commit 9bc95570175a7fbca29d86d22c54bbf399f4ad5a
author Rob Clark <robdclark@chromium.org> Mon Jul 26 07:46:50 2021 -0700
committer Rob Clark <robdclark@chromium.org> Tue Jul 27 17:54:36 2021 -0700
tree 4bcffcdc786e9334c787be5b7abe8e81507328d6
parent 552fce98b06f702b041432ad650db3e50c380ce3

drm/msm: Devfreq tuning

This adds a few things to try and make frequency scaling better match
the workload:

1) Longer polling interval to avoid whip-lashing between too-high and
   too-low frequencies in certain workloads, like mobile games which
   throttle themselves to 30fps.

   Previously our polling interval was short enough to let things
   ramp down to minimum freq in the "off" frame, but long enough to
   not react quickly enough when rendering started on the next frame,
   leading to uneven frame times.  (Ie. rather than a consistent 33ms
   it would alternate between 16/33/48ms.)

2) Awareness of when the GPU is active vs idle.  Since we know when
   the GPU is active vs idle, we can clamp the frequency down to the
   minimum while it is idle.  (If it is idle for long enough, then
   the autosuspend delay will eventually kick in and power down the
   GPU.)

   Since devfreq has no knowledge of powered-but-idle, this takes a
   small bit of trickery to maintain a "fake" frequency while idle.
   This, combined with the longer polling period allows devfreq to
   arrive at a reasonable "active" frequency, while still clamping
   to minimum freq when idle to reduce power draw.

3) Boost.  Because simple_ondemand needs to see a certain threshold
   of busyness to ramp up, we could end up needing multiple polling
   cycles before it reacts appropriately on interactive workloads
   (ex. scrolling a web page after reading for some time), on top
   of the already lengthened polling interval, when we see a idle
   to active transition after a period of idle time we boost the
   frequency that we return to.

Signed-off-by: Rob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20210726144653.2180096-4-robdclark@gmail.com
Signed-off-by: Rob Clark <robdclark@chromium.org>

drivers/gpu/drm/msm/msm_gpu.c

diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index fedbd78..c4ed869 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -680,6 +680,14 @@ static void retire_submit(struct msm_gpu *gpu, struct msm_ringbuffer *ring,
 	list_del(&submit->node);
 	spin_unlock_irqrestore(&ring->submit_lock, flags);
 
+	/* Update devfreq on transition from active->idle: */
+	mutex_lock(&gpu->active_lock);
+	gpu->active_submits--;
+	WARN_ON(gpu->active_submits < 0);
+	if (!gpu->active_submits)
+		msm_devfreq_idle(gpu);
+	mutex_unlock(&gpu->active_lock);
+
 	msm_gem_submit_put(submit);
 }
 
@@ -781,6 +789,13 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	list_add_tail(&submit->node, &ring->submits);
 	spin_unlock_irqrestore(&ring->submit_lock, flags);
 
+	/* Update devfreq on transition from idle->active: */
+	mutex_lock(&gpu->active_lock);
+	if (!gpu->active_submits)
+		msm_devfreq_active(gpu);
+	gpu->active_submits++;
+	mutex_unlock(&gpu->active_lock);
+
 	gpu->funcs->submit(gpu, submit);
 	priv->lastctx = submit->queue->ctx;
 
@@ -866,6 +881,7 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	sched_set_fifo_low(gpu->worker->task);
 
 	INIT_LIST_HEAD(&gpu->active_list);
+	mutex_init(&gpu->active_lock);
 	kthread_init_work(&gpu->retire_work, retire_worker);
 	kthread_init_work(&gpu->recover_work, recover_worker);
 	kthread_init_work(&gpu->fault_work, fault_worker);