From owner-freebsd-current@FreeBSD.ORG Tue Jun 3 14:24:56 2003 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A993A37B401; Tue, 3 Jun 2003 14:24:56 -0700 (PDT) Received: from gravy.homeunix.net (pool-151-197-48-53.phil.east.verizon.net [151.197.48.53]) by mx1.FreeBSD.org (Postfix) with ESMTP id B0C8F43F85; Tue, 3 Jun 2003 14:24:55 -0700 (PDT) (envelope-from bleez@verizon.net) Received: from gravy.homeunix.net (gravy.homeunix.net [192.168.1.2]) by gravy.homeunix.net (8.12.9/8.12.9) with ESMTP id h53LOpGK000925; Tue, 3 Jun 2003 17:24:51 -0400 (EDT) (envelope-from bleez@verizon.net) Date: Tue, 3 Jun 2003 17:24:51 -0400 (EDT) From: Bryan Liesner To: Scott Long In-Reply-To: <3EDD04CF.80101@freebsd.org> Message-ID: <20030603171044.U911@gravy.homeunix.net> References: <20030602230746.U69975-100000@mail.chesapeake.net> <20030603115912.Q648@gravy.homeunix.net> <3EDCCB27.5020707@freebsd.org> <20030603132913.D761@gravy.homeunix.net> <3EDD04CF.80101@freebsd.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: Bryan Liesner cc: current@freebsd.org cc: Jeff Roberson Subject: Re: umtx/libthr SMP fixes. X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 03 Jun 2003 21:24:56 -0000 On Tue, 3 Jun 2003, Scott Long wrote: > It's very hard to imagine Jeff's patches causing a problem at the point > that the PR mentions. Have you confirmed the problem in a kernel that > was build in a totally clean environment? > > Scott If you mean a kernel build with standard optimizations, yes I have seen it. No local patches: cc -O -pipe -mcpu=pentiumpro The panic happens before running init. Sometimes I'm "lucky" and the kernel doesn't panic during boot, but it will panic shortly before or after login. It locks up tight afterwards, and doesn't do a dump. All this happened after kern_umtx.c v1.2 When I revert back, no panic... It won't panic if I turn off the USB drive, or compile a kernel without atapicam. Possible problem with EHCI? The driver finds an OHCI device, calls it umass0, detaches it, finds an EHCI device and calls it umass1. I'm sure it's not Jeffs patches that are the problem, but other potential problems are exacerbated by them.