cosmopolitan/libc/intrin/fenv.S
Justine Tunney 8f522cb702
Make improvements
This change progresses our AARCH64 support:

- The AARCH64 build and tests are now passing
- Add 128-bit floating-point support to printf()
- Fix clone() so it initializes cosmo's x28 TLS register
- Fix TLS memory layout issue with aarch64 _Alignas vars
- Revamp microbenchmarking tools so they work on aarch64
- Make some subtle improvements to aarch64 crash reporting
- Make kisdangerous() memory checks more accurate on aarch64
- Remove sys_open() since it's not available on Linux AARCH64

This change makes general improvements to Cosmo and Redbean:

- Introduce GetHostIsa() function in Redbean
- You can now feature check using pledge(0, 0)
- You can now feature check using unveil("",0)
- Refactor some more x86-specific asm comments
- Refactor and write docs for some libm functions
- Make the mmap() API behave more similar to Linux
- Fix WIFSIGNALED() which wrongly returned true for zero
- Rename some obscure cosmo keywords from noFOO to dontFOO
2023-06-03 08:12:22 -07:00

211 lines
5.6 KiB
ArmAsm

/*-*- mode:unix-assembly; indent-tabs-mode:t; tab-width:8; coding:utf-8 -*-│
vi: set et ft=asm ts=8 tw=8 fenc=utf-8 :vi
Musl Libc
Copyright © 2005-2014 Rich Felker, et al.
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/
#include "libc/macros.internal.h"
// Clears floating point exception status, e.g.
//
// feclearexcept(FE_ALL_EXCEPT);
//
// @param excepts may bitwise-or the following:
// - `FE_INVALID`
// - `FE_DIVBYZERO`
// - `FE_OVERFLOW`
// - `FE_UNDERFLOW`
// - `FE_INEXACT`
// - `FE_ALL_EXCEPT` (all of the above)
// @return 0 on success, or nonzero on error
feclearexcept:
#ifdef __x86_64__
// maintain exceptions in the sse mxcsr, clear x87 exceptions
mov %edi,%ecx
and $0x3f,%ecx
fnstsw %ax
test %eax,%ecx
jz 1f
fnclex
1: stmxcsr -8(%rsp)
and $0x3f,%eax
or %eax,-8(%rsp)
test %ecx,-8(%rsp)
jz 1f
not %ecx
and %ecx,-8(%rsp)
ldmxcsr -8(%rsp)
1: xor %eax,%eax
ret
#elif defined(__aarch64__)
and w0,w0,#0x1f
mrs x1,fpsr
bic w1,w1,w0
msr fpsr,x1
mov w0,#0
ret
#else
#error "unsupported architecture"
#endif
.endfn feclearexcept,globl
// Checks for floating point exception.
//
// This function may be used to check the thread-local
// floating-point exception status bits, e.g.
//
// feclearexcept(FE_ALL_EXCEPT);
// volatile double x = 0, y = 1 / x;
// assert(fetestexcept(FE_ALL_EXCEPT) == FE_DIVBYZERO);
//
// @param excepts may bitwise-or the following:
// - `FE_INVALID`
// - `FE_DIVBYZERO`
// - `FE_OVERFLOW`
// - `FE_UNDERFLOW`
// - `FE_INEXACT`
// - `FE_ALL_EXCEPT` (all of the above)
// @return mask of which exception status codes are currently set,
// or zero if there aren't any floating point exceptions
fetestexcept:
#ifdef __x86_64__
and $0x3f,%edi
push %rax
stmxcsr (%rsp)
pop %rsi
fnstsw %ax
or %esi,%eax
and %edi,%eax
ret
#elif defined(__aarch64__)
and w0,w0,#0x1f
mrs x1,fpsr
and w0,w0,w1
ret
#endif
.endfn fetestexcept,globl
feraiseexcept:
#ifdef __x86_64__
and $0x3f,%edi
stmxcsr -8(%rsp)
or %edi,-8(%rsp)
ldmxcsr -8(%rsp)
xor %eax,%eax
ret
#elif defined(__aarch64__)
and w0,w0,#0x1f
mrs x1,fpsr
orr w1,w1,w0
msr fpsr,x1
mov w0,#0
ret
#endif
.endfn feraiseexcept,globl
__fesetround:
#ifdef __x86_64__
push %rax
xor %eax,%eax
mov %edi,%ecx
fnstcw (%rsp)
andb $0xf3,1(%rsp)
or %ch,1(%rsp)
fldcw (%rsp)
stmxcsr (%rsp)
shl $3,%ch
andb $0x9f,1(%rsp)
or %ch,1(%rsp)
ldmxcsr (%rsp)
pop %rcx
ret
#elif defined(__aarch64__)
mrs x1,fpcr
bic w1,w1,#0xc00000
orr w1,w1,w0
msr fpcr,x1
mov w0,#0
ret
#endif
.endfn __fesetround,globl,hidden
fegetround:
#ifdef __x86_64__
push %rax
stmxcsr (%rsp)
pop %rax
shr $3,%eax
and $0xc00,%eax
ret
#elif defined(__aarch64__)
mrs x0,fpcr
and w0,w0,#0xc00000
ret
#endif
.endfn fegetround,globl
fegetenv:
#ifdef __x86_64__
xor %eax,%eax
fnstenv (%rdi)
stmxcsr 28(%rdi)
ret
#elif defined(__aarch64__)
mrs x1,fpcr
mrs x2,fpsr
stp w1,w2,[x0]
mov w0,#0
ret
#endif
.endfn fegetenv,globl
fesetenv:
#ifdef __x86_64__
xor %eax,%eax
inc %rdi
jz 1f
fldenv -1(%rdi)
ldmxcsr 27(%rdi)
ret
1: push %rax
push %rax
pushq $0xffff
pushq $0x37f
fldenv (%rsp)
pushq $0x1f80
ldmxcsr (%rsp)
add $40,%rsp
ret
#elif defined(__aarch64__)
mov x1,#0
mov x2,#0
cmn x0,#1
b.eq 1f
ldp w1,w2,[x0]
1: msr fpcr,x1
msr fpsr,x2
mov w0,#0
ret
#endif
.endfn fesetenv,globl