[Intel-gfx] [PATCH] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Chris Wilson]

We dropped calling process_csb prior to handling direct submission in
order to avoid the nesting of spinlocks and lift process_csb() and the
majority of the tasklet out of irq-off. However, we do want to avoid
ksoftirqd latency in the fast path, so try and pull the interrupt-bh
local to direct submission if we can acquire the tasklet's lock.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
Cc: Francisco Jerez <currojerez at riseup.net>
Cc: Tvrtko Ursulin <tvrtko.ursulin at linux.intel.com>
---
 drivers/gpu/drm/i915/gt/intel_lrc.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index f09dd87324b9..996554766aa5 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -2884,6 +2884,11 @@ static void queue_request(struct intel_engine_cs *engine,
 	set_bit(I915_FENCE_FLAG_PQUEUE, &rq->fence.flags);
 }
 
+static bool pending_csb(const struct intel_engine_execlists *execlists)
+{
+	return READ_ONCE(*execlists->csb_write) != execlists->csb_head;
+}
+
 static void __submit_queue_imm(struct intel_engine_cs *engine)
 {
 	struct intel_engine_execlists * const execlists = &engine->execlists;
@@ -2891,10 +2896,19 @@ static void __submit_queue_imm(struct intel_engine_cs *engine)
 	if (reset_in_progress(execlists))
 		return; /* defer until we restart the engine following reset */
 
-	if (execlists->tasklet.func == execlists_submission_tasklet)
-		__execlists_submission_tasklet(engine);
-	else
-		tasklet_hi_schedule(&execlists->tasklet);
+	/* Hopefully we clear execlists->pending[] to let us through */
+	if (execlists->pending[0] && tasklet_trylock(&execlists->tasklet)) {
+		process_csb(engine);
+		tasklet_unlock(&execlists->tasklet);
+	}
+
+	__execlists_submission_tasklet(engine);
+
+	/* Try and pull an interrupt-bh queued on another CPU to here */
+	if (pending_csb(execlists) && tasklet_trylock(&execlists->tasklet)) {
+		process_csb(engine);
+		tasklet_unlock(&execlists->tasklet);
+	}
 }
 
 static void submit_queue(struct intel_engine_cs *engine,
-- 
2.20.1