmpdec/memory.c
+++ b/third_party/python/Modules/_decimal/libmpdec/memory.c
@@ -31,11 +31,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/mpalloc.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/typearith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* Guaranteed minimum allocation for a coefficient. May be changed once
    at program start using mpd_setminalloc(). */
diff --git a/third_party/python/Modules/_decimal/libmpdec/mpdecimal.c b/third_party/python/Modules/_decimal/libmpdec/mpdecimal.c
index e1a818f36..e93eee8a2 100644
--- a/third_party/python/Modules/_decimal/libmpdec/mpdecimal.c
+++ b/third_party/python/Modules/_decimal/libmpdec/mpdecimal.c
@@ -36,11 +36,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/mpalloc.h"
 #include "third_party/python/Modules/_decimal/libmpdec/typearith.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 #define MPD_NEWTONDIV_CUTOFF 1024L
 
diff --git a/third_party/python/Modules/_decimal/libmpdec/notice.c b/third_party/python/Modules/_decimal/libmpdec/notice.c
new file mode 100644
index 000000000..fd4f0fbe1
--- /dev/null
+++ b/third_party/python/Modules/_decimal/libmpdec/notice.c
@@ -0,0 +1,3 @@
+__notice(libmpdec_notice, "\
+libmpdec (BSD-2)\n\
+Copyright 2008-2016 Stefan Krah");
diff --git a/third_party/python/Modules/_decimal/libmpdec/numbertheory.c b/third_party/python/Modules/_decimal/libmpdec/numbertheory.c
index 1551a5968..604d29f5c 100644
--- a/third_party/python/Modules/_decimal/libmpdec/numbertheory.c
+++ b/third_party/python/Modules/_decimal/libmpdec/numbertheory.c
@@ -31,11 +31,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/numbertheory.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /* Bignum: Initialize the Number Theoretic Transform. */
 
diff --git a/third_party/python/Modules/_decimal/libmpdec/sixstep.c b/third_party/python/Modules/_decimal/libmpdec/sixstep.c
index 06e8ba071..16d3e3f28 100644
--- a/third_party/python/Modules/_decimal/libmpdec/sixstep.c
+++ b/third_party/python/Modules/_decimal/libmpdec/sixstep.c
@@ -34,11 +34,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/sixstep.h"
 #include "third_party/python/Modules/_decimal/libmpdec/transpose.h"
 #include "third_party/python/Modules/_decimal/libmpdec/umodarith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 /*
                 Cache Efficient Matrix Fourier Transform
diff --git a/third_party/python/Modules/_decimal/libmpdec/transpose.c b/third_party/python/Modules/_decimal/libmpdec/transpose.c
index 4419b6063..ceb3d9f63 100644
--- a/third_party/python/Modules/_decimal/libmpdec/transpose.c
+++ b/third_party/python/Modules/_decimal/libmpdec/transpose.c
@@ -34,11 +34,7 @@
 #include "third_party/python/Modules/_decimal/libmpdec/constants.h"
 #include "third_party/python/Modules/_decimal/libmpdec/mpdecimal.h"
 #include "third_party/python/Modules/_decimal/libmpdec/typearith.h"
-
-asm(".ident\t\"\\n\\n\
-libmpdec (BSD-2)\\n\
-Copyright 2008-2016 Stefan Krah\"");
-asm(".include \"libc/disclaimer.inc\"");
+__static_yoink("libmpdec_notice");
 
 #define BUFSIZE 4096
 #define SIDE 128
diff --git a/third_party/python/Modules/_randommodule.c b/third_party/python/Modules/_randommodule.c
index 51fea4586..bafdf9fae 100644
--- a/third_party/python/Modules/_randommodule.c
+++ b/third_party/python/Modules/_randommodule.c
@@ -25,10 +25,8 @@
 PYTHON_PROVIDE("_random");
 PYTHON_PROVIDE("_random.Random");
 
-asm(".ident\t\"\\n\\n\
-mt19937 (BSD-3)\\n\
-Copyright 1997-2004 Makoto Matsumoto and Takuji Nishimura\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(python_mt19937_notice, "Python mt19937 (BSD-3)\n\
+Copyright 1997-2004 Makoto Matsumoto and Takuji Nishimura");
 
 /* ------------------------------------------------------------------
    The code in this module was based on a download from:
diff --git a/third_party/python/Modules/_sqlite/cache.c b/third_party/python/Modules/_sqlite/cache.c
index 3076e68f8..f88969fde 100644
--- a/third_party/python/Modules/_sqlite/cache.c
+++ b/third_party/python/Modules/_sqlite/cache.c
@@ -25,11 +25,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/python/Modules/_sqlite/cache.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /* only used internally */
 pysqlite_Node* pysqlite_new_node(PyObject* key, PyObject* data)
 {
diff --git a/third_party/python/Modules/_sqlite/connection.c b/third_party/python/Modules/_sqlite/connection.c
index b1c3bb638..38651adbb 100644
--- a/third_party/python/Modules/_sqlite/connection.c
+++ b/third_party/python/Modules/_sqlite/connection.c
@@ -36,11 +36,6 @@
 
 PYTHON_YOINK("sqlite3.dump");
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #define ACTION_FINALIZE 1
 #define ACTION_RESET 2
 
diff --git a/third_party/python/Modules/_sqlite/cursor.c b/third_party/python/Modules/_sqlite/cursor.c
index cc2c5e8f3..c4fc23f91 100644
--- a/third_party/python/Modules/_sqlite/cursor.c
+++ b/third_party/python/Modules/_sqlite/cursor.c
@@ -27,11 +27,6 @@
 #include "third_party/python/Modules/_sqlite/module.h"
 #include "third_party/python/Modules/_sqlite/util.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 PyObject* pysqlite_cursor_iternext(pysqlite_Cursor* self);
 
 static const char errmsg_fetch_across_rollback[] = "Cursor needed to be reset because of commit/rollback and can no longer be fetched from.";
diff --git a/third_party/python/Modules/_sqlite/microprotocols.c b/third_party/python/Modules/_sqlite/microprotocols.c
index 168b931be..eecada5b3 100644
--- a/third_party/python/Modules/_sqlite/microprotocols.c
+++ b/third_party/python/Modules/_sqlite/microprotocols.c
@@ -27,11 +27,6 @@
 #include "third_party/python/Modules/_sqlite/microprotocols.h"
 #include "third_party/python/Modules/_sqlite/prepare_protocol.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /** the adapters registry **/
 
 PyObject *psyco_adapters;
diff --git a/third_party/python/Modules/_sqlite/module.c b/third_party/python/Modules/_sqlite/module.c
index bf031a97f..63926be5c 100644
--- a/third_party/python/Modules/_sqlite/module.c
+++ b/third_party/python/Modules/_sqlite/module.c
@@ -32,12 +32,11 @@
 #include "third_party/python/Modules/_sqlite/row.h"
 #include "third_party/python/Modules/_sqlite/statement.h"
 
-PYTHON_PROVIDE("_sqlite3");
+__notice(pysqlite_notice, "\
+pysqlite (zlib license)\n\
+Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>");
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
+PYTHON_PROVIDE("_sqlite3");
 
 /* #if SQLITE_VERSION_NUMBER >= 3003003 */
 /* #define HAVE_SHARED_CACHE */
diff --git a/third_party/python/Modules/_sqlite/prepare_protocol.c b/third_party/python/Modules/_sqlite/prepare_protocol.c
index eafb29baf..5b110e9fd 100644
--- a/third_party/python/Modules/_sqlite/prepare_protocol.c
+++ b/third_party/python/Modules/_sqlite/prepare_protocol.c
@@ -25,11 +25,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/python/Modules/_sqlite/prepare_protocol.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 int pysqlite_prepare_protocol_init(pysqlite_PrepareProtocol* self, PyObject* args, PyObject* kwargs)
 {
     return 0;
diff --git a/third_party/python/Modules/_sqlite/row.c b/third_party/python/Modules/_sqlite/row.c
index b23f1afc3..be06b9213 100644
--- a/third_party/python/Modules/_sqlite/row.c
+++ b/third_party/python/Modules/_sqlite/row.c
@@ -26,11 +26,6 @@
 #include "third_party/python/Modules/_sqlite/cursor.h"
 #include "third_party/python/Modules/_sqlite/row.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 void pysqlite_row_dealloc(pysqlite_Row* self)
 {
     Py_XDECREF(self->data);
diff --git a/third_party/python/Modules/_sqlite/statement.c b/third_party/python/Modules/_sqlite/statement.c
index 066a6e870..381d97d63 100644
--- a/third_party/python/Modules/_sqlite/statement.c
+++ b/third_party/python/Modules/_sqlite/statement.c
@@ -30,11 +30,6 @@
 #include "third_party/python/Modules/_sqlite/statement.h"
 #include "third_party/python/Modules/_sqlite/util.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /* prototypes */
 static int pysqlite_check_remaining_sql(const char* tail);
 
diff --git a/third_party/python/Modules/_sqlite/util.c b/third_party/python/Modules/_sqlite/util.c
index 7eae8f8a3..ab53adb08 100644
--- a/third_party/python/Modules/_sqlite/util.c
+++ b/third_party/python/Modules/_sqlite/util.c
@@ -26,11 +26,6 @@
 #include "third_party/python/Modules/_sqlite/connection.h"
 #include "third_party/python/Modules/_sqlite/module.h"
 
