[Bug 100585] [BXT/GLK/KBL/BDW/BSW]gem_mmap/swap-bo causing page allocation failure

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=100585

--- Comment #6 from Jari Tahvanainen <jari.tahvanainen at intel.com> ---
Chris - you said that this is known limitation. So are you saying that one can
use ~95% of memory successfully with this gem_mmap at swap-bo case? If huge object
is bigger than that then i915 (or kernel) mm fails?

On my dev-HSW I found out that the following is the limit (90% of repetitions
the case if pass):
Subtest swap-bo failed.
**** DEBUG ****
(gem_mmap:25393) intel-os-DEBUG: Checking 1 surfaces of size 15 985 541 120
bytes (total 15 985 545 216) against RAM + swap
(gem_mmap:25393) drmtest-DEBUG: Test requirement passed: !(fd<0)
(gem_mmap:25393) igt-debugfs-DEBUG: Opening debugfs directory
'/sys/kernel/debug/dri/0'
(gem_mmap:25393) intel-os-DEBUG: Test requirement passed:
__intel_check_memory(count, size, mode, &required, &total)
(gem_mmap:25393) igt-core-DEBUG: Test requirement passed:
!igt_run_in_simulation()
(gem_mmap:25393) DEBUG: Exercising 15 985 541 120 bytes
(gem_mmap:25393) DEBUG: Total_ram 16 589 520 896 bytes

-- 
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the bug.
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20170605/c17302a6/attachment.html>