Date: Fri, 10 Feb 95 7:11:20 MST From: terry@cs.weber.edu (Terry Lambert) To: peter@bonkers.taronga.com (Peter da Silva) Cc: davidg@Root.COM, jmb@kryten.atinc.com, hackers@FreeBSD.org Subject: Re: MIT SHM X11 extensions? (fwd) Message-ID: <9502101411.AA10760@cs.weber.edu> In-Reply-To: <199502100302.VAA00899@bonkers.taronga.com> from "Peter da Silva" at Feb 9, 95 09:02:36 pm
next in thread | previous in thread | raw e-mail | index | archive | help
This was actually sent before but got stuck in my mail spool (which I cleaned out yesterday when a user took it upon himself to personally mail-bomb Cantor and Seigal. 22438 messages sent from 3 machines, all with the Reply-To: set. > > Uh, why is ETXTBUSY still around? > > What else would you do if you opened a currently running executable for > writing? > > (ETXTBUSY on delete is bogus, and I hope FreeBSD 2.x has dumped that, but > there's really no alternative on open than things that have surprising > consequences) Actually, you could fault the whole image to swap and close the reference to the file -- and just allow the write, with never an ETXTBUSY to be seen. This would involve setting an "affinity" for pages from swap in an executing image. Once this was done, the determination that a file is on a remote file system could cause a pre-load to swap before executing; this would be a rather minor exec change. It would leverage the affinity for swap change to also resolve the NFS-image-overwritten-on-server problem at the same time. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9502101411.AA10760>