All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH 19/38] drm/i915: Identify active requests
Date: Fri, 18 Jan 2019 14:00:50 +0000	[thread overview]
Message-ID: <20190118140109.25261-20-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <20190118140109.25261-1-chris@chris-wilson.co.uk>

To allow requests to forgo a common execution timeline, one question we
need to be able to answer is "is this request running?". To track
whether a request has started on HW, we can emit a breadcrumb at the
beginning of the request and check its timeline's HWSP to see if the
breadcrumb has advanced past the start of this request. (This is in
contrast to the global timeline where we need only ask if we are on the
global timeline and if the timeline has advanced past the end of the
previous request.)

There is still confusion from a preempted request, which has already
started but relinquished the HW to a high priority request. For the
common case, this discrepancy should be negligible. However, for
identification of hung requests, knowing which one was running at the
time of the hang will be much more important.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_request.c     |  1 +
 drivers/gpu/drm/i915/i915_request.h     |  1 +
 drivers/gpu/drm/i915/i915_timeline.c    |  1 +
 drivers/gpu/drm/i915/i915_timeline.h    |  2 ++
 drivers/gpu/drm/i915/intel_engine_cs.c  |  4 +++-
 drivers/gpu/drm/i915/intel_lrc.c        | 23 +++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_ringbuffer.c |  2 ++
 7 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index 0d7b71aff28f..f61cc5c1bf08 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -332,6 +332,7 @@ void i915_request_retire_upto(struct i915_request *rq)
 
 static u32 timeline_get_seqno(struct i915_timeline *tl)
 {
+	tl->seqno += tl->has_initial_breadcrumb;
 	return ++tl->seqno;
 }
 
diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index a16a3b7f7d92..83ce982dcbd9 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -324,6 +324,7 @@ static inline u32 hwsp_seqno(const struct i915_request *rq)
  */
 static inline bool i915_request_started(const struct i915_request *rq)
 {
+	/* Remember: started but may have since been preempted! */
 	return i915_seqno_passed(hwsp_seqno(rq), rq->fence.seqno - 1);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_timeline.c b/drivers/gpu/drm/i915/i915_timeline.c
index 92532f713dd7..189717db11a9 100644
--- a/drivers/gpu/drm/i915/i915_timeline.c
+++ b/drivers/gpu/drm/i915/i915_timeline.c
@@ -124,6 +124,7 @@ int i915_timeline_init(struct drm_i915_private *i915,
 	timeline->i915 = i915;
 	timeline->name = name;
 	timeline->pin_count = 0;
+	timeline->has_initial_breadcrumb = !global_hwsp;
 
 	if (global_hwsp) {
 		timeline->hwsp_ggtt = i915_vma_get(global_hwsp);
diff --git a/drivers/gpu/drm/i915/i915_timeline.h b/drivers/gpu/drm/i915/i915_timeline.h
index 0c3739d53d79..421eb34568de 100644
--- a/drivers/gpu/drm/i915/i915_timeline.h
+++ b/drivers/gpu/drm/i915/i915_timeline.h
@@ -47,6 +47,8 @@ struct i915_timeline {
 	struct i915_vma *hwsp_ggtt;
 	u32 hwsp_offset;
 
+	bool has_initial_breadcrumb;
+
 	/**
 	 * List of breadcrumbs associated with GPU requests currently
 	 * outstanding.
diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index c850d131d8c3..ae455b874c9f 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -1239,7 +1239,9 @@ static void print_request(struct drm_printer *m,
 	drm_printf(m, "%s%x%s [%llx:%llx]%s @ %dms: %s\n",
 		   prefix,
 		   rq->global_seqno,
-		   i915_request_completed(rq) ? "!" : "",
+		   i915_request_completed(rq) ? "!" :
+		   i915_request_started(rq) ? "*" :
+		   "",
 		   rq->fence.context, rq->fence.seqno,
 		   buf,
 		   jiffies_to_msecs(jiffies - rq->emitted_jiffies),
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 041b64fb203c..740e09d3ca26 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1275,6 +1275,25 @@ execlists_context_pin(struct intel_engine_cs *engine,
 	return __execlists_context_pin(engine, ctx, ce);
 }
 
