On 2018-06-15 05:11 PM, Christian König wrote:
> Am 15.06.2018 um 16:42 schrieb Michel Dänzer:
>> From: Michel Dänzer <michel.daenzer at amd.com>
>>
>> Even BOs with AMDGPU_GEM_CREATE_NO_CPU_ACCESS may end up at least
>> partially in CPU visible VRAM, in particular when all VRAM is visible.
>>
>> Cc: stable at vger.kernel.org
>> Signed-off-by: Michel Dänzer <michel.daenzer at amd.com>
>> ---
>> Â drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 26 ++++++++++++++++++--
>> Â 1 file changed, 24 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
>> index 491fd0a789f7..9b34df7017f0 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
>> @@ -106,10 +106,32 @@ static u64 amdgpu_vram_mgr_vis_size(struct
>> amdgpu_device *adev,
>> Â Â */
>> Â u64 amdgpu_vram_mgr_bo_invisible_size(struct amdgpu_bo *bo)
>> Â {
>> -Â Â Â if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
>> +Â Â Â struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
>> +Â Â Â struct amdgpu_vram_mgr *mgr = adev->mman.bdev.man[TTM_PL_VRAM].priv;
>> +Â Â Â struct ttm_mem_reg *mem = &bo->tbo.mem;
>> +Â Â Â struct drm_mm_node *nodes = mem->mm_node;
>> +Â Â Â unsigned pages = mem->num_pages;
>> +Â Â Â u64 usage = 0;
>> +
>> +Â Â Â if (adev->gmc.visible_vram_size == adev->gmc.real_vram_size)
>> +Â Â Â Â Â Â Â return 0;
>
> Andrey just added a helper for that into amdgpu_gmc.h.
Yep, see patch 4 and the cover letter. :)
>> +
>> +Â Â Â if (mem->start >= adev->gmc.visible_vram_size >> PAGE_SHIFT)
>> Â Â Â Â Â Â Â Â Â return amdgpu_bo_size(bo);
>> Â -Â Â Â return 0;
>> +Â Â Â if (!nodes)
>> +Â Â Â Â Â Â Â return 0;
>> +
>> +Â Â Â spin_lock(&mgr->lock);
>
> I actually don't see any need to grab the lock here.
Ah, so amdgpu_vram_mgr_del only grabs it for drm_mm_remove_node?
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X developer
