| Summary: | [915GM] 'xrandr' with VGA plugged in garbles screen, fifo underrun | ||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | xorg | Reporter: | Bryce Harrington <bryce> | ||||||||||||||||||||||||||||||
| Component: | Driver/intel | Assignee: | ykzhao <yakui.zhao> | ||||||||||||||||||||||||||||||
| Status: | RESOLVED WORKSFORME | QA Contact: | Xorg Project Team <xorg-team> | ||||||||||||||||||||||||||||||
| Severity: | major | ||||||||||||||||||||||||||||||||
| Priority: | medium | CC: | dash, francisco, jg, michael.fu | ||||||||||||||||||||||||||||||
| Version: | 7.4 (2008.09) | Keywords: | NEEDINFO | ||||||||||||||||||||||||||||||
| Hardware: | All | ||||||||||||||||||||||||||||||||
| OS: | Linux (All) | ||||||||||||||||||||||||||||||||
| Whiteboard: | |||||||||||||||||||||||||||||||||
| i915 platform: | i915 features: | ||||||||||||||||||||||||||||||||
| Attachments: |
|
||||||||||||||||||||||||||||||||
|
Description
Bryce Harrington
2009-08-03 08:39:26 UTC
Created attachment 28307 [details]
screen-photo.jpg
Created attachment 28308 [details]
xrandr.out
Created attachment 28309 [details]
CurrentDmesg.txt
Created attachment 28310 [details]
XorgLog.txt
Does this laptop has real TV output? What's the last kernel that works? The laptop does not have any TV ports. The last environment that works OK for me, is Ubuntu 9.04 (Jaunty) which according to 'uname -r' runs Linux version 2.6.27-14-generic. I tried some early Karmic alphas, but then X didn't work at all (even before issuing a 'xrandr' command). Do you think that the problem is primarily with the kernel, and not with X? >1. xorg is started >2. The external monitor (a Dell 2001 FP) is plugged into the VGA port >3. 'xrandr' is called without arguments, and instantly the screen is garbled - >mostly black, except for a few pixels at the top of the screen that appears >relatively normal. > >If step 2 is done before step 1, the screen stays normal. looks like there is no EDID at all for VGA after monitor plugged in. Agree with what Michael said in comment #7. It seems that there is no EDID for the VGA monitor. Does this issue still exist when using another external monitor? Will you please add the modedebug option in xorg.conf and attach the xorg.log? Can it work well under UMS mode? Please attach the xorg.log under UMS mode. Will you please also attach the vbios.dump? You can use the following command to get the vbios.dump echo 1 > /sys/devices/pci0000:00/0000:00:02.0/rom cat /sys/devices/pci0000:00/0000:00:02.0/rom >vbios.dump echo 0 > /sys/devices/pci0000:00/0000:00:02.0/rom Thanks. a comparison intel_reg_dump output of good and bad case will also provide good hints. it's in xserver-xorg-video-intel-dbg package in Ubuntu. >1. xorg is started >2. The external monitor (a Dell 2001 FP) is plugged into the VGA port >3. 'xrandr' is called without arguments, and instantly the screen is garbled - >mostly black, except for a few pixels at the top of the screen that appears >relatively normal. please capture a reg snapshot via running intel_reg_dump, name it as bad >If step 2 is done before step 1, the screen stays normal. and in this environment, run intel_reg_dump again, name it as good. please then attach both output here. thanks. Created attachment 28806 [details]
Output from intel_reg_dumper after steps 1, 2, 3
Created attachment 28807 [details]
bad
Created attachment 28808 [details]
good
I noticed some lines in /var/log/syslog that might be relevant. They don't seem to be consistently reproducible, but still... Aug 20 07:18:05 dash-laptop kernel: [78019.257267] integrated sync not supported Aug 20 07:18:05 dash-laptop kernel: [78019.467472] [drm] DAC-6: set mode 25 Aug 20 07:18:08 dash-laptop kernel: [78022.745272] render error detected, EIR: 0x00000010 Aug 20 07:18:08 dash-laptop kernel: [78022.745279] page table error Aug 20 07:18:08 dash-laptop kernel: [78022.745281] PGTBL_ER: 0x00000100 Aug 20 07:18:08 dash-laptop kernel: [78022.745287] [drm:i915_handle_error] *ERROR* EIR stuck: 0x00000010, masking Aug 20 07:18:08 dash-laptop kernel: [78022.745298] render error detected, EIR: 0x00000010 Aug 20 07:18:08 dash-laptop kernel: [78022.745301] page table error Aug 20 07:18:08 dash-laptop kernel: [78022.745303] PGTBL_ER: 0x00000100 *** Bug 23429 has been marked as a duplicate of this bug. *** would you please attach a regdump without any external display attached? thanks. Created attachment 28915 [details]
no-external-display
