[Piglit] [PATCH] util: Use Sleep for Windows.

2016-11-11 Thread Vinson Lee
Fix MinGW warning. piglit-util.c: In function 'piglit_delay_ns': piglit-util.c:634:2: warning: implicit declaration of function 'usleep' [-Wimplicit-function-declaration] usleep(time_ns / 1000); ^~ Fixes: e0048f4940f7 ("util: Add piglit_delay_ns() api") Signed-off-by: Vinson Lee

Re: [Piglit] [PATCH] glsl: Add tests for large version that could be interpreted as -1

2016-11-11 Thread Ian Romanick
On 11/11/2016 12:47 AM, Juan A. Suarez Romero wrote: > On Tue, 2016-11-08 at 11:28 -0800, Ian Romanick wrote: >> From: Ian Romanick >> >> One of the possible fixes to bug #97420 was to use -1 (instead of 0) >> as >> the "version not set" flag. I believe that would have

[Piglit] [PATCH 1/2] polygon-offset: draw tests in two rows of nine

2016-11-11 Thread Brian Paul
Render the 18 tests in two rows of nine, where the bottom row shows the blue prim behind the red prim and the top row shows the blue prim in front of the red prim. This makes it a little easier to figure out which test corresponds to which graphics state. And add some comments. ---

[Piglit] [PATCH 2/2] gl-1.4-polygon-offset: print more info on failure.

2016-11-11 Thread Brian Paul
--- tests/spec/gl-1.4/polygon-offset.c | 4 1 file changed, 4 insertions(+) diff --git a/tests/spec/gl-1.4/polygon-offset.c b/tests/spec/gl-1.4/polygon-offset.c index 08ec6dc..a749071 100644 --- a/tests/spec/gl-1.4/polygon-offset.c +++ b/tests/spec/gl-1.4/polygon-offset.c @@ -497,6 +497,8

[Piglit] [PATCH] Add nv_image_formats tests

2016-11-11 Thread Lionel Landwerlin
Following the compiler tests added by edee46325935, we can also check that no errors are raised with new formats through glBindImageTexture(). Signed-off-by: Lionel Landwerlin --- tests/spec/CMakeLists.txt| 1 +

Re: [Piglit] [PATCH] glsl: Add tests for large version that could be interpreted as -1

2016-11-11 Thread Juan A. Suarez Romero
On Tue, 2016-11-08 at 11:28 -0800, Ian Romanick wrote: > From: Ian Romanick > > One of the possible fixes to bug #97420 was to use -1 (instead of 0) > as > the "version not set" flag.  I believe that would have failed these > tests. > > Signed-off-by: Ian Romanick