-asm(".ident\t\"\\n\\n\
-pysqlite (zlib license)\\n\
-Copyright (C) 2005-2010 Gerhard Häring <gh@ghaering.de>\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 int pysqlite_step(sqlite3_stmt* statement, pysqlite_Connection* connection)
 {
     int rc;
diff --git a/third_party/python/Modules/_sre.c b/third_party/python/Modules/_sre.c
index da31f9cdd..a786dbdb4 100644
--- a/third_party/python/Modules/_sre.c
+++ b/third_party/python/Modules/_sre.c
@@ -39,10 +39,9 @@ PYTHON_PROVIDE("_sre.compile");
 PYTHON_PROVIDE("_sre.getcodesize");
 PYTHON_PROVIDE("_sre.getlower");
 
-asm(".ident\t\"\\n\\n\
-SRE 2.2.2 (Python license)\\n\
-Copyright 1997-2002 Secret Labs AB\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(python_sre_notice, "\
+SRE 2.2.2 (Python license)\n\
+Copyright 1997-2002 Secret Labs AB");
 
 /*
  * Secret Labs' Regular Expression Engine
diff --git a/third_party/python/Modules/expat/xmlparse.c b/third_party/python/Modules/expat/xmlparse.c
index 6bfce3613..a6ba69a5b 100644
--- a/third_party/python/Modules/expat/xmlparse.c
+++ b/third_party/python/Modules/expat/xmlparse.c
@@ -46,11 +46,9 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
-asm(".ident\t\"\\n\\n\
-expat (MIT License)\\n\
-Copyright (c) 1997-2000 Thai Open Source Software Center Ltd\\n\
-Copyright (c) 2000-2017 Expat development team\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(expat_notice, "expat (MIT License)\n\
+Copyright (c) 1997-2000 Thai Open Source Software Center Ltd\n\
+Copyright (c) 2000-2017 Expat development team");
 
 #define XML_BUILDING_EXPAT 1
 
diff --git a/third_party/python/Modules/expat/xmlrole.c b/third_party/python/Modules/expat/xmlrole.c
index 6ce8885db..401329321 100644
--- a/third_party/python/Modules/expat/xmlrole.c
+++ b/third_party/python/Modules/expat/xmlrole.c
@@ -30,12 +30,6 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
-asm(".ident\t\"\\n\\n\
-expat (MIT License)\\n\
-Copyright (c) 1997-2000 Thai Open Source Software Center Ltd\\n\
-Copyright (c) 2000-2017 Expat development team\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #include "third_party/python/Modules/expat/expat_config.h"
 #include "third_party/python/Modules/expat/expat_external.h"
 #include "third_party/python/Modules/expat/internal.inc"
diff --git a/third_party/python/Modules/expat/xmltok.c b/third_party/python/Modules/expat/xmltok.c
index 149b12e10..6f1e6df95 100644
--- a/third_party/python/Modules/expat/xmltok.c
+++ b/third_party/python/Modules/expat/xmltok.c
@@ -30,12 +30,6 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
-asm(".ident\t\"\\n\\n\
-expat (MIT License)\\n\
-Copyright (c) 1997-2000 Thai Open Source Software Center Ltd\\n\
-Copyright (c) 2000-2017 Expat development team\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 #include "third_party/python/Modules/expat/expat_config.h"
 #include "third_party/python/Modules/expat/expat_external.h"
 #include "third_party/python/Modules/expat/internal.inc"
diff --git a/third_party/python/Modules/hashtable.c b/third_party/python/Modules/hashtable.c
index fbed762c3..335ed3b17 100644
--- a/third_party/python/Modules/hashtable.c
+++ b/third_party/python/Modules/hashtable.c
@@ -8,10 +8,9 @@
 #include "third_party/python/Include/pymem.h"
 #include "third_party/python/Modules/hashtable.h"
 
-asm(".ident\t\"\\n\\n\
-cfuhash (bsd-3)\\n\
-Copyright (c) 2005 Don Owens\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(cfuhash_notice, "\
+cfuhash (bsd-3)\n\
+Copyright (c) 2005 Don Owens");
 
 /* The implementation of the hash table (_Py_hashtable_t) is based on the
    cfuhash project:
diff --git a/third_party/python/Modules/parsermodule.c b/third_party/python/Modules/parsermodule.c
index 001054863..cd9c7fa22 100644
--- a/third_party/python/Modules/parsermodule.c
+++ b/third_party/python/Modules/parsermodule.c
@@ -40,12 +40,13 @@ PYTHON_PROVIDE("parser.st2tuple");
 PYTHON_PROVIDE("parser.suite");
 PYTHON_PROVIDE("parser.tuple2st");
 
-asm(".ident\t\"\\n\\n\
-parsermodule (Python license)\\n\
-Copyright 1995-1996 by Virginia Polytechnic Institute & State\\n\
-University, Blacksburg, Virginia, USA, and Fred L. Drake, Jr., Reston,\\n\
-Virginia, USA.  Portions copyright 1991-1995 by Stichting Mathematisch\\n\
-Centrum, Amsterdam, The Netherlands.\"");
+__notice(parsermodule_notice, "\
+parsermodule (Python license)\n\
+Copyright 1995-1996 by Virginia Polytechnic Institute & State\n\
+University, Blacksburg, Virginia, USA, and Fred L. Drake, Jr., Reston,\n\
+Virginia, USA.  Portions copyright 1991-1995 by Stichting Mathematisch\n\
+Centrum, Amsterdam, The Netherlands.");
+
 static const char parser_copyright_string[] =
 "Copyright 1995-1996 by Virginia Polytechnic Institute & State\n\
 University, Blacksburg, Virginia, USA, and Fred L. Drake, Jr., Reston,\n\
diff --git a/third_party/python/Modules/syslogmodule.c b/third_party/python/Modules/syslogmodule.c
index 6d2e4a916..31c66ba9b 100644
--- a/third_party/python/Modules/syslogmodule.c
+++ b/third_party/python/Modules/syslogmodule.c
@@ -59,11 +59,10 @@ PYTHON_PROVIDE("syslog.openlog");
 PYTHON_PROVIDE("syslog.setlogmask");
 PYTHON_PROVIDE("syslog.syslog");
 
-asm(".ident\t\"\\n\\n\
-syslogmodule (mit)\\n\
-Copyright 1994 by Lance Ellinghouse\\n\
-Cathedral City, California Republic, United States of America\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(python_syslogmodule_notice, "\
+syslogmodule (MIT License)\n\
+Copyright 1994 by Lance Ellinghouse\n\
+Cathedral City, California Republic, United States of America");
 
 /***********************************************************
 Copyright 1994 by Lance Ellinghouse,
diff --git a/third_party/python/Modules/unicodedata_3.2.0.c b/third_party/python/Modules/unicodedata_3.2.0.c
index 49e4df3a5..b2d967577 100644
--- a/third_party/python/Modules/unicodedata_3.2.0.c
+++ b/third_party/python/Modules/unicodedata_3.2.0.c
@@ -432,12 +432,10 @@ static const unsigned char _PyUnicode_ChangeIndex_3_2_0_rodata[365+1][2] = { /*
     {  2, 0x02},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_ChangeIndex_3_2_0_init(void) {
     rldecode2(_PyUnicode_ChangeIndex_3_2_0, (void *)_PyUnicode_ChangeIndex_3_2_0_rodata);
 }
-const void *const _PyUnicode_ChangeIndex_3_2_0_ctor[] initarray = {
-    _PyUnicode_ChangeIndex_3_2_0_init,
-};
 
