NetBSD Problem Report #49875

From dieter.r@pandora.be  Sun May  3 12:54:42 2015
Return-Path: <dieter.r@pandora.be>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id F32EEA57FD
	for <gnats-bugs@gnats.NetBSD.org>; Sun,  3 May 2015 12:54:41 +0000 (UTC)
Message-Id: <20150503123932.20BF3BEB50@simult.amelgem.be>
Date: Sun,  3 May 2015 14:39:32 +0200 (CEST)
From: dieter.NetBSD@pandora.be
Reply-To: dieter.NetBSD@pandora.be
To: gnats-bugs@NetBSD.org
Subject: kernel crash with i915drmkms on Intel 82855GM
X-Send-Pr-Version: 3.95

>Number:         49875
>Category:       kern
>Synopsis:       kernel crash with i915drmkms on Intel 82855GM
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    riastradh
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun May 03 12:55:00 +0000 2015
>Closed-Date:    Wed Aug 05 06:47:02 +0000 2015
>Last-Modified:  Wed Aug 05 06:47:02 +0000 2015
>Originator:     dieter roelants
>Release:        NetBSD 7.0_BETA
>Organization:
>Environment:
System: NetBSD origin.amelgem.be 7.0_BETA NetBSD 7.0_BETA (GENERIC.201504302140Z) i386
Architecture: i386
Machine: i386
>Description:
	Starting Xorg (chrooted, but I assume it doesn't matter)
	on my laptop with "Intel 82855GM GMCH Integrated Graphics
	Device", the kernel crashes with following messages and bt:

warning: /usr/src/sys/external/bsd/drm2/dist/drm/i915/i915_gem.c:5689: list_empty(&obj->vma_list    
uvm_fault(0xc3429774, 0, 1) -> 0xe
fatal page fault in supervisor mode
trap type 6 code 0 eip c03d5c7a cs 8 eflags 13246 cr2 54 ilevel 0 esp c322fa00 
curlwp 0xc37fb2c0 pid 741 lid 1 lowest kstack 0xdbfca2c0
kernel: supervisor trap page fault, code=0
Stopped in pid 741.1 (Xorg) at  netbsd:i915_gem_object_pin_to_display_plane+0xb3:       movzbl  54(%
eax),%edx

db{0}> bt
i915_gem_object_pin_to_display_plane() at netbsd:i915_gem_object_pin_to_display_plane+0xb3
intel_pin_and_fence_fb_obj() at netbsd:intel_pin_and_fence_fb_obj+0x99
intel_pipe_set_base() at netbsd:intel_pipe_set_base+0x73
intel_crtc_set_config() at netbsd:intel_crtc_set_config+0x894 
drm_mode_set_config_internal() at netbsd:drm_mode_set_config_internal+0x4d
drm_mode_setcrtc() at netbsd:drm_mode_setcrtc+0xc6
drm_ioctl() at netbsd:drm_ioctl+0xff  
sys_ioctl() at netbsd:sys_ioctl+0x198 
syscall() at netbsd:syscall+0x7f 
--- syscall (number 54) ---
bb77bcc7:

	Kernel messages about the graphics device are:

agp0 at pchb0: i855-family chipset
agp0: detected 8060k stolen memory
agp0: aperture at 0xe8000000, size 0x8000000
i915drmkms0 at pci0 dev 2 function 0: Intel 82855GM GMCH Integrated Graphics Device (rev. 0x02)
drm: Memory usable by graphics device = 128M
drm: Supports vblank timestamp caching Rev 2 (21.10.2013).
drm: Driver supports precise vblank timestamp query.
i915drmkms0: interrupting at irq 9 (i915)
drm: GMBUS [i915 gmbus panel] timed out, falling back to bit banging on pin 3
drm: initialized overlay support
intelfb0 at i915drmkms0
i915drmkms0: info: registered panic notifier
intelfb0: framebuffer at 0xdadd4000, size 1400x1050, depth 32, stride 5632

>How-To-Repeat:
	Boot NetBSD 7.0_BETA, startx.
>Fix:
	The upstream commit 19656430a874132a1d79c56387a6eec1ef9a5689
	(See https://freedesktop.org/patch/26114/) stops the kernel
	from crashing, and results in an error message instead;
	"DRM error in intel_pipe_set_base: pin & fence failed".

