https://bugs.freedesktop.org/show_bug.cgi?id=70388

--- Comment #19 from Linas Vepstas <linasvepstas-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> ---
FWIW, I am hitting a similar or the same bug.  Let me keep the report brief:

-- In my case, its an NV44A
-- OS is Ubuntu 12.04 with this weeks latest, brand-new kernel: 3.16.1
-- X seems to hang/crash once/twice right after boot, and then the following
kern messages start showing up:

[  168.871453] nouveau  [     DRM] 0xD2C1: Parsing digital output script table
[  168.934091] nouveau E[   PFIFO][0000:01:06.0] DMA_PUSHER - ch 1 [Xorg[2278]]
get 0x00010000 put 0x00010090 state 0x80000000 (err: INVALID_CMD) push
0x00000000
[  200.472639] nouveau E[   PFIFO][0000:01:06.0] DMA_PUSHER - ch 2 [Xorg[2278]]
get 0x01033000 put 0x01033088 state 0x80000000 (err: INVALID_CMD) push
0x00000000
[  215.472015] nouveau E[Xorg[2278]] failed to idle channel 0xcccc0001
[Xorg[2278]]
[  215.472066] nouveau E[   PFIFO][0000:01:06.0] DMA_PUSHER - ch 2 [Xorg[2278]]
get 0x01033088 put 0x01033090 state 0x80000000 (err: INVALID_CMD) push
0x00000000
[  230.472014] nouveau E[Xorg[2278]] failed to idle channel 0xcccc0001
[Xorg[2278]]
[  245.472044] nouveau E[Xorg[2278]] failed to idle channel 0xcccc0000
[Xorg[2278]]
[  260.472018] nouveau E[Xorg[2278]] failed to idle channel 0xcccc0000
[Xorg[2278]]
[  261.637808] nouveau  [     DRM] 0xC633: Parsing digital output script table
[  261.787097] nouveau  [     DRM] 0xD2C1: Parsing digital output script table
[  261.849926] nouveau E[   PFIFO][0000:01:06.0] DMA_PUSHER - ch 1 [Xorg[2373]] 

-- Note above is almost 3 minutes after boot!
-- Note the 'fielded to idle channel' are exactly 15 seconds apart.
-- Xorg is still crashing/hanging at this point.
 Basically, it starts, the mouse pointer works fine, the lightdm tries to paint
the login dialog, the painting never completes (just a square of grey), some
timer kicks in, and then things repeat again.  Typically one Xorg hang +
autorestart every 3 minutes, repeating 8 to 12 times (about 1/2 hour) when one
of them finally, magically "just works", after which I can log in and use the
machine as normal.  (OK, I have not yet tried watching youtube or doing 3D
yet...)

During later attempts, I get showers of the PFIFO DMA_PUSHER errors,  then 7
minutes in, I start getting this for the first time:

[  431.120645] nouveau E[   PFIFO][0000:01:06.0] DMA_PUSHER - ch 1 [Xorg[2455]]
get 0xbc882d90 put 0x00010130 state 0xc0000000 (err: MEM_FAULT) push 0x00000000
[  431.120704] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x193c data 0x3e998b7a
[  431.120727] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x1940 data 0x5e07b578
[  431.120751] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x1944 data 0x3df04d6e
[  431.120995] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x1948 data 0xd4a617e5
[  431.121017] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x194c data 0x92b307e9
[  431.121039] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x1950 data 0x05d62b60
[  431.121061] nouveau E[   PFIFO][0000:01:06.0] CACHE_ERROR - ch 1
[Xorg[2455]] subc 7 mthd 0x1954 data 0x42e9cdfa

which goes on a good while. ... then the cycle repeats, as described above. 
Never quite exactly the same way twice, but similar each time.

-- 
You are receiving this mail because:
You are the assignee for the bug.