 unsigned char _PyUnicode_ChangeData_3_2_0[31360];
 static const unsigned char _PyUnicode_ChangeData_3_2_0_rodata[1390+1][2] = { /* 8.8648% profit */
@@ -1833,12 +1831,10 @@ static const unsigned char _PyUnicode_ChangeData_3_2_0_rodata[1390+1][2] = { /*
     { 16, 0x00},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_ChangeData_3_2_0_init(void) {
     rldecode2(_PyUnicode_ChangeData_3_2_0, (void *)_PyUnicode_ChangeData_3_2_0_rodata);
 }
-const void *const _PyUnicode_ChangeData_3_2_0_ctor[] initarray = {
-    _PyUnicode_ChangeData_3_2_0_init,
-};
 
 const _PyUnicode_ChangeRecord *_PyUnicode_GetChange_3_2_0(Py_UCS4 n)
 {
diff --git a/third_party/python/Modules/unicodedata_comp.c b/third_party/python/Modules/unicodedata_comp.c
index 9b2c142fe..95706fd0c 100644
--- a/third_party/python/Modules/unicodedata_comp.c
+++ b/third_party/python/Modules/unicodedata_comp.c
@@ -1152,6 +1152,7 @@ static const unsigned short _PyUnicode_CompIndex_rodata[1146+1][2] = { /* 38.598
     {  1, 0x2a9},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_CompIndex_init(void) {
     int i, j, k;
     for (k = i = 0; i < 1146; ++i) {
@@ -1160,9 +1161,6 @@ static textstartup void _PyUnicode_CompIndex_init(void) {
         }
     }
 }
-const void *const _PyUnicode_CompIndex_ctor[] initarray = {
-    _PyUnicode_CompIndex_init,
-};
 
 const unsigned int _PyUnicode_CompData[1449] = {
     0, 0, 0, 922746880, 17, 9011200, 3758096384, 1101, 0, 50594176, 
diff --git a/third_party/python/Modules/unicodedata_decomp.c b/third_party/python/Modules/unicodedata_decomp.c
index ca8b27add..8fbcf188f 100644
--- a/third_party/python/Modules/unicodedata_decomp.c
+++ b/third_party/python/Modules/unicodedata_decomp.c
@@ -1446,12 +1446,10 @@ static const unsigned char _PyUnicode_DecompIndex1_rodata[159+1][2] = { /* 3.653
     { 39, 0x00},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_DecompIndex1_init(void) {
     rldecode2(_PyUnicode_DecompIndex1, (void *)_PyUnicode_DecompIndex1_rodata);
 }
-const void *const _PyUnicode_DecompIndex1_ctor[] initarray = {
-    _PyUnicode_DecompIndex1_init,
-};
 
 const unsigned short _PyUnicode_DecompIndex2[13312] = {
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
diff --git a/third_party/python/Modules/unicodedata_phrasebook.c b/third_party/python/Modules/unicodedata_phrasebook.c
index dfa0630fc..702ef6089 100644
--- a/third_party/python/Modules/unicodedata_phrasebook.c
+++ b/third_party/python/Modules/unicodedata_phrasebook.c
@@ -11647,12 +11647,10 @@ static const unsigned char _PyUnicode_PhrasebookOffset1_rodata[206+1][2] = { /*
     {253, 0x34},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_PhrasebookOffset1_init(void) {
     rldecode2(_PyUnicode_PhrasebookOffset1, (void *)_PyUnicode_PhrasebookOffset1_rodata);
 }
-const void *const _PyUnicode_PhrasebookOffset1_ctor[] initarray = {
-    _PyUnicode_PhrasebookOffset1_init,
-};
 
 const unsigned int _PyUnicode_PhrasebookOffset2[25056] = {
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 786433, 37748832, 
diff --git a/third_party/python/Modules/unicodedata_records.c b/third_party/python/Modules/unicodedata_records.c
index 9c7b8ba51..7c08e5974 100644
--- a/third_party/python/Modules/unicodedata_records.c
+++ b/third_party/python/Modules/unicodedata_records.c
@@ -674,6 +674,7 @@ static const unsigned short _PyUnicode_RecordsIndex1_rodata[323+1][2] = { /* 7.4
     {  1, 0x10a},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_RecordsIndex1_init(void) {
     int i, j, k;
     for (k = i = 0; i < 323; ++i) {
@@ -682,9 +683,6 @@ static textstartup void _PyUnicode_RecordsIndex1_init(void) {
         }
     }
 }
-const void *const _PyUnicode_RecordsIndex1_ctor[] initarray = {
-    _PyUnicode_RecordsIndex1_init,
-};
 
 unsigned short _PyUnicode_RecordsIndex2[34176];
 static const unsigned short _PyUnicode_RecordsIndex2_rodata[4990+1][2] = { /* 29.2018% profit */
@@ -5680,6 +5678,7 @@ static const unsigned short _PyUnicode_RecordsIndex2_rodata[4990+1][2] = { /* 29
     {  2, 0x00},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_RecordsIndex2_init(void) {
     int i, j, k;
     for (k = i = 0; i < 4990; ++i) {
@@ -5688,7 +5687,3 @@ static textstartup void _PyUnicode_RecordsIndex2_init(void) {
         }
     }
 }
-const void *const _PyUnicode_RecordsIndex2_ctor[] initarray = {
-    _PyUnicode_RecordsIndex2_init,
-};
-
diff --git a/third_party/python/Modules/unicodedata_typerecords.c b/third_party/python/Modules/unicodedata_typerecords.c
index afa05895c..864ba3b4c 100644
--- a/third_party/python/Modules/unicodedata_typerecords.c
+++ b/third_party/python/Modules/unicodedata_typerecords.c
@@ -881,6 +881,7 @@ static const unsigned short _PyUnicode_TypeRecordsIndex1_rodata[370+1][2] = { /*
     {  1, 0x112},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_TypeRecordsIndex1_init(void) {
     int i, j, k;
     for (k = i = 0; i < 370; ++i) {
@@ -889,9 +890,6 @@ static textstartup void _PyUnicode_TypeRecordsIndex1_init(void) {
         }
     }
 }
-const void *const _PyUnicode_TypeRecordsIndex1_ctor[] initarray = {
-    _PyUnicode_TypeRecordsIndex1_init,
-};
 
 unsigned short _PyUnicode_TypeRecordsIndex2[35200];
 static const unsigned short _PyUnicode_TypeRecordsIndex2_rodata[4889+1][2] = { /* 27.7784% profit */
@@ -5786,6 +5784,7 @@ static const unsigned short _PyUnicode_TypeRecordsIndex2_rodata[4889+1][2] = { /
     {  2, 0x00},
     {0},
 };
+__attribute__((__constructor__(99)))
 static textstartup void _PyUnicode_TypeRecordsIndex2_init(void) {
     int i, j, k;
     for (k = i = 0; i < 4889; ++i) {
@@ -5794,7 +5793,3 @@ static textstartup void _PyUnicode_TypeRecordsIndex2_init(void) {
         }
     }
 }
-const void *const _PyUnicode_TypeRecordsIndex2_ctor[] initarray = {
-    _PyUnicode_TypeRecordsIndex2_init,
-};
-
diff --git a/third_party/python/Python/getcopyright.c b/third_party/python/Python/getcopyright.c
index 9ec5beadc..233148714 100644
--- a/third_party/python/Python/getcopyright.c
+++ b/third_party/python/Python/getcopyright.c
@@ -10,16 +10,16 @@
 #include "libc/str/str.h"
 #include "third_party/python/Include/pylifecycle.h"
 
-asm(".ident\t\"\\n\\n\
-Python 3.6 (https://docs.python.org/3/license.html)\\n\
-Copyright (c) 2001-2021 Python Software Foundation.\\n\
-All Rights Reserved.\\n\
-Copyright (c) 2000 BeOpen.com.\\n\
-All Rights Reserved.\\n\
-Copyright (c) 1995-2001 Corporation for National Research Initiatives.\\n\
-All Rights Reserved.\\n\
-Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.\\n\
-All Rights Reserved.\"");
+__notice(python_notice, "\
+Python 3.6 (https://docs.python.org/3/license.html)\n\
+Copyright (c) 2001-2021 Python Software Foundation.\n\
+All Rights Reserved.\n\
+Copyright (c) 2000 BeOpen.com.\n\
+All Rights Reserved.\n\
+Copyright (c) 1995-2001 Corporation for National Research Initiatives.\n\
+All Rights Reserved.\n\
+Copyright (c) 1991-1995 Stichting Mathematisch Centrum, Amsterdam.\n\
+All Rights Reserved.");
 
 const char *
 Py_GetCopyright(void)
@@ -29,7 +29,7 @@ Py_GetCopyright(void)
         char *r = 0;
         const char *p;
         appends(&r, "");
-        for (p = __comment_start; *p; p += strlen(p) + 1) {
+        for (p = __notices; *p; p += strlen(p) + 1) {
             appends(&r, p);
         }
         res = r;
diff --git a/third_party/python/Python/getopt.c b/third_party/python/Python/getopt.c
index c54c7aa68..24aa6b49b 100644
--- a/third_party/python/Python/getopt.c
+++ b/third_party/python/Python/getopt.c
@@ -8,9 +8,9 @@
 #include "libc/str/str.h"
 #include "third_party/python/Include/pygetopt.h"
 
