drm/amdgpu: Add KFD eviction fence

This fence is used by KFD to keep memory resident while user mode
queues are enabled. Trying to evict memory will trigger the
enable_signaling callback, which starts a KFD eviction, which
involves preempting user mode queues before signaling the fence.
There is one such fence per process.

v2:
* Grab a reference to mm_struct
* Dereference fence after NULL check
* Simplify fence release, no need to signal without anyone waiting
* Added signed-off-by Harish, who is the original author of this code

v3:
* update MAINTAINERS file
* change amd_kfd_ prefix to amdkfd_
* remove useless initialization of variable to NULL

v4:
* set amdkfd_fence_ops to be static
* Suggested by: Fengguang Wu <fengguang.wu@intel.com>

Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>

1 files changed, 19 insertions, 2 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index df65c66dc956..b8d3b87fd43e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -31,6 +31,7 @@
 #include <drm/drmP.h>
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
+#include "amdgpu_amdkfd.h"
 
 struct amdgpu_sync_entry {
         struct hlist_node       node;
@@ -85,11 +86,20 @@ static bool amdgpu_sync_same_dev(struct amdgpu_device *adev,
  */
 static void *amdgpu_sync_get_owner(struct dma_fence *f)
 {
-        struct drm_sched_fence *s_fence = to_drm_sched_fence(f);
+        struct drm_sched_fence *s_fence;
+        struct amdgpu_amdkfd_fence *kfd_fence;
+
+        if (!f)
+                return AMDGPU_FENCE_OWNER_UNDEFINED;
 
+        s_fence = to_drm_sched_fence(f);
         if (s_fence)
                 return s_fence->owner;
 
+        kfd_fence = to_amdgpu_amdkfd_fence(f);
+        if (kfd_fence)
+                return AMDGPU_FENCE_OWNER_KFD;
+
         return AMDGPU_FENCE_OWNER_UNDEFINED;
 }
 
@@ -204,11 +214,18 @@ int amdgpu_sync_resv(struct amdgpu_device *adev,
         for (i = 0; i < flist->shared_count; ++i) {
                 f = rcu_dereference_protected(flist->shared[i],
                                               reservation_object_held(resv));
+                /* We only want to trigger KFD eviction fences on
+                 * evict or move jobs. Skip KFD fences otherwise.
+                 */
+                fence_owner = amdgpu_sync_get_owner(f);
+                if (fence_owner == AMDGPU_FENCE_OWNER_KFD &&
+                    owner != AMDGPU_FENCE_OWNER_UNDEFINED)
+                        continue;
+
                 if (amdgpu_sync_same_dev(adev, f)) {
                         /* VM updates are only interesting
                          * for other VM updates and moves.
                          */
-                        fence_owner = amdgpu_sync_get_owner(f);
                         if ((owner != AMDGPU_FENCE_OWNER_UNDEFINED) &&
                             (fence_owner != AMDGPU_FENCE_OWNER_UNDEFINED) &&
                             ((owner == AMDGPU_FENCE_OWNER_VM) !=