Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 on nouveau with SW rendering

Bug #1033533 reported by rajesh on 2012-08-06
932
This bug affects 105 people
Affects Status Importance Assigned to Milestone
xserver-xorg-video-nouveau (Ubuntu)
High
Maarten Lankhorst
Trusty
High
Unassigned

Bug Description

[Errors Bucket]
https://errors.ubuntu.com/bucket/?id=/usr/bin/Xorg:6:memcpy:exaMemcpyBox:exaCopyDirty:exaCopyDirtyToSys:exaPrepareAccessReg_mixed

Crash doing memcpy in EXA on nouveau, with software rendering activated.

#8 <signal handler called>
No symbol table info available.
#9 __memcpy_sse2 () at ../sysdeps/x86_64/multiarch/../memcpy.S:72
No locals.
#10 0x00007f570e63684e in memcpy (__len=23, __src=0x0, __dest=0x7f5713877100) at /usr/include/x86_64-linux-gnu/bits/string3.h:52
No locals.
#11 exaMemcpyBox (pbox=pbox@entry=0x7fffef000de0, src=0x0, src_pitch=64, dst=0x7f5713877100 "", dst_pitch=24, pPixmap=<error reading variable: Unhandled dwarf expression opcode 0xfa>) at ../../exa/exa_migration_classic.c:59
        i = 16
        cpp = <optimized out>
        bytes = 23
#12 0x00007f570e636d03 in exaCopyDirty (migrate=migrate@entry=0x7fffef000e40, pValidDst=0x7f5713dc3e40, pValidSrc=0x7f5713dc3e50, transfer=0x7f570f28b7f0 <nouveau_exa_download_from_screen>, fallback_index=fallback_index@entry=1, sync=0x7f570e635780 <exaWaitSync>) at ../../exa/exa_migration_classic.c:240
        pPixmap = 0x7f5713dc3da0
        pExaPixmap = 0x7f5713dc3e00
        damage = <optimized out>
        CopyReg = {extents = {x1 = 0, y1 = 0, x2 = 23, y2 = 16}, data = 0x0}
        save_use_gpu_copy = 1
        save_pitch = 64
        pBox = 0x7fffef000de0
        nbox = 1
        access_prepared = 1
        need_sync = 0
#13 0x00007f570e636f32 in exaCopyDirtyToSys (migrate=migrate@entry=0x7fffef000e40) at ../../exa/exa_migration_classic.c:285
        pPixmap = <optimized out>
        pExaScr = 0x7f5713877100
        pExaPixmap = 0x0
#14 0x00007f570e63951a in exaPrepareAccessReg_mixed (pPixmap=0x7f5713dc3da0, index=<optimized out>, pReg=0x0) at ../../exa/exa_migration_mixed.c:254
        as_dst = 1
        pixmaps = {{as_dst = 0, as_src = 1, pPix = 0x7f5713dc3da0, pReg = 0x0}}
        pExaPixmap = 0x7f5713dc3e00
        has_gpu_copy = <optimized out>
        success = <optimized out>

[Original Report]
The crash report started immediately after login.

ProblemType: CrashDistroRelease: Ubuntu 12.10
Package: xserver-xorg-core 2:1.12.99.902-0ubuntu1
ProcVersionSignature: Ubuntu 3.5.0-8.8-generic 3.5.0
Uname: Linux 3.5.0-8-generic x86_64
.tmp.unity.support.test.0:

ApportVersion: 2.4-0ubuntu6
Architecture: amd64
CompizPlugins: No value set for `/apps/compiz-1/general/screen0/options/active_plugins'
CompositorRunning: None
CrashCounter: 1
Date: Sun Aug 5 14:46:44 2012
DistUpgraded: Fresh install
DistroCodename: quantal
DistroVariant: ubuntu
DkmsStatus:
 virtualbox, 4.1.18, 3.5.0-6-generic, x86_64: installed
 virtualbox, 4.1.18, 3.5.0-8-generic, x86_64: installed
ExecutablePath: /usr/bin/Xorg
ExtraDebuggingInterest: Yes, if not too technical
GraphicsCard:
 NVIDIA Corporation C61 [GeForce 6100 nForce 400] [10de:03d2] (rev a2) (prog-if 00 [VGA controller])
   Subsystem: NVIDIA Corporation Device [10de:cb84]
InstallationMedia: Ubuntu 12.10 "Quantal Quetzal" - Alpha amd64 (20120519)
Lsusb:
 Bus 001 Device 002: ID 18ec:3399 Arkmicro Technologies Inc.
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
ProcCmdline: /usr/bin/X :0 -auth /var/run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch -background none
ProcEnviron:

ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.5.0-8-generic root=UUID=a80eff25-65c3-4db0-adad-31d8ea95823b ro quiet splash vt.handoff=7
Renderer: Software
Signal: 6SourcePackage: xorg-server
StacktraceTop:
 ?? () from /lib/x86_64-linux-gnu/libc.so.6
 ?? () from /usr/lib/xorg/modules/libexa.so
 ?? () from /usr/lib/xorg/modules/libexa.so
 ?? () from /usr/lib/xorg/modules/libexa.so
 ?? () from /usr/lib/xorg/modules/libexa.so
Title: Xorg crashed with SIGABRT
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

dmi.bios.date: 09/22/2006
dmi.bios.vendor: Phoenix Technologies, LTD
dmi.bios.version: 6.00 PG
dmi.board.name: NF-MCP61
dmi.chassis.type: 3
dmi.modalias: dmi:bvnPhoenixTechnologies,LTD:bvr6.00PG:bd09/22/2006:svn:pn:pvr:rvn:rnNF-MCP61:rvr:cvn:ct3:cvr:
version.compiz: compiz 1:0.9.8+bzr3249-0ubuntu2
version.ia32-libs: ia32-libs N/A
version.libdrm2: libdrm2 2.4.37-0ubuntu2
version.libgl1-mesa-dri: libgl1-mesa-dri 8.0.4-1ubuntu1
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 8.0.4-1ubuntu1
version.xserver-xorg-core: xserver-xorg-core 2:1.12.99.902-0ubuntu1
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev 1:2.7.0-2~ubuntu1
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:6.99.99~git20120713.6ef1ad6a-0ubuntu1
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.20.2-1ubuntu1
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:1.0.1-4~ubuntu1

rajesh (bhatt-rajesh20) wrote :

StacktraceTop:
 memcpy (__len=23, __src=0x0, __dest=0x7f5713877100) at /usr/include/x86_64-linux-gnu/bits/string3.h:52
 exaMemcpyBox (pbox=pbox@entry=0x7fffef000de0, src=0x0, src_pitch=64, dst=0x7f5713877100 "", dst_pitch=24, pPixmap=<error reading variable: Unhandled dwarf expression opcode 0xfa>) at ../../exa/exa_migration_classic.c:59
 exaCopyDirty (migrate=migrate@entry=0x7fffef000e40, pValidDst=0x7f5713dc3e40, pValidSrc=0x7f5713dc3e50, transfer=0x7f570f28b7f0 <nouveau_exa_download_from_screen>, fallback_index=fallback_index@entry=1, sync=0x7f570e635780 <exaWaitSync>) at ../../exa/exa_migration_classic.c:240
 exaCopyDirtyToSys (migrate=migrate@entry=0x7fffef000e40) at ../../exa/exa_migration_classic.c:285
 exaPrepareAccessReg_mixed (pPixmap=0x7f5713dc3da0, index=<optimized out>, pReg=0x0) at ../../exa/exa_migration_mixed.c:254

Changed in xorg-server (Ubuntu):
importance: Undecided → Medium
tags: removed: need-amd64-retrace
Timo Aaltonen (tjaalton) on 2012-08-08
visibility: private → public

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in xorg-server (Ubuntu):
status: New → Confirmed
dino99 (9d9) wrote :

Also get this issue with xserver 904 & kernel 3.5.0.10 i386

tags: added: bugpattern-needed
Bryce Harrington (bryce) on 2012-08-28
Changed in xorg-server (Ubuntu):
importance: Medium → Critical
status: Confirmed → Triaged
description: updated
summary: - Xorg crashed with SIGABRT
+ Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 - "Unhandled dwarf
+ expression opcode 0xfa"
Bryce Harrington (bryce) on 2012-08-28
Changed in xorg-server (Ubuntu):
assignee: nobody → Canonical X.org (canonical-x)
Changed in xorg-server (Ubuntu Precise):
importance: Undecided → Critical
assignee: nobody → Canonical X.org (canonical-x)

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in xorg-server (Ubuntu Precise):
status: New → Confirmed
Bryce Harrington (bryce) on 2012-08-28
description: updated
Bryce Harrington (bryce) on 2012-08-28
no longer affects: xorg-server (Ubuntu Precise)
description: updated
Bryce Harrington (bryce) wrote :

In perusing the dupe bugs, all are nouveau with software rendering. Several are using virtualbox but most are getting sw for some other reason.

Changed in xorg-server (Ubuntu):
importance: Critical → High
summary: - Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 - "Unhandled dwarf
- expression opcode 0xfa"
+ Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 on nouveau with SW
+ rendering
Bryce Harrington (bryce) on 2012-08-28
Changed in xorg-server (Ubuntu):
milestone: none → ubuntu-12.10
Timo Aaltonen (tjaalton) on 2012-09-20
affects: xorg-server (Ubuntu) → xserver-xorg-video-nouveau (Ubuntu)
Changed in xserver-xorg-video-nouveau (Ubuntu):
status: Triaged → Confirmed
Bryce Harrington (bryce) wrote :

We have a lot of dupes on this bug. Maarten, can you also investigate this issue as you have time?

Changed in xserver-xorg-video-nouveau (Ubuntu):
assignee: Canonical X.org (canonical-x) → Maarten Lankhorst (mlankhorst)
Maarten Lankhorst (mlankhorst) wrote :

Someone suspected there was still a virtual memory leak in nouveau, is there any way I can reproduce it sooner or later?

dino99 (9d9) wrote :

on quantal i386, this is no more an issue on my system since a while.

Gaele Strootman (gaele) wrote :

This happened to me several times during the last days (including today with up-to-date Quantal), while scrolling or switching tabs in Firefox. I'm not using Virtualbox.

