[Bug 91577] [HD Graphics 4000] Segfault accessing rq->bo->handle

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=91577

--- Comment #21 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to Dominic from comment #19)
> The patches change the segfault but cannot prevent it. The segault I just
> got did not yield a meaningful backtrace, however. (heap corruption?)
> 
> [   365.632] kgem_cleanup_cache
> [   365.632] kgem_cleanup_cache: sync on cleanup
> [   365.632] __kgem_retire_requests_upto(handle=2)
> [   365.632] __kgem_retire_rq: request 2 complete
> [   365.632] kgem_retire, need_retire?=1
> [   365.632] __kgem_busy: handle=17, busy=1, wedged=0
> [   365.632] kgem_retire__flushing: 18 bo on flushing list
> [   365.632] kgem_retire__requests_ring: ring=0, 0 outstanding requests,
> oldest=0
> [   365.632] __kgem_busy: handle=3, busy=1, wedged=0
> [   365.632] kgem_retire__requests_ring: ring=1, 2 outstanding requests,
> oldest=3
> [   365.632] kgem_retire -- retired=0, need_retire=1
> [   365.632] kgem_bo_is_busy: handle=64390, domain: 0 exec? 0, rq? 0
> [   365.632] gen6_render_retire: resetting idle vbo
> [   365.632] __kgem_bo_clear_busy: handle=3
> [   365.632] __kgem_bo_clear_busy: handle=65024
> [   365.632] kgem_cleanup_cache: need_expire?=0
> [   365.632] __kgem_bo_map__cpu: failed to mmap handle=64326 (prime? 0),
> 65536 bytes, into CPU domain: 12
> [   365.632] kgem_new_batch: unable to map batch bo, mallocing(size=65536)
> [   365.632] _kgem_bo_destroy: handle=64326, proxy? 0
> [   365.632] __kgem_bo_destroy: handle=64326, size=65536
> [   365.632] __kgem_bo_destroy: handle=64326, purged
> [   365.632] kgem_bo_move_to_inactive: moving handle=64326 to inactive

Hmm. That's an odd place for the log to stop. I've tweaked the DBG statements,
maybe that will be clearer?

(In reply to Dominic from comment #20)
> Xorg reaches ~20% MEM at the time of crash - that should be about 4GiB. How
> can I see the object count reliably?

cat /sys/kernel/debug/dri/0/i915_gem_objects

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20150807/ca52db28/attachment.html>