Nouveau - May 2015 - [Bug 90453] New: Desktop periodically freezes when using Nouveau & getting "nouveau E[PDISP]" errors in dmesg

https://bugs.freedesktop.org/show_bug.cgi?id=90453

            Bug ID: 90453
           Summary: Desktop periodically freezes when using Nouveau &
                    getting "nouveau E[PDISP]" errors in dmesg
           Product: xorg
           Version: unspecified
          Hardware: x86-64 (AMD64)
                OS: Linux (All)
            Status: NEW
          Severity: critical
          Priority: medium
         Component: Driver/nouveau
          Assignee: nouveau at lists.freedesktop.org
          Reporter: mhelvens at gmail.com
        QA Contact: xorg-team at lists.x.org

Created attachment 115775
  --> https://bugs.freedesktop.org/attachment.cgi?id=115775&action=edit
output of `dmesg`

Since switching to Nouveau, a couple of hours after booting my system, the
display will freeze up. I can still move the mouse, and I can still hear the
sound of any media-player or YouTube still running. Sometimes, I can still go
to another terminal with Ctrl-Alt-F2, sometimes not. One time when I could, I
restarted X and it would just work again.

(Additionally, the frame rate seems a bit 'stuttery' compared to the
binary
NVidia drivers, especially when moving windows around and such. That may or may
not be a separate issue.)

There are a bunch of nvidia errors in my dmesg output that I do not understand.
I attached this output.

I'd be happy to provide any extra info you need, but would also much
appreciate
any workaround I could use.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/4c2f0799/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #1 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Created attachment 115776
  --> https://bugs.freedesktop.org/attachment.cgi?id=115776&action=edit
output of `lspci -vnn | grep VGA -A 17`

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/6fbf577d/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #2 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Created attachment 115777
  --> https://bugs.freedesktop.org/attachment.cgi?id=115777&action=edit
output of `glxinfo | grep OpenGL  `

I added some attachments with relevant info about my system.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/d97f8245/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #3 from Ilia Mirkin <imirkin at alum.mit.edu> ---
Another user bisected a similar issue to a change that came into 3.16. Can you
try 3.15 and see if that works better for you?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/2a14d57c/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Tobias Klausmann <tobias.klausmann at mni.thm.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Desktop periodically        |[NVE4] Desktop freezes &
                   |freezes when using Nouveau  |PDISP errors
                   |& getting "nouveau          |
                   |E[PDISP]" errors in dmesg   |

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/57e5cf0a/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Tobias Klausmann <tobias.klausmann at mni.thm.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|critical                    |normal

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/1d6eb4f2/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #4 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Created attachment 115779
  --> https://bugs.freedesktop.org/attachment.cgi?id=115779&action=edit
packages I downgraded from v3.16.x

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/f1f2ac2e/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #5 from Michiel Helvensteijn <mhelvens at gmail.com> ---
(In reply to Ilia Mirkin from comment #3)
> Another user bisected a similar issue to a change that came into 3.16. Can
> you try 3.15 and see if that works better for you?
I assume you're referring to GTK/Gnome. OK. I downgraded all gnome-related
packages I could find that had a 3.16-ish version. For almost all of them, I
downgraded to a 3.14 version, as a 3.15 version didn't exist (using Pacman
on
Arch). I attached a list of packages I downgraded.

Anyway, those nouveau errors have disappeared from dmesg! Whether the display
will no longer freeze remains to be seen, but I'm hopeful. Thanks!

Which bug report should I follow to know when it's safe to upgrade again?

