From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48805) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aY8yQ-00087k-FR for qemu-devel@nongnu.org; Tue, 23 Feb 2016 04:09:39 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aY8yP-0005Gq-Mk for qemu-devel@nongnu.org; Tue, 23 Feb 2016 04:09:38 -0500 Date: Tue, 23 Feb 2016 11:09:26 +0200 From: "Michael S. Tsirkin" Message-ID: <20160223110710-mutt-send-email-mst@redhat.com> References: <1456144729-17196-1-git-send-email-mst@redhat.com> <1456151216.29896.63.camel@redhat.com> <20160222220400-mutt-send-email-mst@redhat.com> <1456211237.29896.72.camel@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1456211237.29896.72.camel@redhat.com> Subject: Re: [Qemu-devel] [PATCH RFC] fw-cfg: support writeable blobs List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gerd Hoffmann Cc: Peter Maydell , Xiao Guangrong , Eduardo Habkost , Igor Mammedov , "Gabriel L. Somlo" , qemu-devel@nongnu.org, Michael Walle , Kevin O'Connor , qemu-arm@nongnu.org, Shannon Zhao , Paolo Bonzini , Richard Henderson On Tue, Feb 23, 2016 at 08:07:17AM +0100, Gerd Hoffmann wrote: > On Mo, 2016-02-22 at 22:12 +0200, Michael S. Tsirkin wrote: > > On Mon, Feb 22, 2016 at 03:26:56PM +0100, Gerd Hoffmann wrote: > > > On Mo, 2016-02-22 at 14:41 +0200, Michael S. Tsirkin wrote: > > > > Useful to send guest data back to QEMU. > > > > > > Use case? > > > > > > cheers, > > > Gerd > > > > VM GEN ID at least wants to pass address of some blob > > in guest memory to host. > > > > Apparently, that's also useful for nvdimm. > > > > A reasonable way to do that seems to be to write it into > > fw cfg file, this way it's also migrated automatically. > > Both seem to be about acpi aml code talking to qemu backend. At least not for gen id - there will be a linker patch to initialize that. > Not sure fw_cfg files are reasonable for that, given that you have to > fetch the directory listing, parse it to figure the entry index, ... > > I want see a patch actually using that before going to merge it. > > cheers, > Gerd Absolutely, this makes sense. -- MST