Date: Sun, 26 Nov 2006 22:05:38 +0300 From: "Andrew Pantyukhin" <infofarmer@FreeBSD.org> To: "Kris Kennaway" <kris@obsecurity.org> Cc: FreeBSD Questions <questions@freebsd.org> Subject: Re: Local checkout performance - cvs vs cvsup Message-ID: <cb5206420611261105o64e10324yf183e9770029ef80@mail.gmail.com> In-Reply-To: <20061124224040.GA4125@xor.obsecurity.org> References: <cb5206420611241425x1ae09397q95cf20bd5e070c1e@mail.gmail.com> <20061124224040.GA4125@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 11/25/06, Kris Kennaway <kris@obsecurity.org> wrote: > On Sat, Nov 25, 2006 at 01:25:45AM +0300, Andrew Pantyukhin wrote: > > I use cvsup to maintain a full local copy of the CVS > > repo. I then use plain cvs to check out different branches > > of src from it. The problem is while cvsup only takes a > > minute each time I run it, cvs takes several times more. > > > > I run cvsup with "-gs" and cvs with "-rR" flags. Should > > I run a local cvsupd server or is there another way to > > speed things up? I only need the repo and the branches > > in read-only mode. > > I guess cvs is slower because it needs to check for local diffs > before/while merging in changes. By contrast cvsup will just blow > away any changes you have made. Yay! I set up a local cvsupd and it now takes under 10 seconds to update any branch, as compared to 5-7 minutes cvs took. > There's probably not much you can do apart from not checking out full > trees unless you really need them. You see, I use different branches in jails and on boxes, so it makes sense to get the whole repo over the net and sync to the branches locally.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?cb5206420611261105o64e10324yf183e9770029ef80>