Bug 34437 - [Ironlake LVDS Vaio-Y] blank screen (with backlight) after resuming from suspend or hibernate
Summary: [Ironlake LVDS Vaio-Y] blank screen (with backlight) after resuming from susp...
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Chris Wilson
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on: 36656
Blocks:
  Show dependency tree / graph
 
Reported: 2011-02-18 04:07 UTC by Michel Alexandre Salim
Modified: 2017-07-24 23:05 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:

Attachments
dmesg before suspend (121.60 KB, text/plain)
2011-02-18 04:51 UTC, Michel Alexandre Salim 		no flags Details
dmesg after suspend (121.61 KB, text/plain)
2011-02-18 04:53 UTC, Michel Alexandre Salim 		no flags Details
intel_reg_dumper before suspend (8.77 KB, text/plain)
2011-02-18 04:53 UTC, Michel Alexandre Salim 		no flags Details
intel_reg_dumper after suspend (8.73 KB, text/plain)
2011-02-18 04:54 UTC, Michel Alexandre Salim 		no flags Details
intel_gpu_dump before suspend (part 1) (976.56 KB, application/octet-stream)
2011-02-18 06:09 UTC, Michel Alexandre Salim 		no flags Details
intel_gpu_dump before suspend (part 2) (974.04 KB, application/octet-stream)
2011-02-18 06:10 UTC, Michel Alexandre Salim 		no flags Details
intel_gpu_dump after suspend (part 1) (976.56 KB, application/octet-stream)
2011-02-18 06:11 UTC, Michel Alexandre Salim 		no flags Details
intel_gpu_dump after suspend (part 2) (207.70 KB, application/octet-stream)
2011-02-18 06:12 UTC, Michel Alexandre Salim 		no flags Details
drm/i915: Add quirk to disable SSC on Sony Vaio Y2 (1.13 KB, patch)
2011-07-27 06:29 UTC, Michel Alexandre Salim 		no flags Details | Splinter Review
View All

Description Michel Alexandre Salim 2011-02-18 04:07:23 UTC 
Bug description:
Going into suspend / hibernate works just fine, however, upon resume, regardless of whether the power save action is done from the console or from X, resuming just yield a blank black screen, though the backlight is noticeably on.

KMS is enabled; and this is with the internal panel (LVDS1)
 
System environment: 
-- chipset: Ironlake (Core i5-430UM)
-- system architecture: x86_64
-- xf86-video-intel: 2.14.0 (xorg-x11-drv-intel-2.14.0-2.fc15)
-- xserver: 1.9.99.1 (xorg-x11-server-Xorg-1.9.99.1-4.20101201.fc15)
-- mesa: 7.10-devel (mesa-libGL-7.10-0.26.fc15)
-- libdrm: 2.4.23 (libdrm-2.4.24-0.2.20110106.fc15)
-- kernel: 2.6.38-rc5+ (drm-intel-fixes 811779a0fbd23035872d1fc8f2d1baf550a94f10)
-- Linux distribution: Fedora Rawhide
-- Machine or mobo model: Sony Vaio VPCY2 (2nd generation Y series)
-- Display connector: LVDS

Reproducing steps:
1. Boot to runlevel 3
2. pm-suspend
3. Press a key to resume
4. Verify machine is still alive -- CTRL-ALT-DEL triggers a normal restart
5. pm-hibernate
6. Boot machine again; it should automatically trigger the resume process
7. Plymouth animation is visible, and then the screen is garbled for a bit when
   it appears to be restoring the old video buffer, and then a blank screen
   with backlight on, as with the suspend case 

Additional info:
This machine also suffers from the backlight controls not actually affecting the backlight level -- see 34417
Comment 1 Michel Alexandre Salim 2011-02-18 04:51:31 UTC 
Created attachment 43515 [details]
dmesg before suspend
Comment 2 Michel Alexandre Salim 2011-02-18 04:53:21 UTC 
Created attachment 43516 [details]
dmesg after suspend
Comment 3 Michel Alexandre Salim 2011-02-18 04:53:50 UTC 
Created attachment 43517 [details]
intel_reg_dumper before suspend
Comment 4 Michel Alexandre Salim 2011-02-18 04:54:15 UTC 
Created attachment 43518 [details]
intel_reg_dumper after suspend
Comment 5 Michel Alexandre Salim 2011-02-18 06:09:16 UTC 
Created attachment 43520 [details]
intel_gpu_dump before suspend (part 1)

