netdev
[Top] [All Lists]

Re: sendfile+zerocopy: fairly sexy (nothing to do with ECN)

To: David Lang <dlang@xxxxxxxxxxxxx>
Subject: Re: sendfile+zerocopy: fairly sexy (nothing to do with ECN)
From: "David S. Miller" <davem@xxxxxxxxxx>
Date: Fri, 2 Feb 2001 15:31:41 -0800 (PST)
Cc: Andrew Morton <andrewm@xxxxxxxxxx>, lkml <linux-kernel@xxxxxxxxxxxxxxx>, "netdev@xxxxxxxxxxx" <netdev@xxxxxxxxxxx>
In-reply-to: <Pine.LNX.4.31.0102021511330.1221-100000@xxxxxxxxxxxxxxxxxxx>
References: <14971.15897.432460.25166@xxxxxxxxxxxxxxx> <Pine.LNX.4.31.0102021511330.1221-100000@xxxxxxxxxxxxxxxxxxx>
Sender: owner-netdev@xxxxxxxxxxx
David Lang writes:
 > right, assuming that there is enough sendfile() benifit to overcome the
 > write() penalty from the stuff that can't be cached or sent from a file.
 > 
 > my question was basicly are there enough places where sendfile would
 > actually be used to make it a net gain.

There are non-performance issues as well (really, all of these points
have been mentioned in this thread btw).  One is that since paged
SKBs use only single-order page allocations, the memory allocation
subsystem is stressed less than the current scheme where SLAB
allocates multi-order pages to satisfy allocations of linear SKB data
buffers.

This has consequences and benefits system wide.

Later,
David S. Miller
davem@xxxxxxxxxx

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