Re: New subsystem for acceleration devices

From: Oded Gabbay
Date: Wed Aug 10 2022 - 10:38:56 EST


On Wed, Aug 10, 2022 at 5:10 PM <yuji2.ishikawa@xxxxxxxxxxxxx> wrote:
>
> > -----Original Message-----
> > From: Oded Gabbay <oded.gabbay@xxxxxxxxx>
> > Sent: Wednesday, August 10, 2022 6:42 AM
> > To: Dave Airlie <airlied@xxxxxxxxx>; Greg Kroah-Hartman
> > <gregkh@xxxxxxxxxxxxxxxxxxx>; ishikawa yuji(石川 悠司 ○RDC□AITC○
> > EA開) <yuji2.ishikawa@xxxxxxxxxxxxx>; Jiho Chu <jiho.chu@xxxxxxxxxxx>
> > Cc: dri-devel <dri-devel@xxxxxxxxxxxxxxxxxxxxx>; Arnd Bergmann
> > <arnd@xxxxxxxx>; Linux-Kernel@Vger. Kernel. Org
> > <linux-kernel@xxxxxxxxxxxxxxx>; Jason Gunthorpe <jgg@xxxxxxxxxx>
> > Subject: Re: New subsystem for acceleration devices
> >
> > Hi Jiho, Yuji.
> >
> > I want to update that I'm currently in discussions with Dave to figure out what's
> > the best way to move forward. We are writing it down to do a proper comparison
> > between the two paths (new accel subsystem or using drm). I guess it will take
> > a week or so.
> >
> > In the meantime, I'm putting the accel code on hold. I have only managed to do
> > the very basic infra and add a demo driver that shows how to register and
> > unregister from it.
> > You can check the code at:
> > https://git.kernel.org/pub/scm/linux/kernel/git/ogabbay/linux.git/log/?h=ac
> > cel
> >
> > It has two commits. The first adds the subsystem code and the second adds the
> > demo driver.
> > The subsystem code is basically drm code copied and renamed and slightly
> > modified, but I really only worked on it for a couple of hours so take that into
> > consideration.
> >
> > The important thing is that the demo driver shows the basic steps are really
> > simple. You need to add two function calls in your probe and one function call in
> > your release. Of course you will need to supply some function callbacks, but I
> > haven't got to fill that in the demo driver. Once you register, you get
> > /dev/accel/ac0 and
> > /dev/accel/ac_controlD64 (if you want a control device). If I were to continue
> > this, the next step is to do the open and close part.
> >
> > I will update once we know where things are heading. As I said, I imagine it can
> > take a few weeks.
> >
> > Thanks,
> > Oded
>
> Hi Odded,
> Thank you for uploading the framework as well as a sample.
> It's exciting to see new software is growing up.
>
> Since Visconti DNN is a platform device, I'll write some test code to initialize driver and see if it works.
>
> Regards,
> Yuji

Platform or PCI, it doesn't matter. You just call it from the probe.
But really, this is something I did in a few hours and I stopped
because there were some objections and I wanted to first talk about it
with Dave.
I don't know if it's worth it for you to waste time on it at this point.

Thanks,
Oded