[Piglit] [PATCH] util: Include time.h on all platforms.

[Brian Paul]

On 11/03/2016 03:36 PM, Vinson Lee wrote:
> Fix MinGW build error.
>
> piglit-util.c: In function 'piglit_time_is_monotonic':
> piglit-util.c:583:18: error: storage size of 't' isn't known
>    struct timespec t;
>                    ^
>
> Fixes: e0048f4940f7 ("util: Add piglit_delay_ns() api")
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98486
> Signed-off-by: Vinson Lee <vlee at freedesktop.org>
> ---
>   tests/util/piglit-util.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tests/util/piglit-util.c b/tests/util/piglit-util.c
> index ef955b97dd26..6becfa664490 100644
> --- a/tests/util/piglit-util.c
> +++ b/tests/util/piglit-util.c
> @@ -32,7 +32,6 @@
>   #include <unistd.h>
>   #include <sys/types.h>
>   #include <sys/syscall.h>
> -#include <time.h>
>   #endif
>
>   #include <assert.h>
> @@ -42,6 +41,7 @@
>   #include <string.h>
>   #include <errno.h>
>   #include <inttypes.h>
> +#include <time.h>
>
>   #if defined(PIGLIT_HAS_POSIX_CLOCK_MONOTONIC) && defined(PIGLIT_HAS_POSIX_TIMER_NOTIFY_THREAD)
>   #include <pthread.h>
>

Hmm, I'm not seeing this problem with my MinGW setup here (version 
4.6.3) but your patch does no harm either.

Reviewed-by: Brian Paul <brianp at vmware.com>
Tested-by: Brian Paul <brianp at vmware.com>


BTW, do you see this warning:

piglit/tests/util/piglit-util.c: In function ‘piglit_delay_ns’:
piglit/tests/util/piglit-util.c:634:2: warning: implicit declaration of 
function ‘usleep’ [-Wimplicit-function-declaration]

Using Sleep(time_ns / 1000000) instead of usleep(time_ns / 1000) seems 
to work.  There's only caller of piglit_delay_ns() and it passes in 1 
second so this change would be OK for the time being.  I'm wondering how 
practical ns sleeping is in any case.

-Brian