From: Christoph Hellwig Subject: Re: [PATCH] extend e2fsprogs functionality to add EXT2_FLAG_DIRECT option Date: Tue, 12 Jan 2010 11:38:28 -0500 Message-ID: <20100112163828.GA14633@infradead.org> References: <4B46FCB2.1090308@redhat.com> <4B4B84E2.1050508@redhat.com> <4B4C54DC.4040006@redhat.com> <4B4C6429.6090803@redhat.com> <4B4C67F5.1020009@redhat.com> <20100112122319.GA20596@infradead.org> <4B4C6B70.1050205@redhat.com> <20100112124600.GA7151@infradead.org> <4B4C7297.5030905@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Christoph Hellwig , Ric Wheeler , linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org To: Michal Novotny Return-path: Received: from bombadil.infradead.org ([18.85.46.34]:37389 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751160Ab0ALQi3 (ORCPT ); Tue, 12 Jan 2010 11:38:29 -0500 Content-Disposition: inline In-Reply-To: <4B4C7297.5030905@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: Ok, I looked at the issue. The problem is that the Xen backend drivers are (as expected) utterly braindead and submit bios directly from the virtualization backed without using proper abstractions and thus bypassing all the cache coherency features in the fileystems (the block device nodes are just another mini-filesystem in that respect). So when you first have buffered access in the host pages may stay in cache and get overwritten directly on disk by a Xen guest, and once the guest is down the host may still use the now stale cached data. I would recommend to migrate your cutomers to KVM which uses the proper abtractions and thus doesn't have this problem. There's a reason after all why all the Xen dom0 mess never got merged to mainline.