Date: Fri, 24 Apr 1998 20:54:48 -0700 From: Bill Trost <trost@cloud.rain.com> To: current@FreeBSD.ORG Subject: Re: Bandwidth throttling etc. Message-ID: <908.893476488@cloud.rain.com> In-Reply-To: Your message of Fri, 24 Apr 1998 17:26:48 EDT. <199804242126.RAA10941@khavrinen.lcs.mit.edu> References: <199804242126.RAA10941@khavrinen.lcs.mit.edu> <Pine.GSO.3.96.980424165606.18437A-100000@echonyc.com> <199804241932.VAA22011@labinfo.iet.unipi.it>
next in thread | previous in thread | raw e-mail | index | archive | help
Garrett Wollman writes:
What you Really Want is for each interface to manage its own
allocations. When you want to send a packet, you ask the interface
for a buffer, and it gives you one of an appropriate size and shape
that it knows how to transmit efficiently.....
A lot of the work is not actually difficult, just tedious....
Before anyone spends oodles of time doing this sort of work, has anyone
taken the radical (-: step of actually profiling the current network
stack(s) to see where the time is being eaten? Improving the memory
access behavior may not gain very much, especially if the performance
hits occur in, say, the IP checksum computation (as a random example).
Maybe someone has already looked at this and this is just rehashing old
news, but it just struck me that some hard data would be an important
guide.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?908.893476488>