Date: Tue, 31 Jul 2007 21:59:37 +0200 From: Ulf Lilleengen <lulf@FreeBSD.org> Cc: Perforce Change Reviews <perforce@FreeBSD.org> Subject: Re: PERFORCE change 124344 for review Message-ID: <20070731195937.GA1379@carrot.studby.ntnu.no> In-Reply-To: <20070731175105.GU1092@garage.freebsd.pl> References: <200707301047.l6UAlD8L092463@repoman.freebsd.org> <20070731175105.GU1092@garage.freebsd.pl>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jul 31, 2007 at 07:51:05PM +0200, Pawel Jakub Dawidek wrote: > On Mon, Jul 30, 2007 at 10:47:13AM +0000, Ulf Lilleengen wrote: > > http://perforce.freebsd.org/chv.cgi?CH=124344 > > > > Change 124344 by lulf@lulf_carrot on 2007/07/30 10:47:07 > > > > - Sleep after sending create requests in userland, since the event might > > not yet be processed. If there are noteable issues with this, which is > > a quite unlikely to happen, I have a patch to support suspending > > of the userland process until the events are executed, but it's only > > required in this special case. > > Sleeping that way is, as you noted, not very nice... Can't you put a > loop below errstr check, which will wait for drive creation. You could > sleep for eg. 0.1s between checks. Also not very nice, but even 2s may > be too low on very loaded systems. Hmm, yes I agree. Just committed a fix that does this. I'm more and more considering the suspending patch :) Anyway.. thanks. -- Ulf Lilleengen
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070731195937.GA1379>