Date: Fri, 15 Dec 2006 20:40:44 -0500 From: Tom Rhodes <trhodes@FreeBSD.org> To: John Baldwin <jhb@FreeBSD.org> Cc: src-committers@FreeBSD.org, andre@FreeBSD.org, cvs-src@FreeBSD.org, cvs-all@FreeBSD.org, rrs@FreeBSD.org, julian@elischer.org, bms@FreeBSD.org Subject: Re: cvs commit: src/lib/libc/net Makefile.inc sctp_sys_calls.c src/sys/sys param.h Message-ID: <20061215204044.22d7d0e7.trhodes@FreeBSD.org> In-Reply-To: <200612151551.31355.jhb@freebsd.org> References: <200612151201.kBFC1qEv006825@repoman.freebsd.org> <4582A6C9.8010009@FreeBSD.org> <4582FB5A.4010208@elischer.org> <200612151551.31355.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 15 Dec 2006 15:51:30 -0500 John Baldwin <jhb@FreeBSD.org> wrote: > On Friday 15 December 2006 14:45, Julian Elischer wrote: > > Bruce M. Simpson wrote: > > > Andre Oppermann wrote: > > >> > > >> What makes these sctp_* syscalls so special as opposed to their > > >> generic and protocol agnostic counterparts? > > > They're used for operations which do not have a direct correspondence in > > > the existing functions, i.e. connecting to multihomed peers, and dealing > > > with one-to-many sockets. > > > > > > See Section 9.3-9.12, UNIX Network Programming Vol 1 3e for more info. > > > > > > generally we would use socket ops or ioctls for this sort of thing.. > > syscalls is not how they would normally be done.... > > I'll give a free paper cookie to the first person to actually go _read_ the > committed code and notice that, *tada*, aside from the sctp_send*(), and > sctp_recvmsg() functions, these are indeed library wrapper functions around > getsockopt() and setsockopt(). > Adding useless fodder, I must say ... "John for the win" -- Tom Rhodes
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20061215204044.22d7d0e7.trhodes>