FS#40489 - [chromium] 35.0.1916.114-1 screen corruption on startup

Attached to Project: Arch Linux
Opened by jb (jb.1234abcd) - Thursday, 22 May 2014, 05:40 GMT
Last edited by Evangelos Foutras (foutrelis) - Thursday, 12 June 2014, 16:08 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Evangelos Foutras (foutrelis)
Architecture i686
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 3
Private No

Details

Description:
Chromium corrupts screen on startup.

Additional info:
* package version(s)
chromium] 35.0.1916.114-1

* config and/or log files etc.
$ lspci -v
00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Proces
sor to DRAM Controller (rev 03)
Subsystem: Hewlett-Packard Company NX6110/NC6120
Flags: bus master, fast devsel, latency 0
Capabilities: [e0] Vendor Specific Information: Len=09 <?>
Kernel driver in use: agpgart-intel
Kernel modules: intel_agp

00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Exp
ress Graphics Controller (rev 03) (prog-if 00 [VGA controller])
Subsystem: Hewlett-Packard Company NX6110/NC6120
Flags: bus master, fast devsel, latency 0, IRQ 16
Memory at d0100000 (32-bit, non-prefetchable) [size=512K]
I/O ports at 2000 [size=8]
Memory at c0000000 (32-bit, prefetchable) [size=256M]
Memory at d0180000 (32-bit, non-prefetchable) [size=256K]
Expansion ROM at <unassigned> [disabled]
Capabilities: [d0] Power Management version 2
Kernel driver in use: i915
Kernel modules: i915

Steps to reproduce:
Install chromium and start it.

Note: the same happens with Opera.
This task depends upon

Closed by  Evangelos Foutras (foutrelis)
Thursday, 12 June 2014, 16:08 GMT
Reason for closing:  Fixed
Additional comments about closing:  Fixed by the xf86-video-intel 2.99.912-1 update.
Comment by Doug Newgard (Scimmia) - Thursday, 22 May 2014, 05:46 GMT
Check this thread: https://bbs.archlinux.org/viewtopic.php?id=181823
No idea where the setting would be for Opera, though.
Comment by jb (jb.1234abcd) - Thursday, 22 May 2014, 16:49 GMT
I corrected both, Opera and Chromium corruptions, with "uxa" instead of "sna":
/etc/X11/xorg.conf.d/20-intel.conf

Section "Device"
Identifier "Intel Graphics"
Driver "intel"
Option "AccelMethod" "uxa"
EndSection

I do not know if Arch Chromium (and Opera) maintainer may be involved in sna v uxa problem too via package
building, so I would suggest passing this bug report to him just in case.

Also, Chromium shows this error on startup in xterm:
$ chromium
[5777:5777:0522/183451:ERROR:component_loader.cc(138)] Failed to parse extension manifest.
ATTENTION: default value of option force_s3tc_enable overridden by environment.
Comment by Mathieu Boespflug (mboes) - Sunday, 25 May 2014, 20:38 GMT
The uxa workaround isn't possible for me. At least on my hardware (MBP"13in retina), the screen hangs when X is started (single non-blinking cursor at top of black screen). sna works fine. Here's the content of lspci:

00:02.0 VGA compatible controller: Intel Corporation 3rd Gen Core processor Graphics Controller (rev 09)

xf86-video-intel 2.99.911-2
xorg-server 1.15.1-1

I can confirm the corruption reported in the forum thread. But none of the workarounds posted there work for me.
Comment by André Felício (antigerme) - Thursday, 29 May 2014, 13:48 GMT
See this post.

https://bbs.archlinux.org/viewtopic.php?pid=1417881#p1417881
Comment by Mathieu Boespflug (mboes) - Thursday, 29 May 2014, 13:58 GMT
That workaround from dfeng doesn't work for me.
Comment by jb (jb.1234abcd) - Wednesday, 11 June 2014, 12:40 GMT
As of today, no screen corruption - there was Chromium but no Opera update up to now, so it must have been fixed
thru graphics-related updates. Removed /etc/X11/xorg.conf.d/20-intel.conf .

That Chromium error on startup is still there.
Comment by Evangelos Foutras (foutrelis) - Thursday, 12 June 2014, 10:38 GMT
Are you using the [testing] repository? If so, it's likely that the fix was in the xf86-video-intel 2.99.912-1 update.
Comment by Mathieu Boespflug (mboes) - Thursday, 12 June 2014, 14:26 GMT
Confirmed. The problem has gone away for me as well, but only after enabling [testing].

Loading...