Bug 20866 - [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
Summary: [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled...
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: 7.4 (2008.09)
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Jesse Barnes
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: NEEDINFO
Depends on:
Blocks:
 
Reported: 2009-03-25 16:05 UTC by Arkadiusz Miskiewicz
Modified: 2009-04-07 16:55 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
xorg log (23.28 KB, text/plain)
2009-03-26 00:39 UTC, Arkadiusz Miskiewicz
no flags Details
dmesg (56.86 KB, text/plain)
2009-03-26 00:42 UTC, Arkadiusz Miskiewicz
no flags Details
new xorg log (18.03 KB, text/plain)
2009-03-26 12:30 UTC, Arkadiusz Miskiewicz
no flags Details
new dmesg (72.71 KB, text/plain)
2009-03-26 12:31 UTC, Arkadiusz Miskiewicz
no flags Details

Description Arkadiusz Miskiewicz 2009-03-25 16:05:13 UTC
I'm using:
- xserver 1.6
- kernel 2.6.29
- drm from airlied drm-next branch
- intel driver from git master (from 20 minutes ago)
- UXA enabled, DRI also enabled

When logging out from kde4 kdm X segfaults, unfortunately backtrace is not logged, log just ends with:

(**) Option "xkb_options" "altwin:left_meta_win,lv3:menu_switch"

Backtrace:

and X is restarted.

Few logins/logouts and dmesg contains:
[ 2447.649952] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2636.120552] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2637.532974] X[12404]: segfault at 2 ip 00007f65a14dd0bf sp 00007fffbeb769b0 error 4 in libgcc_s.so.1[7f65a14cd000+16000]
[ 2637.581244] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2708.597008] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2708.618726] mtrr: no MTRR for d0000000,10000000 found
[ 2710.001075] X[12753]: segfault at 2 ip 00007fc5f2c880bf sp 00007fff103c6230 error 4 in libgcc_s.so.1[7fc5f2c78000+16000]
[ 2710.053275] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2761.432990] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
[ 2761.454403] mtrr: no MTRR for d0000000,10000000 found
[ 2771.693794] X[13069]: segfault at 2 ip 00007fe8673580bf sp 00007fff847e3630 error 4 in libgcc_s.so.1[7fe867348000+16000]
[ 2771.742306] [drm:gm45_get_vblank_counter] *ERROR* trying to get vblank count for disabled pipe 0
Comment 1 Gordon Jin 2009-03-25 19:07:02 UTC
Please attach the full Xorg.0.log and dmesg.
Comment 2 Arkadiusz Miskiewicz 2009-03-26 00:39:47 UTC
Created attachment 24252 [details]
xorg log
Comment 3 Arkadiusz Miskiewicz 2009-03-26 00:42:21 UTC
Created attachment 24253 [details]
dmesg

Xorg and dmesg logs attached.

Note that this time I used drm from final 2.6.29 (previously drm-next) and intel driver from git from 2.7 branch.

Now doing the same test again but with drm from drm-next and intel from git from master branch.
Comment 4 Arkadiusz Miskiewicz 2009-03-26 01:05:04 UTC
Unfortunately seems that I cannot reproduce the "trying to get vblank count for disabled pipe 0" easily.

I always get with #20867 triggered even back on intel master and drm from drm-next when trying to trigger "disabled pipe" thing.
Comment 5 Arkadiusz Miskiewicz 2009-03-26 12:30:03 UTC
Created attachment 24284 [details]
new xorg log
Comment 6 Arkadiusz Miskiewicz 2009-03-26 12:31:15 UTC
Created attachment 24286 [details]
new dmesg

Figured out how to reproduce. This vblank thing happens if KMS/modeset is disabled.
Comment 7 Jesse Barnes 2009-03-30 17:54:49 UTC
Can you try updating your Mesa to 7.4ish?  This bug should be fixed there I think.
Comment 8 Arkadiusz Miskiewicz 2009-04-06 06:04:39 UTC
Updated to 7.4, so far unable to reproduce (unforunteatly I never know how to reliably trigger it).
Comment 9 Jesse Barnes 2009-04-07 16:55:20 UTC
Prior to Mesa 7.4 it should have happened everytime a DRI app started...  Anyway thanks for confirming.


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.