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 v2 11/12] drm/i915: Track the previous pinned context inside the request
Date: Tue, 19 Apr 2016 12:40:52 +0100	[thread overview]
Message-ID: <1461066053-30072-12-git-send-email-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <1461066053-30072-1-git-send-email-chris@chris-wilson.co.uk>

As the contexts are accessed by the hardware until the switch is completed
to a new context, the hardware may still be writing to the context object
after the breadcrumb is visible. We must not unpin/unbind/prune that
object whilst still active and so we keep the previous context pinned until
the following request. If we move this tracking onto the request, we can
simplify the code and enable freeing of the request without the
struct_mutex in subsequent patches.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_drv.h  | 11 +++++++++++
 drivers/gpu/drm/i915/intel_lrc.c | 17 ++++++++++-------
 2 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index c0a6fbff4678..ff1aa7363ab6 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2302,6 +2302,17 @@ struct drm_i915_gem_request {
 	struct intel_context *ctx;
 	struct intel_ringbuffer *ringbuf;
 
+	/**
+	 * Context related to the previous request.
+	 * As the contexts are accessed by the hardware until the switch is
+	 * completed to a new context, the hardware may still be writing
+	 * to the context object after the breadcrumb is visible. We must
+	 * not unpin/unbind/prune that object whilst still active and so
+	 * we keep the previous context pinned until the following (this)
+	 * request is retired.
+	 */
+	struct intel_context *previous_context;
+
 	/** Batch buffer related to this request if any (used for
 	    error state dump only) */
 	struct drm_i915_gem_object *batch_obj;
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 2910866611b2..68382e05f720 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -788,12 +788,14 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 	if (intel_engine_stopped(engine))
 		return 0;
 
-	if (engine->last_context != request->ctx) {
-		if (engine->last_context)
-			intel_lr_context_unpin(engine->last_context, engine);
-		intel_lr_context_pin(request->ctx, engine);
-		engine->last_context = request->ctx;
-	}
+	/* We keep the previous context alive until we retire the following
+	 * request. This ensures that any the context object is still pinned
+	 * for any residual writes the HW makes into it on the context switch
+	 * into the next object following the breadcrumb. Otherwise, we may
+	 * retire the context too early.
+	 */
+	request->previous_context = engine->last_context;
+	engine->last_context = request->ctx;
 
 	if (dev_priv->guc.execbuf_client)
 		i915_guc_submit(dev_priv->guc.execbuf_client, request);
@@ -1015,7 +1017,8 @@ void intel_execlists_retire_requests(struct intel_engine_cs *engine)
 	spin_unlock_bh(&engine->execlist_lock);
 
 	list_for_each_entry_safe(req, tmp, &retired_list, execlist_link) {
-		intel_lr_context_unpin(req->ctx, engine);
+		if (req->previous_context)
+			intel_lr_context_unpin(req->previous_context, engine);
 
 		list_del(&req->execlist_link);
 		i915_gem_request_unreference(req);
-- 
2.8.0.rc3

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

  parent reply	other threads:[~2016-04-19 11:41 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-19 11:40 Premature unpinning Chris Wilson
2016-04-19 11:40 ` [PATCH v2 01/12] drm/i915: Mark the current context as lost on suspend Chris Wilson
2016-04-19 14:12   ` [PATCH v2] " Chris Wilson
2016-04-19 14:20     ` Mika Kuoppala
2016-04-19 11:40 ` [PATCH v2 02/12] drm/i915: L3 cache remapping is part of context switching Chris Wilson
2016-04-19 11:40 ` [PATCH v2 03/12] drm/i915: Consolidate L3 remapping LRI Chris Wilson
2016-04-19 11:40 ` [PATCH v2 04/12] drm/i915: Remove early l3-remap Chris Wilson
2016-04-19 11:40 ` [PATCH v2 05/12] drm/i915: Rearrange switch_context to load the aliasing ppgtt on first use Chris Wilson
2016-04-19 11:40 ` [PATCH v2 06/12] drm/i915: Assign every HW context a unique ID Chris Wilson
2016-04-19 11:40 ` [PATCH v2 07/12] drm/i915: Replace the pinned context address with its " Chris Wilson
2016-04-19 11:40 ` [PATCH v2 08/12] drm/i915: Refactor execlists default context pinning Chris Wilson
2016-04-19 12:25   ` Tvrtko Ursulin
2016-04-20 14:07   ` Mika Kuoppala
2016-04-19 11:40 ` [PATCH v2 09/12] drm/i915: Move context initialisation to first-use Chris Wilson
2016-04-19 11:40 ` [PATCH v2 10/12] drm/i915: Move the magical deferred context allocation into the request Chris Wilson
2016-04-19 12:40   ` kbuild test robot
2016-04-19 11:40 ` Chris Wilson [this message]
2016-04-19 11:40 ` [PATCH v2 12/12] drm/i915: Move releasing of the GEM request from free to retire/cancel Chris Wilson
2016-04-19 12:59 ` Update to patches 11 and 12 Chris Wilson
2016-04-19 12:59   ` [PATCH 1/2] drm/i915: Move releasing of the GEM request from free to retire/cancel Chris Wilson
2016-04-20 13:55     ` Tvrtko Ursulin
2016-04-19 12:59   ` [PATCH 2/2] drm/i915: Track the previous pinned context inside the request Chris Wilson
2016-04-20 14:08     ` Tvrtko Ursulin
2016-04-20 14:18       ` Chris Wilson
2016-04-20 14:22       ` Chris Wilson
2016-04-20 16:34         ` Tvrtko Ursulin
2016-04-20 14:24       ` [RFC 1/2] drm/i915: Store LRC hardware id in " Tvrtko Ursulin
2016-04-20 14:24         ` [RFC 2/2] drm/i915: Stop tracking execlists retired requests Tvrtko Ursulin
2016-04-20 18:15         ` [RFC 1/2] drm/i915: Store LRC hardware id in the request Chris Wilson
2016-04-19 13:24 ` ✗ Fi.CI.BAT: failure for series starting with [v2,01/12] drm/i915: Mark the current context as lost on suspend Patchwork
2016-04-19 14:58 ` ✗ Fi.CI.BAT: failure for series starting with [v2] drm/i915: Mark the current context as lost on suspend (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=1461066053-30072-12-git-send-email-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.