Add threaded test of sem_trywait

[Carlos O'Donell]

On 11/7/24 8:49 PM, Joseph Myers wrote:
> All the existing glibc tests of sem_trywait are single-threaded.  Add
> one that calls sem_trywait and sem_post in separate threads.
> 
> Tested for x86_64.

Tests should have a description of intent.

Looking forward to a v2.

> diff --git a/sysdeps/pthread/Makefile b/sysdeps/pthread/Makefile
> index 4c1dc04b20..1a9f2559d9 100644
> --- a/sysdeps/pthread/Makefile
> +++ b/sysdeps/pthread/Makefile
> @@ -256,6 +256,7 @@ tests += \
>    tst-sem14 \
>    tst-sem15 \
>    tst-sem16 \
> +  tst-sem18 \

OK. New test.

>    tst-setuid3 \
>    tst-signal1 \
>    tst-signal2 \
> diff --git a/sysdeps/pthread/tst-sem18.c b/sysdeps/pthread/tst-sem18.c
> new file mode 100644
> index 0000000000..ade311473e
> --- /dev/null
> +++ b/sysdeps/pthread/tst-sem18.c
> @@ -0,0 +1,68 @@
> +/* Test sem_trywait with threads.

OK. One line description.

> +   Copyright (C) 2024 Free Software Foundation, Inc.
> +   This file is part of the GNU C Library.
> +
> +   The GNU C Library is free software; you can redistribute it and/or
> +   modify it under the terms of the GNU Lesser General Public
> +   License as published by the Free Software Foundation; either
> +   version 2.1 of the License, or (at your option) any later version.
> +
> +   The GNU C Library is distributed in the hope that it will be useful,
> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
> +   Lesser General Public License for more details.
> +
> +   You should have received a copy of the GNU Lesser General Public
> +   License along with the GNU C Library; if not, see
> +   <https://www.gnu.org/licenses/>.  */
> +
> +#include <errno.h>
> +#include <pthread.h>
> +#include <semaphore.h>
> +#include <support/check.h>
> +#include <support/xthread.h>
> +

This needs an explanation of the intent of the test.

Suggest:

/* The test uses two threads, the main thread and a newly created
   thread to test the operation of sem_trywait in a threaded
   scenario.  The intent is to test sem_trywait when it would return
   EAGAIN, and then again after the critical section in the new
   thread has posted to the semaphore and the main thread succeeds
   in calling sem_trywait.  It is possible this test fails with
   a timeout if the second thread takes longer than the test timeout
   to acquire the lock, and post.  */

> +static pthread_mutex_t lock = PTHREAD_MUTEX_INITIALIZER;
> +static sem_t sem;
> +
> +static void *
> +tf (void *arg)
> +{
> +  xpthread_mutex_lock (&lock);
> +  sem_post (&sem);
> +  xpthread_mutex_unlock (&lock);
> +  return NULL;
> +}
> +
> +static int
> +do_test (void)
> +{
> +  int ret;
> +
> +  ret = sem_init (&sem, 0, 0);
> +  TEST_VERIFY_EXIT (ret == 0);
> +  xpthread_mutex_lock (&lock);
> +  pthread_t th = xpthread_create (NULL, tf, NULL);
> +  errno = 0;
> +  /* The other thread is waiting on the lock before it calls sem_post,
> +     so sem_trywait should fail.  */
> +  ret = sem_trywait (&sem);

OK. Correct.

> +  TEST_COMPARE (ret, -1);
> +  TEST_COMPARE (errno, EAGAIN);
> +  xpthread_mutex_unlock (&lock);

OK.

> +  /* The other thread now takes the lock, calls sem_post and releases
> +     the lock.  */
> +  for (;;)
> +    {
> +      errno = 0;
> +      ret = sem_trywait (&sem);
> +      if (ret == 0)
> +	break;
> +      TEST_COMPARE (errno, EAGAIN);

OK. This must loop until we succeed or timeout. This *might* timeout if the thread
takes longer than the test timeout to complete, but that's not a problem we can
easily solve and it should work.

> +    }
> +  xpthread_join (th);
> +
> +  return 0;
> +}
> +
> +#include <support/test-driver.c>
> 

-- 
Cheers,
Carlos.