linux-toradex.git/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c, branch v4.10-rc4 Linux kernel for Apalis and Colibri modules drm/amdgpu: remove adev pointer from struct amdgpu_bo v2 2016-10-25T18:38:26+00:00 Christian König christian.koenig@amd.com 2016-09-15T12:58:48+00:00 a7d64de659946e852cd8f2a9691a21ddbb4ebc86 It's completely pointless to have two pointers to the device in the same structure. v2: rename function to amdgpu_ttm_adev, fix typos Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
It's completely pointless to have two pointers to the
device in the same structure.

v2: rename function to amdgpu_ttm_adev, fix typos

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: make amdgpu_mn_get wait for mmap_sem killable 2016-05-24T00:04:14+00:00 Michal Hocko mhocko@suse.com 2016-05-23T23:26:17+00:00 b5637051f1ed48f68370234e879d40d8dd4ab087 amdgpu_mn_get which is called during ioct path relies on mmap_sem for write. If the waiting task gets killed by the oom killer it would block oom_reaper from asynchronous address space reclaim and reduce the chances of timely OOM resolving. Wait for the lock in the killable mode and return with EINTR if the task got killed while waiting. [arnd@arndb.de: use ERR_PTR() to return from amdgpu_mn_get] Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> Reviewed-by: Christian König <christian.koenig@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> 
amdgpu_mn_get which is called during ioct path relies on mmap_sem for
write.  If the waiting task gets killed by the oom killer it would block
oom_reaper from asynchronous address space reclaim and reduce the
chances of timely OOM resolving.  Wait for the lock in the killable mode
and return with EINTR if the task got killed while waiting.

[arnd@arndb.de: use ERR_PTR() to return from amdgpu_mn_get]
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drm/amdgpu: add invalidate_page callback for userptrs 2016-03-21T15:52:14+00:00 Christian König christian.koenig@amd.com 2016-03-18T18:29:52+00:00 ae20f12d2de6629ee6f679ccf22f9b7b209c464d Otherwise we can run into problems with the writeback code. Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
Otherwise we can run into problems with the writeback code.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Revert "remove the userptr rmn->lock" 2016-03-21T15:52:14+00:00 Christian König christian.koenig@amd.com 2016-03-18T18:29:51+00:00 0d2b42b0bdba45c82d29d794ea30a4c90e3f4098 This reverts commit c02196834456f2d5fad334088b70e98ce4967c34. In the meantime we moved get_user_pages() outside of the reservation lock, so that shouldn't be an issue any more Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> 
This reverts commit c02196834456f2d5fad334088b70e98ce4967c34.

In the meantime we moved get_user_pages() outside of the reservation lock,
so that shouldn't be an issue any more

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: Don't call interval_tree_remove in amdgpu_mn_destroy 2016-02-17T21:10:56+00:00 Felix Kuehling Felix.Kuehling@amd.com 2016-02-16T20:31:30+00:00 390be2824fa4211c2e973c69b72e04000559bba3 rbtree_postorder_for_each_entry_safe can skip over some entries if the tree is rebalanced in interval_tree_remove. interval_tree_remove is also redundant when the tree is just about to be freed. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> 
rbtree_postorder_for_each_entry_safe can skip over some entries if
the tree is rebalanced in interval_tree_remove. interval_tree_remove
is also redundant when the tree is just about to be freed.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
drm/amdgpu: Fix race condition in amdgpu_mn_unregister 2016-02-17T21:10:42+00:00 Felix Kuehling Felix.Kuehling@amd.com 2016-02-16T20:29:23+00:00 b8ea3783467586246d0b2b08f2e3f42853219d94 Exchange locking order of adev->mn_lock and mm_sem, so that rmn->mm->mmap_sem can be taken safely, protected by adev->mn_lock, when amdgpu_mn_destroy runs concurrently. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> 
Exchange locking order of adev->mn_lock and mm_sem, so that
rmn->mm->mmap_sem can be taken safely, protected by adev->mn_lock,
when amdgpu_mn_destroy runs concurrently.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
drm/amdgpu: Fix race condition in MMU notifier release 2016-02-16T22:17:14+00:00 Felix Kuehling Felix.Kuehling@amd.com 2016-01-14T05:35:08+00:00 fa5b5000308136ad61a33126d8ad380cf1fef016 The release notifier can get called a second time from mmu_notifier_unregister depending on a race between __mmu_notifier_release and amdgpu_mn_destroy. Use mmu_notifier_unregister_no_release to avoid this. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> 
The release notifier can get called a second time from
mmu_notifier_unregister depending on a race between
__mmu_notifier_release and amdgpu_mn_destroy. Use
mmu_notifier_unregister_no_release to avoid this.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
drm/amdgpu: remove the userptr rmn->lock 2016-02-12T20:32:07+00:00 Christian König christian.koenig@amd.com 2016-02-09T15:13:37+00:00 c41d271d751ea023f67f1ea755e144b49079b6d5 Avoid a lock inversion problem by just using the mmap_sem to protect the entries of the intervall tree. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> 
Avoid a lock inversion problem by just using the mmap_sem to
protect the entries of the intervall tree.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
drm/amdgpu: fix issue with overlapping userptrs 2016-02-10T19:16:43+00:00 Christian König christian.koenig@amd.com 2016-02-08T09:57:22+00:00 d7006964d46d76930a44e14f1aae401ccb1797da Otherwise we could try to evict overlapping userptr BOs in get_user_pages(), leading to a possible circular locking dependency. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> 
Otherwise we could try to evict overlapping userptr BOs in get_user_pages(),
leading to a possible circular locking dependency.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
drm/amdgpu: fix userptr lockup 2015-06-04T01:03:26+00:00 Christian König christian.koenig@amd.com 2015-05-04T11:20:36+00:00 a961ea7349d08d5962b6a619f2c2cca670da41ed Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com> Reviewed-by: Monk Liu <monk.liu@amd.com> 
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>