This patch series implements a bunch of tests for the
ARB_shader_atomic_counters extension -- Patches adding support for it
to the i965 driver have been around for a while on the mesa-dev
mailing list, and they can also be found in my personal branch [1].

54 of these tests pass using the i965 implementation on both Ivy
Bridge and Haswell, 5 are skipped mainly due to the lack of support
for tessellation and geometry shaders from mesa master, none of the
tests fail on i965.

I've pushed the whole series to my personal piglit repository to ease
testing [2].

Thanks.

[1] http://cgit.freedesktop.org/~currojerez/mesa/log/?h=atomic-counters
[2] http://cgit.freedesktop.org/~currojerez/piglit/

[PATCH 01/13] arb_shader_atomic_counters: Import common helper functions.
[PATCH 02/13] arb_shader_atomic_counters: Import GLSL parser tests.
[PATCH 03/13] arb_shader_atomic_counters: Test the atomic counter query 
functions.
[PATCH 04/13] arb_shader_atomic_counters: Test dynamic indexing of atomic 
counter arrays.
[PATCH 05/13] arb_shader_atomic_counters: Test the buffer binding API functions.
[PATCH 06/13] arb_shader_atomic_counters: Test that atomics aren't allocated 
from the default uniform block.
[PATCH 07/13] arb_shader_atomic_counters: Test that atomic ops aren't executed 
for discarded fragments.
[PATCH 08/13] arb_shader_atomic_counters: Test that atomic counters can be 
passed as function arguments.
[PATCH 09/13] arb_shader_atomic_counters: Test the maximum number of supported 
counters, buffers and bindings.
[PATCH 10/13] arb_shader_atomic_counters: Test the minimum maximum limits 
defined by the spec.
[PATCH 11/13] arb_shader_atomic_counters: Test linkage of atomic counters with 
conflicting locations.
[PATCH 12/13] arb_shader_atomic_counters: Test that atomic built-ins execute 
the expected operations on memory.
[PATCH 13/13] arb_shader_atomic_counters: Test the atomicity of atomic built-in 
functions.
_______________________________________________
Piglit mailing list
Piglit@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/piglit

Reply via email to