Re: [PATCH net-next v2 4/5] page_pool: disable sync for cpu for dmabuf memory provider

[Pavel Begunkov]

On 11/8/24 15:58, Stanislav Fomichev wrote:
> On 11/08, Jason Gunthorpe wrote:
> > On Thu, Nov 07, 2024 at 09:23:08PM +0000, Mina Almasry wrote:
> > > dmabuf dma-addresses should not be dma_sync'd for CPU/device. Typically
> > > its the driver responsibility to dma_sync for CPU, but the driver should
> > > not dma_sync for CPU if the netmem is actually coming from a dmabuf
> > > memory provider.
> >
> > This is not completely true, it is not *all* dmabuf, just the parts of
> > the dmabuf that are actually MMIO.
> >
> > If you do this you may want to block accepting dmabufs that have CPU
> > pages inside them.
>
> We still want udmabufs to work, so probably need some new helper to test
> whether a particular netmem is backed by the cpu memory?

Agree. I guess it's fair to assume that page pool is backed either
by one or another, so could be a page pool flag that devmem.c can set
on init.

--
Pavel Begunkov