Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750935AbVLLUf7 (ORCPT ); Mon, 12 Dec 2005 15:35:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751123AbVLLUf6 (ORCPT ); Mon, 12 Dec 2005 15:35:58 -0500 Received: from gate.crashing.org ([63.228.1.57]:28127 "EHLO gate.crashing.org") by vger.kernel.org with ESMTP id S1750935AbVLLUf6 (ORCPT ); Mon, 12 Dec 2005 15:35:58 -0500 Subject: Re: Memory corruption & SCSI in 2.6.15 From: Benjamin Herrenschmidt To: Linus Torvalds Cc: Andrew Morton , Linux Kernel list , Paul Mackerras , Jens Axboe , Brian King In-Reply-To: References: <1134371606.6989.95.camel@gaston> Content-Type: text/plain Date: Tue, 13 Dec 2005 07:33:28 +1100 Message-Id: <1134419609.6989.116.camel@gaston> Mime-Version: 1.0 X-Mailer: Evolution 2.2.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 948 Lines: 25 > If it's easily repeatable, doing a "git bisect" to see when it starts > happening is the obvious big sledge-hammer thing to try. Even if you don't > bisect all the way, just narrowing it down a bit more might help. Sure, though I assumed it had already be tracked down at least partially by Brian ;) > Also, enabling DEBUG_PAGEALLOC might help, but that's not available on > powerpc. Remind me what is needed to get that working ? Unmapping of linear mapping pages ? (I suppose I could do that if I also disable using large pages for it). > There's a raid1 use-after-free bugfix that I just merged and pushed out, > but I doubt that one is relevant. But you might try to update. > > Linus - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/