Changed in xserver-xorg-video-nouveau (Ubuntu):
milestone: ubuntu-12.10 → quantal-updates
Maarten Lankhorst (mlankhorst) wrote :

How can I reproduce this easily?

Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Confirmed → Incomplete
Download full text (5.9 KiB)

This happens on logging in with Gnome (full effects) and Ubuntu and
using for a few minutes, then crashes.

Regards
Alex

On Mon, 2012-10-22 at 12:43 +0000, Maarten Lankhorst wrote:
> How can I reproduce this easily?
>
> ** Changed in: xserver-xorg-video-nouveau (Ubuntu Quantal)
> Status: Confirmed => Incomplete
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (1045115).
> https://bugs.launchpad.net/bugs/1033533
>
> Title:
> Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 on nouveau with
> SW rendering
>
> Status in “xserver-xorg-video-nouveau” package in Ubuntu:
> Confirmed
> Status in “xserver-xorg-video-nouveau” source package in Quantal:
> Incomplete
>
> Bug description:
> Crash doing memcpy in EXA on nouveau, with software rendering
> activated.
>
> #8 <signal handler called>
> No symbol table info available.
> #9 __memcpy_sse2 () at ../sysdeps/x86_64/multiarch/../memcpy.S:72
> No locals.
> #10 0x00007f570e63684e in memcpy (__len=23, __src=0x0, __dest=0x7f5713877100) at /usr/include/x86_64-linux-gnu/bits/string3.h:52
> No locals.
> #11 exaMemcpyBox (pbox=pbox@entry=0x7fffef000de0, src=0x0, src_pitch=64, dst=0x7f5713877100 "", dst_pitch=24, pPixmap=<error reading variable: Unhandled dwarf expression opcode 0xfa>) at ../../exa/exa_migration_classic.c:59
> i = 16
> cpp = <optimized out>
> bytes = 23
> #12 0x00007f570e636d03 in exaCopyDirty (migrate=migrate@entry=0x7fffef000e40, pValidDst=0x7f5713dc3e40, pValidSrc=0x7f5713dc3e50, transfer=0x7f570f28b7f0 <nouveau_exa_download_from_screen>, fallback_index=fallback_index@entry=1, sync=0x7f570e635780 <exaWaitSync>) at ../../exa/exa_migration_classic.c:240
> pPixmap = 0x7f5713dc3da0
> pExaPixmap = 0x7f5713dc3e00
> damage = <optimized out>
> CopyReg = {extents = {x1 = 0, y1 = 0, x2 = 23, y2 = 16}, data = 0x0}
> save_use_gpu_copy = 1
> save_pitch = 64
> pBox = 0x7fffef000de0
> nbox = 1
> access_prepared = 1
> need_sync = 0
> #13 0x00007f570e636f32 in exaCopyDirtyToSys (migrate=migrate@entry=0x7fffef000e40) at ../../exa/exa_migration_classic.c:285
> pPixmap = <optimized out>
> pExaScr = 0x7f5713877100
> pExaPixmap = 0x0
> #14 0x00007f570e63951a in exaPrepareAccessReg_mixed (pPixmap=0x7f5713dc3da0, index=<optimized out>, pReg=0x0) at ../../exa/exa_migration_mixed.c:254
> as_dst = 1
> pixmaps = {{as_dst = 0, as_src = 1, pPix = 0x7f5713dc3da0, pReg = 0x0}}
> pExaPixmap = 0x7f5713dc3e00
> has_gpu_copy = <optimized out>
> success = <optimized out>
>
> [Original Report]
> The crash report started immediately after login.
>
> ProblemType: Crash
> DistroRelease: Ubuntu 12.10
> Package: xserver-xorg-core 2:1.12.99.902-0ubuntu1
> ProcVersionSignature: Ubuntu 3.5.0-8.8-generic 3.5.0
> Uname: Linux 3.5.0-8-generic x86_64
> .tmp.unity.support.test.0:
>
> ApportVersion: 2.4-0ubuntu6
> Architecture: amd64
> CompizPlugins: No value set for `/apps/compiz-1/general/screen0/options/act...

Read more...

Download full text (11.8 KiB)

I have this old Lenovo laptop T61. The graphics have been dysfunctional
during the last 14 days of updating quantal, practically continuously, and a
few times before. Meaning big squares on the screen, no or few clicks
registred, HUD not showing and such. Also some high processor use for
Cheese. So I figured the driver was not working and changed to nvidias
driver. problems gone.

-----Opprinnelig melding-----
From: Alex Buell
Sent: Monday, October 22, 2012 3:17 PM
To: <email address hidden>
Subject: Re: [Bug 1033533] Re: Xorg crashed with SIGABRT: exaMemcpyBox
withsrc=0x0 on nouveau with SW rendering

This happens on logging in with Gnome (full effects) and Ubuntu and
using for a few minutes, then crashes.

