Issue metadata
Sign in to add a comment
|
Serious external display issues on Kevin R67 |
||||||||||||||||||||||
Issue descriptionChrome Version: 67.0.3396.78 (Official Build) beta (32-bit) OS: 10575.54.0 (Official Build) beta-channel kevin My Kevin just got AUed to R67 stable and ever since the external display support seems to be completely broken. When I connect my external display, it says "Mirroring to..." or "Extending to..." with the correct display name, but nothing ever shows up on the external screen. If I then unplug the display again, my internal screen flickers a bunch of times and *stays black*. If I try to reboot (e.g. Refresh+Power), I can see the login screen, but after login my display goes black again and stays there! So somehow this does something so bad to my profile that it survives a logout. Restarting Chrome with Alt+VolUp+X or closing the lid to suspend/resume seems to fix that state. I'm using an HP ZR30w monitor and an ASUS Type-C hub with an HDMI port that I've been using for a year and that have always worked perfectly fine before. This is a very serious regression that essentially prevents me from running my normal work setup, and it is on stable channel. Here's a feedback report taken right after trying this (and force-rebooting) a couple of times: https://feedback.corp.google.com/product/208/neutron?lView=rd&lReport=85501593763
,
Jun 15 2018
Actually, I'm not sure if the black screen after login is even related to the external display. It just seems to happen randomly (which is worse).
,
Jun 15 2018
I definitely noticed the black screen after login this morning. I had to suspend/resume and re-login to get things working. It's _possible_ that the black screen at login might be seen only by people with non-standard chrome://flags. I think people who work at Google might have some chrome://flags pushed to their work logins and I do know that changing chrome://flags causes an extra restart of Chrome when you login... I haven't tested that though.
,
Jun 15 2018
Yeah, I know the problem with flags. I explicitly went to chrome://flags and clicked "Reset all to default". Didn't fix it (in fact it happened again after pressing "Relaunch Now" from that page...).
,
Jun 15 2018
I confirmed that the "blinking" black I've been seeing for a little while now only happens on corp. ...what I saw this morning was that it stayed black for a long period of time. That was new for me. I can't reproduce that anymore now though. ...but I guess, as you said, the more serious problem is that your external display seems broken. Any chance that has to do with all the recent DP training fixes we've made for scarlet?
,
Jun 15 2018
Is this the same bug? https://issuetracker.google.com/109782243
,
Jun 15 2018
Yeah, that sounds likely to be the same issue. Unfortunately, it seems like this has already rolled out to Kevin stable so the ReleaseBlock-Stable tag didn't help anymore. Is there an easy fix or rollback we can get into the next stable push? This really isn't something that we can make users put up with for 6 weeks...
,
Jun 15 2018
It looks like people are debating this one: https://chromium-review.googlesource.com/c/chromium/src/+/1100666 It's not clear whether an alternative is on the horizon, or if that will be the solution for now...
,
Jun 15 2018
Also, IIUC M67 is *not* on stable yet. You're probably on beta channel.
,
Jun 18 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/76904d165231ce06a6f59a4d9cb3b0f04b25e6f2 commit 76904d165231ce06a6f59a4d9cb3b0f04b25e6f2 Author: Eliot Courtney <edcourtney@google.com> Date: Mon Jun 18 16:15:24 2018 ozone/drm: Disable AFBC for modeset buffers AFBC for modeset buffers doesn't work correctly, as we can't fill them with valid AFBC contents. For now, don't use AFBC for modeset buffers. Remove this after we make ozone/drm capable of creating valid AFBC modeset buffers (tracked by crbug.com/852675 ). Bug: 852675 , 853328 , b/109782243 Test: External monitor works on Kevin Change-Id: I902f2615aa78e34251381f87e25c5e76f39b94b5 Reviewed-on: https://chromium-review.googlesource.com/1100666 Commit-Queue: Daniele Castagna <dcastagna@chromium.org> Reviewed-by: Daniele Castagna <dcastagna@chromium.org> Cr-Commit-Position: refs/heads/master@{#568024} [modify] https://crrev.com/76904d165231ce06a6f59a4d9cb3b0f04b25e6f2/ui/ozone/platform/drm/gpu/DEPS [modify] https://crrev.com/76904d165231ce06a6f59a4d9cb3b0f04b25e6f2/ui/ozone/platform/drm/gpu/hardware_display_controller.cc [modify] https://crrev.com/76904d165231ce06a6f59a4d9cb3b0f04b25e6f2/ui/ozone/platform/drm/gpu/hardware_display_controller.h [modify] https://crrev.com/76904d165231ce06a6f59a4d9cb3b0f04b25e6f2/ui/ozone/platform/drm/gpu/screen_manager.cc
,
Jun 18 2018
Issue 853301 has been merged into this issue.