+static int emit_initial_breadcrumb(struct i915_request *rq)
+{
+	u32 *cs;
+
+	GEM_BUG_ON(!rq->timeline->has_initial_breadcrumb);
+
+	cs = intel_ring_begin(rq, 4);
+	if (IS_ERR(cs))
+		return PTR_ERR(cs);
+
+	*cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT;
+	*cs++ = i915_timeline_seqno_address(rq->timeline);
+	*cs++ = 0;
+	*cs++ = rq->fence.seqno - 1;
+
+	intel_ring_advance(rq, cs);
+	return 0;
+}
+
 static int emit_pdps(struct i915_request *rq)
 {
 	const struct intel_engine_cs * const engine = rq->engine;
@@ -1349,6 +1368,10 @@ static int execlists_request_alloc(struct i915_request *request)
 	 * to cancel/unwind this request now.
 	 */
 
+	ret = emit_initial_breadcrumb(request);
+	if (ret)
+		return ret;
+
 	/* Unconditionally invalidate GPU caches and TLBs. */
 	if (i915_vm_is_48bit(&request->gem_context->ppgtt->vm))
 		ret = request->engine->emit_flush(request, EMIT_INVALIDATE);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index bcc700e7037b..2b4beb15a271 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1593,6 +1593,7 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine)
 		err = PTR_ERR(timeline);
 		goto err;
 	}
+	GEM_BUG_ON(timeline->has_initial_breadcrumb);
 
 	ring = intel_engine_create_ring(engine, timeline, 32 * PAGE_SIZE);
 	i915_timeline_put(timeline);
@@ -1946,6 +1947,7 @@ static int ring_request_alloc(struct i915_request *request)
 	int ret;
 
 	GEM_BUG_ON(!request->hw_context->pin_count);
+	GEM_BUG_ON(request->timeline->has_initial_breadcrumb);
 
