Bug 85100 - [NV108] Geforce GT730: X freezes during normal 2d operation.
Summary: [NV108] Geforce GT730: X freezes during normal 2d operation.
Status: NEW
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-10-16 16:22 UTC by nobody-freedesktop.org
Modified: 2014-10-16 16:52 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
Kernel Log (102.71 KB, text/plain)
2014-10-16 16:22 UTC, nobody-freedesktop.org
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description nobody-freedesktop.org 2014-10-16 16:22:24 UTC
Created attachment 107936 [details]
Kernel Log

This is the version of software that I am running:

extra/mesa-dri 10.3.1-1
extra/xf86-video-nouveau 1.0.11-2
core/linux 3.16.4-1
extra/libdrm 2.4.58-1

The system has an integrated ATI GPU onboard which is not used and a PCIe Geforce GT730 which is used in a dual head setup.

Occasionally, X will freeze during normal activity (no 3d, no video playback, just some terminals)

Sometimes the mouse cursor is still movable for a few seconds, afterwards, graphics is completely frozen up, the system is usually still pingable though.
Comment 1 Ilia Mirkin 2014-10-16 16:26:51 UTC
[   70.860202] nouveau E[   PFIFO][0000:02:00.0] PBDMA0: LBREQ
[   70.860218] nouveau E[   PFIFO][0000:02:00.0] PBDMA0: ch 2 [Xorg.bin[390]] subc 5 mthd 0x0000 data 0x0000906e

I guess it's having trouble attaching the software object? Weird...
Comment 2 nobody-freedesktop.org 2014-10-16 16:48:15 UTC
I am not familiar enough with nouveau to understand what exactly is happening there.

However it might be interesting to add that it runs fine from a users perspective while it is outputting these log messages every few seconds.

The freeze just happens at this point:
[  570.936213] nouveau E[   PFIFO][0000:02:00.0] write fault at 0x000029f000 [PTE] from GR/GPC0/PROP_0 on channel 0x007faed000 [Xorg.bin[390]]
[  570.936222] nouveau E[   PFIFO][0000:02:00.0] PGRAPH engine fault on channel 2, recovering...
[  570.937090] nouveau E[    PBUS][0000:02:00.0] MMIO read of 0x00000000 FAULT at 0x000260 [ TIMEOUT ]
[  570.937104] nouveau E[    PBUS][0000:02:00.0] MMIO write of 0xbad0011e FAULT at 0x000260 [ TIMEOUT ]
[  570.937447] nouveau E[    PBUS][0000:02:00.0] MMIO write of 0xbad0011f FAULT at 0x000260 [ TIMEOUT ]
Comment 3 Ilia Mirkin 2014-10-16 16:52:26 UTC
The 0x260 stuff isn't an issue... that register is just gone on the GK208 but we're still writing to it. I think that 3.17 fixes that... if not, 3.18.

The engine faults aren't so great, however. The ones you pasted are from PGRAPH, but the earlier stuff was in attaching the software object to the fifo. The software object is used for vsync/page flipping handling, so failures with it can account for perceived hangs.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.