-asm(".ident\t\"\\n\\n\
-python getopt (isc license)\\n\
-Copyright 1992-1994 David Gottner\"");
+__notice(python_getopt_notice, "\
+python getopt (isc license)\n\
+Copyright 1992-1994 David Gottner");
 
 /*---------------------------------------------------------------------------*
  * <RCS keywords>
diff --git a/third_party/python/Python/sysmodule.c b/third_party/python/Python/sysmodule.c
index 9cf5f5cf1..8b0c6edc7 100644
--- a/third_party/python/Python/sysmodule.c
+++ b/third_party/python/Python/sysmodule.c
@@ -2013,6 +2013,7 @@ _PySys_Init(void)
                          PyBool_FromLong(Py_DontWriteBytecodeFlag));
     SET_SYS_FROM_STRING("api_version",
                         PyLong_FromLong(PYTHON_API_VERSION));
+    // asm("int3");
     SET_SYS_FROM_STRING("copyright",
                         PyUnicode_FromString(Py_GetCopyright()));
     SET_SYS_FROM_STRING("platform",
diff --git a/third_party/regex/notice.inc b/third_party/regex/notice.inc
deleted file mode 100644
index d154d1146..000000000
--- a/third_party/regex/notice.inc
+++ /dev/null
@@ -1,10 +0,0 @@
-asm(".ident\t\"\\n\
-Musl Libc (MIT License)\\n\
-Copyright 2005-2014 Rich Felker\"");
-asm(".include \"libc/disclaimer.inc\"");
-
-asm(".ident\t\"\\n\
-TRE regex (BSD-2 License)\\n\
-Copyright 2001-2009 Ville Laurikari <vl@iki.fi>\\n\
-Copyright 2016 Szabolcs Nagy\"");
-asm(".include \"libc/disclaimer.inc\"");
diff --git a/third_party/regex/regexec.c b/third_party/regex/regexec.c
index 81f722087..fd4b4446f 100644
--- a/third_party/regex/regexec.c
+++ b/third_party/regex/regexec.c
@@ -60,6 +60,11 @@
 #include "libc/limits.h"
 #include "third_party/regex/tre.inc"
 
+__notice(tre_regex, "\
+TRE regex (BSD-2 License)\n\
+Copyright 2001-2009 Ville Laurikari <vl@iki.fi>\n\
+Copyright 2016 Szabolcs Nagy");
+
 static void tre_fill_pmatch(size_t nmatch, regmatch_t pmatch[], int cflags,
                             const tre_tnfa_t *tnfa, regoff_t *tags,
                             regoff_t match_eo);
diff --git a/third_party/regex/tre.inc b/third_party/regex/tre.inc
index 2c26396d9..2f6e39854 100644
--- a/third_party/regex/tre.inc
+++ b/third_party/regex/tre.inc
@@ -60,7 +60,6 @@
 #include "libc/mem/alg.h"
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
-#include "third_party/regex/notice.inc"
 #include "third_party/regex/regex.h"
 
 #undef TRE_MBSTATE
diff --git a/third_party/stb/stb_image.c b/third_party/stb/stb_image.c
index 6852c3adc..f796a034b 100644
--- a/third_party/stb/stb_image.c
+++ b/third_party/stb/stb_image.c
@@ -36,10 +36,10 @@
 #include "third_party/aarch64/arm_neon.internal.h"
 #include "third_party/intel/ammintrin.internal.h"
 
-asm(".ident\t\"\\n\\n\
-stb_image (Public Domain)\\n\
-Credit: Sean Barrett, et al.\\n\
-http://nothings.org/stb\"");
+__notice(stb_image_notice, "\
+stb_image (Public Domain)\n\
+Credit: Sean Barrett, et al.\n\
+http://nothings.org/stb");
 
 #ifdef __x86_64__
 #define STBI_SSE2
diff --git a/third_party/stb/stb_image_resize.c b/third_party/stb/stb_image_resize.c
index 24cb373e5..7fc71a33e 100644
--- a/third_party/stb/stb_image_resize.c
+++ b/third_party/stb/stb_image_resize.c
@@ -23,10 +23,10 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-stb_image_resize (Public Domain)\\n\
-Credit: Jorge L Rodriguez (@VinoBS), Sean Barrett, et al.\\n\
-http://nothings.org/stb\"");
+__notice(stb_image_resize_notice, "\
+stb_image_resize (Public Domain)\n\
+Credit: Jorge L Rodriguez (@VinoBS), Sean Barrett, et al.\n\
+http://nothings.org/stb");
 
 #define STBIR_ASSERT(x)       assert(x)
 #define STBIR_MALLOC(size, c) ((void)(c), malloc(size))
diff --git a/third_party/stb/stb_image_write.c b/third_party/stb/stb_image_write.c
index 1f984bc78..9af55ae36 100644
--- a/third_party/stb/stb_image_write.c
+++ b/third_party/stb/stb_image_write.c
@@ -29,10 +29,10 @@
 #include "libc/str/str.h"
 #include "third_party/zlib/zlib.h"
 
-asm(".ident\t\"\\n\\n\
-stb_image_write (Public Domain)\\n\
-Credit: Sean Barrett, et al.\\n\
-http://nothings.org/stb\"");
+__notice(stb_image_write_notice, "\
+stb_image_write (Public Domain)\n\
+Credit: Sean Barrett, et al.\n\
+http://nothings.org/stb");
 
 #define STBIW_UCHAR(x)                       (unsigned char)((x)&0xff)
 
diff --git a/third_party/stb/stb_rect_pack.c b/third_party/stb/stb_rect_pack.c
index 65fab7afa..eb07e1e0d 100644
--- a/third_party/stb/stb_rect_pack.c
+++ b/third_party/stb/stb_rect_pack.c
@@ -21,10 +21,9 @@
 #include "libc/dce.h"
 #include "libc/mem/alg.h"
 
-asm(".ident\t\"\\n\\n\
-stb_rect_pack (MIT License)\\n\
-Copyright 2017 Sean Barrett\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(stb_rect_pack_notice, "\
+stb_rect_pack (MIT License)\n\
+Copyright 2017 Sean Barrett");
 
 // stb_rect_pack.h - v1.01 - public domain - rectangle packing
 // Sean Barrett 2014
diff --git a/third_party/stb/stb_truetype.c b/third_party/stb/stb_truetype.c
index a59346df3..e1449c11b 100644
--- a/third_party/stb/stb_truetype.c
+++ b/third_party/stb/stb_truetype.c
@@ -36,10 +36,9 @@
 #include "libc/str/str.h"
 #include "third_party/stb/stb_rect_pack.h"
 
-asm(".ident\t\"\\n\\n\
-stb_truetype (MIT License)\\n\
-Copyright 2017 Sean Barrett\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(stb_truetype_notice, "\
+stb_truetype (MIT License)\n\
+Copyright 2017 Sean Barrett");
 
 // stb_truetype.h - v1.26 - public domain
 // authored from 2009-2021 by Sean Barrett / RAD Game Tools
diff --git a/third_party/stb/stb_vorbis.c b/third_party/stb/stb_vorbis.c
index 3b3da48f3..221dee242 100644
--- a/third_party/stb/stb_vorbis.c
+++ b/third_party/stb/stb_vorbis.c
@@ -46,10 +46,10 @@
 #include "libc/mem/mem.h"
 #include "libc/str/str.h"
 
-asm(".ident\t\"\\n\\n\
-stb_vorbis (Public Domain)\\n\
-Credit: Sean Barrett, et al.\\n\
-http://nothings.org/stb\"");
+__notice(stb_vorbis_notice, "\
+stb_vorbis (Public Domain)\n\
+Credit: Sean Barrett, et al.\n\
+http://nothings.org/stb");
 
 // STB_VORBIS_NO_PUSHDATA_API
 //     does not compile the code for the various stb_vorbis_*_pushdata()
diff --git a/third_party/xed/x86features.c b/third_party/xed/x86features.c
index 3d8fca26c..12cebdf6e 100644
--- a/third_party/xed/x86features.c
+++ b/third_party/xed/x86features.c
@@ -18,13 +18,6 @@
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "third_party/xed/x86isa.h"
 
