GLImageIOSurface/GLImageBindTest/4.BindTexImage failing on Mac NVIDIA with 10.13.4 beta |
|||
Issue descriptionThe Mac FYI Experimental Release (NVIDIA) bot was recently upgraded to the 10.13.4 beta in Issue 812954. Previously it had been (accidentally) running 10.13.3. https://ci.chromium.org/buildbot/chromium.gpu.fyi/Mac%20FYI%20Experimental%20Retina%20Release%20%28NVIDIA%29/ I'm not sure whether it coincided with the OS upgrade -- looking more deeply, it doesn't look like it did -- but GLImageIOSurface/GLImageBindTest/4.BindTexImage is now failing reliably on this bot. Here's the first failing build I could find: https://ci.chromium.org/buildbot/chromium.gpu.fyi/Mac%20FYI%20Experimental%20Retina%20Release%20%28NVIDIA%29/14 and the current build fails too: https://ci.chromium.org/buildbot/chromium.gpu.fyi/Mac%20FYI%20Experimental%20Retina%20Release%20%28NVIDIA%29/43 Here's the failure log: [ RUN ] GLImageIOSurface/GLImageBindTest/4.BindTexImage ../../ui/gl/test/gl_test_helper.cc:77: Failure The difference between expected and actual is 32, which exceeds error, where expected evaluates to 48, actual evaluates to 16, and error evaluates to 0. at 0, 0 channel 0 ../../ui/gl/test/gl_test_helper.cc:77: Failure The difference between expected and actual is 32, which exceeds error, where expected evaluates to 16, actual evaluates to 48, and error evaluates to 0. at 0, 0 channel 2 I wonder whether Apple either fixed a bug or introduced one in this OS point release. Miguel, could you please take this? It's the only reason this bot is red right now so we'd like to get it green again somehow. Thanks.
,
Feb 20 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/58070cfbc6e878671375550cbe1582b5e70ce81a commit 58070cfbc6e878671375550cbe1582b5e70ce81a Author: Miguel Casas-Sanchez <mcasas@chromium.org> Date: Tue Feb 20 19:55:47 2018 fix GLImageIOSurface test failing in Mac 10.13.4 w/ Nvidia This CL tentatively fixes GLImageIOSurface/GLImageBindTest/4 on Nvidia + 10.13.4 Mac OS X systems; the error log points to the current workaround swizzling the R and B channels specifically on such system and OS version combination doing the wrong thing, so let's removes it and see if it does the trick. Bug: 813846 Change-Id: I013da016158860da05b1b2bf0c363d29afdbef2f 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 Reviewed-on: https://chromium-review.googlesource.com/926763 Reviewed-by: Kenneth Russell <kbr@chromium.org> Commit-Queue: Miguel Casas <mcasas@chromium.org> Cr-Commit-Position: refs/heads/master@{#537855} [modify] https://crrev.com/58070cfbc6e878671375550cbe1582b5e70ce81a/ui/gl/gl_image_io_surface_unittest.cc
,
Feb 21 2018
#2 turned the bot green: https://ci.chromium.org/buildbot/chromium.gpu.fyi/Mac%20FYI%20Experimental%20Retina%20Release%20%28NVIDIA%29/45
,
Feb 21 2018
Great! Glad the new macOS version fixed the previous bug. +dba as an FYI! |
|||
►
Sign in to add a comment |
|||
Comment 1 by mcasas@chromium.org
, Feb 20 2018