[Bug 103479] [BAT] igt@* - dmesg-warn - *ERROR* Timeout waiting for engines to idle | *ERROR* vecs0 is not idle before parking

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=103479

--- Comment #13 from Chris Wilson <chris at chris-wilson.co.uk> ---
Now even the kernel agrees that this doesn't make sense:

<3>[  534.357705] [drm:intel_engines_park [i915]] *ERROR* vecs0 is not idle
before parking
<7>[  534.359332] intel_engines_park vecs0
<7>[  534.360232] intel_engines_park    current seqno 10, last 10, hangcheck 0
[234603 ms], inflight 0
<7>[  534.360240] intel_engines_park    Reset count: 0
<7>[  534.360295] intel_engines_park    Requests:
<7>[  534.361646] intel_engines_park    RING_START: 0x00000000 [0x00000000]
<7>[  534.361659] intel_engines_park    RING_HEAD:  0x00000000 [0x00000000]
<7>[  534.361669] intel_engines_park    RING_TAIL:  0x00000000 [0x00000000]
<7>[  534.361681] intel_engines_park    RING_CTL:   0x00000000
<7>[  534.361694] intel_engines_park    RING_MODE:  0x00000200 [idle]
<7>[  534.361711] intel_engines_park    ACTHD:  0x00000000_00000000
<7>[  534.361725] intel_engines_park    BBADDR: 0x00000000_00000000
<7>[  534.361737] intel_engines_park    Execlist status: 0x00000301 00000000
<7>[  534.361747] intel_engines_park    Execlist CSB read 1 [1 cached], write 1
[1 from hws], interrupt posted? no
<7>[  534.361755] intel_engines_park            ELSP[0] idle
<7>[  534.361762] intel_engines_park            ELSP[1] idle
<7>[  534.361768] intel_engines_park            HW active? 0x0
<7>[  534.362049] intel_engines_park IRQ? 0x0 (breadcrumbs? no) (execlists? no)
<7>[  534.362069] intel_engines_park Idle? yes
<7>[  534.362075] intel_engines_park 

We still have a race with a tasklet/irq/something here.

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are on the CC list for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20171109/f73bf6e3/attachment.html>