[Intel-gfx] [PATCH 2/2] drm/i915: Track the previous pinned context inside the request

[Chris Wilson]

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 treat execlists/GuC dispatch identically.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/i915_drv.h  | 11 +++++++++++
 drivers/gpu/drm/i915/i915_gem.c  |  8 ++++----
 drivers/gpu/drm/i915/intel_lrc.c | 17 ++++++++---------
 3 files changed, 23 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index c59b2670cc36..be98e9643072 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/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 9b4854a17264..537aacfda3eb 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1413,13 +1413,13 @@ static void i915_gem_request_retire(struct drm_i915_gem_request *request)
 	list_del_init(&request->list);
 	i915_gem_request_remove_from_client(request);
 
-	if (request->ctx) {
+	if (request->previous_context) {
 		if (i915.enable_execlists)
-			intel_lr_context_unpin(request->ctx, request->engine);
-
-		i915_gem_context_unreference(request->ctx);
+			intel_lr_context_unpin(request->previous_context,
+					       request->engine);
 	}
 
+	i915_gem_context_unreference(request->ctx);
 	i915_gem_request_unreference(request);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index ee4e9bb80042..06e013293ec6 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -590,7 +590,6 @@ static void execlists_context_queue(struct drm_i915_gem_request *request)
 	struct drm_i915_gem_request *cursor;
 	int num_elements = 0;
 
-	intel_lr_context_pin(request->ctx, request->engine);
 	i915_gem_request_reference(request);
 
 	spin_lock_bh(&engine->execlist_lock);
@@ -788,12 +787,14 @@ 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);
@@ -1015,8 +1016,6 @@ 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);
-
 		list_del(&req->execlist_link);
 		i915_gem_request_unreference(req);
 	}
-- 
2.8.0.rc3