From: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
In order to time how long a seqno is executed by a ring, we need to
measure both its insertion and its completion. (Using the completion of
the previous seqno as an estimate for when the GPU starts, if busy.) In
order to get an exact completion timestamp, we need irqs. This is
enabled by trace_i915_gem_ring_dispatch, so it makes more sens to pair
the completion event with that rather than the request.
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/i915_gem_execbuffer.c | 2 +-
drivers/gpu/drm/i915/i915_irq.c | 2 +-
drivers/gpu/drm/i915/i915_trace.h | 48 +++++++++++++++---------------
3 files changed, 26 insertions(+), 26 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 3b64b9f..4f90cb3 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1127,7 +1127,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
goto err;
}
- trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
+ trace_i915_gem_ring_dispatch(ring, flags);
i915_gem_execbuffer_move_to_active(&eb->objects, vm, ring);
i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index efac65b..d61bd5b 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -801,7 +801,7 @@ static void notify_ring(struct drm_device *dev,
if (ring->obj == NULL)
return;
- trace_i915_gem_request_complete(ring);
+ trace_i915_gem_ring_complete(ring);
wake_up_all(&ring->irq_queue);
i915_queue_hangcheck(dev);
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index a1797f6..5c8e36a 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -234,8 +234,8 @@ TRACE_EVENT(i915_gem_evict_everything,
);
TRACE_EVENT(i915_gem_ring_dispatch,
- TP_PROTO(struct intel_ring_buffer *ring, u32 seqno, u32 flags),
- TP_ARGS(ring, seqno, flags),
+ TP_PROTO(struct intel_ring_buffer *ring, u32 flags),
+ TP_ARGS(ring, flags),
TP_STRUCT__entry(
__field(u32, dev)
@@ -247,15 +247,35 @@ TRACE_EVENT(i915_gem_ring_dispatch,
TP_fast_assign(
__entry->dev = ring->dev->primary->index;
__entry->ring = ring->id;
- __entry->seqno = seqno;
+ __entry->seqno = intel_ring_get_seqno(ring),
__entry->flags = flags;
- i915_trace_irq_get(ring, seqno);
+ i915_trace_irq_get(ring, __entry->seqno);
),
TP_printk("dev=%u, ring=%u, seqno=%u, flags=%x",
__entry->dev, __entry->ring, __entry->seqno, __entry->flags)
);
+TRACE_EVENT(i915_gem_ring_complete,
+ TP_PROTO(struct intel_ring_buffer *ring),
+ TP_ARGS(ring),
+
+ TP_STRUCT__entry(
+ __field(u32, dev)
+ __field(u32, ring)
+ __field(u32, seqno)
+ ),
+
+ TP_fast_assign(
+ __entry->dev = ring->dev->primary->index;
+ __entry->ring = ring->id;
+ __entry->seqno = ring->get_seqno(ring, false);
+ ),
+
+ TP_printk("dev=%u, ring=%u, seqno=%u",
+ __entry->dev, __entry->ring, __entry->seqno)
+);
+
TRACE_EVENT(i915_gem_ring_flush,
TP_PROTO(struct intel_ring_buffer *ring, u32 invalidate, u32 flush),
TP_ARGS(ring, invalidate, flush),
@@ -304,26 +324,6 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_add,
TP_ARGS(ring, seqno)
);
-TRACE_EVENT(i915_gem_request_complete,
- TP_PROTO(struct intel_ring_buffer *ring),
- TP_ARGS(ring),
-
- TP_STRUCT__entry(
- __field(u32, dev)
- __field(u32, ring)
- __field(u32, seqno)
- ),
-
- TP_fast_assign(
- __entry->dev = ring->dev->primary->index;
- __entry->ring = ring->id;
- __entry->seqno = ring->get_seqno(ring, false);
- ),
-
- TP_printk("dev=%u, ring=%u, seqno=%u",
- __entry->dev, __entry->ring, __entry->seqno)
-);
-
DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
TP_ARGS(ring, seqno)
--
1.8.1.4
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
