From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [Xen-devel] [PATCH] xen_disk: implement BLKIF_OP_FLUSH_DISKCACHE, remove BLKIF_OP_WRITE_BARRIER Date: Wed, 25 Apr 2012 12:20:24 +0200 Message-ID: <20120425102024.GA19800@lst.de> References: <1334595957-12552-1-git-send-email-stefano.stabellini@eu.citrix.com> <20120425084524.GA17537@lst.de> <1335344565.28015.7.camel@zakaz.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1335344565.28015.7.camel@zakaz.uk.xensource.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org Sender: qemu-devel-bounces+gceq-qemu-devel=gmane.org@nongnu.org To: Ian Campbell Cc: "kwolf@redhat.com" , "xen-devel@lists.xensource.com" , Stefano Stabellini , "konrad.wilk@oracle.com" , "qemu-devel@nongnu.org" , Christoph Hellwig List-Id: xen-devel@lists.xenproject.org On Wed, Apr 25, 2012 at 10:02:45AM +0100, Ian Campbell wrote: > The blkif spec was recently much improved, you can find it at > http://xenbits.xen.org/docs/unstable/hypercall/include,public,io,blkif.h.html > > TBH I'm not sure it actually answers your questions wrt > BLKIF_OP_FLUSH_DISKCACHE, if not please let us know and we can see about > improving it. That description in there is overly simple, and does not match any of the implementations known to me on either end. Talking about those: the mainline Linux blkback backend also implements different semantics from what mainline Linux blkfront seems to expect, as well as different from qemu. Looking at these three alone I can't see how Xen ever managed to get data to disk reliably if using the paravirt interface. with the implementations in qemu and the Linux kernel frontend and backends, which > > Ian > ---end quoted text---