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

libc/thread/mktls.c

@@ -16,6 +16,7 @@
 │ TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR             │
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
+#include "ape/sections.internal.h"
 #include "libc/dce.h"
 #include "libc/intrin/asan.internal.h"
 #include "libc/intrin/asancodes.h"
@@ -76,26 +77,25 @@ static char *_mktls_below(struct CosmoTib **out_tib) {
 }
 
 static char *_mktls_above(struct CosmoTib **out_tib) {
-  size_t siz;
+  size_t hiz, siz;
   char *mem, *dtv, *tls;
   struct CosmoTib *tib, *old;
 
   // allocate memory for tdata, tbss, and tib
-  siz = ROUNDUP(sizeof(struct CosmoTib) + 2 * sizeof(void *) + I(_tls_size),
-                TLS_ALIGNMENT);
+  hiz = ROUNDUP(sizeof(*tib) + 2 * sizeof(void *), I(_tls_align));
+  siz = hiz + I(_tls_size);
   mem = memalign(TLS_ALIGNMENT, siz);
   if (!mem) return 0;
 
   // poison memory between tdata and tbss
   if (IsAsan()) {
-    __asan_poison(
-        mem + sizeof(struct CosmoTib) + 2 * sizeof(void *) + I(_tdata_size),
-        I(_tbss_offset) - I(_tdata_size), kAsanProtected);
+    __asan_poison(mem + hiz + I(_tdata_size), I(_tbss_offset) - I(_tdata_size),
+                  kAsanProtected);
   }
 
   tib = (struct CosmoTib *)mem;
   dtv = mem + sizeof(*tib);
-  tls = dtv + 2 * sizeof(void *);
+  tls = mem + hiz;
 
   // set dtv
   ((uintptr_t *)dtv)[0] = 1;

libc/thread/pthread_getname_np.c

@@ -26,6 +26,7 @@
 #include "libc/intrin/atomic.h"
 #include "libc/macros.internal.h"
 #include "libc/str/str.h"
+#include "libc/sysv/consts/at.h"
 #include "libc/sysv/consts/o.h"
 #include "libc/sysv/consts/pr.h"
 #include "libc/thread/posixthread.internal.h"
@@ -52,7 +53,7 @@ static errno_t pthread_getname_impl(pthread_t thread, char *name, size_t size) {
       p = stpcpy(p, "/proc/self/task/");
       p = FormatUint32(p, tid);
       p = stpcpy(p, "/comm");
-      if ((fd = sys_open(path, O_RDONLY | O_CLOEXEC, 0)) == -1) {
+      if ((fd = sys_openat(AT_FDCWD, path, O_RDONLY | O_CLOEXEC, 0)) == -1) {
         rc = errno;
         errno = e;
         return rc;

libc/thread/pthread_setname_np.c

@@ -26,6 +26,7 @@
 #include "libc/intrin/asmflag.h"
 #include "libc/intrin/atomic.h"
 #include "libc/str/str.h"
+#include "libc/sysv/consts/at.h"
 #include "libc/sysv/consts/o.h"
 #include "libc/sysv/consts/pr.h"
 #include "libc/thread/posixthread.internal.h"
@@ -50,7 +51,7 @@ static errno_t pthread_setname_impl(pthread_t thread, const char *name) {
       p = stpcpy(p, "/proc/self/task/");
       p = FormatUint32(p, tid);
       p = stpcpy(p, "/comm");
-      if ((fd = sys_open(path, O_WRONLY | O_CLOEXEC, 0)) == -1) {
+      if ((fd = sys_openat(AT_FDCWD, path, O_WRONLY | O_CLOEXEC, 0)) == -1) {
         rc = errno;
         errno = e;
         return rc;

libc/thread/tls.h

@@ -66,7 +66,11 @@ void __set_tls(struct CosmoTib *);
   })
 #define __adj_tls(tib) (tib)
 #elif defined(__aarch64__)
-#define __get_tls()    ((struct CosmoTib *)__builtin_thread_pointer() - 1)
+#define __get_tls()                          \
+  ({                                         \
+    register struct CosmoTib *_t asm("x28"); \
+    _t - 1;                                  \
+  })
 #define __adj_tls(tib) ((struct CosmoTib *)(tib) + 1)
 #endif