-asm(".ident\t\"\\n\\n\
-Xed (Apache 2.0)\\n\
-Copyright 2018 Intel Corporation\\n\
-Copyright 2019 Justine Alexandra Roberts Tunney\\n\
-Modifications: Trimmed down to 3kb [2019-03-22 jart]\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 /**
  * Mapping of enum XedChip -> bitset<enum XedIsaSet>.
  *
diff --git a/third_party/xed/x86ild.greg.c b/third_party/xed/x86ild.greg.c
index 3e76b71af..5bcb2d927 100644
--- a/third_party/xed/x86ild.greg.c
+++ b/third_party/xed/x86ild.greg.c
@@ -28,12 +28,11 @@
 #include "third_party/xed/private.h"
 #include "third_party/xed/x86.h"
 
-asm(".ident\t\"\\n\\n\
-Xed (Apache 2.0)\\n\
-Copyright 2018 Intel Corporation\\n\
-Copyright 2019 Justine Alexandra Roberts Tunney\\n\
-Modifications: Trimmed down to 3kb [2019-03-22 jart]\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(xed_notice, "\
+Xed (Apache 2.0)\n\
+Copyright 2018 Intel Corporation\n\
+Copyright 2019 Justine Alexandra Roberts Tunney\n\
+Changes: Trimmed Intel's assembler down to 3kb [2019-03-22 jart]");
 
 #define XED_ILD_HASMODRM_IGNORE_MOD 2
 
diff --git a/third_party/xed/x86isa.c b/third_party/xed/x86isa.c
index 9cc96240e..48b3af6bd 100644
--- a/third_party/xed/x86isa.c
+++ b/third_party/xed/x86isa.c
@@ -19,13 +19,6 @@
 #include "third_party/xed/x86.h"
 #include "third_party/xed/x86isa.h"
 
-asm(".ident\t\"\\n\\n\
-Xed (Apache 2.0)\\n\
-Copyright 2018 Intel Corporation\\n\
-Copyright 2019 Justine Alexandra Roberts Tunney\\n\
-Modifications: Trimmed down to 3kb [2019-03-22 jart]\"");
-asm(".include \"libc/disclaimer.inc\"");
-
 bool xed_isa_set_is_valid_for_chip(int isa_set, int chip) {
   unsigned n, r;
   n = isa_set / 64;
diff --git a/third_party/zlib/adler32_simd.c b/third_party/zlib/adler32_simd.c
index f767c49bd..b3d310fbb 100644
--- a/third_party/zlib/adler32_simd.c
+++ b/third_party/zlib/adler32_simd.c
@@ -1,6 +1,4 @@
-asm(".ident\t\"\\n\\n\
-Chromium (BSD-3 License)\\n\
-Copyright 2017 The Chromium Authors\"");
+__static_yoink("chromium_notice");
 
 /* adler32_simd.c
  *
diff --git a/third_party/zlib/crc32_simd.c b/third_party/zlib/crc32_simd.c
index cf38bb94b..a1b35ab56 100644
--- a/third_party/zlib/crc32_simd.c
+++ b/third_party/zlib/crc32_simd.c
@@ -5,10 +5,7 @@
  * found in the Chromium source repository LICENSE file.
  */
 
-asm(".ident\t\"\\n\\n\
-Chromium (BSD-3 License)\\n\
-Copyright 2017 The Chromium Authors\"");
-
+__static_yoink("chromium_notice");
 #include "third_party/intel/x86gprintrin.internal.h"
 #include "third_party/zlib/crc32_simd.internal.h"
 #if defined(CRC32_SIMD_AVX512_PCLMUL)
diff --git a/third_party/zlib/deflate.c b/third_party/zlib/deflate.c
index 6db7b0c28..58f9474e7 100644
--- a/third_party/zlib/deflate.c
+++ b/third_party/zlib/deflate.c
@@ -11,11 +11,7 @@
 #include "third_party/zlib/deflate.internal.h"
 #include "third_party/zlib/internal.h"
 #include "third_party/zlib/zutil.internal.h"
-
-asm(".ident\t\"\\n\\n\
-zlib 1.2.13 (zlib License)\\n\
-Copyright 1995-2022 Jean-loup Gailly and Mark Adler\\n\
-Invented 1990 Phillip Walter Katz\"");
+__static_yoink("zlib_notice");
 
 /*
  *  ALGORITHM
diff --git a/third_party/zlib/inffast_chunk.c b/third_party/zlib/inffast_chunk.c
index 7ab74c011..a5302a49b 100644
--- a/third_party/zlib/inffast_chunk.c
+++ b/third_party/zlib/inffast_chunk.c
@@ -7,10 +7,7 @@
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
-asm(".ident\t\"\\n\\n\
-Chromium (BSD-3 License)\\n\
-Copyright 2017 The Chromium Authors\"");
-
+__static_yoink("chromium_notice");
 #include "third_party/zlib/zutil.internal.h"
 #include "third_party/zlib/inftrees.internal.h"
 #include "third_party/zlib/inflate.internal.h"
diff --git a/third_party/zlib/inflate.c b/third_party/zlib/inflate.c
index 2d5a8c424..75fa6b56e 100644
--- a/third_party/zlib/inflate.c
+++ b/third_party/zlib/inflate.c
@@ -6,10 +6,7 @@
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
-asm(".ident\t\"\\n\\n\
-zlib 1.2.13 (zlib License)\\n\
-Copyright 1995-2022 Jean-loup Gailly and Mark Adler\\n\
-Invented 1990 Phillip Walter Katz\"");
+__static_yoink("zlib_notice");
 
 /*
  * Change history:
diff --git a/third_party/zlib/inftrees.c b/third_party/zlib/inftrees.c
index 4a628af76..c089e77b0 100644
--- a/third_party/zlib/inftrees.c
+++ b/third_party/zlib/inftrees.c
@@ -5,14 +5,10 @@
  * Copyright (C) 1995-2022 Mark Adler
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
+__static_yoink("zlib_notice");
 #include "third_party/zlib/inftrees.internal.h"
 #include "third_party/zlib/zutil.internal.h"
 
-asm(".ident\t\"\\n\\n\
-zlib 1.2.13 (zlib License)\\n\
-Copyright 1995-2022 Jean-loup Gailly and Mark Adler\\n\
-Invented 1990 Phillip Walter Katz\"");
-
 #define MAXBITS 15
 
 /*
diff --git a/third_party/zlib/notice.c b/third_party/zlib/notice.c
new file mode 100644
index 000000000..5f6ca928d
--- /dev/null
+++ b/third_party/zlib/notice.c
@@ -0,0 +1,4 @@
+__notice(zlib_notice, "\
+zlib 1.2.13 (zlib License)\n\
+Copyright 1995-2022 Jean-loup Gailly and Mark Adler\n\
+Invented 1990 Phillip Walter Katz");
diff --git a/tool/build/fixupobj.c b/tool/build/fixupobj.c
index 7cb391bce..a5b713b47 100644
--- a/tool/build/fixupobj.c
+++ b/tool/build/fixupobj.c
@@ -29,7 +29,6 @@
 #include "libc/errno.h"
 #include "libc/fmt/itoa.h"
 #include "libc/fmt/magnumstrs.internal.h"
-#include "libc/intrin/kprintf.h"
 #include "libc/limits.h"
 #include "libc/log/log.h"
 #include "libc/macros.internal.h"
@@ -226,8 +225,56 @@ static void CheckPrivilegedCrossReferences(void) {
   }
 }
 
