Bug 596557 - KMS initialization breaks display [8086:0046]
KMS initialization breaks display [8086:0046]
Status: CLOSED WONTFIX
Product: Fedora
Classification: Fedora
Component: xorg-x11-drv-intel (Show other bugs)
14
All Linux
low Severity high
: ---
: ---
Assigned To: Adam Jackson
Fedora Extras Quality Assurance
RejectedNTH https://fedoraproject.org...
: CommonBugs, Patch, Triaged
: 618481 (view as bug list)
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2010-05-26 19:48 EDT by Adam Williamson
Modified: 2012-08-16 12:56 EDT (History)
11 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2012-08-16 12:56:36 EDT
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:


Attachments (Terms of Use)
lspci output on the affected system (2.61 KB, text/plain)
2010-05-26 19:49 EDT, Adam Williamson
no flags Details
/var/log/messages from the affected system (566.05 KB, text/plain)
2010-05-26 19:55 EDT, Adam Williamson
no flags Details
drm debug messages from a load of i915 (4.15 KB, text/plain)
2010-05-28 16:43 EDT, Adam Williamson
no flags Details


External Trackers
Tracker ID Priority Status Summary Last Updated
FreeDesktop.org 29141 None None None Never
FreeDesktop.org 31988 None None None Never

  None (edit)
Description Adam Williamson 2010-05-26 19:48:07 EDT
I have here a Sony Vaio Z laptop. It's a fairly new model with switchable graphics: it has both a GeForce and an Intel integrated graphics chipset.

I can configure the BIOS so I can enable only one card during any given boot, using the selector switch on the laptop. At this point it ought to be just like dealing with a single adapter for the OS, so we should be able to cope with driving either adapter using nouveau or intel (respectively). In fact, neither work. Here's the report for the intel adapter.

As soon as i915 gets loaded with KMS enabled, the screen goes blank (with the backlight on). If I load i915 without KMS, it succeeds, but obviously this is no use for actually running X now the UMS support is gone from the X driver.

I will attach relevant sections from /var/log/messages , and the lspci -nn output. Xorg.0.log doesn't seem useful in this case as the failure is earlier.

Chip is: 00:02.0 VGA compatible controller [0300]: Intel Corporation Core Processor Integrated Graphics Controller [8086:0046] (rev 02)
Comment 1 Adam Williamson 2010-05-26 19:49:00 EDT
Created attachment 417065 [details]
lspci output on the affected system
Comment 2 Adam Williamson 2010-05-26 19:54:02 EDT
By the looks of the logs, the following appears in /var/log/messages when loading i915 *without* KMS:

May 26 16:39:05 vaioz kernel: [drm] Initialized drm 1.1.0 20060810
May 26 16:39:05 vaioz kernel: pci 0000:00:02.0: power state changed by ACPI to D0
May 26 16:39:05 vaioz kernel: pci 0000:00:02.0: power state changed by ACPI to D0
May 26 16:39:05 vaioz kernel: pci 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
May 26 16:39:05 vaioz kernel: mtrr: no more MTRRs available
May 26 16:39:05 vaioz kernel: [drm] MTRR allocation failed.  Graphics performance may suffer.
May 26 16:39:05 vaioz kernel: acpi device:00: registered as cooling_device4
May 26 16:39:05 vaioz kernel: input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input7
May 26 16:39:05 vaioz kernel: ACPI: Video Device [GFX0] (multi-head: yes  rom: no  post: no)
May 26 16:39:05 vaioz kernel: [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
May 26 16:39:05 vaioz kernel: modprobe used greatest stack depth: 3176 bytes left

when loading i915 *with* KMS, nothing at all appears in the logs. There's nothing in /var/log/messages between where boot is complete, and when the next boot starts up. I will attach the entire /var/log/messages . Here's my timing notes. All refer to today, May 26th:

16:36:28 - boot complete
~16:37 - modprobe i915 , screen blank with backlight on

16:38:35 - second boot complete
~16:39 - modprobe i915 modeset=0, screen fine



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 3 Adam Williamson 2010-05-26 19:55:15 EDT
Created attachment 417066 [details]
/var/log/messages from the affected system
Comment 4 Adam Williamson 2010-05-26 19:55:48 EDT
if you're wondering why the adapter apparently disappears entirely outside of these couple of test boots, to actually *use* the system I boot with the NVIDIA adapter selected, and use the NVIDIA proprietary driver (which works).



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 5 Adam Williamson 2010-05-26 20:00:28 EDT
Final note - this happens with both 2.6.34-2.fc14.x86_64 and 2.6.33.4-95.fc13.x86_64 .



