Allow clearing of VERTEX_ATTRIB_ARRAY_BUFFER_BINDING |
|||||
Issue descriptionAfter spec update https://github.com/KhronosGroup/WebGL/pull/2228 and most recently test update https://github.com/KhronosGroup/WebGL/pull/2731 , Chromium's WebGL implementation needs to be updated to allow clearing out of the VERTEX_ATTRIB_ARRAY_BUFFER_BINDING. https://www.khronos.org/registry/webgl/specs/latest/1.0/#6.2 The specs of vertexAttribPointer and vertexAttribIPointer also should be updated to clarify that this is allowed. This blocks fixing the object lifetime semantics in Issue 739604 , since recent test changes have caused deqp/functional/gles3/lifetime.html to start failing in Chromium.
,
Nov 27
,
Nov 27
,
Dec 21
I have a fix for this.
,
Dec 21
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/d39b023d4d58610875682d63dde7a77caf6fd634 commit d39b023d4d58610875682d63dde7a77caf6fd634 Author: James Darpinian <jdarpinian@chromium.org> Date: Fri Dec 21 04:26:44 2018 gpu: Allow clearing vertex attrib bindings Our check to prohibit the use of client side arrays was too strict. We should allow vertexAttribPointer calls with no buffer bound as long as the pointer is NULL, because this is the only way to clear a vertex attrib binding. See WebGL spec rationale under "No Client Side Arrays": https://www.khronos.org/registry/webgl/specs/latest/1.0/#6.2 Fixes WebGL conformance test deqp/functional/gles3/lifetime.html. Bug: 898350 Change-Id: Idab69b2a41286e6018fdb71203c24b6b60b2f148 Reviewed-on: https://chromium-review.googlesource.com/c/1387460 Reviewed-by: Kenneth Russell <kbr@chromium.org> Commit-Queue: James Darpinian <jdarpinian@chromium.org> Cr-Commit-Position: refs/heads/master@{#618456} [modify] https://crrev.com/d39b023d4d58610875682d63dde7a77caf6fd634/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py [modify] https://crrev.com/d39b023d4d58610875682d63dde7a77caf6fd634/gpu/command_buffer/client/vertex_array_object_manager.cc
,
Dec 21
|
|||||
►
Sign in to add a comment |
|||||
Comment 1 by bugdroid1@chromium.org
, Oct 24