drm/i915/gt: Do not schedule normal requests immediately along virtual

When we push a virtual request onto the HW, we update the rq->engine to
point to the physical engine. A request that is then submitted by the
user that waits upon the virtual engine, but along the physical engine
in use, will then see that it is due to be submitted to the same engine
and take a shortcut (and be queued without waiting for the completion
fence). However, the virtual request may be preempted (either by higher
priority users, or by timeslicing) and removed from the physical engine
to be migrated over to one of its siblings. The dependent normal request
however is oblivious to the removal of the virtual request and remains
queued to execute on HW, believing that once it reaches the head of its
queue all of its predecessors will have completed executing!

v2: Beware restriction of signal->execution_mask prior to submission.

Fixes: 6d06779e ("drm/i915: Load balancing across a virtual engine")
Testcase: igt/gem_exec_balancer/sliced
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.3+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200526090753.11329-2-chris@chris-wilson.co.uk
(cherry picked from commit 511b6d9a)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

drivers/gpu/drm/i915/i915_request.c

@@ -1237,6 +1237,25 @@ i915_request_await_execution(struct i915_request *rq,
 	return 0;
 }
 
+static int
+await_request_submit(struct i915_request *to, struct i915_request *from)
+{
+	/*
+	 * If we are waiting on a virtual engine, then it may be
+	 * constrained to execute on a single engine *prior* to submission.
+	 * When it is submitted, it will be first submitted to the virtual
+	 * engine and then passed to the physical engine. We cannot allow
+	 * the waiter to be submitted immediately to the physical engine
+	 * as it may then bypass the virtual request.
+	 */
+	if (to->engine == READ_ONCE(from->engine))
+		return i915_sw_fence_await_sw_fence_gfp(&to->submit,
+							&from->submit,
+							I915_FENCE_GFP);
+	else
+		return __i915_request_await_execution(to, from, NULL);
+}
+
 static int
 i915_request_await_request(struct i915_request *to, struct i915_request *from)
 {
@@ -1258,10 +1277,8 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 			return ret;
 	}
 
-	if (to->engine == READ_ONCE(from->engine))
-		ret = i915_sw_fence_await_sw_fence_gfp(&to->submit,
-						       &from->submit,
-						       I915_FENCE_GFP);
+	if (is_power_of_2(to->execution_mask | READ_ONCE(from->execution_mask)))
+		ret = await_request_submit(to, from);
 	else
 		ret = emit_semaphore_wait(to, from, I915_FENCE_GFP);
 	if (ret < 0)