Attachment 115222 Details for Bug 90057 – drm/i915: Workaround to avoid lite restore with HEAD==TAIL

[patch] drm/i915: Workaround to avoid lite restore with HEAD==TAIL

0001-drm-i915-Workaround-to-avoid-lite-restore-with-HEAD-.patch (text/plain), 3.90 KB, created by Mika Kuoppala on 2015-04-20 12:22:20 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Mika Kuoppala

Created: 2015-04-20 12:22:20 UTC

Size: 3.90 KB

patch

obsolete

>From cddd726447a36857ef6c5a0068a41e2473c0ac9a Mon Sep 17 00:00:00 2001
>From: Michel Thierry <michel.thierry@intel.com>
>Date: Wed, 15 Apr 2015 18:11:33 +0100
>Subject: [PATCH] drm/i915: Workaround to avoid lite restore with HEAD==TAIL
>
>WaIdleLiteRestore is an execlists-only workaround, and requires the driver
>to ensure that any context always has HEAD!=TAIL when attempting lite
>restore.
>
>Add two extra MI_NOOP instructions at the end of each request, but keep
>the requests tail pointing before the MI_NOOPs. We may not need to
>executed them, and this is why request->tail is sampled before adding
>these extra instructions.
>
>If we submit a context to the ELSP which has previously been submitted,
>move the tail pointer past the MI_NOOPs. This ensures HEAD!=TAIL.
>
>v2: Move overallocation to gen8_emit_request, and added note about
>sampling request->tail in commit message (Chris).
>
>v3: Remove redundant request->tail assignment in __i915_add_request, in
>lrc mode this is already set in execlists_context_queue.
>Do not add wa implementation details inside gem (Chris).
>
>v4: Apply the wa whenever the req has been resubmitted and update
>comment (Chris).
>
>Cc: Chris Wilson <chris@chris-wilson.co.
>Signed-off-by: Thomas Daniel <thomas.daniel@intel.com>
>Signed-off-by: Michel Thierry <michel.thierry@intel.com>
>Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
>---
> drivers/gpu/drm/i915/i915_gem.c  |  3 ++-
> drivers/gpu/drm/i915/intel_lrc.c | 35 ++++++++++++++++++++++++++++++++++-
> 2 files changed, 36 insertions(+), 2 deletions(-)
>
>diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
>index f7b8766..badccf0 100644
>--- a/drivers/gpu/drm/i915/i915_gem.c
>+++ b/drivers/gpu/drm/i915/i915_gem.c
>@@ -2400,10 +2400,11 @@ int __i915_add_request(struct intel_engine_cs *ring,
> 		ret = ring->add_request(ring);
> 		if (ret)
> 			return ret;
>+
>+		request->tail = intel_ring_get_tail(ringbuf);
> 	}
> 
> 	request->head = request_start;
>-	request->tail = intel_ring_get_tail(ringbuf);
> 
> 	/* Whilst this request exists, batch_obj will be on the
> 	 * active_list, and so will hold the active reference. Only when this
>diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
>index a798d75..732fd63 100644
>--- a/drivers/gpu/drm/i915/intel_lrc.c
>+++ b/drivers/gpu/drm/i915/intel_lrc.c
>@@ -416,6 +416,26 @@ static void execlists_context_unqueue(struct intel_engine_cs *ring)
> 		}
> 	}
> 
>+	if (IS_GEN8(ring->dev) || IS_GEN9(ring->dev)) {
>+		/*
>+		 * WaIdleLiteRestore: make sure we never cause a lite
>+		 * restore with HEAD==TAIL
>+		 */
>+		if (req0 && req0->elsp_submitted) {
>+			/*
>+			 * Apply the wa NOOPS to prevent ring:HEAD == req:TAIL
>+			 * as we resubmit the request. See gen8_emit_request()
>+			 * for where we prepare the padding after the end of the
>+			 * request.
>+			 */
>+			struct intel_ringbuffer *ringbuf;
>+
>+			ringbuf = req0->ctx->engine[ring->id].ringbuf;
>+			req0->tail += 8;
>+			req0->tail &= ringbuf->size - 1;
>+		}
>+	}
>+
> 	WARN_ON(req1 && req1->elsp_submitted);
> 
> 	execlists_submit_contexts(ring, req0->ctx, req0->tail,
>@@ -1277,7 +1297,12 @@ static int gen8_emit_request(struct intel_ringbuffer *ringbuf,
> 	u32 cmd;
> 	int ret;
> 
>-	ret = intel_logical_ring_begin(ringbuf, request->ctx, 6);
>+	/*
>+	 * Reserve space for 2 NOOPs at the end of each request to be
>+	 * used as a workaround for not being allowed to do lite
>+	 * restore with HEAD==TAIL (WaIdleLiteRestore).
>+	 */
>+	ret = intel_logical_ring_begin(ringbuf, request->ctx, 8);
> 	if (ret)
> 		return ret;
> 
>@@ -1295,6 +1320,14 @@ static int gen8_emit_request(struct intel_ringbuffer *ringbuf,
> 	intel_logical_ring_emit(ringbuf, MI_NOOP);
> 	intel_logical_ring_advance_and_submit(ringbuf, request->ctx, request);
> 
>+	/*
>+	 * Here we add two extra NOOPs as padding to avoid
>+	 * lite restore of a context with HEAD==TAIL.
>+	 */
>+	intel_logical_ring_emit(ringbuf, MI_NOOP);
>+	intel_logical_ring_emit(ringbuf, MI_NOOP);
>+	intel_logical_ring_advance(ringbuf);
>+
> 	return 0;
> }
> 
>-- 
>1.9.1
>

Actions: View | Splinter Review

Attachments on bug 90057: 115137 | 115212 | 115222 | 115288

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.