Regards
Alex

On Mon, 2012-10-22 at 12:43 +0000, Maarten Lankhorst wrote:
> How can I reproduce this easily?
>
> ** Changed in: xserver-xorg-video-nouveau (Ubuntu Quantal)
> Status: Confirmed => Incomplete
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (1045115).
> https://bugs.launchpad.net/bugs/1033533
>
> Title:
> Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 on nouveau with
> SW rendering
>
> Status in “xserver-xorg-video-nouveau” package in Ubuntu:
> Confirmed
> Status in “xserver-xorg-video-nouveau” source package in Quantal:
> Incomplete
>
> Bug description:
> Crash doing memcpy in EXA on nouveau, with software rendering
> activated.
>
> #8 <signal handler called>
> No symbol table info available.
> #9 __memcpy_sse2 () at ../sysdeps/x86_64/multiarch/../memcpy.S:72
> No locals.
> #10 0x00007f570e63684e in memcpy (__len=23, __src=0x0,
> __dest=0x7f5713877100) at /usr/include/x86_64-linux-gnu/bits/string3.h:52
> No locals.
> #11 exaMemcpyBox (pbox=pbox@entry=0x7fffef000de0, src=0x0, src_pitch=64,
> dst=0x7f5713877100 "", dst_pitch=24, pPixmap=<error reading variable:
> Unhandled dwarf expression opcode 0xfa>) at
> ../../exa/exa_migration_classic.c:59
> i = 16
> cpp = <optimized out>
> bytes = 23
> #12 0x00007f570e636d03 in exaCopyDirty
> (migrate=migrate@entry=0x7fffef000e40, pValidDst=0x7f5713dc3e40,
> pValidSrc=0x7f5713dc3e50, transfer=0x7f570f28b7f0
> <nouveau_exa_download_from_screen>, fallback_index=fallback_index@entry=1,
> sync=0x7f570e635780 <exaWaitSync>) at
> ../../exa/exa_migration_classic.c:240
> pPixmap = 0x7f5713dc3da0
> pExaPixmap = 0x7f5713dc3e00
> damage = <optimized out>
> CopyReg = {extents = {x1 = 0, y1 = 0, x2 = 23, y2 = 16}, data =
> 0x0}
> save_use_gpu_copy = 1
> save_pitch = 64
> pBox = 0x7fffef000de0
> nbox = 1
> access_prepared = 1
> need_sync = 0
> #13 0x00007f570e636f32 in exaCopyDirtyToSys
> (migrate=migrate@entry=0x7fffef000e40) at
> ../../exa/exa_migration_classic.c:285
> pPixmap = <optimized out>
> pExaScr = 0x7f5713877100
> pExaPixmap = 0x0
> #14 0x00007f570e63951a in exaPrepareAccessReg_mixed
> (pPixmap=0x7f5713dc3da0, index=<optimized out>, pReg=0x0) at
> ../../exa/exa_migration_mixed.c:254
> as_dst = 1
> ...

Alex Buell (alex-buell) wrote :
Download full text (18.2 KiB)

If I use Gnome with NO effects, it all works very nicely. I guess there
are still bugs in the Nouveau driver on older chipsets like the N34.
Perhaps 3.7.x will be better?

Regards,
Alex

On Mon, 2012-10-22 at 13:53 +0000, Bengt Sviu wrote:
> I have this old Lenovo laptop T61. The graphics have been dysfunctional
> during the last 14 days of updating quantal, practically continuously, and a
> few times before. Meaning big squares on the screen, no or few clicks
> registred, HUD not showing and such. Also some high processor use for
> Cheese. So I figured the driver was not working and changed to nvidias
> driver. problems gone.
>
> -----Opprinnelig melding-----
> From: Alex Buell
> Sent: Monday, October 22, 2012 3:17 PM
> To: <email address hidden>
> Subject: Re: [Bug 1033533] Re: Xorg crashed with SIGABRT: exaMemcpyBox
> withsrc=0x0 on nouveau with SW rendering
>
> This happens on logging in with Gnome (full effects) and Ubuntu and
> using for a few minutes, then crashes.
>
> Regards
> Alex
>
> On Mon, 2012-10-22 at 12:43 +0000, Maarten Lankhorst wrote:
> > How can I reproduce this easily?
> >
> > ** Changed in: xserver-xorg-video-nouveau (Ubuntu Quantal)
> > Status: Confirmed => Incomplete
> >
> > --
> > You received this bug notification because you are subscribed to a
> > duplicate bug report (1045115).
> > https://bugs.launchpad.net/bugs/1033533
> >
> > Title:
> > Xorg crashed with SIGABRT: exaMemcpyBox with src=0x0 on nouveau with
> > SW rendering
> >
> > Status in “xserver-xorg-video-nouveau” package in Ubuntu:
> > Confirmed
> > Status in “xserver-xorg-video-nouveau” source package in Quantal:
> > Incomplete
> >
> > Bug description:
> > Crash doing memcpy in EXA on nouveau, with software rendering
> > activated.
> >
> > #8 <signal handler called>
> > No symbol table info available.
> > #9 __memcpy_sse2 () at ../sysdeps/x86_64/multiarch/../memcpy.S:72
> > No locals.
> > #10 0x00007f570e63684e in memcpy (__len=23, __src=0x0,
> > __dest=0x7f5713877100) at /usr/include/x86_64-linux-gnu/bits/string3.h:52
> > No locals.
> > #11 exaMemcpyBox (pbox=pbox@entry=0x7fffef000de0, src=0x0, src_pitch=64,
> > dst=0x7f5713877100 "", dst_pitch=24, pPixmap=<error reading variable:
> > Unhandled dwarf expression opcode 0xfa>) at
> > ../../exa/exa_migration_classic.c:59
> > i = 16
> > cpp = <optimized out>
> > bytes = 23
> > #12 0x00007f570e636d03 in exaCopyDirty
> > (migrate=migrate@entry=0x7fffef000e40, pValidDst=0x7f5713dc3e40,
> > pValidSrc=0x7f5713dc3e50, transfer=0x7f570f28b7f0
> > <nouveau_exa_download_from_screen>, fallback_index=fallback_index@entry=1,
> > sync=0x7f570e635780 <exaWaitSync>) at
> > ../../exa/exa_migration_classic.c:240
> > pPixmap = 0x7f5713dc3da0
> > pExaPixmap = 0x7f5713dc3e00
> > damage = <optimized out>
> > CopyReg = {extents = {x1 = 0, y1 = 0, x2 = 23, y2 = 16}, data =
> > 0x0}
> > save_use_gpu_copy = 1
> > save_pitch = 64
> > pBox = 0x7fffef000de0
> > nbox = 1
> > access_prepared = 1
> > need_sync = 0
> > #13 0x00007f570e636f32 in exaCopyD...

