Commit 27a5dcfe authored by Chris Wilson's avatar Chris Wilson Committed by Joonas Lahtinen
Browse files

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: default avatarChris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: default avatarTvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200728152010.30701-1-chris@chris-wilson.co.uk


Signed-off-by: default avatarRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: default avatarJoonas Lahtinen <joonas.lahtinen@linux.intel.com>
parent 3adee4ac

drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c

+0 −13
Original line number Diff line number Diff line
@@ -1928,18 +1928,6 @@ static int eb_parse(struct i915_execbuffer *eb) 
	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);

drivers/gpu/drm/i915/gem/i915_gem_throttle.c

+48 −19
Original line number Diff line number Diff line
@@ -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))

	rcu_read_lock();

	xa_for_each(&file_priv->context_xa, idx, ctx) {

		struct i915_gem_engines_iter it;

		struct intel_context *ce;



		if (!kref_get_unless_zero(&ctx->ref))

			continue;

		rcu_read_unlock();



		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 && xchg(&target->file_priv, NULL))

			list_del(&target->client_link);

				if (time_after(rq->emitted_jiffies,

					       recent_enough))

					continue;



		target = request;

				target = i915_request_get(rq);

				break;

			}

	if (target)

		i915_request_get(target);

	spin_unlock(&file_priv->mm.lock);



			mutex_unlock(&ce->timeline->mutex);

			if (!target)

		return 0;

				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;

}

drivers/gpu/drm/i915/gt/selftest_lrc.c

+2 −3
Original line number Diff line number Diff line
@@ -2729,7 +2729,7 @@ static int create_gang(struct intel_engine_cs *engine, 
	i915_gem_object_put(obj);

	intel_context_put(ce);



	rq->client_link.next = &(*prev)->client_link;

	rq->mock.link.next = &(*prev)->mock.link;

	*prev = rq;

	return 0;
@@ -2970,8 +2970,7 @@ static int live_preempt_gang(void *arg) 
		}



		while (rq) { /* wait for each rq from highest to lowest prio */

			struct i915_request *n =

				list_next_entry(rq, client_link);

			struct i915_request *n = list_next_entry(rq, mock.link);



			if (err == 0 && i915_request_wait(rq, 0, HZ / 5) < 0) {

				struct drm_printer p =

drivers/gpu/drm/i915/i915_drv.c

+0 −1
Original line number Diff line number Diff line
@@ -1119,7 +1119,6 @@ static void i915_driver_postclose(struct drm_device *dev, struct drm_file *file) 
	struct drm_i915_file_private *file_priv = file->driver_priv;



	i915_gem_context_close(file);

	i915_gem_release(dev, file);



	kfree_rcu(file_priv, rcu);

drivers/gpu/drm/i915/i915_drv.h

+0 −6
Original line number Diff line number Diff line
@@ -203,11 +203,6 @@ struct drm_i915_file_private { 
		struct rcu_head rcu;

	};



	struct {

		spinlock_t lock;

		struct list_head request_list;

	} mm;



	struct xarray context_xa;

	struct xarray vm_xa;
@@ -1867,7 +1862,6 @@ void i915_gem_suspend_late(struct drm_i915_private *dev_priv); 
void i915_gem_resume(struct drm_i915_private *dev_priv);



int i915_gem_open(struct drm_i915_private *i915, struct drm_file *file);

void i915_gem_release(struct drm_device *dev, struct drm_file *file);



int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,

				    enum i915_cache_level cache_level);