Re: [Piglit] [PATCH] Optionally capture dmesg changes for each test and report them in a summary

2013-09-20 Thread Marek Olšák
I think your implementation of dmesg logging isn't correct. Note that dmesg is a ring buffer; once it's full, new lines added at the end of dmesg will cause the lines at the beginning to disappear. That's why I had to find and save the timestamp string, and find its position again in a new dmesg

Re: [Piglit] [PATCH 2/2] gl-3.1: Move draw-buffers-errors test from 3.2 to 3.1

2013-09-20 Thread Ian Romanick
On 09/19/2013 06:13 PM, Chad Versace wrote: On 09/19/2013 01:58 PM, Ian Romanick wrote: From: Ian Romanick ian.d.roman...@intel.com The exact same spec language appears in the 3.1 spec. Currently no Mesa driver supports OpenGL 3.2, but there are drivers that support 3.1. This allows the

Re: [Piglit] [PATCH 1/3] piglit-run: add back -c0

2013-09-20 Thread Dylan Baker
On Thursday 19 September 2013 16:02:49 Chad Versace wrote: On 09/17/2013 08:43 AM, Dylan Baker wrote: On Monday 16 September 2013 20:08:33 Marek Olšák wrote: --- piglit-run.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/piglit-run.py b/piglit-run.py

Re: [Piglit] A couple of python bugs found in logic.

2013-09-20 Thread Dylan Baker
On Thursday 19 September 2013 20:12:34 Ken Phillis Jr wrote: I probably should be a little bit more price about issue #3... I do expect the times to be different, but there is a problem of seeing less than one tenth of the total time when I only have four processors. Good example:

Re: [Piglit] [PATCH] Optionally capture dmesg changes for each test and report them in a summary

2013-09-20 Thread Dylan Baker
Doh! On Friday 20 September 2013 16:29:25 Marek Olšák wrote: I think your implementation of dmesg logging isn't correct. Note that dmesg is a ring buffer; once it's full, new lines added at the end of dmesg will cause the lines at the beginning to disappear. That's why I had to find and save

[Piglit] [PATCH] Test accuracy of dFdx and dFdy functions.

2013-09-20 Thread Paul Berry
Mesa/i965 currently (as of commit 1cc3b90) has an inaccurate implementation of dFdy() which produces the same results for every pixel in an aligned 2x2 block. This piglit test exposes the inaccuracy. --- .../execution/fs-dfdx-accuracy.shader_test | 35 ++

[Piglit] [PATCH] egl-1.4: Add test egl-terminate-then-unbind-context

2013-09-20 Thread Chad Versace
This test makes current a context, terminates the context's display, then unbinds the context. According to the EGL 1.4 spec (2011.04.06), Section 3.2 Initialization, no errors should occur. Exposes a use-after-free crash on mesa-9.2 with i965. This is the underlying reason that `./$gles1_test

[Piglit] [PATCH] GLSL: Test that input layout qualifiers cannot be used on variable declarations

2013-09-20 Thread Nicholas Mack
--- ...-layout-qualifiers-with-variable-declarations.geom | 19 +++ ...-layout-qualifiers-with-variable-declarations.geom | 19 +++ 2 files changed, 38 insertions(+) create mode 100644

Re: [Piglit] [PATCH 3/3] Optionally capture dmesg changes for each test and report them in a summary

2013-09-20 Thread Marek Olšák
I'll send a new patch, stay tuned. Marek On Fri, Sep 20, 2013 at 10:32 PM, Dylan Baker baker.dyla...@gmail.com wrote: On Wednesday 18 September 2013 01:21:59 Marek Olšák wrote: On Wed, Sep 18, 2013 at 1:04 AM, Dylan Baker baker.dyla...@gmail.com wrote: On Wednesday 18 September 2013

Re: [Piglit] [PATCH 3/3] Optionally capture dmesg changes for each test and report them in a summary

2013-09-20 Thread Dylan Baker
On Wednesday 18 September 2013 01:21:59 Marek Olšák wrote: On Wed, Sep 18, 2013 at 1:04 AM, Dylan Baker baker.dyla...@gmail.com wrote: On Wednesday 18 September 2013 00:48:45 Marek Olšák wrote: On Tue, Sep 17, 2013 at 6:43 PM, Dylan Baker baker.dyla...@gmail.com wrote: On Monday 16

[Piglit] [PATCH 3/3] Optionally capture dmesg changes for each test and report them in a summary

2013-09-20 Thread Marek Olšák
The Radeon driver writes GPU page faults to dmesg and we need to know which tests caused them. If there is any change in dmesg during a test run, the test result is changed as follows: * pass - dmesg-warn * warn - dmesg-warn * fail - dmesg-fail Dmesg is captured before and after each test and the

[Piglit] [PATCH 2/3] util/egl: Add piglit_egl_bind_api()

2013-09-20 Thread Chad Versace
This is a wrapper for eglBindAPI() that does error-checking. Signed-off-by: Chad Versace chad.vers...@linux.intel.com --- tests/util/piglit-util-egl.c | 31 +++ tests/util/piglit-util-egl.h | 12 2 files changed, 43 insertions(+) diff --git

[Piglit] [PATCH 0/3] egl: Fix tests for context flags and add test for debug flag

2013-09-20 Thread Chad Versace
Chad Versace (3): egl_khr_create_context: Fix tests for invalid flags util/egl: Add piglit_egl_bind_api() egl_khr_create_context: Test the debug flag tests/all.tests| 2 + .../spec/egl_khr_create_context/CMakeLists.gl.txt | 1 +

[Piglit] [PATCH 1/3] egl_khr_create_context: Fix tests for invalid flags

2013-09-20 Thread Chad Versace
According to version 15 of the EGL_KHR_create_context spec, EGL_CONTEXT_OPENGL_DEBUG_BIT_KHR is a valid flag for OpenGL ES contexts. But the test EGL_KHR_create_context/invalid flag GLES verifies that it is an *invalid* flag. Signed-off-by: Chad Versace chad.vers...@linux.intel.com ---

[Piglit] [PATCH 3/3] egl_khr_create_context: Test the debug flag

2013-09-20 Thread Chad Versace
Test eglCreateContext with with EGL_CONTEXT_OPENGL_BIT_KHR set for GL, GLES1, GLES2, and GLES3. If context creation succeeds, then verify the context is really a debug context by verifying GL_CONTEXT_FLAGS contains GL_CONTEXT_FLAG_DEBUG_BIT. If context creation fails, then verify that