Commit f7b02a52 authored by Chris Wilson's avatar Chris Wilson

drm/i915: Skip waking the signaler when enabling before request submission

If we are enabling the breadcrumbs signaling prior to submitting the
request, we know that we cannot have missed the interrupt and can
therefore skip immediately waking the signaler to check.

This reduces a significant chunk of the __i915_gem_request_submit()
overhead for inter-engine synchronisation, for example in gem_exec_whisper.
Signed-off-by: default avatarChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170426080659.28771-1-chris@chris-wilson.co.ukReviewed-by: default avatarTvrtko Ursulin <tvrtko.ursulin@intel.com>
parent 01001863
...@@ -61,7 +61,7 @@ static bool i915_fence_enable_signaling(struct dma_fence *fence) ...@@ -61,7 +61,7 @@ static bool i915_fence_enable_signaling(struct dma_fence *fence)
if (i915_fence_signaled(fence)) if (i915_fence_signaled(fence))
return false; return false;
intel_engine_enable_signaling(to_request(fence)); intel_engine_enable_signaling(to_request(fence), true);
return true; return true;
} }
...@@ -437,7 +437,7 @@ void __i915_gem_request_submit(struct drm_i915_gem_request *request) ...@@ -437,7 +437,7 @@ void __i915_gem_request_submit(struct drm_i915_gem_request *request)
spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING); spin_lock_nested(&request->lock, SINGLE_DEPTH_NESTING);
request->global_seqno = seqno; request->global_seqno = seqno;
if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags)) if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags))
intel_engine_enable_signaling(request); intel_engine_enable_signaling(request, false);
spin_unlock(&request->lock); spin_unlock(&request->lock);
engine->emit_breadcrumb(request, engine->emit_breadcrumb(request,
......
...@@ -649,7 +649,7 @@ static void nested_enable_signaling(struct drm_i915_gem_request *rq) ...@@ -649,7 +649,7 @@ static void nested_enable_signaling(struct drm_i915_gem_request *rq)
trace_dma_fence_enable_signal(&rq->fence); trace_dma_fence_enable_signal(&rq->fence);
spin_lock_nested(&rq->lock, SINGLE_DEPTH_NESTING); spin_lock_nested(&rq->lock, SINGLE_DEPTH_NESTING);
intel_engine_enable_signaling(rq); intel_engine_enable_signaling(rq, true);
spin_unlock(&rq->lock); spin_unlock(&rq->lock);
} }
......
...@@ -667,12 +667,13 @@ static int intel_breadcrumbs_signaler(void *arg) ...@@ -667,12 +667,13 @@ static int intel_breadcrumbs_signaler(void *arg)
return 0; return 0;
} }
void intel_engine_enable_signaling(struct drm_i915_gem_request *request) void intel_engine_enable_signaling(struct drm_i915_gem_request *request,
bool wakeup)
{ {
struct intel_engine_cs *engine = request->engine; struct intel_engine_cs *engine = request->engine;
struct intel_breadcrumbs *b = &engine->breadcrumbs; struct intel_breadcrumbs *b = &engine->breadcrumbs;
struct rb_node *parent, **p; struct rb_node *parent, **p;
bool first, wakeup; bool first;
u32 seqno; u32 seqno;
/* Note that we may be called from an interrupt handler on another /* Note that we may be called from an interrupt handler on another
...@@ -705,7 +706,7 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request) ...@@ -705,7 +706,7 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request)
* If we are the oldest waiter, enable the irq (after which we * If we are the oldest waiter, enable the irq (after which we
* must double check that the seqno did not complete). * must double check that the seqno did not complete).
*/ */
wakeup = __intel_engine_add_wait(engine, &request->signaling.wait); wakeup &= __intel_engine_add_wait(engine, &request->signaling.wait);
/* Now insert ourselves into the retirement ordered list of signals /* Now insert ourselves into the retirement ordered list of signals
* on this engine. We track the oldest seqno as that will be the * on this engine. We track the oldest seqno as that will be the
......
...@@ -672,7 +672,8 @@ bool intel_engine_add_wait(struct intel_engine_cs *engine, ...@@ -672,7 +672,8 @@ bool intel_engine_add_wait(struct intel_engine_cs *engine,
struct intel_wait *wait); struct intel_wait *wait);
void intel_engine_remove_wait(struct intel_engine_cs *engine, void intel_engine_remove_wait(struct intel_engine_cs *engine,
struct intel_wait *wait); struct intel_wait *wait);
void intel_engine_enable_signaling(struct drm_i915_gem_request *request); void intel_engine_enable_signaling(struct drm_i915_gem_request *request,
bool wakeup);
void intel_engine_cancel_signaling(struct drm_i915_gem_request *request); void intel_engine_cancel_signaling(struct drm_i915_gem_request *request);
static inline bool intel_engine_has_waiter(const struct intel_engine_cs *engine) static inline bool intel_engine_has_waiter(const struct intel_engine_cs *engine)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment