netdev
[Top] [All Lists]

Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000

To: David Woodhouse <dwmw2@xxxxxxxxxxxxx>
Subject: Re: Early SPECWeb99 results on 2.5.33 with TSO on e1000
From: Jeff Garzik <jgarzik@xxxxxxxxxxxxxxxx>
Date: Mon, 16 Sep 2002 19:08:15 -0400
Cc: "David S. Miller" <davem@xxxxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, todd-lkml@xxxxxxxxxxxxx, hadi@xxxxxxxxxx, tcw@xxxxxxxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx, pfeather@xxxxxxxxxx
Organization: MandrakeSoft
References: <20020916.154640.78359545.davem@redhat.com> <20020916.125211.82482173.davem@redhat.com> <Pine.LNX.4.44.0209161528140.13850-100000@gp.staff.osogrande.com> <12116.1032216780@redhat.com> <12293.1032217399@redhat.com>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.1) Gecko/20020826
David Woodhouse wrote:
davem@xxxxxxxxxx said:

 Er, surely the same goes for sys_sendfile? Why have a new system
 call rather than just swapping the 'in' and 'out' fds?


There is an assumption that one is a linear stream of output (in this
case a socket) and the other one is a page cache based file.


That's an implementation detail and it's not clear we should be exposing it to the user. It's not entirely insane to contemplate socket->socket or file->file sendfile either -- would we invent new system calls for those too? File descriptors are file descriptors.

I was rather disappointed when file->file sendfile was [purposefully?] broken in 2.5.x...


        Jeff




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