+// Change AMD code to use %gs:0x30 instead of %fs:0
+// We assume -mno-tls-direct-seg-refs has been used
+static void ChangeTlsFsToGs(unsigned char *p, size_t n) {
+  unsigned char *e = p + n - 9;
+  while (p <= e) {
+    // we're checking for the following expression:
+    //   0144 == p[0] &&           // %fs
+    //   0110 == (p[1] & 0373) &&  // rex.w (and ignore rex.r)
+    //   (0213 == p[2] ||          // mov reg/mem → reg (word-sized)
+    //   0003 == p[2]) &&          // add reg/mem → reg (word-sized)
+    //   0004 == (p[3] & 0307) &&  // mod/rm (4,reg,0) means sib → reg
+    //   0045 == p[4] &&           // sib (5,4,0) → (rbp,rsp,0) → disp32
+    //   0000 == p[5] &&           // displacement (von Neumann endian)
+    //   0000 == p[6] &&           // displacement
+    //   0000 == p[7] &&           // displacement
+    //   0000 == p[8]              // displacement
+    uint64_t w = READ64LE(p) & READ64LE("\377\373\377\307\377\377\377\377");
+    if ((w == READ64LE("\144\110\213\004\045\000\000\000") ||
+         w == READ64LE("\144\110\003\004\045\000\000\000")) &&
+        !p[8]) {
+      p[0] = 0145;  // change %fs to %gs
+      p[5] = 0x30;  // change 0 to 0x30
+      p += 9;
+    } else {
+      ++p;
+    }
+  }
+}
+
+static void RewriteTlsCodeAmd64(void) {
+  int i;
+  uint8_t *p;
+  Elf64_Shdr *shdr;
+  for (i = 0; i < elf->e_shnum; ++i) {
+    if (!(shdr = GetElfSectionHeaderAddress(elf, esize, i))) {
+      Die("elf header overflow #1");
+    }
+    if (shdr->sh_type == SHT_PROGBITS &&  //
+        (shdr->sh_flags & SHF_ALLOC) &&   //
+        (shdr->sh_flags & SHF_EXECINSTR)) {
+      if (!(p = GetElfSectionAddress(elf, esize, shdr))) {
+        Die("elf header overflow #2");
+      }
+      ChangeTlsFsToGs(p, shdr->sh_size);
+    }
+  }
+}
+
 // Modify ARM64 code to use x28 for TLS rather than tpidr_el0.
-static void RewriteTlsCode(void) {
+static void RewriteTlsCodeArm64(void) {
   int i;
   Elf64_Shdr *shdr;
   uint32_t *p, *pe;
@@ -594,10 +641,11 @@ static void FixupObject(void) {
     CheckPrivilegedCrossReferences();
     if (mode == O_RDWR) {
       if (elf->e_machine == EM_NEXGEN32E) {
+        RewriteTlsCodeAmd64();
         OptimizePatchableFunctionEntries();
         GenerateIfuncInit();
       } else if (elf->e_machine == EM_AARCH64) {
-        RewriteTlsCode();
+        RewriteTlsCodeArm64();
         if (elf->e_type != ET_REL) {
           UseFreebsdOsAbi();
         }
diff --git a/tool/cosmocc/README.md b/tool/cosmocc/README.md
index 1b3aa124a..d55b9717e 100644
--- a/tool/cosmocc/README.md
+++ b/tool/cosmocc/README.md
@@ -9,12 +9,13 @@ reach a broader audience from the platform(s) of your choosing.
 
 ## What's Included
 
-This toolchain bundles GCC 11.2.0, Cosmopolitan Libc, LLVM LIBCXX, and
-LLVM compiler-rt. Additional libraries were provided by Musl Libc, and
-the venerable BSDs OSes. This lets you benefit from the awesome modern
-GCC compiler with the strongest GPL barrier possible. The preprocessor
-advertises cross compilers as both `__COSMOCC__` and `__COSMOPOLITAN__`
-whereas `cosmocc` additionally defines `__FATCOSMOCC__`.
+This toolchain bundles GCC 12.3.0, Cosmopolitan Libc, LLVM LIBCXX, LLVM
+compiler-rt, and LLVM OpenMP. Additional libraries were provided by Musl
+Libc, and the venerable BSDs OSes. This lets you benefit from the
+awesome modern GCC compiler with the strongest GPL barrier possible. The
+preprocessor advertises cross compilers as both `__COSMOCC__` and
+`__COSMOPOLITAN__` whereas `cosmocc` additionally defines
+`__FATCOSMOCC__`.
 
 ## Getting Started
 
@@ -296,9 +297,9 @@ EINVAL: ... }` in cases where constants like `EINVAL` are linkable
 symbols. Your code will be rewritten in such cases to use a series of if
 statements instead, so that Cosmopolitan Libc's system constants will
 work as expected. Our modifications to GNU GCC are published under the
-ISC license at <https://github.com/ahgamut/gcc/tree/portcosmo-11.2>. The
+ISC license at <https://github.com/ahgamut/gcc/tree/portcosmo-12.3>. The
 binaries you see here were first published at
-<https://github.com/ahgamut/superconfigure/releases/tag/z0.0.30> which
+<https://github.com/ahgamut/superconfigure/releases/tag/z0.0.32> which
 is regularly updated.
 
 ## Legal
diff --git a/tool/cosmocc/bin/cosmocc b/tool/cosmocc/bin/cosmocc
index 719d0b5d6..af8f38f67 100755
--- a/tool/cosmocc/bin/cosmocc
+++ b/tool/cosmocc/bin/cosmocc
@@ -239,7 +239,7 @@ PLATFORM="-D__COSMOPOLITAN__ -D__COSMOCC__ -D__FATCOSMOCC__"
 PREDEF="-include libc/integral/normalize.inc"
 CPPFLAGS="-fno-pie -nostdinc -fno-math-errno -isystem $BIN/../include"
 CFLAGS="-fportcosmo -fno-dwarf2-cfi-asm -fno-unwind-tables -fno-asynchronous-unwind-tables -fno-semantic-interposition"
-LDFLAGS="-static -nostdlib -no-pie -fuse-ld=bfd -Wl,-z,norelro -Wl,--gc-sections"
+LDFLAGS="-static -nostdlib -no-pie -fuse-ld=bfd -Wl,-z,noexecstack -Wl,-z,norelro -Wl,--gc-sections"
 PRECIOUS="-fno-omit-frame-pointer"
 
 if [ x"$OPT" != x"-Os" ] && [ x"$MODE" != x"tiny" ]; then
diff --git a/tool/cosmocc/bin/cosmocross b/tool/cosmocc/bin/cosmocross
index e1f3afd04..ba6d3e54a 100755
--- a/tool/cosmocc/bin/cosmocross
+++ b/tool/cosmocc/bin/cosmocross
@@ -49,7 +49,7 @@ PLATFORM="-D__COSMOPOLITAN__ -D__COSMOCC__"
 PREDEF="-include libc/integral/normalize.inc"
 CFLAGS="-fportcosmo -fno-dwarf2-cfi-asm -fno-unwind-tables -fno-asynchronous-unwind-tables -fno-semantic-interposition"
 CPPFLAGS="-fno-pie -nostdinc -fno-math-errno -isystem $BIN/../include"
-LDFLAGS="-static -no-pie -nostdlib -fuse-ld=bfd"
+LDFLAGS="-static -no-pie -nostdlib -fuse-ld=bfd -Wl,-z,noexecstack"
 APEFLAGS="-Wl,--gc-sections"
 PRECIOUS="-fno-omit-frame-pointer"
 
diff --git a/tool/cosmocc/package.sh b/tool/cosmocc/package.sh
index 63e6bf191..8f43c94fb 100755
--- a/tool/cosmocc/package.sh
+++ b/tool/cosmocc/package.sh
@@ -19,6 +19,7 @@ OUTDIR=${1:-cosmocc}
 APELINK=o/$(mode)/tool/build/apelink.com
 AMD64=${2:-x86_64}
 ARM64=${3:-aarch64}
+GCCVER=12.3.0
 
 make -j32 m= \
   $APELINK
@@ -89,10 +90,10 @@ fetch() {
 OLD=$PWD
 cd "$OUTDIR/"
 if [ ! -x bin/x86_64-linux-cosmo-gcc ]; then
-  fetch https://github.com/ahgamut/superconfigure/releases/download/z0.0.30/aarch64-gcc.zip
+  fetch https://github.com/ahgamut/superconfigure/releases/download/z0.0.32/aarch64-gcc.zip
   unzip aarch64-gcc.zip
   rm -f aarch64-gcc.zip
-  fetch https://github.com/ahgamut/superconfigure/releases/download/z0.0.30/x86_64-gcc.zip
+  fetch https://github.com/ahgamut/superconfigure/releases/download/z0.0.32/x86_64-gcc.zip
   unzip x86_64-gcc.zip
   rm -f x86_64-gcc.zip
 fi
@@ -113,14 +114,14 @@ for arch in aarch64 x86_64; do
   ln -sf $arch-linux-cosmo-objdump bin/$arch-unknown-cosmo-objdump
   ln -sf $arch-linux-cosmo-readelf bin/$arch-unknown-cosmo-readelf
   ln -sf $arch-linux-cosmo-strip bin/$arch-unknown-cosmo-strip
-  cmp -s libexec/gcc/$arch-linux-cosmo/11.2.0/ld.bfd libexec/gcc/$arch-linux-cosmo/11.2.0/ld
-  ln -sf ld.bfd libexec/gcc/$arch-linux-cosmo/11.2.0/ld
-  cmp -s libexec/gcc/$arch-linux-cosmo/11.2.0/ld.bfd bin/$arch-linux-cosmo-ld
-  ln -sf ../libexec/gcc/$arch-linux-cosmo/11.2.0/ld.bfd bin/$arch-linux-cosmo-ld
-  cmp -s libexec/gcc/$arch-linux-cosmo/11.2.0/as bin/$arch-linux-cosmo-as
-  ln -sf ../libexec/gcc/$arch-linux-cosmo/11.2.0/as bin/$arch-linux-cosmo-as
-  cmp -s libexec/gcc/$arch-linux-cosmo/11.2.0/ld.bfd bin/$arch-linux-cosmo-ld.bfd
-  ln -sf ../libexec/gcc/$arch-linux-cosmo/11.2.0/ld.bfd bin/$arch-linux-cosmo-ld.bfd
+  cmp -s libexec/gcc/$arch-linux-cosmo/$GCCVER/ld.bfd libexec/gcc/$arch-linux-cosmo/$GCCVER/ld
+  ln -sf ld.bfd libexec/gcc/$arch-linux-cosmo/$GCCVER/ld
+  cmp -s libexec/gcc/$arch-linux-cosmo/$GCCVER/ld.bfd bin/$arch-linux-cosmo-ld
+  ln -sf ../libexec/gcc/$arch-linux-cosmo/$GCCVER/ld.bfd bin/$arch-linux-cosmo-ld
+  cmp -s libexec/gcc/$arch-linux-cosmo/$GCCVER/as bin/$arch-linux-cosmo-as
+  ln -sf ../libexec/gcc/$arch-linux-cosmo/$GCCVER/as bin/$arch-linux-cosmo-as
+  cmp -s libexec/gcc/$arch-linux-cosmo/$GCCVER/ld.bfd bin/$arch-linux-cosmo-ld.bfd
+  ln -sf ../libexec/gcc/$arch-linux-cosmo/$GCCVER/ld.bfd bin/$arch-linux-cosmo-ld.bfd
 done
 cd "$OLD"
 