,
Jun 18 2018
,
Jun 18 2018
> Also, IIUC M67 is *not* on stable yet. You're probably on beta channel. Right, actually, the version that I pasted in comment #0 also says that. I thought I was on stable because go/cros-schedule lists beta as M68 already. I'm surprised that this only just happened, though, I must have been on M67 beta for almost 6 weeks already. Did we merge something late into M67 that broke this?
,
Jun 18 2018
,
Jun 19 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/19bf5dc1601074925fe0aded324ee5ab2956e4a5 commit 19bf5dc1601074925fe0aded324ee5ab2956e4a5 Author: Eliot Courtney <edcourtney@google.com> Date: Tue Jun 19 04:17:51 2018 ozone/drm: Disable AFBC for modeset buffers AFBC for modeset buffers doesn't work correctly, as we can't fill them with valid AFBC contents. For now, don't use AFBC for modeset buffers. Remove this after we make ozone/drm capable of creating valid AFBC modeset buffers (tracked by crbug.com/852675 ). Bug: 852675 , 853328 , b/109782243 Test: External monitor works on Kevin Change-Id: I902f2615aa78e34251381f87e25c5e76f39b94b5 Reviewed-on: https://chromium-review.googlesource.com/1100666 Commit-Queue: Daniele Castagna <dcastagna@chromium.org> Reviewed-by: Daniele Castagna <dcastagna@chromium.org> Cr-Original-Commit-Position: refs/heads/master@{#568024}(cherry picked from commit 76904d165231ce06a6f59a4d9cb3b0f04b25e6f2) Reviewed-on: https://chromium-review.googlesource.com/1105538 Reviewed-by: Eliot Courtney <edcourtney@chromium.org> Cr-Commit-Position: refs/branch-heads/3440@{#439} Cr-Branched-From: 010ddcfda246975d194964ccf20038ebbdec6084-refs/heads/master@{#561733} [modify] https://crrev.com/19bf5dc1601074925fe0aded324ee5ab2956e4a5/ui/ozone/platform/drm/gpu/DEPS [modify] https://crrev.com/19bf5dc1601074925fe0aded324ee5ab2956e4a5/ui/ozone/platform/drm/gpu/hardware_display_controller.cc [modify] https://crrev.com/19bf5dc1601074925fe0aded324ee5ab2956e4a5/ui/ozone/platform/drm/gpu/hardware_display_controller.h [modify] https://crrev.com/19bf5dc1601074925fe0aded324ee5ab2956e4a5/ui/ozone/platform/drm/gpu/screen_manager.cc
,
Jun 19 2018
DUP of b/109782243? Assume so per #7, but if we can confirm to track in one place. Please confirm ASAP.
,
Jun 19 2018
Highly likely a duplicate. Probably good to track on b/109782243, where verification is already happening.
,
Jun 19 2018
Yes, call this a dup for now. I can't really check whether any other issues remain besides that (e.g. black screen on login) until that fix has rolled out to me. I'll open a new bug in that case.
,
Jun 20 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/f97cb28f57b695fa9195e2f51eff01c239fca5bc commit f97cb28f57b695fa9195e2f51eff01c239fca5bc Author: Eliot Courtney <edcourtney@google.com> Date: Wed Jun 20 01:30:49 2018 ozone/drm: Disable AFBC for modeset buffers AFBC for modeset buffers doesn't work correctly, as we can't fill them with valid AFBC contents. For now, don't use AFBC for modeset buffers. Remove this after we make ozone/drm capable of creating valid AFBC modeset buffers (tracked by crbug.com/852675 ). Bug: 852675 , 853328 , b/109782243 Test: External monitor works on Kevin Change-Id: I902f2615aa78e34251381f87e25c5e76f39b94b5 Reviewed-on: https://chromium-review.googlesource.com/1100666 Commit-Queue: Daniele Castagna <dcastagna@chromium.org> Reviewed-by: Daniele Castagna <dcastagna@chromium.org> Cr-Original-Commit-Position: refs/heads/master@{#568024}(cherry picked from commit 76904d165231ce06a6f59a4d9cb3b0f04b25e6f2) Reviewed-on: https://chromium-review.googlesource.com/1105717 Reviewed-by: Eliot Courtney <edcourtney@chromium.org> Cr-Commit-Position: refs/branch-heads/3396@{#782} Cr-Branched-From: 9ef2aa869bc7bc0c089e255d698cca6e47d6b038-refs/heads/master@{#550428} [modify] https://crrev.com/f97cb28f57b695fa9195e2f51eff01c239fca5bc/ui/ozone/platform/drm/gpu/DEPS [modify] https://crrev.com/f97cb28f57b695fa9195e2f51eff01c239fca5bc/ui/ozone/platform/drm/gpu/hardware_display_controller.cc [modify] https://crrev.com/f97cb28f57b695fa9195e2f51eff01c239fca5bc/ui/ozone/platform/drm/gpu/hardware_display_controller.h [modify] https://crrev.com/f97cb28f57b695fa9195e2f51eff01c239fca5bc/ui/ozone/platform/drm/gpu/screen_manager.cc |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by diand...@chromium.org
, Jun 15 2018