(PS: Another problem, which I first assumed to be related, is not yet solved.
When I enter my login data in GDM, I do not login, but GDM just re-appears. I
can only get into Gnome using startx.)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/9e474183/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #6 from Ilia Mirkin <imirkin at alum.mit.edu> ---
(In reply to Michiel Helvensteijn from comment #5)
> (In reply to Ilia Mirkin from comment #3)
> > Another user bisected a similar issue to a change that came into 3.16.
Can
> > you try 3.15 and see if that works better for you?
> 
> I assume you're referring to GTK/Gnome. OK. I downgraded all
gnome-related
> packages I could find that had a 3.16-ish version. For almost all of them,
I
> downgraded to a 3.14 version, as a 3.15 version didn't exist (using
Pacman
> on Arch). I attached a list of packages I downgraded.
Sorry for the confusion. I meant kernel.
> 
> Anyway, those nouveau errors have disappeared from dmesg! Whether the
> display will no longer freeze remains to be seen, but I'm hopeful.
Thanks!
> 
> Which bug report should I follow to know when it's safe to upgrade
again?
That's actually a bit surprising. Means that the DE is part of what's
tickling
the nouveau bug.
> 
> (PS: Another problem, which I first assumed to be related, is not yet
> solved. When I enter my login data in GDM, I do not login, but GDM just
> re-appears. I can only get into Gnome using startx.)
You'll have to figure out what's going wrong in your stack. Presumably
file a
bug with the gdm people?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/b907a6f2/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #7 from Ilia Mirkin <imirkin at alum.mit.edu> ---
FTR, the other bug is bug 90276.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/a19d16c1/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #8 from Michiel Helvensteijn <mhelvens at gmail.com> ---
(In reply to Ilia Mirkin from comment #6)
> > > Another user bisected a similar issue to a change that came into
3.16. Can
> > > you try 3.15 and see if that works better for you?
> > 
> > I assume you're referring to GTK/Gnome.
> 
> Sorry for the confusion. I meant kernel.
That was my first guess, but 3.16 was already a while ago for me, and I'm
currently running 4.0.2. On the other hand, the latest version of GTK and
friends actually is 3.16, so... Funny coincidence.
> > Anyway, those nouveau errors have disappeared from dmesg! Whether the
> > display will no longer freeze remains to be seen, but I'm hopeful.
Thanks!
> > 
> > Which bug report should I follow to know when it's safe to upgrade
again?
> 
> That's actually a bit surprising. Means that the DE is part of
what's
> tickling the nouveau bug.
Well, let me know if I can provide you with any further info. DE or not, this
still manifests as a nouveau bug.
> You'll have to figure out what's going wrong in your stack.
Presumably file
> a bug with the gdm people?
Yeah, my system's a hot mess right now. Thanks.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/4f53a0b0/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #9 from Michiel Helvensteijn <mhelvens at gmail.com> ---
OK, so... The dmesg errors are gone, but the display still freezes up. Yeah, it
seems they were two independent problems.

In the absence of PDISP errors, do you still suggest I downgrade the kernel to
3.15?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/ab63e09f/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #10 from Michiel Helvensteijn <mhelvens at gmail.com> ---
I was finally able to connect to my pc *after* a freeze-up using SSH. Here is
the coinciding dmesg output:

[ 8970.897260] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[ 8970.897265] nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on channel 6,
recovering...

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/fb4575a4/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Michiel Helvensteijn <mhelvens at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[NVE4] Desktop freezes &    |[NVE4] Desktop freezes
&
                   |PDISP errors                |PDISP errors & PFIFO errors

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/84bbfac0/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #11 from Ilia Mirkin <imirkin at alum.mit.edu> ---
(In reply to Michiel Helvensteijn from comment #10)
> I was finally able to connect to my pc *after* a freeze-up using SSH. Here
> is the coinciding dmesg output:
> 
> [ 8970.897260] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [
CTXSW_TIMEOUT
> ]
> [ 8970.897265] nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on
channel
> 6, recovering...
Hrmph... could well have been fixed by one of the ctxsw changes since 3.15 :(

Sorry, this will require someone who knows what they're talking about (i.e.
not
me) to properly diagnose.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150514/cb698a25/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #12 from Michiel Helvensteijn <mhelvens at gmail.com> ---
(In reply to Ilia Mirkin from comment #11)
> > I was finally able to connect to my pc *after* a freeze-up using SSH.
Here
> > is the coinciding dmesg output:
> > 
> > [ 8970.897260] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [
CTXSW_TIMEOUT]
> > [ 8970.897265] nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on
channel
> > 6, recovering...
> 
> Hrmph... could well have been fixed by one of the ctxsw changes since 3.15
:(
> 
> Sorry, this will require someone who knows what they're talking about
(i.e.
> not me) to properly diagnose.
Then I hope someone like that will come along. This problem is really quite
frustrating. :-)

Thanks for your help Ilia!