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 6 Adam Williamson 2010-05-28 16:42:34 EDT
with thanks to the inhabitants of #fedora-devel, I'll attach the drm debug messages (with drm.debug=0x06 as directed by ajax).



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 7 Adam Williamson 2010-05-28 16:43:00 EDT
Created attachment 417742 [details]
drm debug messages from a load of i915
Comment 8 Adam Williamson 2010-08-20 14:19:46 EDT
This is still the case with the latest F14 kernel (2.6.35.2-9.fc14) - same behaviour, as soon as KMS kicks in, the system hangs with backlight on.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 9 Adam Williamson 2010-08-26 14:12:02 EDT
*** Bug 618481 has been marked as a duplicate of this bug. ***
Comment 10 Adam Williamson 2010-09-28 14:49:07 EDT
Upstream:

https://bugs.freedesktop.org/show_bug.cgi?id=29141

Jesse Barnes apparently now has a matching system and is working on this.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 11 Adam Williamson 2010-10-08 12:29:50 EDT
Jesse's posted fixes to his edp-hacks branch which others have tested with positive results - see the upstream bug. I'm trying to patch the fixes onto an F15 kernel build to test right now.

I'm fairly sure the fixes are the commits:

17 hours ago 	Jesse Barnes	drm/i915/dp: more broken sequencing fixes  edp-hacks 	commit | commitdiff | tree | snapshot
43 hours ago 	Jesse Barnes	drm/i915/dp: broken eDP power sequencing bits 	commit | commitdiff | tree | snapshot
43 hours ago 	Jesse Barnes	drm/i915/dp: wait for panel idle when enabling/disablin... 	commit | commitdiff | tree | snapshot
Comment 13 Adam Williamson 2010-10-08 14:57:02 EDT
Discussed at the blocker/NTH review meeting today, given the ickiness of making changes to panel code this late, we decided not to take it as NTH.
Comment 14 Adam Williamson 2010-10-08 18:28:00 EDT
so I'm sucking at making these patches apply to the f15 kernel and build. But they have gone into drm-intel-next as part of a giant change set (just look at all the commits from Jesse dated together, late on 7th Oct). ajax, could you possibly just pull that whole change set into the f15 kernel (note: f15 not f14) for me to try? thanks!
Comment 15 Adam Williamson 2010-10-09 14:47:01 EDT
OK, so I couldn't get a clean patch against a Fedora kernel, but I just built a kernel straight from the upstream drm-intel-next tree and confirmed that it works great.
Comment 16 Adam Williamson 2010-11-01 21:18:22 EDT

-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 17 Adam Williamson 2010-11-29 13:21:09 EST
bumping this to at least 14 because i don't see it being backported to 13; it may eventually only be fixed in rawhide/15.

it was fixed briefly in rawhide by a backport of drm-intel-next/edp-fixes but that was backed out apparently for causing other problems, so current status is that it's still broken everywhere.
Comment 18 Claes Wallin 2011-01-02 20:13:49 EST
I had similar problems on a Lenovo U160 with integrated Intel graphics. linux-2.6.37-rc8 fixes the problem for me. Have you tried it?
Comment 19 Adam Williamson 2011-01-03 07:26:46 EST
Yes, and no. When it comes to graphics, 'similar' problems are generally not similar at all. I'm already well aware of all relevant changes for this bug. But thanks. The reason this report isn't changing much is that all the interesting stuff is happening on the upstream report.



-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 20 Adam Williamson 2011-01-03 07:28:31 EST

-- 
Fedora Bugzappers volunteer triage team
https://fedoraproject.org/wiki/BugZappers
Comment 21 Fedora End Of Life 2012-08-16 12:56:39 EDT
This message is a notice that Fedora 14 is now at end of life. Fedora 
has stopped maintaining and issuing updates for Fedora 14. It is 
Fedora's policy to close all bug reports from releases that are no 
longer maintained.  At this time, all open bugs with a Fedora 'version'
of '14' have been closed as WONTFIX.

(Please note: Our normal process is to give advanced warning of this 
occurring, but we forgot to do that. A thousand apologies.)

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, feel free to reopen 
this bug and simply change the 'version' to a later Fedora version.

Bug Reporter: Thank you for reporting this issue and we are sorry that 
we were unable to fix it before Fedora 14 reached end of life. If you 
would still like to see this bug fixed and are able to reproduce it 
against a later version of Fedora, you are encouraged to click on 
"Clone This Bug" (top right of this page) and open it against that 
version of Fedora.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events.  Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

The process we are following is described here: 
http://fedoraproject.org/wiki/BugZappers/HouseKeeping

Note You need to log in before you can comment on or make changes to this bug.