dino99 (9d9) wrote :

Reading these comments above, it seems that some users have compiz installed on too low graphic card/chipset models, and got crashes (at least its easy to understand that crashes are the highest possible probability)

So activating compiz on such hardware should not be possible (either compiz should detect graphic hardware before beeing activated, or simply not installable, or nouveau need to deactivate compiz before crashing)

Fpfilko (fpfilko) on 2012-10-23
Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Incomplete → Opinion
status: Opinion → Incomplete
Jens Bremmekamp (nem75) wrote :

I have this crash all the time on my Lenovo X220, which uses intel graphics (so focusing on Noveau may _not_ be the answer here). I don't notice any actual probelms though, everything is displayed correctly.

a2512069@rmqkr.net (a2512069) wrote :

This bug reproduces quite well for me when browsing Internet by different browsers.
I had this crash several times scrolling web pages with graphics using arrow keys in Google Chrome.

Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Incomplete → Confirmed
Maarten Lankhorst (mlankhorst) wrote :

Could you add xserver-xorg-video-nouveau from ppa:mlankhorst/ppa ? I added some debug messages for nouveau, I suspect it runs out of memory, so something should show up in the xserver log with that deb at least..

tags: added: raring
Ubuntu QA Website (ubuntuqa) wrote :

This bug has been reported on the Ubuntu ISO testing tracker.

A list of all reports related to this bug can be found here:
http://iso.qa.ubuntu.com/qatracker/reports/bugs/1033533

tags: added: iso-testing
Alex Peters (alex-peters) wrote :

I've added the ppa:mlankhorst/ppa repository to my sources, but a newer version of the xserver-xorg-video-nouveau package is installed from the main repositories. That newer version still fails with no additional debugging information.

I can easily reproduce this by triggering a nouveau bug by doing suspend/resume with a G92 (NV92) card.

See https://bugs.launchpad.net/bugs/1111884 with the attached https://launchpadlibrarian.net/130058311/nouveau.txt .

I'll be glad to try an alternate xserver-xorg-video-nouveau for raring.

Maarten Lankhorst (mlankhorst) wrote :

That bug is a different one it seems. Not sure why it got marked as dupe, sorry for the noise!

It was marked as a dupe because the bug seems to provoke the same kind of crash in Xorg (same stack trace). I guess I can test the error handling in Xorg at least.

Maarten Lankhorst (mlankhorst) wrote :

Ok, I uploaded a new version of xserver-xorg-video-nouveau to the ppa, could you test against it with the spam there once it finishes building?

Alex Peters (alex-peters) wrote :

Given the following dpkg-query output, I assume I have the new version that you mention:

$ dpkg-query --show xserver-xorg-video-nouveau
xserver-xorg-video-nouveau 1:1.0.6-0ubuntu3+spam~quantal2

The following lines appeared on the screen (intermittently, alternating between that and a screen of pixel garbage every 10-15 seconds):

[drm] nouveau 0000:01:00.0: GPU lockup - switching to software fbcon
(4 seconds pass)
[TTM] Failed to expire sync object before buffer eviction
[drm] nouveau 0000:01:00.0: fail ttm_validate
[drm] nouveau 0000:01:00.0: validate vram_list
[drm] nouveau 0000:01:00.0: validate: -12
(1.5 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(1 second passes)
[drm] nouveau 0000:01:00.0: fail ttm_validate
[drm] nouveau 0000:01:00.0: validate vram_list
[drm] nouveau 0000:01:00.0: validate: -16
(2 seconds pass)
[drm] nouveau 0000:01:00.0: PGRAPH idle timed out with status 0x09800001
(2 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
[drm] nouveau 0000:01:00.0: fail ttm_validate
[drm] nouveau 0000:01:00.0: validate vram_list
[drm] nouveau 0000:01:00.0: validate: -16
(47 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(3 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
(35 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(3 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
(45 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(3 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
(41 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(3 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
(36 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 1.
(3 seconds pass)
[drm] nouveau 0000:01:00.0: Failed to idle channel 2.
...

