Date: Sun, 29 Jun 2003 20:56:06 -0400 (EDT) From: Robert Watson <rwatson@freebsd.org> To: "George V. Neville-Neil" <gnn@neville-neil.com> Cc: freebsd-net@freebsd.org Subject: Re: Another question on locking... Message-ID: <Pine.NEB.3.96L.1030629205430.94924I-100000@fledge.watson.org> In-Reply-To: <87y8zkwj6k.wl@jchurch.neville-neil.com.neville-neil.com>
next in thread | previous in thread | raw e-mail | index | archive | help
The locking in the -CURRENT tree for the socket stack is currently incomplete; something we hope greatly to remedy on the way to 5.2-RELEASE. So currently all this code runs under the Giant lock, which is why it's safe. Robert N M Watson FreeBSD Core Team, TrustedBSD Projects robert@fledge.watson.org Network Associates Laboratories On Sun, 29 Jun 2003, George V. Neville-Neil wrote: > Hi, > > Looking at the code in uipc_socket.c and udp_usrreq.c I am a > bit confused as to how the locking mechanism works between the > sockets and the protocols. The socket calls all occur under > Giant, which the protocols do not deal with, and the protocols > all lock on the inp mutex, which the socket calls do not deal > with. Furthermore the sblock/sbunlock calls are only use by > the socket layer. I'm reading udp_input() as it's simpler to > follow. Have I missed something? What prevents a protocol > from writing to a socket buffer at the wrong time or the > socket code from tweaking it at the wrong time? Is this > handled simply because of the priority of the threads doing > the work? > > Thanks, > George > _______________________________________________ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org" >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96L.1030629205430.94924I-100000>