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/512cd641/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #13 from Michiel Helvensteijn <mhelvens at gmail.com> ---
During the freeze-up just now, dmesg gave me this beauty:

[ 9126.410442] nouveau E[  PGRAPH][0000:01:00.0] TRAP ch 8 [0x007f2f2000
gnome-shell[1419]]
[ 9126.410451] nouveau E[  PGRAPH][0000:01:00.0] GPC1/PROP trap: (unknown enum
0x00000002)
[ 9126.410454] nouveau E[  PGRAPH][0000:01:00.0] x = 4560, y = 160, format = 0,
storage type = 0
[ 9126.410461] nouveau E[  PGRAPH][0000:01:00.0] GPC2/PROP trap: (unknown enum
0x00000002)
[ 9126.410462] nouveau E[  PGRAPH][0000:01:00.0] x = 4576, y = 152, format = 0,
storage type = 0
[ 9130.744302] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[ 9130.744306] nouveau E[   PFIFO][0000:01:00.0] PGRAPH engine fault on channel
9, recovering...
[ 9145.526131] Peer 84.104.35.99:51165/34567 unexpectedly shrunk window
2380020382:2380048967 (repaired)
[ 9146.092788] Peer 84.104.35.99:51165/34567 unexpectedly shrunk window
2380020382:2380048967 (repaired)
[ 9147.229435] Peer 84.104.35.99:51165/34567 unexpectedly shrunk window
2380020382:2380048967 (repaired)
[ 9164.935810] Peer 84.104.35.99:51165/34567 unexpectedly shrunk window
2380247127:2380249895 (repaired)
[ 9165.382470] Peer 84.104.35.99:51165/34567 unexpectedly shrunk window
2380247127:2380249895 (repaired)
[ 9172.099024] nouveau E[systemd-logind[431]] failed to idle channel 0xcccc0000
[systemd-logind[431]]
[ 9187.098777] nouveau E[systemd-logind[431]] failed to idle channel 0xcccc0000
[systemd-logind[431]]
[ 9187.098878] nouveau E[   PFIFO][0000:01:00.0] read fault at 0x0000023000
[PTE] from PBDMA0/HOST_CPU on channel 0x007f541000 [unknown]
[ 9202.351857] nouveau E[gnome-shell[1419]] failed to idle channel 0xcccc0000
[gnome-shell[1419]]
[ 9217.351609] nouveau E[gnome-shell[1419]] failed to idle channel 0xcccc0000
[gnome-shell[1419]]
[ 9232.351361] nouveau E[chromium[2607]] failed to idle channel 0xcccc0000
[chromium[2607]]
[ 9247.351111] nouveau E[chromium[2607]] failed to idle channel 0xcccc0000
[chromium[2607]]

In case it makes a difference, this is after I tried installing the binary
nvidia drivers again, failing, and going back to nouveau. This is also with all
the latest gtk / gnome versions installed again (so the above errors follow the
long string of PDISP errors in the dmesg output).

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/133dd0c0/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Michiel Helvensteijn <mhelvens at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[NVE4] Desktop freezes &    |[NVE4] Desktop freezes
&
                   |PDISP errors & PFIFO errors |PDISP, PFIFO and PGRAPH
                   |                            |errors

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/4de9f622/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #14 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Oh, and by the way, this time my mouse cursor could *not* move like it could
the other times, and eventually the display went back to GDM, but without
responding to any input.

(I'm a bit desperate here. A virtual beer to whomever can help me out with
this!)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/c5309c1e/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #15 from Ilia Mirkin <imirkin at alum.mit.edu> ---
BTW, as per my suggestion in
https://bugs.freedesktop.org/show_bug.cgi?id=90276#c11 you can likely
"fix" the
PDISP errors by changing drivers/gpu/drm/nouveau/nv50_display.c:

-       if (show && nv_crtc->cursor.nvbo)
+       if (show && nv_crtc->base.enabled &&
nv_crtc->cursor.nvbo)

However that didn't fix the underlying issues for the other user, and it is
likely not to help your hang either, I'm afraid. But perhaps worth a shot?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/c4f21d7a/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #16 from Michiel Helvensteijn <mhelvens at gmail.com> ---
(In reply to Ilia Mirkin from comment #15)
> BTW, as per my suggestion in
> https://bugs.freedesktop.org/show_bug.cgi?id=90276#c11 you can likely
"fix"
> the PDISP errors by changing drivers/gpu/drm/nouveau/nv50_display.c:
> 
> -       if (show && nv_crtc->cursor.nvbo)
> +       if (show && nv_crtc->base.enabled &&
nv_crtc->cursor.nvbo)
> 
> However that didn't fix the underlying issues for the other user, and
it is
> likely not to help your hang either, I'm afraid. But perhaps worth a
shot?
Indeed, the PDISP errors are gone. May I assume that this fix will make it into
the official kernel in the near future?