(Errors typed manually; apologies for any typos. Time spans typed manually to avoid transcribing 230 digits.)

Alex Peters (alex-peters) wrote :

This bug pretty much renders my system entirely inoperable.

Steps to completely freeze my system:

1. Boot.
2. Open and close Dash three times, observing pixel corruption (screen shot attached) from second Dash opening. Observe system getting slower and slower with each operation.
3. Attempt to open Dash fourth time.

Maarten Lankhorst (mlankhorst) wrote :

Having the xorg log would be more useful to me. :) Sounds like you are probably hitting a different bug, in which case I would prefer having a separate bug, because it's easier to mark bugs as dupes, as it is to disentangle them later.

Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Confirmed → Incomplete
Changed in xserver-xorg-video-nouveau (Ubuntu Raring):
status: Confirmed → Incomplete
Alex Peters (alex-peters) wrote :

I have attached Xorg logs separately in bug 1161735, as requested.

Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Incomplete → Confirmed
Changed in xserver-xorg-video-nouveau (Ubuntu Raring):
status: Incomplete → Confirmed
Maarten Lankhorst (mlankhorst) wrote :

Ok, it seems that your system hangs first, and nouveau dies here later. At this point you were already crashing, so this crash bug is mostly a result of another bug triggering first. The gpu is already non-responsive at that point, which makes the xorg crash not really a surprise.

Unfortunately in some cases I don't see exactly what's causing it, but in all cases it seems that they have in common that the gpu locks up first, for various reasons..

Maarten Lankhorst (mlankhorst) wrote :

Does this alleviate the xserver crashes? Nouveau might still fail, the underlying reason for the hang is NOT fixed, and there
should probably be separate upstream bug reports for each hang..

tags: added: patch
elhoir (jfarroyo82) wrote :

FYI,

in my case (raring, 3.9.0-rc5 kernel, nouveau 1.0.7) Ubuntu is not completely frozen, i can still log in remotely. Only X are frozen.

Tried the patch; crashes at another place. See #1166522.

Or rather https://bugs.launchpad.net/bugs/1166522 since # didn't do any magic.

elhoir (jfarroyo82) wrote :
Download full text (17.0 KiB)

hello,

hope this is the correct place to post this....

Right now my X Server is frozen. I have logged in remotely, and these are the Xorg.0.log contents:

[ 11826.207] (II) NOUVEAU(0): Modeline "1152x864"x0.0 108.00 1152 1216 1344 1600 864 865 868 900 +hsync +vsync (67.5 kHz e)
(EE) [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
(EE)
(EE) Backtrace:
(EE) 0: /usr/bin/X (xorg_backtrace+0x36) [0x7fbd0da561c6]
(EE) 1: /usr/bin/X (mieqEnqueue+0x26b) [0x7fbd0da374db]
(EE) 2: /usr/bin/X (0x7fbd0d8a6000+0x6d3d2) [0x7fbd0d9133d2]
(EE) 3: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fbd067d8000+0x5f44) [0x7fbd067ddf44]
(EE) 4: /usr/bin/X (0x7fbd0d8a6000+0x96887) [0x7fbd0d93c887]
(EE) 5: /usr/bin/X (0x7fbd0d8a6000+0xc0078) [0x7fbd0d966078]
(EE) 6: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7fbd0c9a9000+0xfbd0) [0x7fbd0c9b8bd0]
(EE) 7: /lib/x86_64-linux-gnu/libc.so.6 (ioctl+0x7) [0x7fbd0b6c6747]
(EE) 8: /usr/lib/x86_64-linux-gnu/libdrm.so.2 (drmIoctl+0x28) [0x7fbd0c7a1338]
(EE) 9: /usr/lib/x86_64-linux-gnu/libdrm.so.2 (drmCommandWrite+0x1b) [0x7fbd0c7a36fb]
(EE) 10: /usr/lib/x86_64-linux-gnu/libdrm_nouveau.so.2 (nouveau_bo_wait+0x89) [0x7fbd0a0461d9]
(EE) 11: /usr/lib/xorg/modules/drivers/nouveau_drv.so (0x7fbd0a24a000+0x718b) [0x7fbd0a25118b]
(EE) 12: /usr/lib/xorg/modules/libexa.so (0x7fbd093eb000+0xb650) [0x7fbd093f6650]
(EE) 13: /usr/bin/X (0x7fbd0d8a6000+0x1a0443) [0x7fbd0da46443]
(EE) 14: /usr/bin/X (0x7fbd0d8a6000+0xe5a60) [0x7fbd0d98ba60]
(EE) 15: /usr/bin/X (0x7fbd0d8a6000+0x55be1) [0x7fbd0d8fbbe1]
(EE) 16: /usr/bin/X (0x7fbd0d8a6000+0x58a21) [0x7fbd0d8fea21]
(EE) 17: /usr/bin/X (0x7fbd0d8a6000+0x474da) [0x7fbd0d8ed4da]
(EE) 18: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0xf5) [0x7fbd0b5f6ea5]
(EE) 19: /usr/bin/X (0x7fbd0d8a6000+0x47821) [0x7fbd0d8ed821]
(EE)
(EE) [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
(EE) [mi] mieq is *NOT* the cause. It is a victim.
(EE) [mi] EQ overflow continuing. 100 events have been dropped.
(EE)
(EE) Backtrace:
(EE) 0: /usr/bin/X (xorg_backtrace+0x36) [0x7fbd0da561c6]
(EE) 1: /usr/bin/X (0x7fbd0d8a6000+0x6d3d2) [0x7fbd0d9133d2]
(EE) 2: /usr/lib/xorg/modules/input/evdev_drv.so (0x7fbd067d8000+0x5f44) [0x7fbd067ddf44]
(EE) 3: /usr/bin/X (0x7fbd0d8a6000+0x96887) [0x7fbd0d93c887]
(EE) 4: /usr/bin/X (0x7fbd0d8a6000+0xc0078) [0x7fbd0d966078]
(EE) 5: /lib/x86_64-linux-gnu/libpthread.so.0 (0x7fbd0c9a9000+0xfbd0) [0x7fbd0c9b8bd0]
(EE) 6: /lib/x86_64-linux-gnu/libc.so.6 (ioctl+0x7) [0x7fbd0b6c6747]
(EE) 7: /usr/lib/x86_64-linux-gnu/libdrm.so.2 (drmIoctl+0x28) [0x7fbd0c7a1338]
(EE) 8: /usr/lib/x86_64-linux-gnu/libdrm.so.2 (drmCommandWrite+0x1b) [0x7fbd0c7a36fb]
(EE) 9: /usr/lib/x86_64-linux-gnu/libdrm_nouveau.so.2 (nouveau_bo_wait+0x89) [0x7fbd0a0461d9]
(EE) 10: /usr/lib/xorg/modules/drivers/nouveau_drv.so (0x7fbd0a24a000+0x718b) [0x7fbd0a25118b]
(EE) 11: /usr/lib/xorg/modules/libexa.so (0x7fbd093eb000+0xb650) [0x7fbd093f6650]
(EE) 12: /usr/bin/X (0x7fbd0d8a6000+0x1a0443) [0x7fbd0da46443]
(EE) 13: /usr/bin/X (0x7fbd0d8a6000+0xe5a60) [0x7fbd0d98ba60]
(EE) 14: /usr/bin/X (0x7fbd0d8a6000+0x55be1) [0x7fbd0d8fbbe1]
(EE) 15: /usr/...