	I also found that neither the crash nor the error (with
	patched kernel) occur when I `dd if=/dev/zeroof=/dev/null
	bs=512m count=1` before starting X, and it seems to be
	running ok. I haven't stressed it, though. Should I file
	a separate PR for this?

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: kern-bug-people->riastradh
Responsible-Changed-By: tsutsui@NetBSD.org
Responsible-Changed-When: Sat, 25 Jul 2015 15:04:49 +0000
Responsible-Changed-Why:
Could you please take a look at this PR?
Is it okay to apply upstream change as mentioned in PR?
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=19656430a874132a1d79c56387a6eec1ef9a5689
Thanks,


From: matthew green <mrg@eterna.com.au>
To: gnats-bugs@NetBSD.org
Cc: riastradh@NetBSD.org, kern-bug-people@netbsd.org,
    netbsd-bugs@netbsd.org, gnats-admin@netbsd.org, tsutsui@NetBSD.org,
    dieter.NetBSD@pandora.be
Subject: re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sun, 26 Jul 2015 07:13:47 +1000

 tsutsui@NetBSD.org writes:
 > Synopsis: kernel crash with i915drmkms on Intel 82855GM
 > 
 > Responsible-Changed-From-To: kern-bug-people->riastradh
 > Responsible-Changed-By: tsutsui@NetBSD.org
 > Responsible-Changed-When: Sat, 25 Jul 2015 15:04:49 +0000
 > Responsible-Changed-Why:
 > Could you please take a look at this PR?
 > Is it okay to apply upstream change as mentioned in PR?
 > http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=19656430a874132a1d79c56387a6eec1ef9a5689
 > Thanks,

 my limited understanding says this is OK to commit.  and to send
 for a pullup...


 .mrg.

From: Taylor R Campbell <riastradh@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, netbsd-bugs@netbsd.org,
	gnats-admin@netbsd.org, tsutsui@NetBSD.org, dieter.NetBSD@pandora.be
Subject: Re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sun, 26 Jul 2015 02:12:14 +0000

    Date: Sat, 25 Jul 2015 15:04:49 +0000 (UTC)
    From: tsutsui@NetBSD.org

    Could you please take a look at this PR?
    Is it okay to apply upstream change as mentioned in PR?
    http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?i=
 d=3D19656430a874132a1d79c56387a6eec1ef9a5689

 We can apply the change, although it's not clear to me that that fixes
 whatever the real problem here is.  Does X successfully start -- with
 an apparently harmless warning in dmesg -- or does it fail in some
 other way with the change?

From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: riastradh@NetBSD.org
Cc: gnats-bugs@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sun, 26 Jul 2015 16:08:38 +0900

 > We can apply the change, although it's not clear to me that that fixes
 > whatever the real problem here is.

 - The real problem is kernel panic (NULL pointer dereference?)
   in i915_gem_object_pin_to_display_plane() on starting Xserver

 > Does X successfully start -- with
 > an apparently harmless warning in dmesg -- or does it fail in some
 > other way with the change?

 - With the upstream change X successfully starts with warnings
   (at least no kernel panic)
 - The warnings are not stable:
  - On the first startup
    "DRM error in intel_pipe_set_base: pin & fence failed"
  - On the second (and further) startups
    "DRM error in i8xx_irq_handler: pipe B underrun"
 - Xserver sometimes renders garbages and hangs when
   multiple heavy GTK applications (firefox + mikutter etc.)
   are running
   (I'm not sure if this is 855GM specific or not, though)

 In short word, "the upstream fix is not perfect but much better".

 ---
 Izumi Tsutsui

From: Taylor R Campbell <riastradh@NetBSD.org>
To: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
Cc: gnats-bugs@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sun, 26 Jul 2015 14:50:57 +0000

    Date: Sun, 26 Jul 2015 16:08:38 +0900
    From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>

    - Xserver sometimes renders garbages and hangs when
      multiple heavy GTK applications (firefox + mikutter etc.)
      are running
      (I'm not sure if this is 855GM specific or not, though)

 So it sounds like there is still a problem here, and it is probably
 855GM specific.  I'll apply the change, but leave this open, unless
 you or the original submitter would like to submit a separate PR for
 the symptoms that the change reveals.

