[opensuse-factory] My GPU seems to be on its last legs
HI I get the occasional desktop freezes that needs a reboot and it looks like its down to a failing GPU or could it just be a bug? I get the following (just a subset) log records for nouveau. I also get various nouveau log records that reference things like kmail or akonadi_xxx - is this usual? May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d0 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d4 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d8 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01dc data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 000e0080 00000000 00000000 00000000 00000000 00000004 00000000 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: 00200000 [] ch 12 [001f0bf000 akonadi_mailfil[12416]] subc 3 class 8597 mthd 13bc data 00000054 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: fb: trapped read at 002027df00 on channel 12 [1f0bf000 akonadi_mailfil[12416]] engine 00 [PGRAPH] client 05 [CCACHE] subclient 00 [CB] reason 00000006 [NULL_DMAOBJ] May 10 15:00:38 LianLi kernel: nouveau 0000:01:00.0: DRM: GPU lockup - switching to software fbcon May 10 15:01:31 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail .... May 10 15:01:55 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH_VSTATUS2: 00000000 [] May 10 15:01:56 LianLi kernel: nouveau 0000:01:00.0: kmail[12243]: failed to idle channel 6 [kmail[12243]] .... May 10 15:05:08 LianLi kernel: nouveau 0000:01:00.0: akonadi_mailfil[12416]: failed to idle channel 12 [akonadi_mailfil[12416]] .... May 10 15:05:10 LianLi kernel: nouveau 0000:01:00.0: X[11876]: failed to idle channel 4 [X[11876]] May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini()! May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 12 [akonadi_mailfil[12416]] unload timeout May 10 15:05:14 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini()! May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 4 [X[11876]] unload timeout May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail ..... Any ideas? Regards Ian -- opensuse:tumbleweed:20170505 Qt: 5.7.1 KDE Frameworks: 5.33.0 KDE Plasma: 5.9.5 kwin 5.9.5 kmail2 5.4.3 akonadiserver 5.4.3 Kernel: 4.10.13-1-default Nouveau: 1.0.15_1.1 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, 10 May 2017 18:21:56 +0100
ianseeks
HI
I get the occasional desktop freezes that needs a reboot and it looks like its down to a failing GPU or could it just be a bug? I get the following (just a subset) log records for nouveau. I also get various nouveau log records that reference things like kmail or akonadi_xxx - is this usual?
May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d0 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d4 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d8 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01dc data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 000e0080 00000000 00000000 00000000 00000000 00000004 00000000 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: 00200000 [] ch 12 [001f0bf000 akonadi_mailfil[12416]] subc 3 class 8597 mthd 13bc data 00000054 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: fb: trapped read at 002027df00 on channel 12 [1f0bf000 akonadi_mailfil[12416]] engine 00 [PGRAPH] client 05 [CCACHE] subclient 00 [CB] reason 00000006 [NULL_DMAOBJ] May 10 15:00:38 LianLi kernel: nouveau 0000:01:00.0: DRM: GPU lockup - switching to software fbcon May 10 15:01:31 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail .... May 10 15:01:55 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH_VSTATUS2: 00000000 [] May 10 15:01:56 LianLi kernel: nouveau 0000:01:00.0: kmail[12243]: failed to idle channel 6 [kmail[12243]] .... May 10 15:05:08 LianLi kernel: nouveau 0000:01:00.0: akonadi_mailfil[12416]: failed to idle channel 12 [akonadi_mailfil[12416]] .... May 10 15:05:10 LianLi kernel: nouveau 0000:01:00.0: X[11876]: failed to idle channel 4 [X[11876]] May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini()! May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 12 [akonadi_mailfil[12416]] unload timeout May 10 15:05:14 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini()! May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 4 [X[11876]] unload timeout May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail .....
Any ideas?
It is normal. Nouveau is broken. You had to agree to it when you installed it. At the moment you see LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] your GPU has locked up. Depending on your hardware, kernel and Xorg version you may have better luck updating one or the other but since you did not say what you are running exactly it is hard to tell. You can also peruse the kernel.org an freedesktop.org bugzilla and search for (probably numerous) reports with similar error messages. HTH Michal -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Michal Suchánek composed on 2017-05-10 19:33 (UTC+0200):
Nouveau is broken. You had to agree to it when you installed it.
By this do you mean xf86-video-nouveau?
At the moment you see LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] your GPU has locked up.
Depending on your hardware, kernel and Xorg version you may have better luck updating one or the other but since you did not say what you are running exactly it is hard to tell. You can also peruse the kernel.org an freedesktop.org bugzilla and search for (probably numerous) reports with similar error messages.
Having neither proprietary NVidia nor xf86-video-nouveau drivers installed in my Leaps and TWs with various GeForce devices seems perfectly fine using the built-into-Xorg modesetting driver. People keep pointing out that the modesetting driver is only 2D, but I can't grok what 3D is supposed to bring to a two dimensional viewing surface. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
El 10-05-2017 a las 15:26, Felix Miata escribió:
Michal Suchánek composed on 2017-05-10 19:33 (UTC+0200):
Nouveau is broken. You had to agree to it when you installed it.
By this do you mean xf86-video-nouveau?
Messages come from the kernel module nouveau... -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, 10 May 2017 15:54:43 -0300
Cristian Rodríguez
El 10-05-2017 a las 15:26, Felix Miata escribió:
Michal Suchánek composed on 2017-05-10 19:33 (UTC+0200):
Nouveau is broken. You had to agree to it when you installed it.
By this do you mean xf86-video-nouveau?
Messages come from the kernel module nouveau...
