drm/i915: Defer transfer onto execution timeline to actual hw submission

Defer the transfer from the client's timeline onto the execution
timeline from the point of readiness to the point of actual submission.
For example, in execlists, a request is finally submitted to hardware
when the hardware is ready, and only put onto the hardware queue when
the request is ready. By deferring the transfer, we ensure that the
timeline is maintained in retirement order if we decide to queue the
requests onto the hardware in a different order than fifo.

v2: Rebased onto distinct global/user timeline lock classes.
v3: Play with the position of the spin_lock().
v4: Nesting finally resolved with distinct sw_fence lock classes.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161114204105.29171-4-chris@chris-wilson.co.uk

1 files changed, 25 insertions, 13 deletions

diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c
index d0f6b9f82636..952d2aec5244 100644
--- a/drivers/gpu/drm/i915/i915_gem_request.c
+++ b/drivers/gpu/drm/i915/i915_gem_request.c
@@ -306,25 +306,16 @@ static u32 timeline_get_seqno(struct i915_gem_timeline *tl)
 	return atomic_inc_return(&tl->next_seqno);
 }
 
-static int __i915_sw_fence_call
-submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
+void __i915_gem_request_submit(struct drm_i915_gem_request *request)
 {
-	struct drm_i915_gem_request *request =
-		container_of(fence, typeof(*request), submit);
 	struct intel_engine_cs *engine = request->engine;
 	struct intel_timeline *timeline;
-	unsigned long flags;
 	u32 seqno;
 
-	if (state != FENCE_COMPLETE)
-		return NOTIFY_DONE;
-
 	/* Transfer from per-context onto the global per-engine timeline */
 	timeline = engine->timeline;
 	GEM_BUG_ON(timeline == request->timeline);
-
-	/* Will be called from irq-context when using foreign DMA fences */
-	spin_lock_irqsave(&timeline->lock, flags);
+	assert_spin_locked(&timeline->lock);
 
 	seqno = timeline_get_seqno(timeline->common);
 	GEM_BUG_ON(!seqno);
@@ -344,15 +335,36 @@ submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
 	GEM_BUG_ON(!request->global_seqno);
 	engine->emit_breadcrumb(request,
 				request->ring->vaddr + request->postfix);
-	engine->submit_request(request);
 
 	spin_lock(&request->timeline->lock);
 	list_move_tail(&request->link, &timeline->requests);
 	spin_unlock(&request->timeline->lock);
 
 	i915_sw_fence_commit(&request->execute);
+}
+
+void i915_gem_request_submit(struct drm_i915_gem_request *request)
+{
+	struct intel_engine_cs *engine = request->engine;
+	unsigned long flags;
 
-	spin_unlock_irqrestore(&timeline->lock, flags);
+	/* Will be called from irq-context when using foreign fences. */
+	spin_lock_irqsave(&engine->timeline->lock, flags);
+
+	__i915_gem_request_submit(request);
+
+	spin_unlock_irqrestore(&engine->timeline->lock, flags);
+}
+
+static int __i915_sw_fence_call
+submit_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
+{
+	if (state == FENCE_COMPLETE) {
+		struct drm_i915_gem_request *request =
+			container_of(fence, typeof(*request), submit);
+
+		request->engine->submit_request(request);
+	}
 
 	return NOTIFY_DONE;
 }