On Thu, Jul 29, 2021 at 9:18 AM Daniel Vetter <daniel@xxxxxxxx> wrote:
>
> On Thu, Jul 29, 2021 at 5:19 PM Rob Clark <robdclark@xxxxxxxxx> wrote:
> >
> > On Thu, Jul 29, 2021 at 12:03 AM Daniel Vetter <daniel@xxxxxxxx> wrote:
> > >
> > > On Wed, Jul 28, 2021 at 10:58:51AM -0700, Rob Clark wrote:
> > > > On Wed, Jul 28, 2021 at 10:23 AM Christian König
> > > > <christian.koenig@xxxxxxx> wrote:
> > > > >
> > > > >
> > > > >
> > > > > Am 28.07.21 um 17:15 schrieb Rob Clark:
> > > > > > On Wed, Jul 28, 2021 at 4:37 AM Christian König
> > > > > > <ckoenig.leichtzumerken@xxxxxxxxx> wrote:
> > > > > >> Am 28.07.21 um 09:03 schrieb Christian König:
> > > > > >>> Am 27.07.21 um 16:25 schrieb Rob Clark:
> > > > > >>>> On Tue, Jul 27, 2021 at 12:11 AM Christian König
> > > > > >>>> <ckoenig.leichtzumerken@xxxxxxxxx> wrote:
> > > > > >>>>> Am 27.07.21 um 01:38 schrieb Rob Clark:
> > > > > >>>>>> From: Rob Clark <robdclark@xxxxxxxxxxxx>
> > > > > >>>>>>
> > > > > >>>>>> Add a way to hint to the fence signaler of an upcoming deadline,
> > > > > >>>>>> such as
> > > > > >>>>>> vblank, which the fence waiter would prefer not to miss. This is to
> > > > > >>>>>> aid
> > > > > >>>>>> the fence signaler in making power management decisions, like boosting
> > > > > >>>>>> frequency as the deadline approaches and awareness of missing
> > > > > >>>>>> deadlines
> > > > > >>>>>> so that can be factored in to the frequency scaling.
> > > > > >>>>>>
> > > > > >>>>>> Signed-off-by: Rob Clark <robdclark@xxxxxxxxxxxx>
> > > > > >>>>>> ---
> > > > > >>>>>> drivers/dma-buf/dma-fence.c | 39
> > > > > >>>>>> +++++++++++++++++++++++++++++++++++++
> > > > > >>>>>> include/linux/dma-fence.h | 17 ++++++++++++++++
> > > > > >>>>>> 2 files changed, 56 insertions(+)
> > > > > >>>>>>
> > > > > >>>>>> diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
> > > > > >>>>>> index ce0f5eff575d..2e0d25ab457e 100644
> > > > > >>>>>> --- a/drivers/dma-buf/dma-fence.c
> > > > > >>>>>> +++ b/drivers/dma-buf/dma-fence.c
> > > > > >>>>>> @@ -910,6 +910,45 @@ dma_fence_wait_any_timeout(struct dma_fence
> > > > > >>>>>> **fences, uint32_t count,
> > > > > >>>>>> }
> > > > > >>>>>> EXPORT_SYMBOL(dma_fence_wait_any_timeout);
> > > > > >>>>>>
> > > > > >>>>>> +
> > > > > >>>>>> +/**
> > > > > >>>>>> + * dma_fence_set_deadline - set desired fence-wait deadline
> > > > > >>>>>> + * @fence: the fence that is to be waited on
> > > > > >>>>>> + * @deadline: the time by which the waiter hopes for the fence to be
> > > > > >>>>>> + * signaled
> > > > > >>>>>> + *
> > > > > >>>>>> + * Inform the fence signaler of an upcoming deadline, such as
> > > > > >>>>>> vblank, by
> > > > > >>>>>> + * which point the waiter would prefer the fence to be signaled
> > > > > >>>>>> by. This
> > > > > >>>>>> + * is intended to give feedback to the fence signaler to aid in power
> > > > > >>>>>> + * management decisions, such as boosting GPU frequency if a periodic
> > > > > >>>>>> + * vblank deadline is approaching.
> > > > > >>>>>> + */
> > > > > >>>>>> +void dma_fence_set_deadline(struct dma_fence *fence, ktime_t
> > > > > >>>>>> deadline)
> > > > > >>>>>> +{
> > > > > >>>>>> + unsigned long flags;
> > > > > >>>>>> +
> > > > > >>>>>> + if (dma_fence_is_signaled(fence))
> > > > > >>>>>> + return;
> > > > > >>>>>> +
> > > > > >>>>>> + spin_lock_irqsave(fence->lock, flags);
> > > > > >>>>>> +
> > > > > >>>>>> + /* If we already have an earlier deadline, keep it: */
> > > > > >>>>>> + if (test_bit(DMA_FENCE_FLAG_HAS_DEADLINE_BIT, &fence->flags) &&
> > > > > >>>>>> + ktime_before(fence->deadline, deadline)) {
> > > > > >>>>>> + spin_unlock_irqrestore(fence->lock, flags);
> > > > > >>>>>> + return;
> > > > > >>>>>> + }
> > > > > >>>>>> +
> > > > > >>>>>> + fence->deadline = deadline;
> > > > > >>>>>> + set_bit(DMA_FENCE_FLAG_HAS_DEADLINE_BIT, &fence->flags);
> > > > > >>>>>> +
> > > > > >>>>>> + spin_unlock_irqrestore(fence->lock, flags);
> > > > > >>>>>> +
> > > > > >>>>>> + if (fence->ops->set_deadline)
> > > > > >>>>>> + fence->ops->set_deadline(fence, deadline);
> > > > > >>>>>> +}
> > > > > >>>>>> +EXPORT_SYMBOL(dma_fence_set_deadline);
> > > > > >>>>>> +
> > > > > >>>>>> /**
> > > > > >>>>>> * dma_fence_init - Initialize a custom fence.
> > > > > >>>>>> * @fence: the fence to initialize
> > > > > >>>>>> diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h
> > > > > >>>>>> index 6ffb4b2c6371..4e6cfe4e6fbc 100644
> > > > > >>>>>> --- a/include/linux/dma-fence.h
> > > > > >>>>>> +++ b/include/linux/dma-fence.h
> > > > > >>>>>> @@ -88,6 +88,7 @@ struct dma_fence {
> > > > > >>>>>> /* @timestamp replaced by @rcu on
> > > > > >>>>>> dma_fence_release() */
> > > > > >>>>>> struct rcu_head rcu;
> > > > > >>>>>> };
> > > > > >>>>>> + ktime_t deadline;
> > > > > >>>>> Mhm, adding the flag sounds ok to me but I'm a bit hesitating adding
> > > > > >>>>> the
> > > > > >>>>> deadline as extra field here.
> > > > > >>>>>
> > > > > >>>>> We tuned the dma_fence structure intentionally so that it is only 64
> > > > > >>>>> bytes.
> > > > > >>>> Hmm, then I guess you wouldn't be a fan of also adding an hrtimer?
> > > > > >>>>
> > > > > >>>> We could push the ktime_t (and timer) down into the derived fence
> > > > > >>>> class, but I think there is going to need to be some extra storage
> > > > > >>>> *somewhere*.. maybe the fence signaler could get away with just
> > > > > >>>> storing the nearest upcoming deadline per fence-context instead?
> > > > > >>> I would just push that into the driver instead.
> > > > > >>>
> > > > > >>> You most likely don't want the deadline per fence anyway in complex
> > > > > >>> scenarios, but rather per frame. And a frame is usually composed from
> > > > > >>> multiple fences.
