From owner-cvs-src@FreeBSD.ORG Wed Jun 25 02:14:23 2003 Return-Path: Delivered-To: cvs-src@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1AE6C37B401; Wed, 25 Jun 2003 02:14:23 -0700 (PDT) Received: from brian.webcom.it (brianlap.inet.it [213.92.1.190]) by mx1.FreeBSD.org (Postfix) with ESMTP id ECF9043FE0; Wed, 25 Jun 2003 02:14:19 -0700 (PDT) (envelope-from andrea@webcom.it) Received: by brian.webcom.it (Postfix, from userid 1000) id 65C8F80; Wed, 25 Jun 2003 11:14:17 +0200 (CEST) Date: Wed, 25 Jun 2003 11:14:17 +0200 From: Andrea Campi To: Nate Lawson Message-ID: <20030625091417.GA656@webcom.it> References: <200306120337.h5C3bSmB075154@repoman.freebsd.org> <20030625001613.GA636@webcom.it> <20030624231534.K71390@root.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20030624231534.K71390@root.org> X-Echelon: BND CIA NSA Mossad KGB MI6 IRA detonator nuclear assault strike User-Agent: Mutt/1.5.4i cc: cvs-src@FreeBSD.org cc: src-committers@FreeBSD.org cc: Warner Losh cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/dev/pccbb pccbb.c pccbbvar.h X-BeenThere: cvs-src@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Jun 2003 09:14:23 -0000 On Tue, Jun 24, 2003 at 11:18:15PM -0700, Nate Lawson wrote: > > Um, did you cvsup to 1.78? I found an error in the original commit where > it locks Giant twice (instead of unlocking) which would give exactly these > symptoms. > That's what I get for not being more careful with my diffs... Yes, I still had trouble with 1.78; I initially had the problem with a kernel compiled from sources cvsup'ed on Jun 20 and then Jun 24, that's why I started tracking it down. I've just cvsup'd again to the HEAD, applied my patch and I'm running a make buildkernel just to be extra sure, but I think this is real. Anyway, I'm curious - are you sure the symptoms for double locking Giant would only be buffers not emptying, and not actually a freeze? Where would the Giant be unlocked on the way out from the intr handler? (not a rethoric question, I really didn't check). I'll try playing with it. Now that I think of it, I should also recompile with WITNESS et al. I'll get back when I know more. Bye, Andrea -- Where do you think you're going today?