Nouveau - Aug 2014 - [Bug 82152] New: any OpenGL application crashes X, locks up machine with nouveau and PRIME

https://bugs.freedesktop.org/show_bug.cgi?id=82152

          Priority: medium
            Bug ID: 82152
          Assignee: nouveau at lists.freedesktop.org
           Summary: any OpenGL application crashes X, locks up machine
                    with nouveau and PRIME
          Severity: normal
    Classification: Unclassified
                OS: Linux (All)
          Reporter: celticmadman at gmail.com
          Hardware: x86-64 (AMD64)
            Status: NEW
           Version: 10.2
         Component: Drivers/DRI/nouveau
           Product: Mesa

Created attachment 104017
  --> https://bugs.freedesktop.org/attachment.cgi?id=104017&action=edit
Full output of journalctl, including kernel logs, between system boot and
poweroff after crash.

Any OpenGL application, even as minor as glxinfo, either crashes Xorg or locks
up the machine entirely (no network, magic sysrq fails) when started with
DRI_PRIME=1. Has happened across multiple Mesa and kernel versions, most
recently with Mesa 10.2.4 and Linux 3.15.8 on Arch Linux.

-- 
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/20140804/88650e3c/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #1 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104018
  --> https://bugs.freedesktop.org/attachment.cgi?id=104018&action=edit
Output from crashing glxinfo.

-- 
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/20140804/fe975f89/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

Patrick Burroughs <celticmadman at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #104017|0                           |1
        is obsolete|                            |

--- Comment #2 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104019
  --> https://bugs.freedesktop.org/attachment.cgi?id=104019&action=edit
Kernel logs filtered from journal 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/20140804/f2894144/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #3 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104020
  --> https://bugs.freedesktop.org/attachment.cgi?id=104020&action=edit
Xorg.0.log from crash.

-- 
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/20140804/3a8b0dff/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #4 from Ilia Mirkin <imirkin at alum.mit.edu> ---
There are two issues:

(a) The null deref in the kernel when putting the card back to sleep
(b) The fact that init of the card fails

To mitigate the first, you could boot with "nouveau.runpm=0". However
you still
wouldn't get working accel with nouveau.

The claim by NVIDIA was that the graph-not-powered-up problem was restricted to
GK104/GK106. But looking at the latest code, it seems like it runs on GK107 as
well (not in Ben's repo anymore, but still in linux-3.16) and perhaps has
the
reverse effect there.

I wonder if a patch like

diff --git a/nvkm/engine/graph/nve4.c b/nvkm/engine/graph/nve4.c
index 51e0c07..4dd376e 100644
--- a/nvkm/engine/graph/nve4.c
+++ b/nvkm/engine/graph/nve4.c
@@ -350,7 +350,7 @@ nve4_graph_oclass = &(struct nvc0_graph_oclass) {
                .ctor = nvc0_graph_ctor,
                .dtor = nvc0_graph_dtor,
                .init = nve4_graph_init,
-               .fini = nve4_graph_fini,
+               .fini = _nouveau_graph_fini,
        },
        .cclass = &nve4_grctx_oclass,
        .sclass = nve4_graph_sclass,

will help you out. (You'll need to apply it with care... cd into
drivers/gpu/drm/nouveau/core and apply it with patch -p2 )

-- 
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/20140804/40c3f003/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

Ilia Mirkin <imirkin at alum.mit.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|any OpenGL application      |[NVE7] NULL deref when
                   |crashes X, locks up machine |putting card back to sleep
                   |with nouveau and PRIME      |after unsuccessful init
                   |                            |(HUB_INIT timeout)

-- 
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/20140804/f60ee786/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #5 from Patrick Burroughs <celticmadman at gmail.com> ---
I get the same crash and HUB_INIT timeout after the patch. Attaching 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/20140804/ce9a2dde/attachment-0001.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #6 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104026
  --> https://bugs.freedesktop.org/attachment.cgi?id=104026&action=edit
dmesg output post initial patch.

-- 
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/20140804/aeeb9dcc/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #7 from Tobias Klausmann <tobias.klausmann at mni.thm.de> ---
If i look at the system + the kernel bug, this looks similar to a problem i was
facing some weeks ago:

so i'd suggest to try DRI3 with the whole package:

Update your packages:
xf86-video-intel
mesa
(all dependencies of course)
Remove:
xf86-video-nouveau (with DRI3 you wont need it to do: DRI_PRIME=1 myprog)


you'll need a kernel with rendernodes enabled (boot with drm.rnodes=1)
you may need to add a file to /etc/udev/rules.d/ containing:

SUBSYSTEM=="drm", IMPORT{builtin}="path_id" 

to get ID_PATH tags for rendernodes.

-- 
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/20140804/43c6a7bf/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #8 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104031
  --> https://bugs.freedesktop.org/attachment.cgi?id=104031&action=edit
dmesg output using DRI3

Using DRI3 defers all errors until after attempting to run an OpenGL
application with DRI_PRIME=1, and prevents the crash from bringing down X or
the kernel.

-- 
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/20140805/d460e825/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #9 from Patrick Burroughs <celticmadman at gmail.com> ---
Created attachment 104033
  --> https://bugs.freedesktop.org/attachment.cgi?id=104033&action=edit
dmesg output using firmware ripped from the blob

Using DRI3 and the firmware from the blob I still get crashes, but finally have
a different error message 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/20140805/e1e9a8e9/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #10 from Patrick Burroughs <celticmadman at gmail.com> ---
Tried again with Linux 3.19.3 and Mesa 10.5.2, no changes.

-- 
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/20150406/bd8bee21/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

--- Comment #11 from Patrick Burroughs <celti at celti.name> ---
Created attachment 122992
  --> https://bugs.freedesktop.org/attachment.cgi?id=122992&action=edit
dmesg output with errors from successful load

With Linux 4.5.0-ARCH, Mesa 11.1.2-3, DRI3, and using modesetting_drv instead
of intel_drv for the main display (not sure if that's relevant)...
everything
works! (If "everything" consists of glxinfo, glxgears, and a few
minutes of
Darwinia.)

I do still get errors in dmesg, though, as attached. I'll be happy to follow
along and do whatever digging is necessary to eradicate them, if someone wants
to take up that task.

-- 
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/20160416/3c9a73d3/attachment.html>

https://bugs.freedesktop.org/show_bug.cgi?id=82152

GitLab Migration User <gitlab-migration at fdo.invalid> changed:

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

--- Comment #12 from GitLab Migration User <gitlab-migration at
fdo.invalid> ---
-- 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/mesa/mesa/issues/1066.

-- 
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/20190918/1fe79e53/attachment.html>