Bug 107759

This has been introduced in drmtip_90, and has failed consistently since.

I wrote it off as being another GTT coherency issue; but no what it actually is comparing is the results of a gpu copy into a prime mmap. So this is much more likely to be a bcs full-ppgtt issue.

Best, hopeful, guess: https://patchwork.freedesktop.org/series/48813/

Hoping resolved by

commit 70b73f9ac113983f9c7db9887447f1344ac5b69b (HEAD -> drm-intel-next-queued, drm-intel/drm-intel-next-queued)
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Aug 30 17:10:42 2018 +0100

    drm/i915/ringbuffer: Delay after invalidating gen6+ xcs
    
    During stress testing of full-ppgtt (on Baytrail at least), we found
    that the invalidation around a context/mm switch was insufficient (writes
    would go astray). Adding a second MI_FLUSH_DW barrier prevents this, but
    it is unclear as to whether this is merely a delaying tactic or if it is
    truly serialising with the TLB invalidation. Either way, it is
    empirically required.
    
    v2: Avoid the loop for readability;
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107715
    References: https://bugs.freedesktop.org/show_bug.cgi?id=107759
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
    Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
    Cc: Matthew Auld <matthew.william.auld@gmail.com>
    Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
    Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
    Link: https://patchwork.freedesktop.org/patch/msgid/20180830161042.29193-1-chris@chris-wilson.co.uk

Still happening:

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_100/fi-byt-clapper/igt@prime_mmap_coherency@ioctl-errors.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_100/fi-byt-j1900/igt@prime_mmap_coherency@ioctl-errors.html

(prime_mmap_coherency:1395) CRITICAL: Test assertion failure function blit_and_cmp, file ../tests/prime_mmap_coherency.c:231:
(prime_mmap_coherency:1395) CRITICAL: Failed assertion: !(ptr2_cpu[i] != 0x11111111)
(prime_mmap_coherency:1396) CRITICAL: Found 0x00000000 at offset 0x00000000

Same bcs full-ppgtt issue, the gem_persistent_relocs one seems easy to hit (even if may take some time)

*** This bug has been marked as a duplicate of bug 107769 ***

This issue used to appear in every round of drmtip, Last seen this issue in drmtip_109. I assume this bug is fixed as the original issue is closed.