[Bug 97128] Kernel hang when running out of memory

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97128

Bas Nieuwenhuizen  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |FIXED

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 97128] Kernel hang when running out of memory

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97128

--- Comment #2 from Bas Nieuwenhuizen  ---
I can confirm that patch fixes the hang for me.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 97128] Kernel hang when running out of memory

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97128

--- Comment #1 from Alex Deucher  ---
Should be fixed in this patch:
https://lists.freedesktop.org/archives/amd-gfx/2016-July/000686.html

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 97128] Kernel hang when running out of memory

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=97128

Bug ID: 97128
   Summary: Kernel hang when running out of memory
   Product: DRI
   Version: XOrg git
  Hardware: Other
OS: All
Status: NEW
  Severity: normal
  Priority: medium
 Component: DRM/AMDgpu
  Assignee: dri-devel at lists.freedesktop.org
  Reporter: bas at basnieuwenhuizen.nl

If I allocate too much memory I get a kernel hang using the drm-next-4.8 branch
from Alex Deucher. SSH stops working too.

This has been tests by allocating GTT buffers in a loop using the winsys for
the new radv vulkan driver. Just as radeonsi, the winsys has a 3 step
allocation process: 1) allocate the buffer 2) Allocate an address range 3) Map
the buffer to address range.

This hang is a regression. Bisection pointed to the following patch as culprit:

commit 089f16c55baacd5e8ae3745625efa82899b4b217
Author: Christian König 
Date:   Mon Jun 6 10:17:50 2016 +0200

drm/ttm: cleanup ttm_tt_(unbind|destroy)

ttm_tt_destroy should be the only one unbinding the object.

Reviewed-by: Alex Deucher 
Signed-off-by: Christian König 
Signed-off-by: Alex Deucher 

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: