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 10/12] drm/i915: Flush idle barriers when waiting
Date: Tue, 22 Oct 2019 23:38:29 +0100	[thread overview]
Message-ID: <20191022223831.22677-10-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <20191022223831.22677-1-chris@chris-wilson.co.uk>

If we do find ourselves with an idle barrier inside our active while
waiting, attempt to flush it by emitting a pulse using the kernel
context.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_active.c           | 21 ++++++++-
 drivers/gpu/drm/i915/selftests/i915_active.c | 46 ++++++++++++++++++++
 2 files changed, 65 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index 07d39f22a2c3..4c190a548ba7 100644
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -6,6 +6,7 @@
 
 #include <linux/debugobjects.h>
 
+#include "gt/intel_engine_heartbeat.h"
 #include "gt/intel_engine_pm.h"
 
 #include "i915_drv.h"
@@ -435,6 +436,21 @@ static void enable_signaling(struct i915_active_fence *active)
 	dma_fence_put(fence);
 }
 
+static int flush_barrier(struct active_node *it)
+{
+	struct intel_engine_cs *engine;
+
+	if (!is_barrier(&it->base))
+		return 0;
+
+	engine = __barrier_to_engine(it);
+	smp_rmb(); /* serialise with add_active_barriers */
+	if (!is_barrier(&it->base))
+		return 0;
+
+	return intel_engine_flush_barriers(engine);
+}
+
 int i915_active_wait(struct i915_active *ref)
 {
 	struct active_node *it, *n;
@@ -448,8 +464,9 @@ int i915_active_wait(struct i915_active *ref)
 	/* Flush lazy signals */
 	enable_signaling(&ref->excl);
 	rbtree_postorder_for_each_entry_safe(it, n, &ref->tree, node) {
-		if (is_barrier(&it->base)) /* unconnected idle barrier */
-			continue;
+		err = flush_barrier(it); /* unconnected idle barrier? */
+		if (err)
+			break;
 
 		enable_signaling(&it->base);
 	}
diff --git a/drivers/gpu/drm/i915/selftests/i915_active.c b/drivers/gpu/drm/i915/selftests/i915_active.c
index 96513a7d4739..aca107cb6176 100644
--- a/drivers/gpu/drm/i915/selftests/i915_active.c
+++ b/drivers/gpu/drm/i915/selftests/i915_active.c
@@ -193,11 +193,57 @@ static int live_active_retire(void *arg)
 	return err;
 }
 
+static int live_active_barrier(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+	struct intel_engine_cs *engine;
+	struct live_active *active;
+	int err = 0;
+
+	/* Check that we get a callback when requests retire upon waiting */
+
+	active = __live_alloc(i915);
+	if (!active)
+		return -ENOMEM;
+
+	err = i915_active_acquire(&active->base);
+	if (err)
+		goto out;
+
+	for_each_uabi_engine(engine, i915) {
+		err = i915_active_acquire_preallocate_barrier(&active->base,
+							      engine);
+		if (err)
+			break;
+
+		i915_active_acquire_barrier(&active->base);
+	}
+
+	i915_active_release(&active->base);
+
+	if (err == 0)
+		err = i915_active_wait(&active->base);
+
+	if (err == 0 && !READ_ONCE(active->retired)) {
+		pr_err("i915_active not retired after flushing barriers!\n");
+		err = -EINVAL;
+	}
+
+out:
+	__live_put(active);
+
+	if (igt_flush_test(i915))
+		err = -EIO;
+
+	return err;
+}
+
 int i915_active_live_selftests(struct drm_i915_private *i915)
 {
 	static const struct i915_subtest tests[] = {
 		SUBTEST(live_active_wait),
 		SUBTEST(live_active_retire),
+		SUBTEST(live_active_barrier),
 	};
 
 	if (intel_gt_is_wedged(&i915->gt))
-- 
2.24.0.rc0

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

  parent reply	other threads:[~2019-10-22 22:39 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-22 22:38 [PATCH 01/12] drm/i915/gt: Expose engine properties via sysfs Chris Wilson
2019-10-22 22:38 ` [PATCH 02/12] drm/i915/gt: Expose engine->mmio_base " Chris Wilson
2019-10-22 22:38 ` [PATCH 03/12] drm/i915/gt: Expose timeslice duration to sysfs Chris Wilson
2019-10-22 22:38 ` [PATCH 04/12] drm/i915/gt: Expose reset stop timeout via sysfs Chris Wilson
2019-10-22 22:38 ` [PATCH 05/12] drm/i915/execlists: Force preemption Chris Wilson
2019-10-22 22:38 ` [PATCH 06/12] drm/i915/execlists: Cancel banned contexts on schedule-out Chris Wilson
2019-10-22 22:38 ` [PATCH 07/12] drm/i915/gem: Cancel contexts when hangchecking is disabled Chris Wilson
2019-10-22 22:38 ` [PATCH 08/12] drm/i915: Replace hangcheck by heartbeats Chris Wilson
2019-10-22 22:38 ` [PATCH 09/12] drm/i915/gem: Make context persistence optional Chris Wilson
2019-10-22 22:38 ` Chris Wilson [this message]
2019-10-22 22:38 ` [PATCH 11/12] drm/i915: Allow userspace to specify ringsize on construction Chris Wilson
2019-10-22 22:38 ` [PATCH 12/12] drm/i915/gem: Honour O_NONBLOCK before throttling execbuf submissions Chris Wilson

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=20191022223831.22677-10-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.