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: [Intel-gfx] [PATCH 04/20] drm/i915/perf: Wait for lrc_reconfigure on disable
Date: Thu, 27 Feb 2020 08:57:07 +0000	[thread overview]
Message-ID: <20200227085723.1961649-4-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <20200227085723.1961649-1-chris@chris-wilson.co.uk>

Wait for the last request (and so waits for all context updates) when
disabling OA. This prevents a rather bizarre error seen on Skylake
where the context is subsequently corrupted. Let's play safe and assume
it may impact all.

Reported-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 drivers/gpu/drm/i915/i915_perf.c | 23 +++++++++++++++++------
 1 file changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 2334c45f1d08..20c68b5dea63 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -2191,7 +2191,8 @@ static int gen8_modify_context(struct intel_context *ce,
 }
 
 static int gen8_modify_self(struct intel_context *ce,
-			    const struct flex *flex, unsigned int count)
+			    const struct flex *flex, unsigned int count,
+			    bool sync)
 {
 	struct i915_request *rq;
 	int err;
@@ -2204,7 +2205,12 @@ static int gen8_modify_self(struct intel_context *ce,
 
 	err = gen8_load_flex(rq, ce, flex, count);
 
+	i915_request_get(rq);
 	i915_request_add(rq);
+	if (sync && i915_request_wait(rq, 0, HZ) < 0)
+		err = -ETIME;
+	i915_request_put(rq);
+
 	return err;
 }
 
@@ -2281,7 +2287,7 @@ static int gen12_configure_oar_context(struct i915_perf_stream *stream, bool ena
 		return err;
 
 	/* Apply regs_lri using LRI with pinned context */
-	return gen8_modify_self(ce, regs_lri, ARRAY_SIZE(regs_lri));
+	return gen8_modify_self(ce, regs_lri, ARRAY_SIZE(regs_lri), false);
 }
 
 /*
@@ -2311,7 +2317,8 @@ static int gen12_configure_oar_context(struct i915_perf_stream *stream, bool ena
  */
 static int oa_configure_all_contexts(struct i915_perf_stream *stream,
 				     struct flex *regs,
-				     size_t num_regs)
+				     size_t num_regs,
+				     bool enable)
 {
 	struct drm_i915_private *i915 = stream->perf->i915;
 	struct intel_engine_cs *engine;
@@ -2368,7 +2375,7 @@ static int oa_configure_all_contexts(struct i915_perf_stream *stream,
 
 		regs[0].value = intel_sseu_make_rpcs(i915, &ce->sseu);
 
-		err = gen8_modify_self(ce, regs, num_regs);
+		err = gen8_modify_self(ce, regs, num_regs, !enable);
 		if (err)
 			return err;
 	}
@@ -2386,7 +2393,9 @@ static int gen12_configure_all_contexts(struct i915_perf_stream *stream,
 		},
 	};
 
-	return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs));
+	return oa_configure_all_contexts(stream,
+					 regs, ARRAY_SIZE(regs),
+					 oa_config);
 }
 
 static int lrc_configure_all_contexts(struct i915_perf_stream *stream,
@@ -2423,7 +2432,9 @@ static int lrc_configure_all_contexts(struct i915_perf_stream *stream,
 	for (i = 2; i < ARRAY_SIZE(regs); i++)
 		regs[i].value = oa_config_flex_reg(oa_config, regs[i].reg);
 
-	return oa_configure_all_contexts(stream, regs, ARRAY_SIZE(regs));
+	return oa_configure_all_contexts(stream,
+					 regs, ARRAY_SIZE(regs),
+					 oa_config);
 }
 
