netdev
[Top] [All Lists]

Re: TCP IP Offloading Interface

To: Alan Cox <alan@xxxxxxxxxxxxxxxxxxx>
Subject: Re: TCP IP Offloading Interface
From: Jeff Garzik <jgarzik@xxxxxxxxx>
Date: Sun, 13 Jul 2003 12:58:01 -0400
Cc: Roland Dreier <roland@xxxxxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxx>, Alan Shih <alan@xxxxxxxxxxxxxxxx>, Linux Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>, linux-net@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <1058113895.554.7.camel@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
Organization: none
References: <ODEIIOAOPGGCDIKEOPILCEMBCMAA.alan@xxxxxxxxxxxxxxxx> <20030713004818.4f1895be.davem@xxxxxxxxxx> <52u19qwg53.fsf@xxxxxxxxxxx> <1058113895.554.7.camel@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20021213 Debian/1.2.1-2.bunk
Alan Cox wrote:
Finally if you are streaming objects by non mapped references (eg
sendfile or see LM's paper from long ago on splice()) then the problem
goes away.


As an aside, I really like sendfile's semantics except for

* People occasionally want to add a receivefile(2). I disagree... sendfile(2) interface should be really be considered a universal "fdcopy" interface, regardless of what the 'to' and 'from' file descriptors are attached to. File to socket. Socket to file. File to file. socket to socket. All should be supported, even if the fallback is a stupid (but small!) in-kernel copy loop.

* Copy-until-EOF semantics are either undefined, or, unclear to me personally.

        Jeff




<Prev in Thread] Current Thread [Next in Thread>