> > > > > > Right, I ended up keeping track of the nearest deadline in patch 5/4
> > > > > > which added drm/msm support:
> > > > > >
> > > > > > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatchwork.freedesktop.org%2Fpatch%2F447138%2F&data=04%7C01%7Cchristian.koenig%40amd.com%7Cce6ace85263d448bbc9f08d951d9f06c%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637630819606427306%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=ameszAOlClaZNeUDlYr37ZdIytVXNgiEUKuctjXLqZ0%3D&reserved=0
> > > > > >
> > > > > > But if we do have the ktime_t in dma_fence in dma_fence, we can add
> > > > > > some checks and avoid calling back to the driver if a later deadline
> > > > > > is set on a fence that already has an earlier deadline. OTOH I
> > > > > > suppose I can push all that back to the driver to start, and we can
> > > > > > revisit once we have more drivers implementing deadline support.
> > > > >
> > > > > I still think that all of this is rather specific to your use case and
> > > > > have strong doubt that anybody else will implement that.
> > > >
> > > > i915 does already have a similar thing in it's hand-rolled atomic
> > > > commit path. So I think msm won't be the only one. It should be also
> > > > useful to the other mobile GPUs with a gpu vs kms driver split,
> > > > although looking at the other gpu devfreq implementations, I don't
> > > > think they've yet gotten to this point in the fine tuning..
> > >
> > > Yeah I have a dream that maybe i915 will use the atomic commit helpers, I
> > > originally wrote them with i915 in mind :-) even had patches!
> > >
> > > I also think we'll need this eventually in other areas, Android also has
> > > some hacks like this to make sure idle->first touch doesn't suck and
> > > similar things.
> >
> > input-boost is another thing I have on my roadmap.. part of the solution is:
> >
> > commit 9bc95570175a7fbca29d86d22c54bbf399f4ad5a
> > Author: Rob Clark <robdclark@xxxxxxxxxxxx>
> > AuthorDate: Mon Jul 26 07:46:50 2021 -0700
> > Commit: Rob Clark <robdclark@xxxxxxxxxxxx>
> > CommitDate: Tue Jul 27 17:54:36 2021 -0700
> >
> > drm/msm: Devfreq tuning
> >
> > which gives the freq a bit of a nudge if the GPU has been idle for
> > longer than a certain threshold.
> >
> > But the other part is that if the GPU has been idle for more than 66ms
> > (typical autosuspend delay for adreno) it will suspend. For modern
> > adreno's it takes ~2ms to "boot up" the GPU from suspend. Which is
> > something you want to take out of the submit/execbuf path if you are
> > trying to reduce input-to-pageflip latency.
> >
> > We have a downstream patch that boosts the CPUs on input events (with
> > a cooldown period to prevent spacebar-heater) and I have been thinking
> > of something along those lines to trigger resuming the GPU.. it is
> > straightforward enough for touch based devices, but gets more
> > complicated with keyboard input. In particular, some keys you want to
> > trigger boost on key-release. Ie. modifier keys (ctrl/shift/alt/etc..
> > the "search" key on chromebooks, etc) you want to boost on
> > key-release, not on key-press because unless you type *really* fast
> > you'll be in the cooldown period when the key-release event happens.
> > Unfortunately the kernel doesn't really know this "policy" sort of
> > information about which keys should boost on press vs release. So I
> > think the long-term/upstream solution is to do input-boost in
> > userspace.. sysfs already has all the knobs that a userspace
> > input-boost daemon would need to twiddle, so no real need for this to
> > be in the kernel. I guess the only drawback is the sysfs knobs are a
> > bit less standardized on the "desktop GPUs" which don't use devfreq.
>
> I think we could do a standard interface for this, either on the drm
> owner/master or somewhere in sysfs. Essentially "I expect to use the
> gpu for the very next frame, get it going". Across all hw there's a
> lot of things we can do. I think abuse is pretty easy to prevent with
> a cooldown or similar.
The userspace input-boost ends up needing to be either part of the
compositor, or a privileged process, in order to sniff input events..
so I don't think the kernel needs to try to prevent abuse here (but
the userspace part defn wants a cooldown period)
BR,
-R
> -Daniel
>
> >
> > BR,
> > -R
> >
> > > -Daniel
> > >
> > > >
> > > > BR,
> > > > -R
> > > >
> > > > > >> Thinking more about it we could probably kill the spinlock pointer and
> > > > > >> make the flags 32bit if we absolutely need that here.
