From owner-freebsd-arm@FreeBSD.ORG Fri Aug 25 14:23:17 2006 Return-Path: X-Original-To: freebsd-arm@freebsd.org Delivered-To: freebsd-arm@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 589BE16A4DD for ; Fri, 25 Aug 2006 14:23:17 +0000 (UTC) (envelope-from mlfbsd@dong.ci0.org) Received: from dong.ci0.org (cognet.ci0.org [80.65.224.102]) by mx1.FreeBSD.org (Postfix) with ESMTP id 264F543D8B for ; Fri, 25 Aug 2006 14:23:09 +0000 (GMT) (envelope-from mlfbsd@dong.ci0.org) Received: from dong.ci0.org (localhost.ci0.org [127.0.0.1]) by dong.ci0.org (8.13.6/8.13.4) with ESMTP id k7PEXvLW060908; Fri, 25 Aug 2006 16:33:57 +0200 (CEST) (envelope-from mlfbsd@dong.ci0.org) Received: (from mlfbsd@localhost) by dong.ci0.org (8.13.7/8.13.4/Submit) id k7PEXvVl060907; Fri, 25 Aug 2006 16:33:57 +0200 (CEST) (envelope-from mlfbsd) Date: Fri, 25 Aug 2006 16:33:56 +0200 From: Olivier Houchard To: ticso@cicely.de Message-ID: <20060825143356.GA60471@ci0.org> References: <20060825093816.GD49178@cicely12.cicely.de> <20060825111209.GA59325@ci0.org> <20060825121341.GG49178@cicely12.cicely.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20060825121341.GG49178@cicely12.cicely.de> User-Agent: Mutt/1.4.1i Cc: freebsd-arm@freebsd.org Subject: Re: Sleeping thread (tid 100017, pid 27) owns a non-sleepable lock X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting FreeBSD to the StrongARM Processor List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 25 Aug 2006 14:23:17 -0000 On Fri, Aug 25, 2006 at 02:13:41PM +0200, Bernd Walter wrote: > On Fri, Aug 25, 2006 at 01:12:10PM +0200, Olivier Houchard wrote: > > > > Here, something is deadly wrong. You get a page fault while scheduling the > > interrupt. Most likely something writes where it shouldn't and corrupts the > > ithread structures, or you're mixing INTR_FAST and non-INTR_FAST drivers, but > > reading at your dmesg it doesn't seem so. > > Can you reproduce it, or does it just happen randomly ? > > It is reproduceable. > And contrary to the original boot problem I get excactly the same panic > when turning off one SDRAM chip. > But the panic is gone if I use a kernel without my rlswitch driver. > Now ukphy gets connected to all 6 PHY mappings. > Can't imagine at which place rlswitch could have corrupted memory, but > maybe it happens because of refusing multiple attachment. No I don't think your rlswitch driver is responsible for your problem. I remember we had a a similar problem when trying to use iic. We blame iic at that time, but now I think we have something somewhere code which writes at random location, and it just happen to be noticable when the iic or your rlswitch driver are in the kernel. Or maybe it malloc() some chunk of code, and write before/after the allocated chunk. That's just a guess of course. I thought memguard could help us on this one, but it doesn't seem so. That's really an hard to track problem I fear. Olivier