From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: riastradh@NetBSD.org
Cc: gnats-bugs@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sat, 1 Aug 2015 13:17:10 +0900

 riastradh@ wrote:

 > So it sounds like there is still a problem here, and it is probably
 > 855GM specific.  I'll apply the change, but leave this open, unless
 > you or the original submitter would like to submit a separate PR for
 > the symptoms that the change reveals.

 I can file a new PR for 855GM issue with details, but
 will you commit the upstream fix, or should I commit it first?
  http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=19656430a874132a1d79c56387a6eec1ef9a5689

 ---
 Izumi Tsutsui

From: Taylor R Campbell <riastradh@NetBSD.org>
To: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
Cc: gnats-bugs@NetBSD.org, tsutsui@ceres.dti.ne.jp
Subject: Re: kern/49875 (kernel crash with i915drmkms on Intel 82855GM)
Date: Sat, 1 Aug 2015 04:24:45 +0000

    Date: Sat, 1 Aug 2015 13:17:10 +0900
    From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>

    riastradh@ wrote:

    > So it sounds like there is still a problem here, and it is probably
    > 855GM specific.  I'll apply the change, but leave this open, unless
    > you or the original submitter would like to submit a separate PR for
    > the symptoms that the change reveals.

    I can file a new PR for 855GM issue with details, but
    will you commit the upstream fix, or should I commit it first?
     http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?=
 id=3D19656430a874132a1d79c56387a6eec1ef9a5689

 Sorry, I didn't get around to it.  Please feel free to commit the
 change!

From: "Izumi Tsutsui" <tsutsui@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49875 CVS commit: src/sys/external/bsd/drm2/dist/drm/i915
Date: Sat, 1 Aug 2015 13:42:32 +0000

 Module Name:	src
 Committed By:	tsutsui
 Date:		Sat Aug  1 13:42:32 UTC 2015

 Modified Files:
 	src/sys/external/bsd/drm2/dist/drm/i915: i915_gem.c

 Log Message:
 Pull upstream fix to avoid kernel panic on starting X on Intel 855GM machines.

 Reported in PR kern/49875, and ok'ed to commit by riastradh@.

 Should be pulled up to netbsd-7.


 To generate a diff of this commit:
 cvs rdiff -u -r1.31 -r1.32 src/sys/external/bsd/drm2/dist/drm/i915/i915_gem.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

State-Changed-From-To: open->pending-pullups
State-Changed-By: tsutsui@NetBSD.org
State-Changed-When: Sun, 02 Aug 2015 12:35:46 +0000
State-Changed-Why:
pullup-7 #931


From: "Soren Jacobsen" <snj@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49875 CVS commit: [netbsd-7] src/sys/external/bsd/drm2/dist/drm/i915
Date: Tue, 4 Aug 2015 17:18:40 +0000

 Module Name:	src
 Committed By:	snj
 Date:		Tue Aug  4 17:18:40 UTC 2015

 Modified Files:
 	src/sys/external/bsd/drm2/dist/drm/i915 [netbsd-7]: i915_gem.c

 Log Message:
 Pull up following revision(s) (requested by tsutsui in ticket #931):
 	sys/external/bsd/drm2/dist/drm/i915/i915_gem.c: revision 1.32
 Pull upstream fix to avoid kernel panic on starting X on Intel 855GM machines.
 Reported in PR kern/49875, and ok'ed to commit by riastradh@.


 To generate a diff of this commit:
 cvs rdiff -u -r1.14.2.10 -r1.14.2.11 \
     src/sys/external/bsd/drm2/dist/drm/i915/i915_gem.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

State-Changed-From-To: pending-pullups->closed
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Wed, 05 Aug 2015 06:47:02 +0000
State-Changed-Why:
fixed and pulled up, thanks


>Unformatted:

Home	PR Database Search