drm/i915: Use HW semaphores for inter-engine synchronisation on gen8+

Having introduced per-context seqno, we now have a means to identity
progress across the system without feel of rollback as befell the
global_seqno. That is we can program a MI_SEMAPHORE_WAIT operation in
advance of submission safe in the knowledge that our target seqno and
address is stable.

However, since we are telling the GPU to busy-spin on the target address
until it matches the signaling seqno, we only want to do so when we are
sure that busy-spin will be completed quickly. To achieve this we only
submit the request to HW once the signaler is itself executing (modulo
preemption causing us to wait longer), and we only do so for default and
above priority requests (so that idle priority tasks never themselves
hog the GPU waiting for others).

As might be reasonably expected, HW semaphores excel in inter-engine
synchronisation microbenchmarks (where the 3x reduced latency / increased
throughput more than offset the power cost of spinning on a second ring)
and have significant improvement (can be up to ~10%, most see no change)
for single clients that utilize multiple engines (typically media players
and transcoders), without regressing multiple clients that can saturate
the system or changing the power envelope dramatically.

v3: Drop the older NEQ branch, now we pin the signaler's HWSP anyway.
v4: Tell the world and include it as part of scheduler caps.

Testcase: igt/gem_exec_whisper
Testcase: igt/benchmarks/gem_wsim
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190301170901.8340-3-chris@chris-wilson.co.uk

1 files changed, 136 insertions, 2 deletions

diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index d354967d6ae8..59e30b8c4ee9 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -22,8 +22,9 @@
  *
  */
 
-#include <linux/prefetch.h>
 #include <linux/dma-fence-array.h>
+#include <linux/irq_work.h>
+#include <linux/prefetch.h>
 #include <linux/sched.h>
 #include <linux/sched/clock.h>
 #include <linux/sched/signal.h>
@@ -32,9 +33,16 @@
 #include "i915_active.h"
 #include "i915_reset.h"
 
+struct execute_cb {
+	struct list_head link;
+	struct irq_work work;
+	struct i915_sw_fence *fence;
+};
+
 static struct i915_global_request {
 	struct kmem_cache *slab_requests;
 	struct kmem_cache *slab_dependencies;
+	struct kmem_cache *slab_execute_cbs;
 } global;
 
 static const char *i915_fence_get_driver_name(struct dma_fence *fence)
@@ -325,6 +333,69 @@ void i915_request_retire_upto(struct i915_request *rq)
 	} while (tmp != rq);
 }
 
+static void irq_execute_cb(struct irq_work *wrk)
+{
+	struct execute_cb *cb = container_of(wrk, typeof(*cb), work);
+
+	i915_sw_fence_complete(cb->fence);
+	kmem_cache_free(global.slab_execute_cbs, cb);
+}
+
+static void __notify_execute_cb(struct i915_request *rq)
+{
+	struct execute_cb *cb;
+
+	lockdep_assert_held(&rq->lock);
+
+	if (list_empty(&rq->execute_cb))
+		return;
+
+	list_for_each_entry(cb, &rq->execute_cb, link)
+		irq_work_queue(&cb->work);
+
+	/*
+	 * XXX Rollback on __i915_request_unsubmit()
+	 *
+	 * In the future, perhaps when we have an active time-slicing scheduler,
+	 * it will be interesting to unsubmit parallel execution and remove
+	 * busywaits from the GPU until their master is restarted. This is
+	 * quite hairy, we have to carefully rollback the fence and do a
+	 * preempt-to-idle cycle on the target engine, all the while the
+	 * master execute_cb may refire.
+	 */
+	INIT_LIST_HEAD(&rq->execute_cb);
+}
+
+static int
+i915_request_await_execution(struct i915_request *rq,
+			     struct i915_request *signal,
+			     gfp_t gfp)
+{
+	struct execute_cb *cb;
+
+	if (i915_request_is_active(signal))
+		return 0;
+
+	cb = kmem_cache_alloc(global.slab_execute_cbs, gfp);
+	if (!cb)
+		return -ENOMEM;
+
+	cb->fence = &rq->submit;
+	i915_sw_fence_await(cb->fence);
+	init_irq_work(&cb->work, irq_execute_cb);
+
+	spin_lock_irq(&signal->lock);
+	if (i915_request_is_active(signal)) {
+		i915_sw_fence_complete(cb->fence);
+		kmem_cache_free(global.slab_execute_cbs, cb);
+	} else {
+		list_add_tail(&cb->link, &signal->execute_cb);
+	}
+	spin_unlock_irq(&signal->lock);
+
+	return 0;
+}
+
 static void move_to_timeline(struct i915_request *request,
 			     struct i915_timeline *timeline)
 {
@@ -361,6 +432,8 @@ void __i915_request_submit(struct i915_request *request)
 	    !i915_request_enable_breadcrumb(request))
 		intel_engine_queue_breadcrumbs(engine);
 