In the meantime, I'll keep my expectations low w.r.t. my hang. ;-)

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150515/48b5a2c0/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #17 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Indeed, problem not solved. Here's a new set of errors produced during a
freeze:

[  690.870744] nouveau E[     PGR][0000:01:00.0] TRAP ch 8 [0x007f2f2000
gnome-shell[2652]]
[  690.870754] nouveau E[     PGR][0000:01:00.0] GPC1/PROP trap:
ZETA_STORAGE_TYPE_MISMATCH
[  690.870756] nouveau E[     PGR][0000:01:00.0] x = 1232, y = 56, format = 0,
storage type = fe
[  690.870763] nouveau E[     PGR][0000:01:00.0] GPC2/PROP trap:
ZETA_STORAGE_TYPE_MISMATCH
[  690.870765] nouveau E[     PGR][0000:01:00.0] x = 1056, y = 328, format = 0,
storage type = fe
[  695.179294] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[  695.179298] nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on channel
10, recovering...

This one happened earlier than usual.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/009b2f92/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Michiel Helvensteijn <mhelvens at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[NVE4] Desktop freezes &    |[NVE4] Desktop freezes
&
                   |PDISP, PFIFO and PGRAPH     |PDISP, PFIFO and PGRAPH,
                   |errors                      |PGR errors

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/8118ea10/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Michiel Helvensteijn <mhelvens at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[NVE4] Desktop freezes &    |[NVE4] Desktop freezes
&
                   |PDISP, PFIFO and PGRAPH,    |PDISP, PFIFO, PGRAPH and
                   |PGR errors                  |PGR errors

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/7979b24f/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #18 from Michiel Helvensteijn <mhelvens at gmail.com> ---
This latest instance is also interesting, and might help diagnose the problem.
I was watching a YouTube video, and this time only the Chromium browser froze,
and I got a 'page unresponsive' popup. For a while, I could still move
the
mouse and interact with other windows. After about 30 seconds, everything froze
once again. dmesg output:

[ 7583.945899] nouveau E[     PGR][0000:01:00.0] TRAP ch 8 [0x007f2f2000
gnome-shell[1189]]
[ 7583.945909] nouveau E[     PGR][0000:01:00.0] GPC2/PROP trap: (unknown enum
0x00000002)
[ 7583.945912] nouveau E[     PGR][0000:01:00.0] x = 3488, y = 104, format = 0,
storage type = 0
[ 7588.249980] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[ 7588.249985] nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on channel
10, recovering...

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/64836987/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #19 from Michiel Helvensteijn <mhelvens at gmail.com> ---
Created attachment 115828
  --> https://bugs.freedesktop.org/attachment.cgi?id=115828&action=edit
long term dmesg output

I left my pc on for a while after the last freeze-up, and after about 45
minutes, more stuff came pouring down dmesg. I don't know if this
information
is useful at all. I just thought I'd share it in case.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/1b64b1bd/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #20 from Marcin Slusarz <marcin.slusarz at gmail.com> ---
It seems gnome-shell triggers some bug in our OpenGL driver, which causes the
hardware to raise errors (GPC1/PROP traps), which kernel fails to ack/cleanup
and eventually this leads to context switching failure.

If we'll fix the OpenGL driver, the other bugs won't matter (and they
are much
harder to fix). So the first step is figuring out how to reproduce this bug in
a matter of minutes and generating an apitrace and/or mmt trace.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/6967a78a/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #21 from Michiel Helvensteijn <mhelvens at gmail.com> ---
(In reply to Marcin Slusarz from comment #20)
> If we'll fix the OpenGL driver, the other bugs won't matter (and
they are
> much harder to fix). So the first step is figuring out how to reproduce
this
> bug in a matter of minutes and generating an apitrace and/or mmt trace.
In order to get some work done, I switched back to the binary nvidia drivers
for now.

