Re: [PATCH net-next v15 02/14] net: netdev netlink api to bind dma-buf to a net device

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, 1 Jul 2024 at 20:05, Mina Almasry <almasrymina@xxxxxxxxxx> wrote:
>
> On Fri, Jun 28, 2024 at 3:10 AM Donald Hunter <donald.hunter@xxxxxxxxx> wrote:
> >
> > Mina Almasry <almasrymina@xxxxxxxxxx> writes:
> > > +  -
> > > +    name: bind-dmabuf
> > > +    attributes:
> > > +      -
> > > +        name: ifindex
> > > +        doc: netdev ifindex to bind the dma-buf to.
> >
> > Minor nit:
> >
> > The series uses a mix of dmabuf and dma-buf but the doc additions
> > (devmem.rst) consistently uses dmabuf. I think it would be helpful to be
> > consistent here and say 'devmem dmabuf' in the docstring to highlight
> > whos dmabuf it is and keep the generated netdev docs in alignment.
>
> To be honest, even the dmabuf docs mixes 'dma-buf' and 'dmabuf', to my eye:
>
> https://docs.kernel.org/driver-api/dma-buf.html
>
> I can edit these docs I'm adding so these are consistent.
>
> But on 'devmem dmabuf', not sure to be honest. Technically all dmabufs
> are supported, even non-devmem ones. I'm not sure non-devmem dmabufs
> are common at all, the only example I can think of is udmabuf whose
> primary user is qemu and testing, so it's somewhat implied that the
> dmabuf is devmem, and even if it isn't, it would be supported. I
> prefer to keep the docs saying just 'dmabuf' as technically all are
> supported. Maybe I should add a note about this somewhere in the
> dedicated docs.

That's a fair point. If you could mention it in the docs, that would be great.

Thanks,
Donald.





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux