Issue metadata
Sign in to add a comment
|
"Context3D not available" on Chrome OS Flash Player
Reported by
timok...@gmail.com,
Apr 19 2017
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2987.133 Safari/537.36 Steps to reproduce the problem: 1. Open a site with Flash content on ChromeOS 57.0.2987.146 and Flash Player 25.0.0.148 What is the expected behavior? The Flash content plays. What went wrong? On some ChromeOS devices (reporoduced on an Asus Chromebit) the Flash fails and reports "Context3D not available". Did this work before? Yes I'm assuming version prior to 57.0.2987.146, because the issues started after this version got published Does this work in other browsers? Yes Chrome version: 57.0.2987.133 Channel: stable OS Version: OS X 10.10.5 Flash Version: Shockwave Flash 25.0 r0 More information here: https://forums.adobe.com/message/9450125 https://forum.starling-framework.org/topic/context3d-not-available-on-chrome-os-since-yesterday
,
Apr 21 2017
The issue is on Chrome OS. I've submitted the issue on Mac, hence the Mac information, sorry about that...
,
Apr 21 2017
Thank you for providing more feedback. Adding requester "sureshkumari@chromium.org" to the cc list and removing "Needs-Feedback" label. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Apr 24 2017
,
Apr 24 2017
Hi, I have some details from a device on which I reproduced the issue. Here's a Google Photos album which contains: https://goo.gl/photos/nshMFtn41qDx2JB67 - Pics of the "About Chrome OS" screen & build info (57.0.2987.146) - Flash version (25.0.0.148) - Samsung XE303C12 Chromebook - A video of refreshing a couple times, showing it's intermittent (randomly works, randomly fails) I notice that both this Chromebook and the Chromebit mentioned above are ARM devices with Mali GPU's (and this issue seems to be GPU-related.) Workaround in the meanwhile: the problem is intermittent and refreshing a few times may get you in a working state.
,
Apr 26 2017
Another note: on my test Samsung 303C device, I obtained it when it was Chrome 53.0.2785.154, and didn't see the error. I updated to 57.0.2987.146 and started seeing the error. Perhaps unrelated, but there is a difference in the Chrome flags related to PPAPI GPU between the two builds.
,
Apr 26 2017
Vinayak, could you see if there is a repro? This seems to affect various EDU sites like https://www.wootmath.com/, but I can't recognize the exact link in the video. Jeff, thanks for keeping track of the flags in the side-by-side picture! This change is harmless. But if you could attach the output of about:gpu and about:flash when that problem happens, that could help narrow the issue.
,
Apr 26 2017
Thanks for the reply. The feathers UI demo exhibits the problem (screenshot attached). Just like in the WootMath video, I refresh a few times and sometimes it works, sometimes it doesn't: https://feathersui.com/examples/transitions-explorer/ As for the log outputs, I pasted them here: about:gpu --> http://hastebin.com/ixixacerid.pas about:flash --> http://hastebin.com/azitesiwaq.pas Let me know if I can provide any other details. Thanks, -Jeff
,
Apr 26 2017
Nothing interesting in about:flash. Attaching about:gpu so we don't lose it. It is full of errors like: 11441:11441:0426/095445.137463:ERROR:gles2_cmd_decoder.cc(2459)] : [GroupMarkerNotSet( crbug.com/242999 )!:6823FAB8]GL ERROR :GL_INVALID_FRAMEBUFFER_OPERATION : BackFramebuffer::Create: <- error from previous GL command [11441:11441:0426/095445.144384:ERROR:texture_manager.cc(3224)] : [.DisplayCompositor-0xb8627900]GL ERROR :GL_INVALID_FRAMEBUFFER_OPERATION : glTexImage2D: <- error from previous GL command [11441:11458:0426/095445.586725:VERBOSE1:drm_util.cc(204)] : Unable to get cursor width capability: Invalid argument [11441:11458:0426/095445.587013:VERBOSE1:drm_util.cc(337)] : Failed to get EDID blob for connector 20 [11441:11458:0426/095445.596700:VERBOSE1:drm_display.cc(102)] : DRM configuring: device=/sys/devices/platform/exynos-drm/drm/card0 crtc=27 connector=20 origin=0,0 size=1366x768 [11441:11458:0426/095445.597683:VERBOSE1:drm_util.cc(204)] : Unable to get cursor width capability: Invalid argument [11441:11441:0426/100604.957444:ERROR:gles2_cmd_decoder.cc(2459)] : [GroupMarkerNotSet( crbug.com/242999 )!:681EFAB8]GL ERROR :GL_INVALID_FRAMEBUFFER_OPERATION : BackFramebuffer::Create: <- error from previous GL command
,
May 2 2017
Vinayak, did you have a chance to repro?
,
May 2 2017
FYI, this affects any Flash content running with Stage3D (GPU acceleration), including Facebook Angry Birds Friends. I've added a video of Angry Birds (refreshing to show how it's intermittent) to the shared album: https://goo.gl/photos/nshMFtn41qDx2JB67 Let me know if it would help to ship you this device, Samsung XE303C12.
,
May 2 2017
Joseph, could you please try to repro this issue?
,
May 2 2017
issue reproduce on 9460.20.0/59.0.3071.33 with Kevin device while launching Facebook Angry Birds Friends game. Feedback: Report ID: 59223276363
,
May 3 2017
The original report is for daisy, but apparently also kevin. Both ARM Chromebooks. I couldn't repro on alex or sentry, both Intel Chromebooks. I do have daisy and kevin available, so will try next here.
