Re: [PATCH v2 00/25] AMDKFD kernel driver

[Oded Gabbay <oded.gabbay@amd.com>]

On 22/07/14 10:40, Daniel Vetter wrote:
> On Tue, Jul 22, 2014 at 09:28:51AM +0200, Daniel Vetter wrote:
>> On Mon, Jul 21, 2014 at 03:03:07PM -0400, Jerome Glisse wrote:
>>> On Mon, Jul 21, 2014 at 09:41:29PM +0300, Oded Gabbay wrote:
>>>> On 21/07/14 21:22, Daniel Vetter wrote:
>>>>> On Mon, Jul 21, 2014 at 7:28 PM, Oded Gabbay <oded.gabbay@amd.com> wrote:
>>>>>>> I'm not sure whether we can do the same trick with the hw scheduler. But
>>>>>>> then unpinning hw contexts will drain the pipeline anyway, so I guess we
>>>>>>> can just stop feeding the hw scheduler until it runs dry. And then unpin
>>>>>>> and evict.
>>>>>> So, I'm afraid but we can't do this for AMD Kaveri because:
>>>>>
>>>>> Well as long as you can drain the hw scheduler queue (and you can do
>>>>> that, worst case you have to unmap all the doorbells and other stuff
>>>>> to intercept further submission from userspace) you can evict stuff.
>>>>
>>>> I can't drain the hw scheduler queue, as I can't do mid-wave preemption.
>>>> Moreover, if I use the dequeue request register to preempt a queue
>>>> during a dispatch it may be that some waves (wave groups actually) of
>>>> the dispatch have not yet been created, and when I reactivate the mqd,
>>>> they should be created but are not. However, this works fine if you use
>>>> the HIQ. the CP ucode correctly saves and restores the state of an
>>>> outstanding dispatch. I don't think we have access to the state from
>>>> software at all, so it's not a bug, it is "as designed".
>>>>
>>>
>>> I think here Daniel is suggesting to unmapp the doorbell page, and track
>>> each write made by userspace to it and while unmapped wait for the gpu to
>>> drain or use some kind of fence on a special queue. Once GPU is drain we
>>> can move pinned buffer, then remap the doorbell and update it to the last
>>> value written by userspace which will resume execution to the next job.
>>
>> Exactly, just prevent userspace from submitting more. And if you have
>> misbehaving userspace that submits too much, reset the gpu and tell it
>> that you're sorry but won't schedule any more work.
>>
>> We have this already in i915 (since like all other gpus we're not
>> preempting right now) and it works. There's some code floating around to
>> even restrict the reset to _just_ the offending submission context, with
>> nothing else getting corrupted.
>>
>> You can do all this with the doorbells and unmapping them, but it's a
>> pain. Much easier if you have a real ioctl, and I haven't seen anyone with
>> perf data indicating that an ioctl would be too much overhead on linux.
>> Neither in this thread nor internally here at intel.
>
> Aside: Another reason why the ioctl is better than the doorbell is
> integration with other drivers. Yeah I know this is about compute, but
> sooner or later someone will want to e.g. post-proc video frames between
> the v4l capture device and the gpu mpeg encoder. Or something else fancy.
>
> Then you want to be able to somehow integrate into a cross-driver fence
> framework like android syncpts, and you can't do that without an ioctl for
> the compute submissions.
> -Daniel
>

I assume you talk about interop between graphics and compute. For that, we have 
a module that is now being tested, and indeed uses an ioctl to map a graphic 
object to compute process address space. However, after the translation is done, 
the work is done only in userspace.

	Oded