From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 8060 invoked by alias); 19 Apr 2008 16:55:23 -0000 Received: (qmail 8044 invoked by uid 22791); 19 Apr 2008 16:55:22 -0000 X-Spam-Check-By: sourceware.org Received: from bombadil.infradead.org (HELO bombadil.infradead.org) (18.85.46.34) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sat, 19 Apr 2008 16:54:56 +0000 Received: from shinybook-bcm.infradead.org ([2001:8b0:10b:1:20a:95ff:fef3:9992]) by bombadil.infradead.org with esmtpsa (Exim 4.68 #1 (Red Hat Linux)) id 1JnGKv-0004ZX-R8; Sat, 19 Apr 2008 16:54:50 +0000 Subject: Re: Official GCC git repository From: David Woodhouse To: NightStrike Cc: Daniel Berlin , "Frank Ch. Eigler" , overseers@gcc.gnu.org, bernie@codewiz.org, GCC Development , Samuel Tardieu , "Kirill A. Shutemov" , harvey.harrison@gmail.com, Tobias Grosser In-Reply-To: References: <4aca3dc20803261138s20fb11famceefc20905543ef@mail.gmail.com> <4aca3dc20804180845l7037b734u1881e69d3799019e@mail.gmail.com> <2008-04-18-18-15-50+trackit+sam@rfc1149.net> <4aca3dc20804181022v19855d9cm6911e2123aae7df8@mail.gmail.com> <2008-04-18-19-40-44+trackit+sam@rfc1149.net> <4aca3dc20804181156l355ef3ag5366f8f06a4094ff@mail.gmail.com> <2008-04-18-21-07-20+trackit+sam@rfc1149.net> <20080419025819.GA11270@ednor.casa.cgf.cx> <4aca3dc20804190832q22b3808co7b7f505926d58359@mail.gmail.com> Content-Type: text/plain Date: Sat, 19 Apr 2008 16:55:00 -0000 Message-Id: <1208624090.14973.40.camel@shinybook.infradead.org> Mime-Version: 1.0 X-Mailer: Evolution 2.22.1 (2.22.1-1.fc9) Content-Transfer-Encoding: 7bit X-SRS-Rewrite: SMTP reverse-path rewritten from by bombadil.infradead.org See http://www.infradead.org/rpr.html Mailing-List: contact overseers-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: , Sender: overseers-owner@sourceware.org X-SW-Source: 2008-q2/txt/msg00048.txt.bz2 On Sat, 2008-04-19 at 11:34 -0400, NightStrike wrote: > On 4/19/08, Daniel Berlin wrote: > > The problem with commits is that the average is not what matters. > > Commits are bursty. > > People make 5 commits to different branches in the course of a minute > > or two, then there is nothing for another 15 minutes, etc. > > Oh. Well... I still had fun writing the script :) Yeah, if performance is an issue, it'd probably help to play with responses to the commit trigger -- maybe wait for 2 minutes for the repo to become idle, and then prevent another update within 5 minutes (picking numbers out of my wossname). -- dwmw2