Mark A. Baldridge (reiki33) wrote :

My version of the crash is an initial [drm] nouveau 0000:01:00.0: GPU lockup - switching to fbcom \n and similar repeated messages of "Failed to idle channel 1. (or 2), all interleaved with full screen patterns of color block patches.

It is interesting that while I see mp4 videos from edx.org classes, what really exacerbates the problem is alt-tab between screens. My first real indication of a problem is when alt-tab produces a solid triangle of color on the bottom right half of the display. the diagonal goes from the bottom left of the left monitor to the top right of the right monitor. Shortly after that the system becomes unusable, and power button for reboot.

I have not gotten to the point of opening up port 22 to remotely connect to the system.

This is a clean install of 12.10, and all of the updates are current.

Peter Hurley (phurley) wrote :

For me, this bug is easily reproducible (happens within 1 min. of user session start) on xserver-xorg-video-nouveau 1.0.2

My user X session is compiz + classic + upstream gnome-panel + Radiance theme on NV84.

The stack trace is nearly identical to the original reporter's (I do get 'Failed to idle' or 'Unexpected page flip' at other times but those seem unrelated to this bug). This bug has been low-priority for me because it doesn't seem to affect stability.

This bug is _not_ reproducible with ppa:mlankhorst/ppa xserver-xorg-video-nouveau 1.0.6+spam~quantal2. Let me know if bisect is required.

Dan Kegel (dank) wrote :

This is 100% reproducible here. All I have to do is run 'make check' in wine, tell gdb to attach to X and ignore SIGPIPE,
and in a few minutes, gdb says e.g.

Program received signal SIGSEGV, Segmentation fault.
0x00007f74ae864bd5 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
(gdb) bt
#0 0x00007f74ae864bd5 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x00007f74ac53985e in memcpy (__len=2700, __src=0x0, __dest=0x7f74b2ba40d0)
    at /usr/include/x86_64-linux-gnu/bits/string3.h:51
#2 exaMemcpyBox (pbox=pbox@entry=0x7fffd5a2fbc0, src=0x0, src_pitch=2752, dst=0x7f74b2ba40d0 "h\377\255\256t\177",
    dst_pitch=2700, pPixmap=0x7f74b2b80e50) at ../../exa/exa_migration_classic.c:59
...

This is on a stock install of 13.04 daily from April 21 or so.

Maarten Lankhorst (mlankhorst) wrote :

It's a generic lockup bug, userspace will always fail in this way when nouveau locks up, but the real problem is in the kernel somewhere. Why nouveau locks up? Presumably something upset the card, causing it to hang. In case of make check, I would imagine one of the d3d9 tests could cause it. More specific information about the card and test would be helpful, and I would recommend filing a new bug for it against xserver-xorg-video-nouveau.

