Bug 8542 - Suspend broke between Xorg 7.0 / i810 1.5.1 and Xorg 7.1 / i810 1.6.1
Summary: Suspend broke between Xorg 7.0 / i810 1.5.1 and Xorg 7.1 / i810 1.6.1
Status: RESOLVED FIXED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: 7.1 (2006.05)
Hardware: x86 (IA32) Linux (All)
: high major
Assignee: Alan Hourihane
QA Contact:
URL:
Whiteboard:
Keywords:
: 8544 (view as bug list)
Depends on:
Blocks:
 
Reported: 2006-10-07 04:44 UTC by Lars Noschinski
Modified: 2006-10-13 09:11 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
"dmesg" log across a resume, i810 driver v1.5.1 (30.40 KB, text/plain)
2006-10-08 15:16 UTC, Patrick Reynolds
no flags Details
Xorg.0.log across a resume, i810 driver v1.5.1 (69.36 KB, text/plain)
2006-10-08 15:16 UTC, Patrick Reynolds
no flags Details
xdpyinfo for i810 driver v1.5.1 (5.87 KB, text/plain)
2006-10-08 15:17 UTC, Patrick Reynolds
no flags Details
"dmesg" log across a resume, i810 driver v1.6.5 (22.95 KB, text/plain)
2006-10-08 15:17 UTC, Patrick Reynolds
no flags Details
Xorg.0.log across a resume, i810 driver v1.6.5 (98.94 KB, text/plain)
2006-10-08 15:17 UTC, Patrick Reynolds
no flags Details
xdpyinfo for i810 driver v1.6.5 (5.87 KB, text/plain)
2006-10-08 15:18 UTC, Patrick Reynolds
no flags Details
Xorg.0.log and dmesg for working and non-working suspend (40.42 KB, application/gzip)
2006-10-10 01:47 UTC, Lars Noschinski
no flags Details
Xorg.0.log and dmesg for working and non-working suspend, part 2 (150.00 KB, application/gzip)
2006-10-10 02:02 UTC, Lars Noschinski
no flags Details
Logs for new test 1.7.0 (Xorg 7.0.0) around suspend (15.68 KB, application/gzip)
2006-10-10 08:56 UTC, Patrick Reynolds
no flags Details
Xorg.0.log with test-driver (218.02 KB, application/octet-stream)
2006-10-12 09:23 UTC, Lars Noschinski
no flags Details

Description Lars Noschinski 2006-10-07 04:44:39 UTC
Updating Xorg to 7.1 with xf86-video-i810 between 1.6.1 and 1.7.0 breaks suspend
to RAM for me: The display stays black, the backlight off, reviving it with
vbetool is possible. Xorg 7.0 with i810 1.5.1 works flawlessy in this regard.

See also bugs.debian.org/cgi-bin/bugreport.cgi?bug=390913 (im using the Debian
packages)
Comment 1 Alan Hourihane 2006-10-08 13:56:03 UTC
Logs comparing 1.5.1 with 1.7.0 would be useful.
Comment 2 Alan Hourihane 2006-10-08 14:09:44 UTC
*** Bug 8544 has been marked as a duplicate of this bug. ***
Comment 3 Patrick Reynolds 2006-10-08 15:16:18 UTC
Created attachment 7292 [details]
"dmesg" log across a resume, i810 driver v1.5.1
Comment 4 Patrick Reynolds 2006-10-08 15:16:46 UTC
Created attachment 7293 [details]
Xorg.0.log across a resume, i810 driver v1.5.1
Comment 5 Patrick Reynolds 2006-10-08 15:17:08 UTC
Created attachment 7294 [details]
xdpyinfo for i810 driver v1.5.1
Comment 6 Patrick Reynolds 2006-10-08 15:17:38 UTC
Created attachment 7295 [details]
"dmesg" log across a resume, i810 driver v1.6.5
Comment 7 Patrick Reynolds 2006-10-08 15:17:58 UTC
Created attachment 7296 [details]
Xorg.0.log across a resume, i810 driver v1.6.5
Comment 8 Patrick Reynolds 2006-10-08 15:18:17 UTC
Created attachment 7297 [details]
xdpyinfo for i810 driver v1.6.5
Comment 9 Patrick Reynolds 2006-10-08 15:22:57 UTC
I've attached "dmesg" logs and X logs for i810 drivers v1.5.1 (under X.org
7.0.0) and and v1.6.5 (under X.org 7.1.1), across an S3 sleep and resume.  I
marked the point where the resume occured with "SUSPEND HAPPENS HERE."

