Re: [LTP] [POSIX][PATCH]Fix bug of "pthread_rwlock_timedwrlock" test

[Hannu H. <ha...@ik...>]

On 24/03/11 11:18 +0800, Bian Naimeng wrote:
> Specifying a passed abs_timeout to pthread_rwlock_timedwrlock.
> 
> Signed-off-by: Bian Naimeng <bi...@cn...>
> 
> ---
>  .../interfaces/pthread_rwlock_timedwrlock/2-1.c    |    7 ++-----
>  1 files changed, 2 insertions(+), 5 deletions(-)
> 
> diff --git a/testcases/open_posix_testsuite/conformance/interfaces/pthread_rwlock_timedwrlock/2-1.c b/testcases/open_posix_testsuite/conformance/interfaces/pthread_rwlock_timedwrlock/2-1.c
> index b837fac..3ce260c 100644
> --- a/testcases/open_posix_testsuite/conformance/interfaces/pthread_rwlock_timedwrlock/2-1.c
> +++ b/testcases/open_posix_testsuite/conformance/interfaces/pthread_rwlock_timedwrlock/2-1.c
> @@ -34,7 +34,6 @@
>  
>  static pthread_rwlock_t rwlock;
>  static int thread_state;
> -static int currsec1, currsec2;
>  static int expired;
>  
>  /* thread_state indicates child thread state:
> @@ -54,7 +53,7 @@ static void* fn(void *arg)
>  	thread_state = ENTERED_THREAD;
>  
>  	/* Absolute time, not relative. */
> -	abs_timeout.tv_sec = currsec1 - TIMEOUT;
> +	abs_timeout.tv_sec = time(NULL) - TIMEOUT;
>  	abs_timeout.tv_nsec = 0;
>  
>  	printf("thread: attempt timed write-lock\n");
> @@ -81,8 +80,6 @@ static void* fn(void *arg)
>  		exit(PTS_UNRESOLVED);
>  	}
>  
> -	/* Get time after the mutex timed out in locking. */
> -	currsec2 = time(NULL);
>  	thread_state = EXITING_THREAD;
>  	pthread_exit(0);
>  	return NULL;
> @@ -167,4 +164,4 @@ int main()
>  
>  	printf("Test PASSED\n");
>  	return PTS_PASS;
> -}
> \ No newline at end of file
> +}
> -- 
> 1.7.1
> 

Hi,

looks sane to me, you could consider using gettimeofday() just for having a
more accuracy by using also tv_usec (microsecs).

br,
Hannu