If you have an actual plan of attack, I'll happily switch back to try it
out.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150516/31353c8e/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #22 from Argyris Megalios <argymeg at gmail.com> ---
I originally added my problem to bug 89912, as my original error message was
not seen here.

kernel: nouveau E[   PFIFO][0000:01:00.0] write fault at 0x00002a0000 [PTE]
from GR/GPC0/PROP_0 on channel 0x007f96c000 [Xorg[383]]
kernel: nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on channel 2,
recovering...

However, after also getting the following:
kernel: nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
kernel: nouveau E[   PFIFO][0000:01:00.0] PGR engine fault on channel 10,
recovering...

I am inclined to thing the bugs are duplicates.
Now, while playing a game (which actually played fine yesterday) I have had two
consecutive freezes with the message:

kernel: nouveau E[     DRM] GPU lockup - switching to software fbcon

I, too, will be switching to the nvidia driver for the moment, however I will
be happy to assist with any fix tests.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150614/71598150/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #23 from Mario Barrera <mbarrera at gmx.com> ---
Same issue over here, it is a NVE4 too. I will be using the proprietary driver
from now on until I know better how to help diagnose this.

[ 7104.402763] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[ 7104.402770] nouveau E[   PFIFO][0000:01:00.0] PGRAPH engine fault on channel
7, recovering...

...and video freezes forever.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150803/c5b00e2e/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #24 from Timothy Pearson <kb9vqf at pearsoncomputing.net> ---
(In reply to Mario Barrera from comment #23)
> Same issue over here, it is a NVE4 too. I will be using the proprietary
> driver from now on until I know better how to help diagnose this.
> 
> [ 7104.402763] nouveau E[   PFIFO][0000:01:00.0] SCHED_ERROR [
CTXSW_TIMEOUT
> ]
> [ 7104.402770] nouveau E[   PFIFO][0000:01:00.0] PGRAPH engine fault on
> channel 7, recovering...
> 
> ...and video freezes forever.
Unfortunately I'm seeing the same problem.  I have been able to reproduce it
very consistently under two circumstances:
1.) Starting a new X11 session when one is already running (using TDM, a fork
of the older KDE 3.5 KDM)
2.) Trying to select an OpenGL screensaver.  If you go through the list of
OpenGL screensavers, occasionally selecting one for full-screen preview, the
GPU will hang after going through no more than around a dozen savers.

Reclocking status does not matter; fault occurs on both low and high clocks.

Kernel 4.1:
[  914.928160] nouveau E[   PFIFO][0000:01:00.0] read fault at 0x0000490000
[PTE] from GR/SCC on channel 0x00fec92000 [matrixview[27252]]
[  914.928169] nouveau E[   PFIFO][0000:01:00.0] PGRAPH engine fault on channel
7, recovering..

Going back to the proprietary drivers is not an option here due to a serious
bug in those drivers, so if there's anything I can do to help diagnose this
issue please let me know.

Thanks!

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150830/5871b84c/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #25 from Arthur Heymans <arthurphilippeheymans at gmail.com>
---
Hi 
I also have this bug (screen freeze and mouse still usable) on a my system
using a nvidia 660ti [NVE4]:
dmesg gives
[  112.780941] nouveau E[   PFIFO][0000:03:00.0] SCHED_ERROR [ CTXSW_TIMEOUT ]
[  112.780947] nouveau E[   PFIFO][0000:03:00.0] PGRAPH engine fault on channel
8, recovering...

or 

[  294.274092] nouveau E[   PFIFO][0000:03:00.0] write fault at 0x000a002000
[PTE] from GR/GPC0/PROP_0 on channel 0x007f6ab000 [systemd-logind[426]]
[  294.274103] nouveau E[   PFIFO][0000:03:00.0] PGRAPH engine fault on channel
5, recovering...

I might be the same bug as I reported earlier
https://bugs.freedesktop.org/show_bug.cgi?id=90276

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.freedesktop.org/archives/nouveau/attachments/20150831/83361672/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #26 from Lucas Ribeiro <lucasout at gmail.com> ---
Now on kernel 4.6 I experienced 0 freezes so far. Something improved on the
660ti NVE4 side.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/nouveau/attachments/20160529/5b2e079a/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

--- Comment #27 from Anthony Foiani <anthony.foiani at gmail.com> ---
Seeing something very similar on my GTX 650.  Most common trigger is
`xscreensaver`, but since I have set to randomly pick one, unsure whether there
is a pattern.  It also tends to trigger on full-screen video (display is
2560x1440 Dell 2713HM).