> > > > > > If we had a 'struct dma_fence_context' we could push the spinlock, ops
> > > > > > pointer, and u64 context into that and replace with a single
> > > > > > dma_fence_context ptr, fwiw
> > > > >
> > > > > That won't work. We have a lot of use cases where you can't allocate
> > > > > memory, but must allocate a context.
> > > > >
> > > > > Christian.
> > > > >
> > > > > >
> > > > > > BR,
> > > > > > -R
> > > > > >
> > > > > >> But I still don't see the need for that, especially since most drivers
> > > > > >> probably won't implement it.
> > > > > >>
> > > > > >> Regards,
> > > > > >> Christian.
> > > > > >>
> > > > > >>> Regards,
> > > > > >>> Christian.
> > > > > >>>
> > > > > >>>> BR,
> > > > > >>>> -R
> > > > > >>>>
> > > > > >>>>> Regards,
> > > > > >>>>> Christian.
> > > > > >>>>>
> > > > > >>>>>> u64 context;
> > > > > >>>>>> u64 seqno;
> > > > > >>>>>> unsigned long flags;
> > > > > >>>>>> @@ -99,6 +100,7 @@ enum dma_fence_flag_bits {
> > > > > >>>>>> DMA_FENCE_FLAG_SIGNALED_BIT,
> > > > > >>>>>> DMA_FENCE_FLAG_TIMESTAMP_BIT,
> > > > > >>>>>> DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT,
> > > > > >>>>>> + DMA_FENCE_FLAG_HAS_DEADLINE_BIT,
> > > > > >>>>>> DMA_FENCE_FLAG_USER_BITS, /* must always be last member */
> > > > > >>>>>> };
> > > > > >>>>>>
> > > > > >>>>>> @@ -261,6 +263,19 @@ struct dma_fence_ops {
> > > > > >>>>>> */
> > > > > >>>>>> void (*timeline_value_str)(struct dma_fence *fence,
> > > > > >>>>>> char *str, int size);
> > > > > >>>>>> +
> > > > > >>>>>> + /**
> > > > > >>>>>> + * @set_deadline:
> > > > > >>>>>> + *
> > > > > >>>>>> + * Callback to allow a fence waiter to inform the fence
> > > > > >>>>>> signaler of an
> > > > > >>>>>> + * upcoming deadline, such as vblank, by which point the
> > > > > >>>>>> waiter would
> > > > > >>>>>> + * prefer the fence to be signaled by. This is intended to
> > > > > >>>>>> give feedback
> > > > > >>>>>> + * to the fence signaler to aid in power management
> > > > > >>>>>> decisions, such as
> > > > > >>>>>> + * boosting GPU frequency.
> > > > > >>>>>> + *
> > > > > >>>>>> + * This callback is optional.
> > > > > >>>>>> + */
> > > > > >>>>>> + void (*set_deadline)(struct dma_fence *fence, ktime_t deadline);
> > > > > >>>>>> };
> > > > > >>>>>>
> > > > > >>>>>> void dma_fence_init(struct dma_fence *fence, const struct
> > > > > >>>>>> dma_fence_ops *ops,
> > > > > >>>>>> @@ -586,6 +601,8 @@ static inline signed long dma_fence_wait(struct
> > > > > >>>>>> dma_fence *fence, bool intr)
> > > > > >>>>>> return ret < 0 ? ret : 0;
> > > > > >>>>>> }
> > > > > >>>>>>
> > > > > >>>>>> +void dma_fence_set_deadline(struct dma_fence *fence, ktime_t
> > > > > >>>>>> deadline);
> > > > > >>>>>> +
> > > > > >>>>>> struct dma_fence *dma_fence_get_stub(void);
> > > > > >>>>>> struct dma_fence *dma_fence_allocate_private_stub(void);
> > > > > >>>>>> u64 dma_fence_context_alloc(unsigned num);
> > > > >
> > >
> > > --
> > > Daniel Vetter
> > > Software Engineer, Intel Corporation
> > > http://blog.ffwll.ch
>
>
>
> --
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch
