From owner-p4-releng Tue Sep 17 12:59:13 2002 Delivered-To: p4-releng@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 32767) id 74C1237B404; Tue, 17 Sep 2002 12:59:11 -0700 (PDT) Delivered-To: perforce@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0E2D437B401; Tue, 17 Sep 2002 12:59:11 -0700 (PDT) Received: from canning.wemm.org (canning.wemm.org [192.203.228.65]) by mx1.FreeBSD.org (Postfix) with ESMTP id 966DF43E72; Tue, 17 Sep 2002 12:59:10 -0700 (PDT) (envelope-from peter@wemm.org) Received: from wemm.org (localhost [127.0.0.1]) by canning.wemm.org (Postfix) with ESMTP id 743D42A7D6; Tue, 17 Sep 2002 12:59:10 -0700 (PDT) (envelope-from peter@wemm.org) X-Mailer: exmh version 2.5 07/13/2001 with nmh-1.0.4 To: John Baldwin Cc: "David O'Brien" , Perforce Change Reviews Subject: Re: PERFORCE change 17622 for review In-Reply-To: Date: Tue, 17 Sep 2002 12:59:10 -0700 From: Peter Wemm Message-Id: <20020917195910.743D42A7D6@canning.wemm.org> Sender: owner-p4-releng@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG John Baldwin wrote: > > On 17-Sep-2002 David O'Brien wrote: > > On Tue, Sep 17, 2002 at 02:01:23PM -0400, John Baldwin wrote: > >> > >> On 17-Sep-2002 Peter Wemm wrote: > >> > John Baldwin wrote: > >> >> http://people.freebsd.org/~peter/p4db/chv.cgi?CH=17622 > >> >> > >> >> Change 17622 by jhb@jhb_zion on 2002/09/17 10:07:07 > >> >> > >> >> Turn off PSE and PG_G for DP2 so that it has a chance of > >> >> !sucking. :-P > >> > > >> > Except that the last I heard from mbr, DISABLE_PG_G and DISABLE_PSE no > >> > longer solve his problem anymore after the first gcc-3.2.1 import. The only > >> > effect it has is to make things 5%-10% slower. > >> > >> Other people have problems on Althons and other P4's that go away > >> (corruption type problems) with these changes. > > > > With up-to-date sources? Peter made some pmap changes a few weeks ago > > that fixed the problem these options address for me and it seemed > > everyone else. Can you point to some reports of this? > > I think there was one person with an Athlon on current@ yesterday: > > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=252354+0+current/freebsd-current > > This guy has a Pentium 3(!) and has issues: > > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=245979+0+current/freebsd-current > > These e-mails were actually today. And I just got private email from a guy who's seeing this happen on -stable too. :-/ I'm trying to get more details. I'm getting to the point now that I suspect we're having trouble on cpus with hardware prefetch. ie: Pentium4's, newer athlons, newer P3/celeron. Cheers, -Peter -- Peter Wemm - peter@wemm.org; peter@FreeBSD.org; peter@yahoo-inc.com "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe p4-releng" in the body of the message