Re: kernel > 2.1.36 & nfs

Alan Cox (alan@lxorguk.ukuu.org.uk)
Wed, 4 Jun 1997 08:08:31 +0100 (BST)


> What cross-cache copy?
>
> When we're talking really high performance, we're talking DMA and cards
> that do the checksum on the card themselves. In that case the data never
> even touches the caches on the CPU that takes the interrupt - or at least
> only the headers for finding out what the packet is all about.

The headers that means, and for real network controllers (the ones that
actually come in computers for normal people) the checksums.

> > The header can be in multiple parts. Also stuff like the NFS rpc code would
> > need a rewrite to cope with fragmented rpc headers
>
> I certainly agree on the need to have the headers in one fragment, but
> that still doesn't mean we need to reassemble _all_ fragments.

No it means we need something akin to BSD m_pullup()