 static int gen8_enable_metric_set(struct i915_perf_stream *stream)
-- 
2.25.1

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

  parent reply	other threads:[~2020-02-27  8:57 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-27  8:57 [Intel-gfx] [PATCH 01/20] drm/i915: Skip barriers inside waits Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 02/20] drm/i915/perf: Mark up the racy use of perf->exclusive_stream Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 03/20] drm/i915/perf: Manually acquire engine-wakeref around use of kernel_context Chris Wilson
2020-02-28 11:53   ` Mika Kuoppala
2020-02-28 11:56     ` Chris Wilson
2020-02-28 12:18       ` Mika Kuoppala
2020-02-27  8:57 ` Chris Wilson [this message]
2020-02-27 11:17   ` [Intel-gfx] [PATCH] drm/i915/perf: Wait for lrc_reconfigure on disable Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 05/20] drm/i915/gem: Consolidate ctx->engines[] release Chris Wilson
2020-02-27  9:51   ` [Intel-gfx] [PATCH] " Chris Wilson
2020-02-27 11:01   ` Chris Wilson
2020-02-28 12:08     ` Tvrtko Ursulin
2020-02-28 12:19       ` Chris Wilson
2020-03-02 14:20         ` Tvrtko Ursulin
2020-02-27  8:57 ` [Intel-gfx] [PATCH 06/20] drm/i915/gt: Prevent allocation on a banned context Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 07/20] drm/i915/gem: Check that the context wasn't closed during setup Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 08/20] drm/i915/selftests: Disable heartbeat around manual pulse tests Chris Wilson
2020-02-27 22:51   ` Andi Shyti
2020-02-27  8:57 ` [Intel-gfx] [PATCH 09/20] drm/i915/gt: Reset queue_priority_hint after wedging Chris Wilson
2020-02-28 12:10   ` Tvrtko Ursulin
2020-02-28 12:31     ` Chris Wilson
2020-02-28 12:59       ` Tvrtko Ursulin
2020-02-28 13:10         ` Chris Wilson
2020-02-28 13:20           ` Tvrtko Ursulin
2020-02-28 13:34             ` Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 10/20] drm/i915/gt: Pull marking vm as closed underneath the vm->mutex Chris Wilson
2020-02-28 12:12   ` Tvrtko Ursulin
2020-02-27  8:57 ` [Intel-gfx] [PATCH 11/20] drm/i915: Protect i915_request_await_start from early waits Chris Wilson
2020-02-28 12:41   ` Tvrtko Ursulin
2020-02-27  8:57 ` [Intel-gfx] [PATCH 12/20] drm/i915/selftests: Verify LRC isolation Chris Wilson
2020-02-28 11:30   ` Mika Kuoppala
2020-02-28 11:52     ` Chris Wilson
2020-02-28 12:13       ` Mika Kuoppala
2020-02-27  8:57 ` [Intel-gfx] [PATCH 13/20] drm/i915/selftests: Check recovery from corrupted LRC Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 14/20] drm/i915/selftests: Wait for the kernel context switch Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 15/20] drm/i915/selftests: Be a little more lenient for reset workers Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 16/20] drm/i915/selftests: Add request throughput measurement to perf Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 17/20] drm/i915/gt: Declare when we enabled timeslicing Chris Wilson
2020-02-28 12:45   ` Tvrtko Ursulin
2020-02-28 13:14     ` Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 18/20] drm/i915/gt: Yield the timeslice if caught waiting on a user semaphore Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 19/20] drm/i915/execlists: Check the sentinel is alone in the ELSP Chris Wilson
2020-02-27  8:57 ` [Intel-gfx] [PATCH 20/20] drm/i915/execlists: Reduce preempt-to-busy roundtrip delay Chris Wilson
2020-02-27  9:14 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [01/20] drm/i915: Skip barriers inside waits Patchwork
2020-02-27  9:45 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2020-02-27 15:06 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [01/20] drm/i915: Skip barriers inside waits (rev4) Patchwork
2020-02-27 15:37 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2020-02-27 22:38 ` [Intel-gfx] [PATCH 01/20] drm/i915: Skip barriers inside waits Andi Shyti
2020-02-28 11:53 ` Tvrtko Ursulin
2020-02-28 12:08   ` Chris Wilson
2020-02-28 16:33 ` [Intel-gfx] ✗ Fi.CI.IGT: failure for series starting with [01/20] drm/i915: Skip barriers inside waits (rev4) 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=20200227085723.1961649-4-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.