,
May 3 2017
Repro on minnie. Not related to HBD. Flash doesn't really doing anything to the wmode, which is provided by the website. The string is provided by the app, which probably is starling based. Maybe it is failing to get 3D? But I don't see a serious graphics failure. That said I don't see this repro with the debug build, only release builds (even -O0). A race of sorts?
,
May 3 2017
This readily reproduces with 23.0.0.207 and 21.0.0.216 binaries from the archive (from one year ago), so it is not a change within the pepper-flash binary that triggers the issue. I will try a really old Chrome OS image next.
,
May 4 2017
Daisy R57-9153.0.0 with 24.0.0.186 shows the issue on Angry Birds. Daisy R47-7482.0.0 with 19.0.0.185 does not (separate DUTs, same account, same time).
,
May 5 2017
It looks like a regression in the Chrome binary between R56 and R57.
,
May 8 2017
Robert, please bisect as discussed.
,
May 9 2017
According to my observation, the issue can be reproduced in build 9033.0.0 (in ~50% cases), but it's not reproducible in version 9032.0.0. Both versions use same version of chrome: 57.0.2933.0. According to ihf@, the only major difference between two these versions is a different gfx driver. Attaching (about:gpu) for both versions of chrome OS.
,
Jun 5 2017
Is there a status update for this issue? Need to update customers regarding it.
,
Jun 29 2017
This issue is impacting Woot Math students on affected Chromebooks. Could you please provide a status update? Last reproduced on build 59.0.3071.113 Thank you. Krista Marks, CEO Woot Math
,
Jul 19 2017
What can we do to resolve this issue? There are thousands of students that are affected in our service, and we have no workaround at the moment. What do we do?
,
Jul 21 2017
The issue is reasonably well understood (starling not well behaved, graphics driver maybe a bit too strict). We will need to update the ARM graphics driver. This has been taking longer than expected, but will hopefully happen soon.
,
Jul 25 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/vendor/mali-ddk/+/9be91c854714f810c74d0e5e513ade92fa9ed795 commit 9be91c854714f810c74d0e5e513ade92fa9ed795 Author: Anders Frostad Pedersen <anders.pedersen@arm.com> Date: Tue Jul 25 03:32:47 2017
,
Jul 25 2017
Works fine with Minnie and AB now. We need this in M60 to make the new school year.
,
Jul 25 2017
This bug requires manual review: Request affecting a post-stable build 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
,
Jul 25 2017
,
Jul 26 2017
Your change meets the bar and is auto-approved for M61. Please go ahead and merge the CL to branch 3163 manually. Please contact 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
,
Jul 26 2017
,
Jul 26 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/vendor/mali-ddk/+/b7d8dd0f810385fc282d07a1f74278642c15e83b commit b7d8dd0f810385fc282d07a1f74278642c15e83b Author: Anders Frostad Pedersen <anders.pedersen@arm.com> Date: Wed Jul 26 22:43:40 2017
,
Jul 26 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/vendor/mali-ddk/+/d8d029519e3a411c1a810b866abb52a16b04d90e commit d8d029519e3a411c1a810b866abb52a16b04d90e Author: Anders Frostad Pedersen <anders.pedersen@arm.com> Date: Wed Jul 26 22:43:43 2017
,
Jul 26 2017
Not sure if we still need to bump the ebuild. Checking.
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-veyron-private/+/779910e4be97448079379021d2bde814ae3c511c commit 779910e4be97448079379021d2bde814ae3c511c Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:16:38 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-peach-private/+/d1a0cc6ba5a780a6f03be1036121f3cbdd4473ad commit d1a0cc6ba5a780a6f03be1036121f3cbdd4473ad Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:16:50 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-daisy-private/+/80929b0566ebab60e362a77592b0b8b4d80e1db6 commit 80929b0566ebab60e362a77592b0b8b4d80e1db6 Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:17:00 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/baseboard-gru-private/+/828ae6e668a7739c5f4a597ea8cb3591efa96116 commit 828ae6e668a7739c5f4a597ea8cb3591efa96116 Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:17:13 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/baseboard-gru-private/+/5173e9fe3034ed602fc31dee887cc42516fe3b23 commit 5173e9fe3034ed602fc31dee887cc42516fe3b23 Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:17:56 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-peach-private/+/d9a2baee9e97fc61322b1bf65640346fb28394da commit d9a2baee9e97fc61322b1bf65640346fb28394da Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:18:26 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-veyron-private/+/cd2c62f8f0c97989556a38ec1df74f6fdbe7c4d7 commit cd2c62f8f0c97989556a38ec1df74f6fdbe7c4d7 Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:18:58 2017
,
Jul 28 2017
The following revision refers to this bug: https://chrome-internal.googlesource.com/chromeos/overlays/overlay-daisy-private/+/cccbe64497719e88ca82051be1312548c7dba1ce commit cccbe64497719e88ca82051be1312548c7dba1ce Author: Ilja H. Friedel <ihf@chromium.org> Date: Fri Jul 28 04:19:45 2017
,
Jul 28 2017
,
Jul 28 2017
Should be fixed on R60 with 9592.73.0, R61 with 9765.10.0 and R62 with 9777.0.0.
,
Aug 11 2017
Verified on 9592.82.0/60.3112.101 and 9765.21.0/61.0.3163.38
,
Sep 8 2017
Thanks for getting this fixed for the start of the school year!
,
Sep 11 2017
Indeed, great to see this taken care of, thanks heaps! :)
,
Sep 11 2017
No problem, thank you for confirming! |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by sureshkumari@chromium.org
, Apr 21 2017Labels: Needs-Feedback