drm/i915: Track the context's seqno in its own timeline HWSP

Now that we have allocated ourselves a cacheline to store a breadcrumb,
we can emit a write from the GPU into the timeline's HWSP of the
per-context seqno as we complete each request. This drops the mirroring
of the per-engine HWSP and allows each context to operate independently.
We do not need to unwind the per-context timeline, and so requests are
always consistent with the timeline breadcrumb, greatly simplifying the
completion checks as we no longer need to be concerned about the
global_seqno changing mid check.

One complication though is that we have to be wary that the request may
outlive the HWSP and so avoid touching the potentially danging pointer
after we have retired the fence. We also have to guard our access of the
HWSP with RCU, the release of the obj->mm.pages should already be RCU-safe.

At this point, we are emitting both per-context and global seqno and
still using the single per-engine execution timeline for resolving
interrupts.

v2: s/fake_complete/mark_complete/

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128181812.22804-5-chris@chris-wilson.co.uk

1 files changed, 18 insertions, 1 deletions

diff --git a/drivers/gpu/drm/i915/selftests/mock_engine.c b/drivers/gpu/drm/i915/selftests/mock_engine.c
index 919c89fd6ee5..95e890d7f58b 100644
--- a/drivers/gpu/drm/i915/selftests/mock_engine.c
+++ b/drivers/gpu/drm/i915/selftests/mock_engine.c
@@ -30,6 +30,17 @@ struct mock_ring {
 	struct i915_timeline timeline;
 };
 
+static void mock_timeline_pin(struct i915_timeline *tl)
+{
+	tl->pin_count++;
+}
+
+static void mock_timeline_unpin(struct i915_timeline *tl)
+{
+	GEM_BUG_ON(!tl->pin_count);
+	tl->pin_count--;
+}
+
 static struct intel_ring *mock_ring(struct intel_engine_cs *engine)
 {
 	const unsigned long sz = PAGE_SIZE / 2;
@@ -76,6 +87,8 @@ static void advance(struct mock_request *request)
 {
 	list_del_init(&request->link);
 	mock_seqno_advance(request->base.engine, request->base.global_seqno);
+	i915_request_mark_complete(&request->base);
+	GEM_BUG_ON(!i915_request_completed(&request->base));
 }
 
 static void hw_delay_complete(struct timer_list *t)
@@ -108,6 +121,7 @@ static void hw_delay_complete(struct timer_list *t)
 
 static void mock_context_unpin(struct intel_context *ce)
 {
+	mock_timeline_unpin(ce->ring->timeline);
 	i915_gem_context_put(ce->gem_context);
 }
 
@@ -129,6 +143,7 @@ mock_context_pin(struct intel_engine_cs *engine,
 		 struct i915_gem_context *ctx)
 {
 	struct intel_context *ce = to_intel_context(ctx, engine);
+	int err = -ENOMEM;
 
 	if (ce->pin_count++)
 		return ce;
@@ -139,13 +154,15 @@ mock_context_pin(struct intel_engine_cs *engine,
 			goto err;
 	}
 
+	mock_timeline_pin(ce->ring->timeline);
+
 	ce->ops = &mock_context_ops;
 	i915_gem_context_get(ctx);
 	return ce;
 
 err:
 	ce->pin_count = 0;
-	return ERR_PTR(-ENOMEM);
+	return ERR_PTR(err);
 }
 
 static int mock_request_alloc(struct i915_request *request)