drm/i915/gem: Remove disordered per-file request list for throttling

I915_GEM_THROTTLE dates back to the time before contexts where there was
just a single engine, and therefore a single timeline and request list
globally. That request list was in execution/retirement order, and so
walking it to find a particular aged request made sense and could be
split per file.

That is no more. We now have many timelines with a file, as many as the
user wants to construct (essentially per-engine, per-context). Each of
those run independently and so make the single list futile. Remove the
disordered list, and iterate over all the timelines to find a request to
wait on in each to satisfy the criteria that the CPU is no more than 20ms
ahead of its oldest request.

It should go without saying that the I915_GEM_THROTTLE ioctl is no
longer used as the primary means of throttling, so it makes sense to push
the complication into the ioctl where it only impacts upon its few
irregular users, rather than the execbuf/retire where everybody has to
pay the cost. Fortunately, the few users do not create vast amount of
contexts, so the loops over contexts/engines should be concise.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200728152010.30701-1-chris@chris-wilson.co.uk
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

2 files changed, 48 insertions, 32 deletions

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 02b1630f513e..90d77cdc6f66 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -1928,18 +1928,6 @@ err:
 	return err;
 }
 
-static void
-add_to_client(struct i915_request *rq, struct drm_file *file)
-{
-	struct drm_i915_file_private *file_priv = file->driver_priv;
-
-	rq->file_priv = file_priv;
-
-	spin_lock(&file_priv->mm.lock);
-	list_add_tail(&rq->client_link, &file_priv->mm.request_list);
-	spin_unlock(&file_priv->mm.lock);
-}
-
 static int eb_submit(struct i915_execbuffer *eb, struct i915_vma *batch)
 {
 	int err;
@@ -2772,7 +2760,6 @@ i915_gem_do_execbuffer(struct drm_device *dev,
 	trace_i915_request_queue(eb.request, eb.batch_flags);
 	err = eb_submit(&eb, batch);
 err_request:
-	add_to_client(eb.request, file);
 	i915_request_get(eb.request);
 	eb_request_add(&eb);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
index 540ef0551789..1929d6cf4150 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c
@@ -9,6 +9,7 @@
 #include <drm/drm_file.h>
 
 #include "i915_drv.h"
+#include "i915_gem_context.h"
 #include "i915_gem_ioctls.h"
 #include "i915_gem_object.h"
 
@@ -35,9 +36,10 @@ int
 i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file)
 {
+	const unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
 	struct drm_i915_file_private *file_priv = file->driver_priv;
-	unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES;
-	struct i915_request *request, *target = NULL;
+	struct i915_gem_context *ctx;
+	unsigned long idx;
 	long ret;
 
 	/* ABI: return -EIO if already wedged */
@@ -45,27 +47,54 @@ i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	spin_lock(&file_priv->mm.lock);
-	list_for_each_entry(request, &file_priv->mm.request_list, client_link) {
-		if (time_after_eq(request->emitted_jiffies, recent_enough))
-			break;
+	rcu_read_lock();
+	xa_for_each(&file_priv->context_xa, idx, ctx) {
+		struct i915_gem_engines_iter it;
+		struct intel_context *ce;
 
-		if (target && xchg(&target->file_priv, NULL))
-			list_del(&target->client_link);
+		if (!kref_get_unless_zero(&ctx->ref))
+			continue;
+		rcu_read_unlock();
 
-		target = request;
-	}
-	if (target)
-		i915_request_get(target);
-	spin_unlock(&file_priv->mm.lock);
+		for_each_gem_engine(ce,
+				    i915_gem_context_lock_engines(ctx),
+				    it) {
+			struct i915_request *rq, *target = NULL;
+
+			if (!ce->timeline)
+				continue;
+
+			mutex_lock(&ce->timeline->mutex);
+			list_for_each_entry_reverse(rq,
+						    &ce->timeline->requests,
+						    link) {
+				if (i915_request_completed(rq))
+					break;
 
-	if (!target)
-		return 0;
+				if (time_after(rq->emitted_jiffies,
+					       recent_enough))
+					continue;
 
-	ret = i915_request_wait(target,
-				I915_WAIT_INTERRUPTIBLE,
-				MAX_SCHEDULE_TIMEOUT);
-	i915_request_put(target);
+				target = i915_request_get(rq);
+				break;
+			}
+			mutex_unlock(&ce->timeline->mutex);
+			if (!target)
+				continue;
+
+			ret = i915_request_wait(target,
+						I915_WAIT_INTERRUPTIBLE,
+						MAX_SCHEDULE_TIMEOUT);
+			i915_request_put(target);
+			if (ret < 0)
+				break;
+		}
+		i915_gem_context_unlock_engines(ctx);
+		i915_gem_context_put(ctx);
+
+		rcu_read_lock();
+	}
+	rcu_read_unlock();
 
 	return ret < 0 ? ret : 0;
 }