OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

virtio-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [virtio-comment] [PATCH requirements 7/7] net-features: Add header data split requirements


> From: Michael S. Tsirkin <mst@redhat.com>
> Sent: Tuesday, June 6, 2023 6:42 PM


> > +## 3.6 Header data split for the receive virtqueue 1. The device
> > +should be able to DMA the packet header and data to two different
> > +   memory locations, this enables driver and networking stack to perform
> zero
> > +   copy to application buffer(s).
> 
> what is header here? which layer?
> 
Usually L2 to L4. To be defined further.

> > +2. The driver should be able to configure maximum header buffer size per
> > +   virtqueue.
> > +3. The header buffer to be in a physically contiguous memory per
> > +virtqueue
> 
> this is a requirement why?
> 
Because fetching individual buffer memory pointer is sub-optimal.

> > +4. The device should be able to indicate header data split in the receive
> > +   completion.
> > +5. The device should be able to zero pad the header buffer when the
> received
> > +   header is shorter than cpu cache line size.
> 
> weirdly specific.
> how does device even know the cache line?
> 
A driver can tell how much padding to be done by keeping the cache line specific details in the driver, but only telling about padding size.


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]