Bug 103772 - Display corruption with Linux kernel commit 616d9cee4fdc4 ("First try the previous execbuffer location")
Summary: Display corruption with Linux kernel commit 616d9cee4fdc4 ("First try the pre...
Status: CLOSED WORKSFORME
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: DRI git
Hardware: x86-64 (AMD64) Linux (All)
: high critical
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords: bisected, regression
Depends on:
Blocks:
 
Reported: 2017-11-16 02:23 UTC by Kenneth C
Modified: 2018-04-20 14:57 UTC (History)
2 users (show)

See Also:
i915 platform: HSW
i915 features: display/Other

Attachments
sudo lspci -vv output (36.51 KB, text/plain)
2017-11-16 02:23 UTC, Kenneth C 		no flags Details
/proc/config.gz (22.27 KB, application/gzip)
2017-11-16 02:24 UTC, Kenneth C 		no flags Details
/proc/cpuinfo (3.89 KB, text/plain)
2017-11-16 02:25 UTC, Kenneth C 		no flags Details
My take on reverting it, wasn't a clean revert, but this seems to be right, and it hasn't blown up so far (2.91 KB, patch)
2017-11-16 02:26 UTC, Kenneth C 		no flags Details | Splinter Review
Xorg.0.log file (working-kernel version however, but the userspace is still the same) (43.48 KB, application/x-trash)
2017-11-17 03:16 UTC, Kenneth C 		no flags Details
Reverting patch (539 bytes, patch)
2018-01-09 15:44 UTC, Mariusz Białończyk 		no flags Details | Splinter Review
View All

Description Kenneth C 2017-11-16 02:23:57 UTC 
Created attachment 135505 [details]
sudo lspci -vv output

On any kernel with this commit (commit 616d9cee4fdc4a377c03be8fd6efa5df4fcd0d81 Author: Chris Wilson <chris@chris-wilson.co.uk>  Date:   Fri Jun 16 15:05:21 2017 +0100 drm/i915: First try the previous execbuffer location), I get screen corruption once X starts (text does not update, flickering).

I'm currently running Linux kernel 4.14.0 with this commit reverted and not seeing the issue (which I'd determined by bisection between kernel versions 4.12.8 and 4.13.0).

I have attached my "lspci -vv" output. Feel free to contact me if you need more information.
Comment 1 Kenneth C 2017-11-16 02:24:42 UTC 
Created attachment 135506 [details]
/proc/config.gz
Comment 2 Kenneth C 2017-11-16 02:25:42 UTC 
Created attachment 135507 [details]
/proc/cpuinfo
Comment 3 Kenneth C 2017-11-16 02:26:51 UTC 
Created attachment 135508 [details] [review]
My take on reverting it, wasn't a clean revert, but this seems to be right, and it hasn't blown up so far
Comment 4 Chris Wilson 2017-11-16 09:49:29 UTC 
We need details on the broken userspace.
Comment 5 Kenneth C 2017-11-17 03:13:07 UTC 
> We need details on the broken userspace.

Certainly. What would you like to see? I'll attach an Xorg.0.log file from around the time the failure occurred.
Comment 6 Kenneth C 2017-11-17 03:16:45 UTC 
Created attachment 135535 [details]
Xorg.0.log file (working-kernel version however, but the userspace is still the same)
Comment 7 Chris Wilson 2017-11-20 08:26:14 UTC 
(In reply to Kenneth C from comment #6)
> Created attachment 135535 [details]
> Xorg.0.log file (working-kernel version however, but the userspace is still
> the same)

And the log for the broken version, since I'm looking for something in particular in there that might explain the type of breakage. Its absence will also be indicative.
Comment 8 Kenneth C 2017-12-02 02:39:46 UTC 
This is still in "Needinfo" state ... It's going to be hard to regress these machines back to a failing kernel at the moment- do you still need the Xorg.0.log file, or is there sufficient information to at least understand the cause for this bug?
Comment 9 Mariusz Białończyk 2018-01-09 15:44:35 UTC 
Created attachment 136632 [details] [review]
Reverting patch

FWIW I am providing a one-liner which is sufficient to revert the commit and have it working back again (I am currently testing it on vanilla linux).

I know that it is only for disabling the commit and not fixing the problem.
Comment 10 Jani Saarinen 2018-03-29 07:11:03 UTC 
First of all. Sorry about spam.
This is mass update for our bugs. 

Sorry if you feel this annoying but with this trying to understand if bug still valid or not.
If bug investigation still in progress, please ignore this and I apologize!

If you think this is not anymore valid, please comment to the bug that can be closed.
If you haven't tested with our latest pre-upstream tree(drm-tip), can you do that also to see if issue is valid there still and if you cannot see issue there, please comment to the bug.
Comment 11 Jani Saarinen 2018-04-20 14:57:48 UTC 
Closing, please re-open if still occurs.

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.