Most nouveau GPU acceleration to be precise. This is delivered in concert by the kernel dri driver, mesa dri driver, and the X11 nouveau driver. If you remove some of these bits you lose the ability to accelerate rendering projection of 3d models on your 2d screen surface and the ability of the window manager to wobble your windows as you drag them across your screen but the stability of your graphics hardware will likely increase because most of its features will go unused. HTH Michal -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, 10 May 2017 14:26:59 -0400
Felix Miata
Michal Suchánek composed on 2017-05-10 19:33 (UTC+0200):
Nouveau is broken. You had to agree to it when you installed it.
By this do you mean xf86-video-nouveau?
At the moment you see LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] your GPU has locked up.
Depending on your hardware, kernel and Xorg version you may have better luck updating one or the other but since you did not say what you are running exactly it is hard to tell. You can also peruse the kernel.org an freedesktop.org bugzilla and search for (probably numerous) reports with similar error messages.
Having neither proprietary NVidia nor xf86-video-nouveau drivers installed in my Leaps and TWs with various GeForce devices seems perfectly fine using the built-into-Xorg modesetting driver. People keep pointing out that the modesetting driver is only 2D, but I can't grok what 3D is supposed to bring to a two dimensional viewing surface.
Actually, modesetting driver is supposed to give 3D acceleration in recent X. That unfortunately decreases its stability. For me Quadro K620 works fine with the default nouveau driver in Leap. On the other hand a Quadro NVS 295 crashes (almost) immediately with nouveau and somewhat works with modesetting. Compared to a HD 5670 NVS 295 is super slow and super glitchy and crashes after a few minutes of running anything related to 3D. Which means the card is useless since it crashes your system randomly. Anyway, modesetting driver is what allows experiencing the slowness and glitchiness at all and seems to be preferred for recent cards https://bugzilla.redhat.com/show_bug.cgi?id=1446000 https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-Vs-Modesetting http://pkgs.fedoraproject.org/cgit/rpms/xorg-x11-server.git/tree/0001-xfree8... Thanks Michal -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wednesday, 10 May 2017 18:33:01 BST Michal Suchánek wrote:
On Wed, 10 May 2017 18:21:56 +0100
ianseeks
wrote: HI
I get the occasional desktop freezes that needs a reboot and it looks like its down to a failing GPU or could it just be a bug? I get the following (just a subset) log records for nouveau. I also get various nouveau log records that reference things like kmail or akonadi_xxx - is this usual?
May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d0 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d4 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01d8 data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: fifo: CACHE_ERROR - ch 12 [akonadi_mailfil[12416]] subc 3 mthd 01dc data beef0201 May 10 15:00:25 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 000e0080 00000000 00000000 00000000 00000000 00000004 00000000 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: gr: 00200000 [] ch 12 [001f0bf000 akonadi_mailfil[12416]] subc 3 class 8597 mthd 13bc data 00000054 May 10 15:00:26 LianLi kernel: nouveau 0000:01:00.0: fb: trapped read at 002027df00 on channel 12 [1f0bf000 akonadi_mailfil[12416]] engine 00 [PGRAPH] client 05 [CCACHE] subclient 00 [CB] reason 00000006 [NULL_DMAOBJ] May 10 15:00:38 LianLi kernel: nouveau 0000:01:00.0: DRM: GPU lockup - switching to software fbcon May 10 15:01:31 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail .... May 10 15:01:55 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH_VSTATUS2: 00000000 [] May 10 15:01:56 LianLi kernel: nouveau 0000:01:00.0: kmail[12243]: failed to idle channel 6 [kmail[12243]] .... May 10 15:05:08 LianLi kernel: nouveau 0000:01:00.0: akonadi_mailfil[12416]: failed to idle channel 12 [akonadi_mailfil[12416]] .... May 10 15:05:10 LianLi kernel: nouveau 0000:01:00.0: X[11876]: failed to idle channel 4 [X[11876]] May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini() ! May 10 15:05:12 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 12 [akonadi_mailfil[12416]] unload timeout May 10 15:05:14 LianLi kernel: nouveau 0000:01:00.0: timeout at ../drivers/ gpu/drm/nouveau/nvkm/engine/fifo/chang84.c:111/g84_fifo_chan_engine_fini() ! May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: fifo: channel 4 [X[11876]] unload timeout May 10 15:05:16 LianLi kernel: nouveau 0000:01:00.0: gr: PGRAPH TLB flush idle timeout fail .....
Any ideas?
It is normal. Nouveau is broken. You had to agree to it when you installed it.
I know its not well at times but the card is quite old now and i've been considering moving to AMD because of the consistent issues but i'm not sure if its the card failing or just nouveau having an issue
At the moment you see LianLi kernel: nouveau 0000:01:00.0: gr: TRAP_CCACHE 00000001 [FAULT] your GPU has locked up. Yep, i saw that. I thought the other messages referencing things like kmail and akonadi on nouveau logs records as strange
Depending on your hardware, kernel and Xorg version you may have better luck updating one or the other but since you did not say what you are running exactly it is hard to tell. You can also peruse the kernel.org an freedesktop.org bugzilla and search for (probably numerous) reports with similar error messages. I think i am as up-to-date with tumbleweed as i can be.
HTH
Michal Thanks
-- opensuse:tumbleweed:20170505 Qt: 5.7.1 KDE Frameworks: 5.33.0 KDE Plasma: 5.9.5 kwin 5.9.5 kmail2 5.4.3 akonadiserver 5.4.3 Kernel: 4.10.13-1-default Nouveau: 1.0.15_1.1 -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (4)
-
Cristian Rodríguez
-
Felix Miata
-
ianseeks
-
Michal Suchánek