[openib-general] Dropping NETIF_F_SG since no checksum feature.
Steven Whitehouse
steve at chygwyn.com
Wed Oct 11 02:09:26 PDT 2006
Hi,
On Wed, Oct 11, 2006 at 11:05:04AM +0200, Michael S. Tsirkin wrote:
> Quoting r. David Miller <davem at davemloft.net>:
> > Subject: Re: Dropping NETIF_F_SG since no checksum feature.
> >
> > From: "Michael S. Tsirkin" <mst at mellanox.co.il>
> > Date: Wed, 11 Oct 2006 02:13:38 +0200
> >
> > > Maybe I can patch linux to allow SG without checksum?
> > > Dave, maybe you could drop a hint or two on whether this is worthwhile
> > > and what are the issues that need addressing to make this work?
> > >
> > > I imagine it's not just the matter of changing net/core/dev.c :).
> >
> > You can't, it's a quality of implementation issue. We sendfile()
> > pages directly out of the filesystem page cache without any
> > blocking of modifications to the page contents, and the only way
> > that works is if the card computes the checksum for us.
> >
> > If we sendfile() a page directly, we must compute a correct checksum
> > no matter what the contents. We can't do this on the cpu before the
> > data hits the device because another thread of execution can go in and
> > modify the page contents which would invalidate the checksum and thus
> > invalidating the packet. We cannot allow this.
> >
> > Blocking modifications is too expensive, so that's not an option
> > either.
> >
>
I would argue that SG does make sense without checksum for protocols that
don't need/use a checksum. DECnet for example could do zero-copy without
caring about the checksum since it doesn't have one. One of these days
I'll get around to writing that bit of code :-)
> But copying still works fine, does it not?
> Dave, could you clarify this please?
>
> ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
> size_t size, int flags)
> {
> ssize_t res;
> struct sock *sk = sock->sk;
>
> if (!(sk->sk_route_caps & NETIF_F_SG) ||
> !(sk->sk_route_caps & NETIF_F_ALL_CSUM))
> return sock_no_sendpage(sock, page, offset, size, flags);
>
>
> So, it seems that if I set NETIF_F_SG but clear NETIF_F_ALL_CSUM,
> data will be copied over rather than sent directly.
> So why does dev.c have to force set NETIF_F_SG to off then?
>
I agree with that analysis,
Steve.
More information about the general
mailing list