Summary: | [KMS] sometimes freezes when resuming from suspend-to-disk | ||||||
---|---|---|---|---|---|---|---|
Product: | DRI | Reporter: | Vincent Danjean <vdanjean> | ||||
Component: | DRM/Intel | Assignee: | Wang Zhenyu <zhenyu.z.wang> | ||||
Status: | CLOSED FIXED | QA Contact: | |||||
Severity: | major | ||||||
Priority: | medium | CC: | jcristau | ||||
Version: | XOrg git | ||||||
Hardware: | x86-64 (AMD64) | ||||||
OS: | Linux (All) | ||||||
Whiteboard: | |||||||
i915 platform: | i915 features: | ||||||
Attachments: |
|
Description
Vincent Danjean
2009-09-10 04:01:04 UTC
Created attachment 30459 [details]
IRC log about a new occurence of this bug
Here is some new info about this bug. You can see that there has really been memory corruption. The kernel is the last from Debian (ie based on 2.6.31.4)
Could you try recent 2.6.32-rc6 kernel? Jesse has several resume fixes might be relate with this one. I cannot try for now: the laptop I use is my work laptop and I do not have time for now to take the risk of a data corruption (and a reinstallation from backup). For my part, new tests will probably needs to wait until January. Regards, Vincent lower severity. I've been asked to test with 2.6.32-rc6. I tested with 2.6.32-rc8 and 2.6.32 (Debian packages). In both case, I can boot with KMS until gdm screen but my keyboard does not seems to work correctly. The place for the login is black (instead of white as usual). When I try to login, gdm seems to freeze and then restart. Note that 2.6.32 (or 2.6.32) without KMS does not work either, nor 2.6.31 with KMS (whereas it was working, at least when I do not suspend to disk, when I wrote this initial bug report) So I think there have been new bugs in related sofware (probably the X server). My Xserver is the one from Debian unstable (package xserver-xorg 1:7.4+4). For now, I've to stick to 2.6.31 without KMS. Debian has xserver-xorg 1:7.5~3 in experimental but I cannot test it because xserver-xorg-video-intel is not (yet ?) recompiled for this server. Regards, Vincent PS: most of the time, there is nothing in the log. Sometimes (I think this is 2.6.32 without KMS but I'm not sure), I can see Fatal server error: Failed to map batchbuffer: Input/output error Please consult the The X.Org Foundation support at http://wiki.x.org for help. Please also check the log file at "/var/log/Xorg.0.log" for additional information. Any news on this bug ? I still suffer from it (kernel 2.6.32 and 2.6.34) as I just explain on the debian bug report ( http://bugs.debian.org/545517 ). Being required to power-off my laptop instead of suspending it is very painful. We think we found the cause: commit 985b823b919273fe1327d56d2196b4f92e5d0fae Author: Linus Torvalds <torvalds@linux-foundation.org> Date: Fri Jul 2 10:04:42 2010 +1000 drm/i915: fix hibernation since i915 self-reclaim fixes Since commit 4bdadb9785696439c6e2b3efe34aa76df1149c83 ("drm/i915: Selectively enable self-reclaim"), we've been passing GFP_MOVABLE to the i915 page allocator where we weren't before due to some over-eager removal of the page mapping gfp_flags games the code used to play. This caused hibernate on Intel hardware to result in a lot of memory corruptions on resume. See for example http://bugzilla.kernel.org/show_bug.cgi?id=13811 Reported-by: Evengi Golov (in bugzilla) Signed-off-by: Dave Airlie <airlied@redhat.com> Tested-by: M. Vefa Bicakci <bicave@superonline.com> Cc: stable@kernel.org Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Which means that there was actually a window in the 2.6.33 where hibernation would have worked... Vincent, hibernation should finally be usable in the current tip of Linus's tree. 2.6.35-rc6 appears to have fixed the hibernate regression. |
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.