I used split for this since the original file size exceeds Bz limits. Just concat the file to get the original back
Comment 6 Michel Alexandre Salim 2011-02-18 06:10:04 UTC 
Created attachment 43521 [details]
intel_gpu_dump before suspend (part 2)
Comment 7 Michel Alexandre Salim 2011-02-18 06:11:31 UTC 
Created attachment 43522 [details]
intel_gpu_dump after suspend (part 1)
Comment 8 Michel Alexandre Salim 2011-02-18 06:12:06 UTC 
Created attachment 43523 [details]
intel_gpu_dump after suspend (part 2)
Comment 9 Chris Wilson 2011-02-24 02:19:57 UTC 
*** Bug 34417 has been marked as a duplicate of this bug. ***
Comment 10 Chris Wilson 2011-02-24 02:20:10 UTC 
This should be fixed in rc6 with:

commit ed764e7ca042dbf4cc1c7f4e12cd842c7789f133
Author: Michael Karcher <kernel@mkarcher.dialup.fu-berlin.de>
Date:   Sat Feb 12 01:40:16 2011 +0100

    ACPI / Video: Probe for output switch method when searching video devices.
    
    This patch reverts one hunk of 677bd810eedce61edf15452491781ff046b92edc
    "ACPI video: remove output switching control", namely the removal of
    probing for _DOS/_DOD when searching for video devices.
    
along with bug 34417.
Comment 11 Michel Alexandre Salim 2011-02-24 20:17:40 UTC 
I'm still seeing this with both Fedora's 2.6.38-0.rc6.git2.1 and drm-intel-fixes 011b9910bdaf2e52c48c012490ab444fceea1959 (post-RC6) -- should I reopen this and my other bug?
Comment 12 Michel Alexandre Salim 2011-02-28 05:59:45 UTC 
Reported fix does not actually work
Comment 13 Michel Alexandre Salim 2011-03-18 18:28:10 UTC 
The brightness issue (34417) is still unresolved, but the kernel bug tracker report referenced in the Git commit does contain a user report of an additional kernel switch that I need to enable to at least get the display restored when resuming:

i915.lvds_use_ssc=0
Comment 14 Michel Alexandre Salim 2011-07-27 00:17:51 UTC 
Now that there's a quirk infrastructure for disabling SSC on given models, perhaps this particular Vaio Y should be blacklisted as well (see the bug this depends on and the posted Git commit fixing it)

Where do I get the subsystem_vendor and subsystem_device numbers that I need to add my machine to the blacklist?
Comment 15 Michel Alexandre Salim 2011-07-27 00:24:52 UTC 
From lspci -nnvv, it looks like 104d:9076 needs to be added to the quirk list.

Would blacklisting this affect other Sony devices, or are these numbers unique to my particular model?

00:02.0 VGA compatible controller [0300]: Intel Corporation Core Processor Integrated Graphics Controller [8086:0046] (rev 02) (prog-if 00 [VGA controller])
        Subsystem: Sony Corporation Device [104d:9076]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 41
        Region 0: Memory at d0000000 (64-bit, non-prefetchable) [size=4M]
        Region 2: Memory at c0000000 (64-bit, prefetchable) [size=256M]
        Region 4: I/O ports at 5050 [size=8]
        Expansion ROM at <unassigned> [disabled]
        Capabilities: <access denied>
        Kernel driver in use: i915
        Kernel modules: i915
Comment 16 Michel Alexandre Salim 2011-07-27 06:29:53 UTC 
Created attachment 49615 [details] [review]
drm/i915: Add quirk to disable SSC on Sony Vaio Y2

Patch fixing this, to be applied to the 3.0 source tree, as sent to the mailing lists:

  http://article.gmane.org/gmane.linux.kernel/1172092
Comment 17 Florian Mickler 2011-08-08 01:52:48 UTC 
A patch referencing this bug report has been merged in Linux v3.1-rc1:

commit 070d329ae52e2fde341771d753a5b728145881f4
Author: Michel Alexandre Salim <salimma@fedoraproject.org>
Date:   Thu Jul 28 18:52:06 2011 +0200

    drm/i915: Add quirk to disable SSC on Sony Vaio Y2

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.