New issue
Advanced search Search tips
Note: Color blocks (like █ or █) mean that a user may not be available. Tooltip shows the reason.

Issue 645298 link

Starred by 1 user
Status: Verified
Owner: ----
Closed: Dec 2016
Cc:
█ kbr@chromium.org
kainino@chromium.org
piman@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Mac
Pri: 2
Type: Bug



Sign in to add a comment

Mac AMD handles instanced rendering incorrectly under certain situations

Project Member Reported by zmo@chromium.org, Sep 8 2016 
It seems like when
1) set up instanced rendering, draw*Instanced()
2) either switch context or readPixels (maybe flush() will also do)
3) switch back, set up exactly the same instanced rendering, draw*Instanced()

In step 3, the draw behaves like the divisor is infinite, i.e., it doesn't move on to the next.

However, if we change the divisor value in step 3, then everything works correctly

Comment 1 by zmo@chromium.org, Sep 8 2016 

A standalone test is added in here: https://github.com/KhronosGroup/WebGL/pull/2017

The failing cases are marked as skip, and the standalone test is marked as 2.0.1

This is really a corner case, because usually we use the same setup to draw multiple times, or we use the different set up.  We seldom set up the same a second time and draw.

Comment 2 by kbr@chromium.org, Sep 9 2016 

To clarify: if the vertex attribute divisor changes from draw call to draw call, it looks like the draw call is performed properly; but if it doesn't, then the second draw call is performed incorrectly if the vertex attribute divisor is greater than 1.

Comment 3 by kbr@chromium.org, Sep 9 2016 

Filed as Radar 28217585.

Comment 4 by zmo@chromium.org, Nov 5 2016

Blocking: -295792 -429053 662644
Mac driver bug, move to 2.0.1

Comment 5 by zmo@chromium.org, Dec 19 2016

Status: Fixed (was: Available)
Fixed in 10.12.2.  I already resolved the radar.

Comment 6 by zmo@chromium.org, Dec 20 2016

Status: Verified (was: Fixed)
Project Member

Comment 7 by bugdroid1@chromium.org, Jun 8 2017 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/1733d120eb79e2d8d49b18892e60ee1297f646f7

commit 1733d120eb79e2d8d49b18892e60ee1297f646f7
Author: zmo <zmo@chromium.org>
Date: Thu Jun 08 23:24:48 2017

Update WebGL2 conformance test expectations for Mac bots.

BUG= 598930 , 617290 , 618464 ,630800, 641149 , 643866 , 645298 , 646182 , 654187 , 663188 ,665197, 665656 , 676848 , 679682 , 679684 , 679686 , 679687 , 679689 , 679690 , 679691 , 680278 , 684903 
TEST=mac bots on GPU FYI waterfall
TBR=kbr@chromium.org,kainino@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2931993002
Cr-Commit-Position: refs/heads/master@{#478121}

[modify] https://crrev.com/1733d120eb79e2d8d49b18892e60ee1297f646f7/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 8 by lafo...@chromium.org, Jun 20 2017

Components: -Internals>GPU>WebGL Blink>WebGL

Comment 9 by jdarpinian@chromium.org, Nov 27

Blocking: -662644
►

Sign in to add a comment