Issue metadata
Sign in to add a comment
|
[Nyan Family devices] Seen UI glitch and unable to login existing user accounts. |
|||||||||||||||||||||||||||||||||||||
Issue descriptionChrome Version: 60.0.3092.0 Chrome OS Version: M60-9534.0.0 Chrome OS Platform: nyan blaze Network info: Please specify Cr-* of the system to which this bug/feature applies (add the label below). Steps To Reproduce: (1)AUed to M60-9534.0.0 (2)Restart device after successful update (3) Expected Result: No UI glitch should be seen. Actual Result: Seen UI glitches and unable to login existing user accounts. How frequently does this problem reproduce? (Always, sometimes, hard to reproduce?) What is the impact to the user, and is there a workaround? If so, what is it? Please provide any additional information below. Attach a screen shot or log if possible. For graphics-related bugs, please copy/paste the contents of the about:gpu page at the end of this report.
,
May 9 2017
,
May 9 2017
+Dave and Daniele Since we haven't touched the nvidia drivers in forever, this is likely to be a chrome-side regression.
,
May 10 2017
Like Daniele's fence change. Can someone with access to one of these devices do a bisect and confirm that this change caused the problem?
,
May 11 2017
What is the change? we can trace back the version when landed and test before/after Alternatively, due severity of this issue we can also revert and try after that
,
May 11 2017
The suspected change is https://codereview.chromium.org/2858693002 Unfortunately, that can't be cleanly reverted, they already tried and failed.
,
May 16 2017
Daniele, are you working on alternate solution if revert is not an option?
,
May 16 2017
Daniele, are you working on alternate solution if revert is not an option?
,
May 16 2017
josafat, has anyone verified that specific change is causing the issue?
,
May 23 2017
Commit was initially landed in 60.0.3088.0 which was initially revved in 9524.0.0 abodeti@, can you guys see if repro with 9523.0.0/60.0.3086.0?
,
May 23 2017
Issue is not reproducible with 9523.0.0/60.03086.0. UI looks fine and was able to login
,
May 24 2017
Reprod on Chrome OS R60-9582.0.0 / 60.0.3107.0
,
May 30 2017
Based on c#10/c#11, it seems like the suspected CL may be the cause since no repro before that dcastagna/reveman@ any recommendation on how to proceed here?
,
May 30 2017
Still reproducible in ChromeOS: 9592.0.0, 60.0.3112.0.
,
May 30 2017
Has anyone verified that this change caused this to break? Looking at these screenshots I have a hard time seeing how that could be the case.
,
May 30 2017
abodeti@, Is this repro in 9524.0.0?
,
May 30 2017
Rep#17. yes, reproduced on cros:9524.0.0
,
May 31 2017
Based on testing results we know the regression started between https://chromium.googlesource.com/chromium/src/+log/60.0.3086.0..60.0.3088.0?pretty=fuller&n=10000 Here is where the suspected CL landed: https://codereview.chromium.org/2858693002 dcastagna@, Are there any other suspected CLs in the list?
,
May 31 2017
josafat@, there a few CLs touching ozone code in that range. I'm not excluding https://codereview.chromium.org/2858693002 could be the culprit. We don't have a nyan blaze here in NYC. It'd be great if we could confirm that the suspect CL is actually the issue before having one shipped to us. If it can help, I can put up a Chromium CL that simulates the old behavior.
,
Jun 1 2017
This issue is also observed on chrome device Kitty with M60 60.0.3112.10 / 9592.3.0 dev. Updating the bug Summary.
,
Jun 8 2017
Still reproducible in 9592.13.0 60.0.3112.23 on blaze which blocks full release testing on blaze.
,
Jun 9 2017
Issue 731684 has been merged into this issue.
,
Jun 9 2017
Stéphane, anyone who could help debug this in MTV? here is the suspected CL -> https://codereview.chromium.org/2858693002 and the range: https://chromium.googlesource.com/chromium/src/+log/60.0.3086.0..60.0.3088.0?pretty=fuller&n=10000
,
Jun 9 2017
This is a Chrome regression, it should probably be handled by Chrome folks. The best I could do is a revert here...
,
Jun 15 2017
,
Jun 15 2017
,
Jun 16 2017
Could anyone try to apply https://chromium-review.googlesource.com/c/538599/ and see if the problem goes away? Unfortunately we don't have an Nvidia device here. We suspect that we were incorrectly relying on calling glFlush before submitting a buffer.
,
Jun 16 2017
abodeti@ tried https://chromium-review.googlesource.com/c/538599/ and confirmed the UI issues go away with that patch. We can land that while we investigate why eglImageFlushExternalEXT and eglClientWaitSyncKHR are not working as expected.
,
Jun 16 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/f388d9e52d945a112d57890f83a207e54d10721f commit f388d9e52d945a112d57890f83a207e54d10721f Author: Daniele Castagna <dcastagna@chromium.org> Date: Fri Jun 16 23:21:09 2017 ozone: Call glFlush before submitting a buffer. Recently we removed glFlush from gbm_surfacless before submitting a buffer, since relying completely on eglImageFlushExternalEXT and eglClientWaitSyncKHR should suffice. Unfortunately UI artifacts showed up on nyan boards after that change. This CL re-adds a glFlush before submitting a buffer. Bug: 720045 Change-Id: I1a6f5fc143fcb2d9c15e177de2b4914d383e7d44 Reviewed-on: https://chromium-review.googlesource.com/538599 Reviewed-by: Robert Kroeger <rjkroege@chromium.org> Reviewed-by: David Reveman <reveman@chromium.org> Commit-Queue: Daniele Castagna <dcastagna@chromium.org> Cr-Commit-Position: refs/heads/master@{#480216} [modify] https://crrev.com/f388d9e52d945a112d57890f83a207e54d10721f/ui/ozone/platform/drm/gpu/gbm_surfaceless.cc
,
Jun 17 2017
,
Jun 17 2017
This bug requires manual review: M60 has already been promoted to the beta branch, so this requires manual review Please contact the milestone owner if you have questions. Owners: amineer@(Android), cmasso@(iOS), josafat@(ChromeOS), bustamante@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Jun 17 2017
https://chromium-review.googlesource.com/538277 is probably the proper fix that doesn't require a glFlush. I'll test that next week. I'd suggest to still proceed and merge https://chromium-review.googlesource.com/538599 to 60, while we test a better solution on ToT.
,
Jun 19 2017
+1,, Considering M60 is on Beta now I'd prefer to go with the more conservative option and do the proper fix on M61 (providing more testing will be required there) Approving https://chromium-review.googlesource.com/538599 for merge to M60
,
Jun 21 2017
#CBC-RS/TC-watchlist
,
Jun 21 2017
Still reproducible in ChromeOS: 9592.29.0,60.0.3112.41 on both big and blaze.
,
Jun 21 2017
https://codereview.chromium.org/2945893002/ is the merge to M60, so I'd expect this issue not to be in 60.0.3112.31. hoegsberg@ tried ToT with the fix and confirmed it worked fine. helenzhang@: Can you post a screenshot to confirm what you're observing is the same issue?
,
Jun 21 2017
I can confirm that ChromeOS ToT (R61 9669) with a git master chromium (5695df5c6ce500881e7a8eacb36f1425e5fd0f98) works here. I did see the artifacts before on R61, but it's gone now. The nature of the artifact was consistent with the fix here. Are you sure you're running the right build of chrome?
,
Jun 21 2017
per #38, reproducible with 9592.29.0,60.0.3112.41 (so it sounds like https://chromium-review.googlesource.com/538599 worked, whereas https://codereview.chromium.org/2945893002/ did not).
,
Jun 21 2017
,
Jun 21 2017
the repros in c#29 and c#40 look like they may be related to crbug 734074 instead of this one
,
Jun 22 2017
@44, that seems reasonable. All the patches to fix 734074 just landed in 59 and 60 today. Let's wait for https://codereview.chromium.org/2951193002 to make it into cros and we shouldn't observe similar issues at that point.
,
Jun 22 2017
Both big and blaze are good in Chrome OS 9675.0.0, 61.0.3136.5.