diff --git a/tool/emacs/c.lang b/tool/emacs/c.lang
index 99f7d7c08..a2dad3636 100644
--- a/tool/emacs/c.lang
+++ b/tool/emacs/c.lang
@@ -121,7 +121,6 @@ Keywords={
 "hasatleast",
 "nodebuginfo",
 "noreturn",
-"initarray",
 "mayalias",
 "dontinstrument",
 "interruptfn",
diff --git a/tool/emacs/cosmo-asm-mode.el b/tool/emacs/cosmo-asm-mode.el
index 70fd0edd2..b948e12af 100644
--- a/tool/emacs/cosmo-asm-mode.el
+++ b/tool/emacs/cosmo-asm-mode.el
@@ -65,6 +65,7 @@
            "pltoff"
            "gotpcrel"
            "progbits"
+           "note"
            "nobits"
            "init_array"
            "fini_array"
diff --git a/tool/emacs/cosmo-c-builtins.el b/tool/emacs/cosmo-c-builtins.el
index 1a9a69330..3dee199be 100644
--- a/tool/emacs/cosmo-c-builtins.el
+++ b/tool/emacs/cosmo-c-builtins.el
@@ -204,6 +204,7 @@
          '("DebugBreak"
            "__veil"
            "__conceal"
+           "__notice"
            "__expropriate"
            "__yoink"
            "__dll_import"
diff --git a/tool/emacs/cosmo-c-keywords.el b/tool/emacs/cosmo-c-keywords.el
index 7fd7cdd30..3f4f691c5 100644
--- a/tool/emacs/cosmo-c-keywords.el
+++ b/tool/emacs/cosmo-c-keywords.el
@@ -87,7 +87,6 @@
            "nomsan"
            "dontubsan"
            "nostackprotector"
-           "initarray"
            "mayalias"
            "dontinstrument"
            "interruptfn"
diff --git a/tool/emacs/key.py b/tool/emacs/key.py
index 43a8b7c73..0d1a56ca1 100644
--- a/tool/emacs/key.py
+++ b/tool/emacs/key.py
@@ -335,7 +335,6 @@ cosmo_kws = frozenset([
   "forcealignargpointer",
   "forceinline",
   "hasatleast",
-  "initarray",
   "interruptfn",
   "mallocesque",
   "mayalias",
@@ -394,7 +393,6 @@ cosmo_kws = frozenset([
   "forcealignargpointer",
   "forceinline",
   "hasatleast",
-  "initarray",
   "interruptfn",
   "mallocesque",
   "mayalias",
diff --git a/tool/hello/BUILD.mk b/tool/hello/BUILD.mk
index 3d2ac96e9..6047047bf 100644
--- a/tool/hello/BUILD.mk
+++ b/tool/hello/BUILD.mk
@@ -39,7 +39,7 @@ o/$(MODE)/tool/hello/hello.com.dbg:				\
 
 # uses apelink to turn it into an ape executable
 # support vector is set to all operating systems
-o/$(MODE)/tool/hello/hello.com:					\
+o/$(MODE)/tool/hello/hello.ape:					\
 		o/$(MODE)/tool/hello/hello.com.dbg		\
 		o/$(MODE)/tool/build/apelink.com		\
 		o/$(MODE)/tool/build/pecheck.com		\
@@ -49,7 +49,7 @@ o/$(MODE)/tool/hello/hello.com:					\
 
 # uses apelink to generate elf-only executable
 # support vector = linux/freebsd/openbsd/netbsd/metal
-o/$(MODE)/tool/hello/hello-elf.com:				\
+o/$(MODE)/tool/hello/hello-elf.ape:				\
 		o/$(MODE)/tool/hello/hello.com.dbg		\
 		o/$(MODE)/tool/build/apelink.com		\
 		o/$(MODE)/ape/ape.elf
@@ -59,7 +59,7 @@ o/$(MODE)/tool/hello/hello-elf.com:				\
 # support vector = macos/linux/freebsd/openbsd/netbsd
 # - great way to avoid attention from bad virus scanners
 # - creates tinier executable by reducing alignment requirement
-o/$(MODE)/tool/hello/hello-unix.com:				\
+o/$(MODE)/tool/hello/hello-unix.ape:				\
 		o/$(MODE)/tool/hello/hello.com.dbg		\
 		o/$(MODE)/tool/build/apelink.com		\
 		o/$(MODE)/ape/ape.elf
@@ -71,7 +71,7 @@ o/$(MODE)/tool/hello/hello-unix.com:				\
 o/$(MODE)/tool/hello/hello-pe.com.dbg:				\
 		o/$(MODE)/tool/hello/hello-pe.o
 	@$(COMPILE) -ALINK.elf $(LINK) $(LINKARGS) $(OUTPUT_OPTION) -q -e WinMain
-o/$(MODE)/tool/hello/hello-pe.com:				\
+o/$(MODE)/tool/hello/hello-pe.ape:				\
 		o/$(MODE)/tool/hello/hello-pe.com.dbg		\
 		o/$(MODE)/tool/build/elf2pe.com
 	@$(COMPILE) -AELF2PE o/$(MODE)/tool/build/elf2pe.com -o $@ $<
@@ -80,7 +80,7 @@ o/$(MODE)/tool/hello/hello-pe.com:				\
 o/$(MODE)/tool/hello/life-pe.com.dbg:				\
 		o/$(MODE)/tool/hello/life-pe.o
 	@$(COMPILE) -ALINK.elf $(LINK) $(LINKARGS) $(OUTPUT_OPTION) -q -e WinMain
-o/$(MODE)/tool/hello/life-pe.com:				\
+o/$(MODE)/tool/hello/life-pe.ape:				\
 		o/$(MODE)/tool/hello/life-pe.com.dbg		\
 		o/$(MODE)/tool/build/elf2pe.com
 	@$(COMPILE) -AELF2PE o/$(MODE)/tool/build/elf2pe.com -o $@ $<
@@ -89,7 +89,7 @@ o/$(MODE)/tool/hello/life-pe.com:				\
 o/$(MODE)/tool/hello/wait-pe.com.dbg:				\
 		o/$(MODE)/tool/hello/wait-pe.o
 	@$(COMPILE) -ALINK.elf $(LINK) $(LINKARGS) $(OUTPUT_OPTION) -q -e WinMain
-o/$(MODE)/tool/hello/wait-pe.com:				\
+o/$(MODE)/tool/hello/wait-pe.ape:				\
 		o/$(MODE)/tool/hello/wait-pe.com.dbg		\
 		o/$(MODE)/tool/build/elf2pe.com
 	@$(COMPILE) -AELF2PE o/$(MODE)/tool/build/elf2pe.com -R 64kb -S 4kb -o $@ $<
diff --git a/tool/net/largon2.c b/tool/net/largon2.c
index e4a6a5c6f..6b5cd51c6 100644
--- a/tool/net/largon2.c
+++ b/tool/net/largon2.c
@@ -32,10 +32,9 @@
 #include "third_party/lua/lua.h"
 #include "third_party/lua/lualib.h"
 