I'm using Fedora 25, and apparently Wayland:

$ echo $WAYLAND_DISPLAY
wayland-0

Let me know if I can provide any more information.  Thanks!

Linux version: `uname -a`

Linux hum 4.9.5-200.fc25.x86_64 #1 SMP Fri Jan 20 12:24:16 UTC 2017 x86_64
x86_64 x86_64 GNU/Linux

Kernel log:

-- at boot:
Jan 28 18:24:03 hum kernel: fb: switching to nouveaufb from EFI VGA
Jan 28 18:24:03 hum kernel: Console: switching to colour dummy device 80x25
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: NVIDIA GK107 (0e7060a2)
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: bios: version 80.07.35.00.04
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: fb: 1024 MiB GDDR5
Jan 28 18:24:03 hum kernel: [TTM] Zone  kernel: Available graphics memory:
8191232 kiB
Jan 28 18:24:03 hum kernel: [TTM] Zone   dma32: Available graphics memory:
2097152 kiB
Jan 28 18:24:03 hum kernel: [TTM] Initializing pool allocator
Jan 28 18:24:03 hum kernel: [TTM] Initializing DMA pool allocator
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: VRAM: 1024 MiB
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: GART: 1048576 MiB
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: TMDS table version 2.0
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB version 4.0
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB outp 00: 01000f02
00020030
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB outp 01: 02000f00
00000000
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB outp 02: 08011f82
00020030
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB outp 03: 02022f62
00020010
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB conn 00: 00001030
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB conn 01: 00002131
Jan 28 18:24:03 hum kernel: nouveau 0000:01:00.0: DRM: DCB conn 02: 00010263
Jan 28 18:24:03 hum kernel: [drm] Supports vblank timestamp caching Rev 2
(21.10.2013).
Jan 28 18:24:03 hum kernel: [drm] Driver supports precise vblank timestamp
query.
Jan 28 18:24:04 hum kernel: nouveau 0000:01:00.0: DRM: MM: using COPY for
buffer copies
Jan 28 18:24:04 hum kernel: nouveau 0000:01:00.0: DRM: allocated 2560x1440 fb:
0x60000, bo ffff95135afd4000
Jan 28 18:24:04 hum kernel: fbcon: nouveaufb (fb0) is primary device
Jan 28 18:24:04 hum kernel: nouveau 0000:01:00.0: fb0: nouveaufb frame buffer
device
Jan 28 18:24:04 hum kernel: [drm] Initialized nouveau 1.3.1 20120801 for
0000:01:00.0 on minor 0

-- when it locked up:
Jan 29 13:05:45 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 06 []
Jan 29 13:05:45 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:05:48 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 06 []
Jan 29 13:05:48 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:05:49 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 06 []
Jan 29 13:05:49 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:10:13 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 06 []
Jan 29 13:10:13 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:11:55 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []
Jan 29 13:11:55 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []
Jan 29 13:12:29 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []
Jan 29 13:13:27 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []
Jan 29 13:13:27 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000004
Jan 29 13:16:12 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000004
Jan 29 13:16:13 hum kernel: nouveau 0000:01:00.0: fifo: SCHED_ERROR 08 []
Jan 29 13:16:13 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000004
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: FB_FLUSH_TIMEOUT
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: PBDMA0: 00000004
[MEMACK_EXTRA] ch 1 [003fe13000 DRM] subc 0 mthd 0020 data 00000000
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
Jan 29 13:16:39 hum kernel: nouveau 0000:01:00.0: fifo: CHSW_ERROR 00000002
[That repeats hundreds of times per second.]