,
Jun 22 2017
,
Jun 23 2017
This issue has been approved for a merge. Please merge the fix to any appropriate branches as soon as possible! If all merges have been completed, please remove any remaining Merge-Approved labels from this issue. Thanks for your time! To disable nags, add the Disable-Nags label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Jun 23 2017
,
Jun 24 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/4e06ceca760fefd5a551237cb57d6aeb8f48c03a commit 4e06ceca760fefd5a551237cb57d6aeb8f48c03a Author: David Reveman <reveman@chromium.org> Date: Sat Jun 24 03:26:11 2017 ui: Use EGL_EXT_image_flush_external correctly when creating EGLImage. EGL_IMAGE_EXTERNAL_FLUSH_EXT attribute should be set to EGL_TRUE when creating an image that support eglImageFlushExternalEXT. BUG= 720045 TEST=gl_unittests 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 Change-Id: I7e01f5cdb964da5c1599a0fe62efb8440b765852 Reviewed-on: https://chromium-review.googlesource.com/538277 Reviewed-by: Kenneth Russell <kbr@chromium.org> Reviewed-by: Robert Kroeger <rjkroege@chromium.org> Reviewed-by: Daniele Castagna <dcastagna@chromium.org> Commit-Queue: David Reveman <reveman@chromium.org> Cr-Commit-Position: refs/heads/master@{#482134} [modify] https://crrev.com/4e06ceca760fefd5a551237cb57d6aeb8f48c03a/third_party/khronos/EGL/eglext.h [modify] https://crrev.com/4e06ceca760fefd5a551237cb57d6aeb8f48c03a/ui/gl/gl_image_native_pixmap.cc [modify] https://crrev.com/4e06ceca760fefd5a551237cb57d6aeb8f48c03a/ui/gl/test/gl_image_test_template.h [modify] https://crrev.com/4e06ceca760fefd5a551237cb57d6aeb8f48c03a/ui/ozone/gl/gl_image_ozone_native_pixmap_unittest.cc
,
Jun 28 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/24d239973fefa7bfd729aac06898e88ce92c90bb commit 24d239973fefa7bfd729aac06898e88ce92c90bb Author: David Reveman <reveman@chromium.org> Date: Wed Jun 28 16:36:28 2017 Revert "ui: Use EGL_EXT_image_flush_external correctly when creating EGLImage." This reverts commit 4e06ceca760fefd5a551237cb57d6aeb8f48c03a. Reason for revert: Causing GLImage create failure on Minnie. BUG= 737235 Original change's description: > ui: Use EGL_EXT_image_flush_external correctly when creating EGLImage. > > EGL_IMAGE_EXTERNAL_FLUSH_EXT attribute should be set to EGL_TRUE > when creating an image that support eglImageFlushExternalEXT. > > BUG= 720045 > TEST=gl_unittests > > 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 > Change-Id: I7e01f5cdb964da5c1599a0fe62efb8440b765852 > Reviewed-on: https://chromium-review.googlesource.com/538277 > Reviewed-by: Kenneth Russell <kbr@chromium.org> > Reviewed-by: Robert Kroeger <rjkroege@chromium.org> > Reviewed-by: Daniele Castagna <dcastagna@chromium.org> > Commit-Queue: David Reveman <reveman@chromium.org> > Cr-Commit-Position: refs/heads/master@{#482134} TBR=rjkroege@chromium.org,reveman@chromium.org,kbr@chromium.org,dcastagna@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: 720045 Change-Id: I9ffde8b9f48de0d7fc2d6288aad8bc823705120a 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/551181 Reviewed-by: David Reveman <reveman@chromium.org> Reviewed-by: Kenneth Russell <kbr@chromium.org> Reviewed-by: Robert Kroeger <rjkroege@chromium.org> Commit-Queue: David Reveman <reveman@chromium.org> Cr-Commit-Position: refs/heads/master@{#483014} [modify] https://crrev.com/24d239973fefa7bfd729aac06898e88ce92c90bb/third_party/khronos/EGL/eglext.h [modify] https://crrev.com/24d239973fefa7bfd729aac06898e88ce92c90bb/ui/gl/gl_image_native_pixmap.cc [modify] https://crrev.com/24d239973fefa7bfd729aac06898e88ce92c90bb/ui/gl/test/gl_image_test_template.h [modify] https://crrev.com/24d239973fefa7bfd729aac06898e88ce92c90bb/ui/ozone/gl/gl_image_ozone_native_pixmap_unittest.cc
,
Jun 29 2017
Issue 738206 has been merged into this issue.
,
Jun 30 2017
Both big and blaze are bad in 9701.0.0/61.0.3142.0. It looks like #29 or #40.
,
Jul 7 2017
I just built 61.0.3151.0 on nyan_big (Acer CB5-311 with 1920x1080) and put it on an USB stick. If I boot from that stick in developer mode, I can configure network, account, start a browser etc. - all no problem. If I either - click on "enable debugging" and reboot - use a VT and do /usr/sbin/chromeos-install and reboot - or simply switch off the device and boot again I get graphical glitches or a blinking screen with diagonal and vertical lines. The only (corresponding) entry in the log is ERROR:drm_display.cc(179) Failed to set color correction for display: crtc_id = 5
,
Jul 12 2017
Issue 739266 has been merged into this issue.
,
Jul 13 2017
This issue is happening again in ToT. In some way it regressed again.
,
Jul 14 2017
,
Jul 14 2017
Reassigning to Kristian since he said he might have a few cycles to determine why/how it regressed again. Since the regression happened only in ToT, we should consider merging f388d9e52d945a112d57890f83a207e54d10721f to 60 and fix beta, in the meanwhile we can figure out the issue on ToT.
,
Jul 17 2017
Adding M-61 as the issue seems to show in Tot now For M-60, changes were already approved for merge, what are the other changes needed?