+	__notify_execute_cb(request);
+
 	spin_unlock(&request->lock);
 
 	engine->emit_fini_breadcrumb(request,
@@ -608,6 +681,7 @@ i915_request_alloc(struct intel_engine_cs *engine, struct i915_gem_context *ctx)
 	}
 
 	INIT_LIST_HEAD(&rq->active_list);
+	INIT_LIST_HEAD(&rq->execute_cb);
 
 	tl = ce->ring->timeline;
 	ret = i915_timeline_get_seqno(tl, rq, &seqno);
@@ -697,6 +771,52 @@ err_unreserve:
 }
 
 static int
+emit_semaphore_wait(struct i915_request *to,
+		    struct i915_request *from,
+		    gfp_t gfp)
+{
+	u32 hwsp_offset;
+	u32 *cs;
+	int err;
+
+	GEM_BUG_ON(!from->timeline->has_initial_breadcrumb);
+	GEM_BUG_ON(INTEL_GEN(to->i915) < 8);
+
+	/* We need to pin the signaler's HWSP until we are finished reading. */
+	err = i915_timeline_read_hwsp(from, to, &hwsp_offset);
+	if (err)
+		return err;
+
+	/* Only submit our spinner after the signaler is running! */
+	err = i915_request_await_execution(to, from, gfp);
+	if (err)
+		return err;
+
+	cs = intel_ring_begin(to, 4);
+	if (IS_ERR(cs))
+		return PTR_ERR(cs);
+
+	/*
+	 * Using greater-than-or-equal here means we have to worry
+	 * about seqno wraparound. To side step that issue, we swap
+	 * the timeline HWSP upon wrapping, so that everyone listening
+	 * for the old (pre-wrap) values do not see the much smaller
+	 * (post-wrap) values than they were expecting (and so wait
+	 * forever).
+	 */
+	*cs++ = MI_SEMAPHORE_WAIT |
+		MI_SEMAPHORE_GLOBAL_GTT |
+		MI_SEMAPHORE_POLL |
+		MI_SEMAPHORE_SAD_GTE_SDD;
+	*cs++ = from->fence.seqno;
+	*cs++ = hwsp_offset;
+	*cs++ = 0;
+
+	intel_ring_advance(to, cs);
+	return 0;
+}
+
+static int
 i915_request_await_request(struct i915_request *to, struct i915_request *from)
 {
 	int ret;
@@ -717,6 +837,9 @@ i915_request_await_request(struct i915_request *to, struct i915_request *from)
 		ret = i915_sw_fence_await_sw_fence_gfp(&to->submit,
 						       &from->submit,
 						       I915_FENCE_GFP);
+	} else if (intel_engine_has_semaphores(to->engine) &&
+		   to->gem_context->sched.priority >= I915_PRIORITY_NORMAL) {
+		ret = emit_semaphore_wait(to, from, I915_FENCE_GFP);
 	} else {
 		ret = i915_sw_fence_await_dma_fence(&to->submit,
 						    &from->fence, 0,
@@ -1208,14 +1331,23 @@ int __init i915_global_request_init(void)
 	if (!global.slab_requests)
 		return -ENOMEM;
 
+	global.slab_execute_cbs = KMEM_CACHE(execute_cb,
+					     SLAB_HWCACHE_ALIGN |
+					     SLAB_RECLAIM_ACCOUNT |
+					     SLAB_TYPESAFE_BY_RCU);
+	if (!global.slab_execute_cbs)
+		goto err_requests;
+
 	global.slab_dependencies = KMEM_CACHE(i915_dependency,
 					      SLAB_HWCACHE_ALIGN |
 					      SLAB_RECLAIM_ACCOUNT);
 	if (!global.slab_dependencies)
-		goto err_requests;
+		goto err_execute_cbs;
 
 	return 0;
 
+err_execute_cbs:
+	kmem_cache_destroy(global.slab_execute_cbs);
 err_requests:
 	kmem_cache_destroy(global.slab_requests);
 	return -ENOMEM;
@@ -1224,11 +1356,13 @@ err_requests:
 void i915_global_request_shrink(void)
 {
 	kmem_cache_shrink(global.slab_dependencies);
+	kmem_cache_shrink(global.slab_execute_cbs);
 	kmem_cache_shrink(global.slab_requests);
 }
 
 void i915_global_request_exit(void)
 {
 	kmem_cache_destroy(global.slab_dependencies);
+	kmem_cache_destroy(global.slab_execute_cbs);
 	kmem_cache_destroy(global.slab_requests);
 }