`sudo lspci -vnn`:

01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GK107 [GeForce GTX
650] [10de:0fc6] (rev a1) (prog-if 00 [VGA controller])
        Subsystem: Device [196e:0973]
        Flags: bus master, fast devsel, latency 0, IRQ 44
        Memory at f3000000 (32-bit, non-prefetchable) [size=16M]
        Memory at d0000000 (64-bit, prefetchable) [size=256M]
        Memory at e0000000 (64-bit, prefetchable) [size=32M]
        I/O ports at e000 [size=128]
        Expansion ROM at 000c0000 [disabled] [size=128K]
        Capabilities: [60] Power Management version 3
        Capabilities: [68] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [78] Express Endpoint, MSI 00
        Capabilities: [b4] Vendor Specific Information: Len=14 <?>
        Capabilities: [100] Virtual Channel
        Capabilities: [128] Power Budgeting <?>
        Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024
<?>
        Capabilities: [900] #19
        Kernel driver in use: nouveau
        Kernel modules: nouveau

Packages:

Installed Packages
abrt-addon-xorg.x86_64                          2.9.0-1.fc25                   
          @fedora                  
xorg-x11-drv-ati.x86_64                        
7.7.1-1.20160928git3fc839ff.fc25          @fedora                  
xorg-x11-drv-evdev.x86_64                       2.10.5-1.fc25                  
          @updates                 
xorg-x11-drv-fbdev.x86_64                       0.4.3-25.fc25                  
          @fedora                  
xorg-x11-drv-intel.x86_64                       2.99.917-26.20160929.fc25      
          @fedora                  
xorg-x11-drv-libinput.x86_64                    0.23.0-2.fc25                  
          @updates                 
xorg-x11-drv-nouveau.x86_64                     1:1.0.13-1.fc25                
          @fedora                  
xorg-x11-drv-openchrome.x86_64                  0.5.0-2.fc25                   
          @fedora                  
xorg-x11-drv-qxl.x86_64                        
0.1.4-10.20160929gite13d28ee.fc25         @fedora                  
xorg-x11-drv-synaptics.x86_64                   1.9.0-1.fc25                   
          @updates                 
xorg-x11-drv-vesa.x86_64                        2.3.2-25.fc25                  
          @fedora                  
xorg-x11-drv-vmware.x86_64                      13.2.1-1.fc25                  
          @updates                 
xorg-x11-drv-wacom.x86_64                       0.34.0-1.fc25                  
          @updates                 
xorg-x11-font-utils.x86_64                      1:7.5-32.fc25                  
          @fedora                  
xorg-x11-fonts-ISO8859-1-100dpi.noarch          7.5-16.fc24                    
          @fedora                  
xorg-x11-proto-devel.noarch                     7.7-20.fc25                    
          @fedora                  
xorg-x11-resutils.x86_64                        7.5-13.fc24                    
          @fedora                  
xorg-x11-server-Xorg.x86_64                     1.19.1-2.fc25                  
          @updates                 
xorg-x11-server-Xwayland.x86_64                 1.19.1-2.fc25                  
          @updates                 
xorg-x11-server-common.x86_64                   1.19.1-2.fc25                  
          @updates                 
xorg-x11-server-utils.x86_64                    7.7-20.fc25                    
          @fedora                  
xorg-x11-utils.x86_64                           7.5-21.fc24                    
          @fedora                  
xorg-x11-xauth.x86_64                           1:1.0.9-5.fc24                 
          @fedora                  
xorg-x11-xinit.x86_64                           1.3.4-13.fc25                  
          @updates                 
xorg-x11-xkb-utils.x86_64                       7.7-17.fc24                    
          @fedora

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/nouveau/attachments/20170130/a6d534fe/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Jeremy Booker <jeremy.booker at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jeremy.booker at gmail.com

--- Comment #28 from Jeremy Booker <jeremy.booker at gmail.com> ---
I may have reported a duplicate/related bug here:
https://bugs.freedesktop.org/show_bug.cgi?id=100567

I'm also running Fedora 25 with Wayland/Gnome3 and had this crash under both
the 4.9 and 4.10 kernels using the nouveau drivers. I have an nVidia NVS 510
card with 3 monitors attached.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/nouveau/attachments/20170406/afd9cd79/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=90453

Martin Peres <martin.peres at free.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |MOVED
             Status|NEW                         |RESOLVED

--- Comment #29 from Martin Peres <martin.peres at free.fr> ---
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been
closed from further activity.

You can subscribe and participate further through the new bug through this link
to our GitLab instance:
https://gitlab.freedesktop.org/xorg/driver/xf86-video-nouveau/issues/189.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/nouveau/attachments/20191204/c261196b/attachment.html>