Introduce pthread_condattr_setclock()

This is one of the few POSIX APIs that was missing. It lets you choose a
monotonic clock for your condition variables. This might improve perf on
some platforms. It might also grant more flexibility with NTP configs. I
know Qt is one project that believes it needs this. To introduce this, I
needed to change some the *NSYNC APIs, to support passing a clock param.
There's also new benchmarks, demonstrating Cosmopolitan's supremacy over
many libc implementations when it comes to mutex performance. Cygwin has
an alarmingly bad pthread_mutex_t implementation. It is so bad that they
would have been significantly better off if they'd used naive spinlocks.
This commit is contained in:
Justine Tunney 2024-09-02 23:37:50 -07:00
parent 79516bf08e
commit 3c61a541bd
No known key found for this signature in database
GPG key ID: BE714B4575D6E328
55 changed files with 449 additions and 155 deletions

View file

@ -26,6 +26,7 @@
#include "libc/intrin/describeflags.h"
#include "libc/intrin/dll.h"
#include "libc/intrin/strace.h"
#include "libc/sysv/consts/clock.h"
#include "libc/thread/posixthread.internal.h"
#include "libc/thread/thread2.h"
#include "libc/thread/tls.h"
@ -74,7 +75,8 @@ static errno_t _pthread_wait(atomic_int *ctid, struct timespec *abstime) {
if (!(err = pthread_testcancel_np())) {
BEGIN_CANCELATION_POINT;
while ((x = atomic_load_explicit(ctid, memory_order_acquire))) {
e = nsync_futex_wait_(ctid, x, !IsWindows() && !IsXnu(), abstime);
e = nsync_futex_wait_(ctid, x, !IsWindows() && !IsXnu(), CLOCK_REALTIME,
abstime);
if (e == -ECANCELED) {
err = ECANCELED;
break;