[PATCH 6/6] drm/msm: add OCMEM driver

[Rob Clark]

On Thu, Oct 1, 2015 at 8:59 PM, Stephen Boyd <sboyd at codeaurora.org> wrote:
> On 10/01, Rob Clark wrote:
>> On Thu, Oct 1, 2015 at 3:25 PM, Rob Clark <robdclark at gmail.com> wrote:
>> > On Thu, Oct 1, 2015 at 2:00 PM, Stephen Boyd <sboyd at codeaurora.org> wrote:
>> >>
>> >> The simplest thing to do is to split the memory between GPU and
>> >> vidc statically. The other use cases with preemption and eviction
>> >> and DMA add a lot of complexity that we can explore at a later
>> >> time if need be.
>> >
>> > true, as long as one of the clients is the static gpu client, I guess
>> > we could reasonably easily support up to two clients reasonably
>> > easily...
>>
>> btw, random thought..  drm_mm is a utility in drm that serves a
>> similar function to genalloc for graphics drivers to manage their
>> address space(s) (used for everything from mmap virtual address space
>> of buffers allocated against device to managing vram and gart
>> allocations, etc... when vram carveout is used w/ drm/msm (due to no
>> iommu) I use it to manage allocations from the carveout).  It has some
>> potentially convenient twists, like supporting allocation from top of
>> the "address space" instead of bottom.  I'm thinking in particular of
>> allocating "narrow mode" allocations from top and "wide mode" from
>> bottom since wide vs narrow can only be set per region and not per
>> macro within the region.  (It can also search by first-fit or
>> best-fit.. although not sure if that is useful to us, since OCMEM size
>> is relatively constrained.)
>>
>> Not that I really want to keep ocmem allocator in drm.. I'd really
>> rather it be someone else's headache once it gets to implementing the
>> crazy stuff for all the random use-cases of other OCMEM users, since
>> gpu's use of OCMEM is rather simple/static..
>>
>> The way downstream driver does this is with a bunch of extra
>> bookkeeping on top of genalloc so it can do a dummy allocation to
>> force "from the top" allocation (and then immediately freeing the
>> dummy allocation)..  Maybe it just makes sense to teach genalloc how
>> to do from-top vs from-bottom allocations?  Not really sure..
>>
>
> It looks like video and GPU both use wide mode, if I'm reading
> the code correctly. So it seems that we don't need to do anything
> special for allocations, just hand the GPU and vidc a chunk of
> the memory in wide mode and then let the GPU and vidc drivers
> manage buffers within their chunk however they see fit.

right, I think it is just audio that uses narrow-mode..  and at that
point I think we need to use rpm for configuring things, since I guess
audio wants to do some of that from dsp for low-power audio playback
type use-cases??  Maybe we can ignore this for a while.. although when
we get to the point of fixing it, spiffing out genalloc sounds more
interesting than layering a bunch of bookkeeping on top of genalloc..

> One pitfall is going to be power management. ocmem is closely
> tied to the GPU power domains, so when video is using its chunk
> of memory we're going to need to keep ocmem powered up even if
> the GPU is off. I suppose for now we can just leave it always
> powered on once the driver probes, but if we ever want to turn it
> off we're going to need some tracking mechanism to make sure we
> don't turn it off when buffers are in use.

part of the problem that right now drm/msm is going to hold onto it's
allocation even when it clks down gpu.  I suppose not terribly
difficult to fix (although one reason why I didn't bother making
things terribly dynamic w/ clk mgmt or macro mode state in current
ocmem patch).. at this point I'm still more at the 'just try to make
it work in the first place' stage, rather than worrying too much about
optimizing for power..

BR,
-R