Date: Thu, 7 Sep 2000 09:55:53 +0000 From: Tony Finch <dot@dotat.at> To: Alfred Perlstein <alfred@FreeBSD.org> Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/kern uipc_accf.c src/sys/netinet accf_http.c src/sys/sys socketvar.h Message-ID: <20000907095553.A6767@hand.dotat.at> In-Reply-To: <20000907022131.V18862@fw.wintelcom.net> References: <200009061849.LAA34906@freefall.freebsd.org> <20000907023342.G656@hand.dotat.at> <20000907022131.V18862@fw.wintelcom.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Alfred Perlstein <alfred@FreeBSD.org> wrote: >* Tony Finch <dot@dotat.at> [000906 19:33] wrote: >> Alfred Perlstein <alfred@FreeBSD.org> wrote: >> > >> > Accept filter maintainance >> > >> > 2) Check if a socket has filled and dump to the listener >> >> Cool, but I think it would be better if you could configure an accept >> filter limit smaller than the socket buffer size. > >Afaik this can be achieved by using setsockopt to set the highwater >on the socket. But you want to set that high so that you can fill a long fat pipe, surely. (Although Apache doesn't do that so I guess it doesn't matter so much -- unless you expect lots of uploads.) Also if I understand things right I think that mbuf-based allocation can cause a multiplication of the memory used for socket receive buffers by quite a large factor. >> Also, what chance on timeouts for filtered connections? > >That's possible, but right now I'd like to fix the way listen() >overflows are handled. (make them FIFO) Yes. Tony. -- en oeccget g mtcaa f.a.n.finch v spdlkishrhtewe y dot@dotat.at eatp o v eiti i d. fanf@covalent.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000907095553.A6767>