-asm(".ident\t\"\\n\\n\
-largon2 (MIT License)\\n\
-Copyright 2016 Thibault Charbonnier\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(largon2_notice, "\
+largon2 (MIT License)\n\
+Copyright 2016 Thibault Charbonnier");
 
 // clang-format off
 /***
diff --git a/tool/net/lsqlite3.c b/tool/net/lsqlite3.c
index 945560538..f46eee72e 100644
--- a/tool/net/lsqlite3.c
+++ b/tool/net/lsqlite3.c
@@ -38,10 +38,9 @@
 #include "third_party/sqlite3/sqlite3.h"
 // clang-format off
 
-asm(".ident\t\"\\n\\n\
-lsqlite3 (MIT License)\\n\
-Copyright 2002-2016 Tiago Dionizio, Doug Currie\"");
-asm(".include \"libc/disclaimer.inc\"");
+__notice(lsqlite3_notice, "\
+lsqlite3 (MIT License)\n\
+Copyright 2002-2016 Tiago Dionizio, Doug Currie");
 
 // LOCAL CHANGES
 //
diff --git a/usr/share/zoneinfo/Anchorage b/usr/share/zoneinfo/Anchorage
deleted file mode 100644
index cdf0572be..000000000
Binary files a/usr/share/zoneinfo/Anchorage and /dev/null differ
diff --git a/usr/share/zoneinfo/Anchorage b/usr/share/zoneinfo/Anchorage
new file mode 120000
index 000000000..cafb24b4f
--- /dev/null
+++ b/usr/share/zoneinfo/Anchorage
@@ -0,0 +1 @@
+US/Alaska
\ No newline at end of file
diff --git a/usr/share/zoneinfo/Boulder b/usr/share/zoneinfo/Boulder
deleted file mode 100644
index 7fc669171..000000000
Binary files a/usr/share/zoneinfo/Boulder and /dev/null differ
diff --git a/usr/share/zoneinfo/Boulder b/usr/share/zoneinfo/Boulder
new file mode 120000
index 000000000..8b727a113
--- /dev/null
+++ b/usr/share/zoneinfo/Boulder
@@ -0,0 +1 @@
+US/Mountain
\ No newline at end of file
diff --git a/usr/share/zoneinfo/Chicago b/usr/share/zoneinfo/Chicago
deleted file mode 100644
index 3dd8f0fa8..000000000
Binary files a/usr/share/zoneinfo/Chicago and /dev/null differ
diff --git a/usr/share/zoneinfo/Chicago b/usr/share/zoneinfo/Chicago
new file mode 120000
index 000000000..0c6fef91f
--- /dev/null
+++ b/usr/share/zoneinfo/Chicago
@@ -0,0 +1 @@
+US/Central
\ No newline at end of file
diff --git a/usr/share/zoneinfo/GMT b/usr/share/zoneinfo/GMT
deleted file mode 100644
index c63474664..000000000
Binary files a/usr/share/zoneinfo/GMT and /dev/null differ
diff --git a/usr/share/zoneinfo/GMT b/usr/share/zoneinfo/GMT
new file mode 120000
index 000000000..1ed082089
--- /dev/null
+++ b/usr/share/zoneinfo/GMT
@@ -0,0 +1 @@
+UTC
\ No newline at end of file
diff --git a/usr/share/zoneinfo/GST b/usr/share/zoneinfo/GST
deleted file mode 100644
index c0ce4402f..000000000
Binary files a/usr/share/zoneinfo/GST and /dev/null differ
diff --git a/usr/share/zoneinfo/GST b/usr/share/zoneinfo/GST
new file mode 120000
index 000000000..e2ebd1750
--- /dev/null
+++ b/usr/share/zoneinfo/GST
@@ -0,0 +1 @@
+US/Pacific
\ No newline at end of file
diff --git a/usr/share/zoneinfo/Honolulu b/usr/share/zoneinfo/Honolulu
deleted file mode 100644
index d0c2595f9..000000000
Binary files a/usr/share/zoneinfo/Honolulu and /dev/null differ
diff --git a/usr/share/zoneinfo/Honolulu b/usr/share/zoneinfo/Honolulu
new file mode 120000
index 000000000..16c5c6023
--- /dev/null
+++ b/usr/share/zoneinfo/Honolulu
@@ -0,0 +1 @@
+US/Hawaii
\ No newline at end of file
diff --git a/usr/share/zoneinfo/Israel b/usr/share/zoneinfo/Israel
index 4992a7929..4c49bbf52 100644
Binary files a/usr/share/zoneinfo/Israel and b/usr/share/zoneinfo/Israel differ
diff --git a/usr/share/zoneinfo/New_York b/usr/share/zoneinfo/New_York
deleted file mode 100644
index 7553fee37..000000000
Binary files a/usr/share/zoneinfo/New_York and /dev/null differ
diff --git a/usr/share/zoneinfo/New_York b/usr/share/zoneinfo/New_York
new file mode 120000
index 000000000..b8d5363be
--- /dev/null
+++ b/usr/share/zoneinfo/New_York
@@ -0,0 +1 @@
+US/Eastern
\ No newline at end of file
diff --git a/usr/share/zoneinfo/Singapore b/usr/share/zoneinfo/Singapore
index 785836666..dbbdea3c8 100644
Binary files a/usr/share/zoneinfo/Singapore and b/usr/share/zoneinfo/Singapore differ
diff --git a/usr/share/zoneinfo/US/Alaska b/usr/share/zoneinfo/US/Alaska
new file mode 100644
index 000000000..cdf0572be
Binary files /dev/null and b/usr/share/zoneinfo/US/Alaska differ
diff --git a/usr/share/zoneinfo/US/Aleutian b/usr/share/zoneinfo/US/Aleutian
new file mode 100644
index 000000000..b1497bda6
Binary files /dev/null and b/usr/share/zoneinfo/US/Aleutian differ
diff --git a/usr/share/zoneinfo/US/Arizona b/usr/share/zoneinfo/US/Arizona
new file mode 100644
index 000000000..c2bd2f949
Binary files /dev/null and b/usr/share/zoneinfo/US/Arizona differ
diff --git a/usr/share/zoneinfo/US/Central b/usr/share/zoneinfo/US/Central
new file mode 100644
index 000000000..b01688065
Binary files /dev/null and b/usr/share/zoneinfo/US/Central differ
diff --git a/usr/share/zoneinfo/US/East-Indiana b/usr/share/zoneinfo/US/East-Indiana
new file mode 100644
index 000000000..6b08d15bd
Binary files /dev/null and b/usr/share/zoneinfo/US/East-Indiana differ
diff --git a/usr/share/zoneinfo/US/Eastern b/usr/share/zoneinfo/US/Eastern
new file mode 100644
index 000000000..2b6c2eea1
Binary files /dev/null and b/usr/share/zoneinfo/US/Eastern differ
diff --git a/usr/share/zoneinfo/US/Hawaii b/usr/share/zoneinfo/US/Hawaii
new file mode 100644
index 000000000..40e3d492e
Binary files /dev/null and b/usr/share/zoneinfo/US/Hawaii differ
diff --git a/usr/share/zoneinfo/US/Indiana-Starke b/usr/share/zoneinfo/US/Indiana-Starke
new file mode 100644
index 000000000..b187d5f8c
Binary files /dev/null and b/usr/share/zoneinfo/US/Indiana-Starke differ
diff --git a/usr/share/zoneinfo/US/Michigan b/usr/share/zoneinfo/US/Michigan
new file mode 100644
index 000000000..6eb3ac46e
Binary files /dev/null and b/usr/share/zoneinfo/US/Michigan differ
diff --git a/usr/share/zoneinfo/US/Mountain b/usr/share/zoneinfo/US/Mountain
new file mode 100644
index 000000000..09e54e5c7
Binary files /dev/null and b/usr/share/zoneinfo/US/Mountain differ
diff --git a/usr/share/zoneinfo/US/Pacific b/usr/share/zoneinfo/US/Pacific
new file mode 100644
index 000000000..aaf07787a
Binary files /dev/null and b/usr/share/zoneinfo/US/Pacific differ
diff --git a/usr/share/zoneinfo/US/Samoa b/usr/share/zoneinfo/US/Samoa
new file mode 100644
index 000000000..001289cee
Binary files /dev/null and b/usr/share/zoneinfo/US/Samoa differ
diff --git a/usr/share/zoneinfo/UTC b/usr/share/zoneinfo/UTC
index c3b97f1a1..157573b1d 100644
Binary files a/usr/share/zoneinfo/UTC and b/usr/share/zoneinfo/UTC differ