From owner-freebsd-current@FreeBSD.ORG Sat Apr 3 07:57:33 2004 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 39DDC16A4CE for ; Sat, 3 Apr 2004 07:57:33 -0800 (PST) Received: from spider.deepcore.dk (cpe.atm2-0-53484.0x50a6c9a6.abnxx9.customer.tele.dk [80.166.201.166]) by mx1.FreeBSD.org (Postfix) with ESMTP id E545043D1D for ; Sat, 3 Apr 2004 07:57:31 -0800 (PST) (envelope-from sos@DeepCore.dk) Received: from DeepCore.dk (sos.deepcore.dk [194.192.25.130]) by spider.deepcore.dk (8.12.11/8.12.10) with ESMTP id i33Fv7Ul040311; Sat, 3 Apr 2004 17:57:12 +0200 (CEST) (envelope-from sos@DeepCore.dk) Message-ID: <406EDED3.90708@DeepCore.dk> Date: Sat, 03 Apr 2004 17:57:07 +0200 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= User-Agent: Mozilla Thunderbird 0.5 (X11/20040329) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Doug Ambrisko References: <200404030612.i336CRmA019492@ambrisko.com> In-Reply-To: <200404030612.i336CRmA019492@ambrisko.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-mail-scanned: by DeepCore Virus & Spam killer v1.4 cc: current@freebsd.org Subject: Re: Intel SATA ICH5/5R 6300ESB support patches 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: Sat, 03 Apr 2004 15:57:33 -0000 Doug Ambrisko wrote: > Here are some patches to deal with SATA devices causing errors and > resulting in wedged systems. Deals with drives removed/powered down, > drives powered up or plugged in and media errors. We need to > read the SATA registers and twiddle the port enable on device > departures. If we don't then the ata code execution will result in > a lock-up. > > I put in a patch for geom for bio_taskqueue_remove. Since ata code > schedules bio_task it need to be cancelled when we abort and call > biodone. If we don't cancel this task then when the task is > run later we get a double free in UMA since we have cleaned up > twice and called biodone twice for the same request. It seems > like in biodone we should clean up tasks there. Suggestions > appreciated. This band-aids it for: > 1) dd if=/dev/ad2 of=/dev/null > 2) then power down the drive. > > There are other issues on media recovery that I've hit but > I'm going to do that separate to this HW support. Promise > patches to deal with SATA issues should be coming soon. > > These are based on my -stable patches but are enhanced a little to > deal with some other cases. > > This should make -current systems more stable with SATA drives when > something goes wrong with them. > > I'd like to commit them to -current after a review. Well, looks promising, but unfortunatly they collide with the WIP I have here in the lab, so I'd appreciate if you would not commit it right now. I'll look into getting the confilicting areas merged into what I have here, then have you test it on the ICH5 as I dont have any of those around... -- -Søren