btw, you only attached a screenshot of the garbled screen on the laptop panel, how about the VGA monitor? Is it ok when the laptop panel is garbled? thanks. (In reply to comment #8) > Agree with what Michael said in comment #7. It seems that there is no EDID for > the VGA monitor. > Does this issue still exist when using another external monitor? > Will you please add the modedebug option in xorg.conf and attach the > xorg.log? Can it work well under UMS mode? Please attach the xorg.log under UMS > mode. > Will you please also attach the vbios.dump? You can use the following > command to get the vbios.dump > echo 1 > /sys/devices/pci0000:00/0000:00:02.0/rom > cat /sys/devices/pci0000:00/0000:00:02.0/rom >vbios.dump > echo 0 > /sys/devices/pci0000:00/0000:00:02.0/rom > > Thanks. > Dag, yakui's request in comment# 17 is also helpful for root-cause the bug. please kindly follow up to provide the logs. fyi, when you need to use UMS mode, add i915.modeset=0 to kernel boot parameters. thanks. The external monitor is also garbled. I'll attach a photo. The dots in the middle of the screen seems to be a weird rendering of the mouse cursor, as it moves along with the mouse. I only answered yakui's request at the Ubuntu bug tracker, where I saw it first. Sorry for the confusion I caused for not keeping the two bug trackers in sync, I'll make amends and upload the files here as well. The gist of it is that I can't reproduce the bug under UMS (booting with i915.modeset=0), so it is probably KMS related. Created attachment 28936 [details]
external-screen-photo.jpg
Created attachment 28937 [details]
Xorg.log-ModeDebug-KMS
Created attachment 28938 [details]
Xorg.log-ModeDebug-UMS
Created attachment 28939 [details]
vbios.dump
Btw, I see that the title of the report is modified with an "VGA (without EDID)". I'll just mention that the Dell 2001 FP normally reports EDID info just fine, so any missing EDID info is probably the fault of X or the kernel. I'll attach a copy of the EDID info acquired via the nvidia-settings tool (from another computer using the DVI connection), in case that is of any help. Created attachment 28940 [details]
Dell_2001FP_edid.bin
Created attachment 28942 [details] [review] clear it when the underflow occurs after changing mode Will you please try the debug patch on the latest kernel(2.6.31-rc7) and see whether the issue still exists? Thanks. Hi, Dag
From the log in comment #21 it seems that the EDID can also be obtained for VGA monitor.
But from the register dump in comment #11/12 it seems that the register of PIPEASTAT is different.
-(II): PIPEASRC: 0x031f0257 (800, 600)
-(II): PIPEASTAT: 0x00000303 (status: VSYNC_INT_STATUS DLINE_COMPARE_STATUS VBLANK_INT_STATUS OREG_UPDATE_STATUS)
+(II): PIPEASRC: 0x063f04af (1600, 1200)
+(II): PIPEASTAT: 0x80000303 (status: FIFO_UNDERRUN VSYNC_INT_STATUS DLINE_COMPARE_STATUS VBLANK_INT_STATUS OREG_UPDATE_STATUS)
I am not sure whether the issue is related with the PIPEA underflow.
Will you please try the debug patch in comment #26 on the latest kernel(2.6.31-rc7) and see whether the issue still exists?
When the problem happens, please switch to other resolution and see whether the problem still exits. For example: 1280x1024, 1024x768.
Thanks.
Will you please try to set the following modes for VGA/LVDS under UMS mode and see whether it still works well? LVDS: 1024x768 VGA: 1600x1200 After the test, please attach the register dump and Xorg.log. Thanks. I have been sidetracked by some other problems (my wireless router died, and an upgrade to karmic had some unwanted side effects), so I haven't been able to test this thoroughly yet. But I have compiled a new kernel with the patch, and the obvious effect is that the screen goes black when I issue the 'xrandr' command. More details and log files will follow. (In reply to comment #29) > I have been sidetracked by some other problems (my wireless router died, and > an upgrade to karmic had some unwanted side effects), so I haven't been able > to test this thoroughly yet. But I have compiled a new kernel with the patch, > and the obvious effect is that the screen goes black when I issue the 'xrandr' > command. More details and log files will follow. > logs? Dag? please try this patch as well: http://bugs.freedesktop.org/show_bug.cgi?id=22921#c5, _without_ patch in comment# 26. ping~ Dag. ping~ dag... ping Dag... Another user (fiacobelli) who says he has the same problem has tested the kernel 2.6.31-rc7, both without any patches and with each of the patches in comments #26 and #31 without making it better. What information are you interested in at this time? I suppose it is best if he files a separate bug report? Can you try the following patch set on 2.6.33-rc5 kernel and see whether
the issue still exists?
>http://lists.freedesktop.org/archives/intel-gfx/2010-January/005505.html
BTW: the patch 1 can be skipped as it is already shipped in 2.6.33-rc5 kernel.
Thanks.
Yakui
Closing as losing communications with the reporter. Geir, you could ask fiacobelli to file a separate bug. Thanks. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.