Bug 36888 - [IronLake] testdisplay sets mode with eDP port will cause system hang
Summary: [IronLake] testdisplay sets mode with eDP port will cause system hang
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: high critical
Assignee: Jesse Barnes
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-05-05 23:19 UTC by bo.b.wang
Modified: 2017-09-04 10:06 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
load the LUT earlier when modesetting (1.20 KB, patch)
2011-06-15 14:34 UTC, Jesse Barnes
no flags Details | Splinter Review

Description bo.b.wang 2011-05-05 23:19:16 UTC
System Environment:
--------------------------
(drm-intel-fixes)commit:dabc8ce34938f9fa2929f51d9fab730e78bd2184
eDP port

Bug detailed description:
-------------------------
Using "testdisplay -f modeline"  will cause system hang

Reproduce steps:
----------------
1. with eDP port
2. testdisplay -f modeline(eg. 
    ./testdisplay -f 38.21,800,832,976,1008,600,612,618,631
Comment 1 Chris Wilson 2011-05-09 22:57:10 UTC
Jesse, only you can help us now!
Comment 2 Jesse Barnes 2011-05-13 11:22:55 UTC
I don't see this on regular DP, so I guess it's related to our VBT or mode handling for eDP... checking an eDP system now.
Comment 3 Jesse Barnes 2011-06-15 13:13:42 UTC
I've reproduced this on my E6510, checking it out now
Comment 4 Jesse Barnes 2011-06-15 14:04:34 UTC
The last function I see called before the hang is ironlake_crtc_enable (I get a vblank wait timeout in enable plane then see the hang).
Comment 5 Jesse Barnes 2011-06-15 14:34:11 UTC
Created attachment 48015 [details] [review]
load the LUT earlier when modesetting

Can you confirm that this patch fixes the problem?
Comment 6 bo.b.wang 2011-06-15 23:38:07 UTC
(In reply to comment #5)
> Created an attachment (id=48015) [details]
> load the LUT earlier when modesetting
> 
> Can you confirm that this patch fixes the problem?

Hello Jesse: this patch have fixed the problem.
Tested-by: wang bo bo.b.wang@intel.com
Comment 7 Jim Gettys 2011-07-01 14:00:19 UTC
(In reply to comment #5)
> Created an attachment (id=48015) [details]
> load the LUT earlier when modesetting
> 
> Can you confirm that this patch fixes the problem?

System (hp 2540p) no longer hangs when patched. I tried all the usual resolutions.

But the screen brightness goes low and won't respond until I switch to a text console and back again.

It also seemed like initial startup of the X server might have been slower than usual, but I haven't really confirmed this: take this comment with a large block of salt.
Comment 8 Jesse Barnes 2011-07-28 17:40:51 UTC
drm/i915: load the LUT before pipe enable on ILK+
author	Jesse Barnes <jbarnes@virtuousgeek.org>	
	Wed, 15 Jun 2011 21:32:33 +0000 (23:32 +0200)
committer	Keith Packard <keithp@keithp.com>	
	Fri, 22 Jul 2011 20:37:00 +0000 (13:37 -0700)
commit	9c54c0dd948d715ccfd79e97d852f80eeb53254a
tree	49f18f67a736684bfee4b6c7f38175cee7e27632	tree | snapshot
parent	f3234706a77bd6e1592ae71fb3268e04cb030dba	commit | diff
drm/i915: load the LUT before pipe enable on ILK+

Per the specs and to address
https://bugs.freedesktop.org/show_bug.cgi?id=36888.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: stable@kernel.org
Signed-off-by: Keith Packard <keithp@keithp.com>

from drm-intel-fixes
Comment 9 cc 2011-07-28 19:27:36 UTC
The commit is on the fixed branch.
Comment 10 cc 2011-07-28 23:14:02 UTC
The commit is on the fixed branch.
Comment 11 Jari Tahvanainen 2017-09-04 10:06:38 UTC
Closing old verified+fixed.


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.