,
Jul 18 2017
The change that re-added glFlush made it before the branch (my bad in #35 for suggesting it needed to be merged) and is already in M-60: https://chromium.googlesource.com/chromium/src/+log/60.0.3112.71/ui/ozone/platform/drm/gpu/gbm_surfaceless.cc AFAIU that should be enough to fix the first regression. I'm not sure if we regressed again before or after the branch. Can anyone check if we still have issues on nyan on 60? If, not, we should focus on fixing the issue on 61.
,
Jul 18 2017
,
Jul 18 2017
I tested chrome 60.0.3112.71 on a Nyan Blaze and Nyan Big and both work correctly. R61 images have the flicker artifact.
,
Jul 20 2017
Removing M60 as per verification in c#62 Leaving bug open for M-61
,
Jul 21 2017
Kristian tried to revert https://chromium-review.googlesource.com/546117 (gpu: Disable EGL_EXT_image_flush_external on NVIDIA.) and the issue seems to go away. Maybe EGL_EXT_image_flush_external was not completely broken after all?
,
Jul 21 2017
Sounds like we just need to change the workaround so it allows the extension to be used incorrectly on these devices. Thanks for testing. I'll create a patch asap.
,
Jul 21 2017
,
Jul 25 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/b12d845345769e78046ba8cca67953ad63f07f6e commit b12d845345769e78046ba8cca67953ad63f07f6e Author: David Reveman <reveman@chromium.org> Date: Tue Jul 25 18:03:28 2017 gpu: Adjust EGL_EXT_image_flush_external workaround on NVIDIA. Instead of disabling the extension completely, allow it to be used without creating EGLImages with correct attributes. This will allow us to enable correct use of this extension in a follow patch. BUG= 720045 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 Change-Id: Ib3d9093a6d319f6b1a62e3f5e313610665e30ab8 Reviewed-on: https://chromium-review.googlesource.com/581229 Reviewed-by: Antoine Labour <piman@chromium.org> Commit-Queue: David Reveman <reveman@chromium.org> Cr-Commit-Position: refs/heads/master@{#489366} [modify] https://crrev.com/b12d845345769e78046ba8cca67953ad63f07f6e/gpu/config/gpu_driver_bug_list.json [modify] https://crrev.com/b12d845345769e78046ba8cca67953ad63f07f6e/gpu/config/gpu_driver_bug_workaround_type.h
,
Aug 4 2017
Ping for an update; we're trying to close out M61 Beta blockers; thanks bunches.
,
Aug 8 2017
Per reveman on IM: that last patch should have fixed the issue. Maybe we need a merge if it didn't land before the branch cut
,
Aug 8 2017
M61 branch position is #488528 so that fix needs to be merged.
,
Aug 8 2017
This bug requires manual review: Reverts referenced in bugdroid comments after merge request. Please contact the milestone owner if you have questions. Owners: amineer@(Android), cmasso@(iOS), ketakid@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Aug 8 2017
Approving merge to M61 Chrome OS.
,
Aug 8 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/7e89d0d4e2ab80e92dbaf0fdcec414d3d8a2c8ce commit 7e89d0d4e2ab80e92dbaf0fdcec414d3d8a2c8ce Author: David Reveman <reveman@chromium.org> Date: Tue Aug 08 23:54:49 2017 gpu: Adjust EGL_EXT_image_flush_external workaround on NVIDIA. Instead of disabling the extension completely, allow it to be used without creating EGLImages with correct attributes. This will allow us to enable correct use of this extension in a follow patch. BUG= 720045 TBR=reveman@chromium.org (cherry picked from commit b12d845345769e78046ba8cca67953ad63f07f6e) 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 Change-Id: Ib3d9093a6d319f6b1a62e3f5e313610665e30ab8 Reviewed-on: https://chromium-review.googlesource.com/581229 Reviewed-by: Antoine Labour <piman@chromium.org> Commit-Queue: David Reveman <reveman@chromium.org> Cr-Original-Commit-Position: refs/heads/master@{#489366} Reviewed-on: https://chromium-review.googlesource.com/607547 Reviewed-by: David Reveman <reveman@chromium.org> Cr-Commit-Position: refs/branch-heads/3163@{#396} Cr-Branched-From: ff259bab28b35d242e10186cd63af7ed404fae0d-refs/heads/master@{#488528} [modify] https://crrev.com/7e89d0d4e2ab80e92dbaf0fdcec414d3d8a2c8ce/gpu/config/gpu_driver_bug_list.json [modify] https://crrev.com/7e89d0d4e2ab80e92dbaf0fdcec414d3d8a2c8ce/gpu/config/gpu_driver_bug_workaround_type.h
,
Aug 8 2017
,
Jan 22 2018
|
||||||||||||||||||||||||||||||||||||||
►
Sign in to add a comment |
||||||||||||||||||||||||||||||||||||||
Comment 1 by pgangishetty@google.com
, May 9 2017770 KB
770 KB View Download
72.7 KB
72.7 KB View Download