 	/*
 	 * Flush enough space to reduce the likelihood of waiting after
-- 
2.20.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2019-01-18 14:01 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-18 14:00 Keeping Tvrtko busy Chris Wilson
2019-01-18 14:00 ` [PATCH 01/38] drm/i915/execlists: Store the highest priority context Chris Wilson
2019-01-18 14:00 ` [PATCH 02/38] drm/i915: Make all GPU resets atomic Chris Wilson
2019-01-18 14:22   ` Mika Kuoppala
2019-01-18 14:00 ` [PATCH 03/38] drm/i915/guc: Disable global reset Chris Wilson
2019-01-18 14:00 ` [PATCH 04/38] drm/i915: Remove GPU reset dependence on struct_mutex Chris Wilson
2019-01-18 14:00 ` [PATCH 05/38] drm/i915/selftests: Trim struct_mutex duration for set-wedged selftest Chris Wilson
2019-01-18 14:29   ` Mika Kuoppala
2019-01-18 14:00 ` [PATCH 06/38] drm/i915: Issue engine resets onto idle engines Chris Wilson
2019-01-18 14:00 ` [PATCH 07/38] drm/i915: Stop tracking MRU activity on VMA Chris Wilson
2019-01-18 16:03   ` Tvrtko Ursulin
2019-01-18 16:06     ` Chris Wilson
2019-01-22 14:19     ` Chris Wilson
2019-01-25 10:46       ` Tvrtko Ursulin
2019-01-25 13:38         ` Chris Wilson
2019-01-25 13:46           ` Chris Wilson
2019-01-25 14:08             ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 08/38] drm/i915: Pull VM lists under the VM mutex Chris Wilson
2019-01-18 16:04   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 09/38] drm/i915: Move vma lookup to its own lock Chris Wilson
2019-01-18 16:14   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 10/38] drm/i915/selftests: Allocate mock ring/timeline per context Chris Wilson
2019-01-18 16:26   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 11/38] drm/i915: Always allocate an object/vma for the HWSP Chris Wilson
2019-01-18 14:00 ` [PATCH 12/38] drm/i915: Move list of timelines under its own lock Chris Wilson
2019-01-18 16:28   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 13/38] drm/i915: Introduce concept of per-timeline (context) HWSP Chris Wilson
2019-01-18 14:00 ` [PATCH 14/38] drm/i915: Enlarge vma->pin_count Chris Wilson
2019-01-18 14:00 ` [PATCH 15/38] drm/i915: Allocate a status page for each timeline Chris Wilson
2019-01-21 11:18   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 16/38] drm/i915: Share per-timeline HWSP using a slab suballocator Chris Wilson
2019-01-18 14:00 ` [PATCH 17/38] drm/i915: Keep all partially allocated HWSP on a freelist Chris Wilson
2019-01-18 14:00 ` [PATCH 18/38] drm/i915: Track the context's seqno in its own timeline HWSP Chris Wilson
2019-01-18 14:00 ` Chris Wilson [this message]
2019-01-18 14:00 ` [PATCH 20/38] drm/i915: Remove the intel_engine_notify tracepoint Chris Wilson
2019-01-18 14:00 ` [PATCH 21/38] drm/i915: Replace global breadcrumbs with per-context interrupt tracking Chris Wilson
2019-01-18 14:00 ` [PATCH 22/38] drm/i915: Drop fake breadcrumb irq Chris Wilson
2019-01-18 14:00 ` [PATCH 23/38] drm/i915: Replace global_seqno with a hangcheck heartbeat seqno Chris Wilson
2019-01-18 14:00 ` [PATCH 24/38] drm/i915: Avoid presumption of execution ordering for kernel context switching Chris Wilson
2019-01-18 14:00 ` [PATCH 25/38] drm/i915/pmu: Always sample an active ringbuffer Chris Wilson
2019-01-22  9:20   ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 26/38] drm/i915: Remove the global per-engine execution timeline Chris Wilson
2019-01-18 14:00 ` [PATCH 27/38] drm/i915: Introduce the i915_user_extension_method Chris Wilson
2019-01-22  9:31   ` Tvrtko Ursulin
2019-01-22 10:47     ` Chris Wilson
2019-01-22 11:05       ` Tvrtko Ursulin
2019-01-18 14:00 ` [PATCH 28/38] drm/i915: Create/destroy VM (ppGTT) for use with contexts Chris Wilson
2019-01-23 11:30   ` Tvrtko Ursulin
2019-01-23 11:51     ` Chris Wilson
2019-01-23 12:03       ` Tvrtko Ursulin
2019-01-24 15:58     ` [PATCH v3] " Chris Wilson
2019-01-18 14:01 ` [PATCH 29/38] drm/i915: Expose user control over the ppGTT associated with a context Chris Wilson
2019-01-23 12:00   ` Tvrtko Ursulin
2019-01-23 12:15     ` Chris Wilson
2019-01-18 14:01 ` [PATCH 30/38] drm/i915: Extend CONTEXT_CREATE to set parameters upon construction Chris Wilson
2019-01-18 14:01 ` [PATCH 31/38] drm/i915: Allow contexts to share a single timeline across all engines Chris Wilson
2019-01-24 17:35   ` Tvrtko Ursulin
2019-01-18 14:01 ` [PATCH 32/38] drm/i915: Fix I915_EXEC_RING_MASK Chris Wilson
2019-01-18 14:01 ` [PATCH 33/38] drm/i915: Remove last traces of exec-id (GEM_BUSY) Chris Wilson
2019-01-18 14:01 ` [PATCH 34/38] drm/i915: Re-arrange execbuf so context is known before engine Chris Wilson
2019-01-18 14:01 ` [PATCH 35/38] drm/i915: Allow a context to define its set of engines Chris Wilson
2019-01-18 14:01 ` [PATCH 36/38] drm/i915/execlists: Refactor out can_merge_rq() Chris Wilson
2019-01-18 14:01 ` [PATCH 37/38] drm/i915: Store the BIT(engine->id) as the engine's mask Chris Wilson
2019-01-18 14:01 ` [PATCH 38/38] drm/i915: Load balancing across a virtual engine Chris Wilson
2019-01-18 14:17 ` ✗ Fi.CI.BAT: failure for series starting with [01/38] drm/i915/execlists: Store the highest priority context Patchwork
2019-01-24 16:28 ` ✗ Fi.CI.BAT: failure for series starting with [01/38] drm/i915/execlists: Store the highest priority context (rev2) Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190118140109.25261-20-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.