On 2016å¹´08æ??18æ?¥ 19:58, Christian König wrote:
> Am 18.08.2016 um 13:17 schrieb Chunming Zhou:
>> V2:
>> 1. don't directly submit to many jobs at the same time.
>> 2. delete unrelated printk.
>>
>> Change-Id: I963598ba6eb44bc8620d70e026c0175d1a1de120
>> Signed-off-by: Chunming Zhou <David1.Zhou at amd.com>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 58
>> +++++++++++++++++++++++++++++-
>> 1 file changed, 57 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> index c9b15c0..0bedb26 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
>> @@ -2120,6 +2120,35 @@ bool amdgpu_need_backup(struct amdgpu_device
>> *adev)
>> return amdgpu_lockup_timeout > 0 ? true : false;
>> }
>> +static int amdgpu_recover_vram_from_shadow(struct amdgpu_device
>> *adev,
>> + struct amdgpu_ring *ring,
>> + struct amdgpu_bo *bo,
>> + struct fence **fence)
>> +{
>> + uint32_t domain;
>> + int r;
>> +
>> + if (!bo->shadow)
>> + return 0;
>> +
>> + r = amdgpu_bo_reserve(bo, false);
>> + if (r)
>> + return r;
>> + domain = amdgpu_mem_type_to_domain(bo->tbo.mem.mem_type);
>> + /* if bo has been evicted, then no need to recover */
>> + if (domain == AMDGPU_GEM_DOMAIN_VRAM) {
>> + r = amdgpu_bo_restore_from_shadow(adev, ring, bo,
>> + NULL, fence, true);
>> + if (r) {
>> + DRM_ERROR("recover page table failed!\n");
>> + goto err;
>> + }
>> + }
>> +err:
>> + amdgpu_bo_unreserve(bo);
>> + return r;
>> +}
>> +
>> /**
>> * amdgpu_gpu_reset - reset the asic
>> *
>> @@ -2202,13 +2231,40 @@ retry:
>> if (r) {
>> dev_err(adev->dev, "ib ring test failed (%d).\n", r);
>> r = amdgpu_suspend(adev);
>> + need_full_reset = true;
>> goto retry;
>> }
>> -
>> + /**
>> + * recovery vm page tables, since we cannot depend on VRAM is
>> + * consistent after gpu full reset.
>> + */
>> + if (need_full_reset && amdgpu_need_backup(adev)) {
>> + struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
>> + struct amdgpu_bo *bo, *tmp;
>> + struct fence *fence = NULL, *next = NULL;
>> +
>> + DRM_INFO("recover vram bo from shadow\n");
>> + mutex_lock(&adev->shadow_list_lock);
>> + list_for_each_entry_safe(bo, tmp, &adev->shadow_list,
>> shadow_list) {
>> + amdgpu_recover_vram_from_shadow(adev, ring, bo, &next);
>> + if (fence)
>> + fence_wait(fence, false);
>
> You should check the return code here as well and abort the loop if
> anything fails.
>
> With that fixed the patch is Reviewed-by: Christian König
> <christian.koenig at amd.com> as well.
>
> Finally done with that!
Agree, Thanks very much for review. Next will be 'recover gart table':)
Cheers,
David Zhou
>
> Cheers,
> Christian.
>
>> + fence_put(fence);
>> + fence = next;
>> + }
>> + mutex_unlock(&adev->shadow_list_lock);
>> + if (fence) {
>> + r = fence_wait(fence, false);
>> + if (r)
>> + WARN(r, "recovery from shadow isn't comleted\n");
>> + }
>> + fence_put(fence);
>> + }
>> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
>> struct amdgpu_ring *ring = adev->rings[i];
>> if (!ring)
>> continue;
>> +
>> amd_sched_job_recovery(&ring->sched);
>> kthread_unpark(ring->sched.thread);
>> }
>
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
