drm/amdkfd: Output migrate end event if migrate failed

If page migration failed, also output migrate end event to match with
migrate start event, with failure error_code added to the end of the
migrate message macro. This will not break uAPI because application uses
old message macro sscanf drop and ignore the error_code.

Output GPU page fault restore end event if migration failed.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdkfd/kfd_migrate.c

@@ -445,14 +445,13 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
 			 mpages, cpages, migrate.npages);
 
-	kfd_smi_event_migration_end(node, p->lead_thread->pid,
-				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    0, node->id, trigger);
-
 	svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages);
 
 out_free:
 	kvfree(buf);
+	kfd_smi_event_migration_end(node, p->lead_thread->pid,
+				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
+				    0, node->id, trigger, r);
 out:
 	if (!r && mpages) {
 		pdd = svm_range_get_pdd_by_node(prange, node);
@@ -751,14 +750,13 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	kfd_smi_event_migration_end(node, p->lead_thread->pid,
-				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
-				    node->id, 0, trigger);
-
 	svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages);
 
 out_free:
 	kvfree(buf);
+	kfd_smi_event_migration_end(node, p->lead_thread->pid,
+				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
+				    node->id, 0, trigger, r);
 out:
 	if (!r && cpages) {
 		mpages = cpages - upages;

drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c

@@ -292,12 +292,13 @@ void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
 
 void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
 				 unsigned long start, unsigned long end,
-				 uint32_t from, uint32_t to, uint32_t trigger)
+				 uint32_t from, uint32_t to, uint32_t trigger,
+				 int error_code)
 {
 	kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_END,
 			  KFD_EVENT_FMT_MIGRATE_END(
 			  ktime_get_boottime_ns(), pid, start, end - start,
-			  from, to, trigger));
+			  from, to, trigger, error_code));
 }
 
 void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid,

drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h

@@ -44,7 +44,8 @@ void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid,
 			     uint32_t trigger);
 void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid,
 			     unsigned long start, unsigned long end,
-			     uint32_t from, uint32_t to, uint32_t trigger);
+			     uint32_t from, uint32_t to, uint32_t trigger,
+			     int error_code);
 void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid,
 				  uint32_t trigger);
 void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid);

drivers/gpu/drm/amd/amdkfd/kfd_svm.c

@@ -3085,8 +3085,6 @@ retry_write_locked:
 	start = max_t(unsigned long, ALIGN_DOWN(addr, size), prange->start);
 	last = min_t(unsigned long, ALIGN(addr + 1, size) - 1, prange->last);
 	if (prange->actual_loc != 0 || best_loc != 0) {
-		migration = true;
-
 		if (best_loc) {
 			r = svm_migrate_to_vram(prange, best_loc, start, last,
 					mm, KFD_MIGRATE_TRIGGER_PAGEFAULT_GPU);
@@ -3109,7 +3107,9 @@ retry_write_locked:
 		if (r) {
 			pr_debug("failed %d to migrate svms %p [0x%lx 0x%lx]\n",
 				 r, svms, start, last);
-			goto out_unlock_range;
+			goto out_migrate_fail;
+		} else {
+			migration = true;
 		}
 	}
 
@@ -3119,6 +3119,7 @@ retry_write_locked:
 		pr_debug("failed %d to map svms 0x%p [0x%lx 0x%lx] to gpus\n",
 			 r, svms, start, last);
 
+out_migrate_fail:
 	kfd_smi_event_page_fault_end(node, p->lead_thread->pid, addr,
 				     migration);
 

include/uapi/linux/kfd_ioctl.h

@@ -609,6 +609,7 @@ struct kfd_ioctl_smi_events_args {
  *    migrate_update: GPU page fault is recovered by 'M' for migrate, 'U' for update
  *    rw: 'W' for write page fault, 'R' for read page fault
  *    rescheduled: 'R' if the queue restore failed and rescheduled to try again
+ *    error_code: migrate failure error code, 0 if no error
  */
 #define KFD_EVENT_FMT_UPDATE_GPU_RESET(reset_seq_num, reset_cause)\
 		"%x %s\n", (reset_seq_num), (reset_cause)
@@ -630,9 +631,9 @@ struct kfd_ioctl_smi_events_args {
 		"%lld -%d @%lx(%lx) %x->%x %x:%x %d\n", (ns), (pid), (start), (size),\
 		(from), (to), (prefetch_loc), (preferred_loc), (migrate_trigger)
 
-#define KFD_EVENT_FMT_MIGRATE_END(ns, pid, start, size, from, to, migrate_trigger)\
+#define KFD_EVENT_FMT_MIGRATE_END(ns, pid, start, size, from, to, migrate_trigger, error_code) \
-		"%lld -%d @%lx(%lx) %x->%x %d\n", (ns), (pid), (start), (size),\
+		"%lld -%d @%lx(%lx) %x->%x %d %d\n", (ns), (pid), (start), (size),\
-		(from), (to), (migrate_trigger)
+		(from), (to), (migrate_trigger), (error_code)
 
 #define KFD_EVENT_FMT_QUEUE_EVICTION(ns, pid, node, evict_trigger)\
 		"%lld -%d %x %d\n", (ns), (pid), (node), (evict_trigger)