Leaw-ning (leaw-ning) wrote :

In my case, udevd crashes sometimes also. Does this information help?

Since this bug:

- Is valid.
- Is well described.
- Is reported in the upstream project.
- Is ready to be worked on by a developer.

It's already triaged.

Changed in xserver-xorg-video-nouveau (Ubuntu):
status: Confirmed → Triaged
Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Confirmed → Triaged
Changed in xserver-xorg-video-nouveau (Ubuntu Raring):
status: Confirmed → Triaged
tags: added: saucy
description: updated
Shahar Or (mightyiam) wrote :

Hi,
Bug #1263303 has occurred in my system and is occurring predictably the following way:
1. Suspend
2. Resume
3. Resume fails and system seems to hang
4. Hardware shutdown
5. Turn on and login
6. System wants to report this bug

It got marked as a duplicate of this bug. Patrick at comment #24 seems similar symptom. Hope this helps somehow. Is there a way to discover what causes this pseudo-bug?

Anthony Russello (arussello) wrote :

I am experiencing the same issues even when using the versions from the PPA included in the comments above. I am not experiencing this issue if I do a fresh install of Ubuntu 12.04 and select to download updates during install. This occurs only after doing all system updates, post installation.

Anthony Russello (arussello) wrote :

I have had relative success (though not tested long term) by enabling noapic and nomodeset as kernel command line options. Without those two, my X would freeze up almost instantly at the login screen. With those two, I was able to boot, log in, and run through multiple windows (including one full 1080p video playback).

However, all of this was accomplished on Ubuntu 13.10.

Edit /etc/default/grub and include:

    GRUB_CMDLINE_LINUX_DEFAULT="quiet splash noapic nomodeset"

Then run sudo update-grub and reboot

Since then, things seem to be running quite well.

Here are the nvidia packages I have installed:

   iB nvidia-319-updates - NVIDIA binary Xorg driver, kernel module a
   i A nvidia-settings-319-updates - Tool for configuring the NVIDIA graphics d

Chaffra Affouda (chaffra) wrote :

Yes but I cannnot run my dual screen with nomodeset. Only my default screen is activated.

tags: added: trusty
Ignacio (igmoweb) wrote :

Hi, I'm new so sorry if I don't describe it very well.

For me this always happens when I suspend the system, I just cannot do it. As soon as I try to resume it, screen gets corrupted, Corrupted pixels and squares are floating around.

Sometimes I get the same error in other ways. Today I was watching a video in IMDB while a Virtuabox + Vagrant + several other tabs were opened in Chrome.

Is there any workaround to avoid this? I have an Nvidia card but Gallium 0.4 on NVA5 driver is installed.

I don't know if this helps. Thanks.

tags: added: utopic
Sam Segers (sam-sgrs) wrote :

My bug report https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/1335626 was marked as duplicate to this one.
With me, it only happend since less than a week on Utopic with the ATI drivers.
I also get artifacts in QtCreator

dino99 (9d9) wrote :
Changed in xserver-xorg-video-nouveau (Ubuntu Quantal):
status: Triaged → Invalid
Changed in xserver-xorg-video-nouveau (Ubuntu Raring):
status: Triaged → Invalid
tags: removed: quantal raring
Chris J Arges (arges) on 2014-07-03
Changed in xserver-xorg-video-nouveau (Ubuntu Trusty):
importance: Undecided → High
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in xserver-xorg-video-nouveau (Ubuntu Trusty):
status: New → Confirmed
Adolfo Jayme (fitojb) on 2014-08-05
Changed in xserver-xorg-video-nouveau (Ubuntu):
milestone: quantal-updates → none
no longer affects: xserver-xorg-video-nouveau (Ubuntu Quantal)
no longer affects: xserver-xorg-video-nouveau (Ubuntu Raring)
Changed in xserver-xorg-video-nouveau (Ubuntu Trusty):
status: Confirmed → Triaged

Same here ubuntu 14.04, xserver-xorg-core 2:1 15.1-0ubuntu2.1 / Xorg crashed with SIGABRT / ApportVersion 2.14.1-0ubuntu3.5 / amd64 / Compositor runnun compiz / CompositorUnredirectDriverBlacklist '(nouveau|Intel).*Mesa 8.0' / Fresh install / NVIDIA GT218M [GeForce 315M][10de:0a7a] (reva2)(prog-if 00 [VGA controller]) Subsystem: Samsung Electronics GeForce 315M [144d:c581] / Ubuntu 3.13.0-36.6-generic 3.13.11.6 / Signal 6 / Tags trusty ubuntu compiz-0.9 ubuntu compiz-0.9 / xserver errors Failed to load module "nvidia" (module does not exits, 0) / xserver 2:1 15.1-0Ubuntu2.1 / xserver video driver nouveau

macJack (macjack) on 2015-06-08
information type: Public → Public Security
information type: Public Security → Private Security
information type: Private Security → Public
information type: Public → Public Security
information type: Public Security → Public
Clon (fillip1) wrote :

Same error (apport say) with integrate Intel (82865G) GPU.

dino99 (9d9) on 2015-10-07
tags: removed: saucy utopic
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Related questions