I've also attached xdpyinfo output for both drivers.

The Composite extension was disabled for all tests.  Enabling it doesn't change
the results.  I have logs for 1.6.5/7.1.1 with Composite on if anyone needs them.
Comment 10 Alan Hourihane 2006-10-09 11:18:22 UTC
can you try a test 1.7.1 driver from http://www.fairlite.demon.co.uk/intel.html
and let me know if that works for you.
Comment 11 Lars Noschinski 2006-10-10 01:44:30 UTC
(In reply to comment #10)
> can you try a test 1.7.1 driver from http://www.fairlite.demon.co.uk/intel.html
> and let me know if that works for you.

Hm, http://www.fairlite.demon.co.uk/i810_drv.so seems to be 1.7.0? Anyway, the
problem occurs also with this driver (and Xorg 7.0).
Comment 12 Lars Noschinski 2006-10-10 01:47:40 UTC
Created attachment 7317 [details]
Xorg.0.log and dmesg for working and non-working suspend

This file contains Xorg logfile and dmesg for Xorg 7.0, i810 1.5.1, working
resume and Xorg 7.1, i810 1.7.0, non-working suspend.
Comment 13 Lars Noschinski 2006-10-10 02:02:18 UTC
Created attachment 7318 [details]
Xorg.0.log and dmesg for working and non-working suspend, part 2

Same as above, but for xorg 7.0 with i810 1.7.0.
Comment 14 Alan Hourihane 2006-10-10 03:22:01 UTC
The driver at my homepage is just a modified 1.7.0 driver.

Now, I've just uploaded another one to try, let me know if this helps.
Comment 15 Patrick Reynolds 2006-10-10 08:56:49 UTC
Created attachment 7327 [details]
Logs for new test 1.7.0 (Xorg 7.0.0) around suspend

The test version of i810 1.7.0 (as of now) works fine with Xorg 7.0.0,
including suspend and resume.  dmesg and X logs for it are attached as a
.tar.gz.

The test version of i810 1.7.0 does not work with Xorg 7.1.1 because of an ABI
mismatch.  If you compile a 7.1.1 version, I can try that across a
suspend/resume.  Let me know.
Comment 16 Alan Hourihane 2006-10-11 08:56:34 UTC
I've also uploaded a new driver which emits some extra debug information for me.

Can you try it, and upload a new log.

Addtionally, the -ignoreABI flag to the Xserver will make it work with Xorg 7.1
Comment 17 Lars Noschinski 2006-10-12 09:23:21 UTC
Created attachment 7388 [details]
Xorg.0.log with test-driver

This log was done with the test driver (downloaded 2006-10-12, 18:05). Still no
working resume.

BTW: I don't know, if this is useful: With this driver, restoring the state
with 'vbetool vbestate restore < state' gave me some nice graphical effects I
hadn't got before (display fading from black over yellow, red to white, but not
homogenous).
Comment 18 Alan Hourihane 2006-10-12 09:41:01 UTC
Sorry, there was a typo in the last test driver. I've reuploaded a new one to try.
Comment 19 Alan Hourihane 2006-10-12 09:42:16 UTC
Oh, and don't use vbetool. It breaks the driver.
Comment 20 Alan Hourihane 2006-10-13 04:54:45 UTC
Committed fixes to driver. The fixes will appear in the 1.7.1 release of the driver.
Comment 21 Lars Noschinski 2006-10-13 09:11:47 UTC
(In reply to comment #20)
> Committed fixes to driver. The fixes will appear in the 1.7.1 release of the
driver.

The driver I just downloaded (18:00 CEST) works for me. Thanks a lot.


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.