Messages in this thread | | | Date | Wed, 5 May 1999 08:41:34 -0400 (EDT) | From | "Richard B. Johnson" <> | Subject | Re: Mark Russinovich's reponse Was: [OT] Comments to WinNT Mag !! (fwd) |
| |
On Tue, 4 May 1999, David Miller wrote:
> From: alan@lxorguk.ukuu.org.uk (Alan Cox) > Date: Tue, 4 May 1999 19:01:21 +0100 (BST) > > > BTW is it really true, that NT transmitfile() does zero copy? I > > strongly suspect, it does not. > > NT5 beta claims to > > They can avoid the extraneous copy, but what they cannot do with most > PC networking cards is avoid touching the data since most cards do not > provide a hardware checksumming facility. > > Most of this would suggest that their existing architecture passes > mbuf-chain-like buffers to the networking drivers in NT, or some other > kind of scatter-gather list like scheme. This is the only way they > could do zero-copy without driver updates from all the networking card > vendors. > > Later, > David S. Miller > davem@redhat.com >
They can't. Many of the driver cards that NT uses use PIO. This, in anybody's book, is a copy operation. There is one Intel board, it has an i960 on it (don't remember the name), plus the regular Intel chipset, that was designed for 'zero-copy' operation. There is a NT driver for it. Even this is not 'zero-copy' although the i960 does the copy instead of the ix86.
In the 'real world' about the only zero-copy that could work is for datagrams, i.e., connectionless. If you are going to assemble out-of-sequence and missing packets in 'user-land' I don't see the advantage of zero-copy at all.
Cheers, Dick Johnson ***** FILE SYSTEM WAS MODIFIED ***** Penguin : Linux version 2.2.6 on an i686 machine (400.59 BogoMips). Warning : It's hard to remain at the trailing edge of technology.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |