drm/i915: Reduce test_and_set_bit to set_bit in i915_request_submit()

Avoid the full blown memory barrier of test_and_set_bit() by noting the
completed request and removing it from the lists.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210114135612.13210-5-chris@chris-wilson.co.uk

drivers/gpu/drm/i915/i915_request.c

@@ -551,8 +551,10 @@ bool __i915_request_submit(struct i915_request *request)
 	 * dropped upon retiring. (Otherwise if resubmit a *retired*
 	 * request, this would be a horrible use-after-free.)
 	 */
-	if (__i915_request_is_complete(request))
+	if (__i915_request_is_complete(request)) {
-		goto xfer;
+		list_del_init(&request->sched.link);
+		goto active;
+	}
 
 	if (unlikely(intel_context_is_banned(request->context)))
 		i915_request_set_error_once(request, -EIO);
@@ -587,11 +589,11 @@ bool __i915_request_submit(struct i915_request *request)
 	engine->serial++;
 	result = true;
 
-xfer:
+	GEM_BUG_ON(test_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags));
-	if (!test_and_set_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags)) {
+	list_move_tail(&request->sched.link, &engine->active.requests);
-		list_move_tail(&request->sched.link, &engine->active.requests);
+active:
-		clear_bit(I915_FENCE_FLAG_PQUEUE, &request->fence.flags);
+	clear_bit(I915_FENCE_FLAG_PQUEUE, &request->fence.flags);
-	}
+	set_bit(I915_FENCE_FLAG_ACTIVE, &request->fence.flags);
 
 	/*
 	 * XXX Rollback bonded-execution on __i915_request_unsubmit()?