Date: Tue, 20 Jun 2000 12:32:06 -0700 From: Paul Saab <paul@mu.org> To: Alfred Perlstein <bright@wintelcom.net> Cc: Dag-Erling Smorgrav <des@flood.ping.uio.no>, arch@FreeBSD.ORG, jlemon@FreeBSD.ORG, msmith@FreeBSD.ORG, nate@FreeBSD.ORG Subject: Re: MORE: Re: kblob discussion. Message-ID: <20000620123206.A56823@elvis.mu.org> In-Reply-To: <20000620093526.Q17420@fw.wintelcom.net>; from bright@wintelcom.net on Tue, Jun 20, 2000 at 09:35:26AM -0700 References: <20000619111309.E26801@fw.wintelcom.net> <xzpk8fktu2x.fsf@flood.ping.uio.no> <20000620093526.Q17420@fw.wintelcom.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Alfred Perlstein (bright@wintelcom.net) wrote: > * Dag-Erling Smorgrav <des@flood.ping.uio.no> [000620 05:28] wrote: > > Alfred Perlstein <bright@wintelcom.net> writes: > > > oops, you can find the preliminary kblob diffs over here: > > > > > > http://people.freebsd.org/~alfred/kblob/kblob-final.diff > > > > > > It's a new syscall, less overhead than sendfile and useful for > > > serving small chunks of data very quickly. > > > > Do kblobs do anything that can't be easily done with sendfile() by > > mmapping suitably large chunks of /dev/null? > > Sendfile's vm tricks and additional overhead per-page sized chunks > make it unsuitable for sending smaller chunks of data. Paul Saab > noticed that for some workloads it was actually slower than just > copying the data while working at Hotmail. I was talking with dg last week and we may have found a flaw in the way that I measured the speed. I am fairly convinced now that sendfile is the prefered way to send data from a file. -- Paul Saab Technical Yahoo paul@mu.org - ps@yahoo-inc.com - ps@freebsd.org Do You .. uhh .. Yahoo!? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000620123206.A56823>