Bug 22529 - [855GM] can't bring up LVDS in latest git
Summary: [855GM] can't bring up LVDS in latest git
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: git
Hardware: Other All
: highest critical
Assignee: Jesse Barnes
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords: regression
Depends on:
Blocks:
 
Reported: 2009-06-29 03:45 UTC by Daniel Vetter
Modified: 2009-06-30 12:02 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
Xorg.log of the defunct server (19.49 KB, text/x-log)
2009-06-29 03:45 UTC, Daniel Vetter
no flags Details
VBIOS rom as requested (64.00 KB, application/octet-stream)
2009-06-29 10:18 UTC, Daniel Vetter
no flags Details
use DVO timing block instead (1.07 KB, patch)
2009-06-30 10:20 UTC, Jesse Barnes
no flags Details | Splinter Review

Description Daniel Vetter 2009-06-29 03:45:29 UTC
Created attachment 27217 [details]
Xorg.log of the defunct server

I've just upgraded to the latest xf86-video-intel git. The server can't start up anymore and throws the following error into Xorg.log:

Fatal server error:
Couldn't find PLL settings for mode!

Full logfile attached. git bisect tracked it down to the following commit

commit 15af8ea6ab6998bbab9f4eeda227565c409da229
Author: Jesse Barnes <jbarnes@virtuousgeek.org>
Date:   Mon Jun 22 11:11:06 2009 -0700

    Fix LFP data block fetch

    Apparently the proper way to do this is to use the LFP data pointer block to

    Fixes fdo bug #19450.

    Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>

Reverting this on latest master (2.7.99.901-21-gf53b323) works around the problem.

Machine is a Thinkpad X40 running 32bit Debian unstable.
Comment 1 Jesse Barnes 2009-06-29 09:20:03 UTC
Can you attach your VBIOS to this bug?

# cd /sys/devices/pci0000\:00/0000\:00\:02.0/
# echo 1 > rom
# cat rom > /tmp/rom.bin
# echo 0 > rom
Comment 2 Daniel Vetter 2009-06-29 10:18:00 UTC
Created attachment 27223 [details]
VBIOS rom as requested
Comment 3 Michael Fu 2009-06-29 19:54:31 UTC
Jesse, looks like you are parsing fp_timing as dvo_timing in your patch...

+    timing_ptr = (unsigned char *)&entry->fp_timing;

Comment 4 Jesse Barnes 2009-06-30 10:20:55 UTC
Created attachment 27267 [details] [review]
use DVO timing block instead

Care to try this patch?
Comment 5 Jesse Barnes 2009-06-30 10:24:39 UTC
Thanks Michael, somehow I got the kernel side right but the UMS side wrong.
Comment 6 Daniel Vetter 2009-06-30 11:49:19 UTC
On Tue, Jun 30, 2009 at 10:20:56AM -0700, bugzilla-daemon@freedesktop.org wrote:
> Created an attachment (id=27267)
>  --> (http://bugs.freedesktop.org/attachment.cgi?id=27267)
> use DVO timing block instead
> 
> Care to try this patch?
Tested and works. Thanks for the speedy fix.

-Daniel
Comment 7 Jesse Barnes 2009-06-30 12:02:54 UTC
Fix pushed, thanks for testing.

commit 53e248af4365232416d2143a791a07c6751f8319
Author: Jesse Barnes <jbarnes@virtuousgeek.org>
Date:   Tue Jun 30 12:00:28 2009 -0700

    Use DVO timing block instead of fp_timing when parsing LFP data


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.