From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48897) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dS7Nb-0004gZ-6R for qemu-devel@nongnu.org; Mon, 03 Jul 2017 15:51:32 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dS7NW-0000Ed-6o for qemu-devel@nongnu.org; Mon, 03 Jul 2017 15:51:31 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51978) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dS7NV-0000DV-Vn for qemu-devel@nongnu.org; Mon, 03 Jul 2017 15:51:26 -0400 Date: Mon, 3 Jul 2017 22:51:21 +0300 From: "Michael S. Tsirkin" Message-ID: <20170703224820-mutt-send-email-mst@kernel.org> References: <20170629132310.18865-1-marcandre.lureau@redhat.com> <20170629132310.18865-2-marcandre.lureau@redhat.com> <20170703213528-mutt-send-email-mst@kernel.org> <20170703185052.GT12152@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20170703185052.GT12152@localhost.localdomain> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH 1/7] vmgenid: replace x-write-pointer-available hack List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eduardo Habkost Cc: =?iso-8859-1?Q?Marc-Andr=E9?= Lureau , qemu-devel@nongnu.org, imammedo@redhat.com, anderson@redhat.com, berrange@redhat.com, lersek@redhat.com, Ben Warren On Mon, Jul 03, 2017 at 03:50:52PM -0300, Eduardo Habkost wrote: > On Mon, Jul 03, 2017 at 09:38:52PM +0300, Michael S. Tsirkin wrote: > > On Thu, Jun 29, 2017 at 03:23:04PM +0200, Marc-Andr=E9 Lureau wrote: > > > This compat property sole function is to prevent the device from be= ing > > > instantiated. Instead of requiring an extra compat property, check = if > > > fw_cfg has DMA enabled. > > >=20 > > > This has the additional benefit of handling other cases properly, l= ike: > > >=20 > > > $ qemu-system-x86_64 -device vmgenid -machine none > > > qemu-system-x86_64: -device vmgenid: vmgenid requires DMA write s= upport in fw_cfg, which this machine type does not provide > > > $ qemu-system-x86_64 -device vmgenid -machine pc-i440fx-2.9 -glob= al fw_cfg.dma_enabled=3Doff > > > qemu-system-x86_64: -device vmgenid: vmgenid requires DMA write s= upport in fw_cfg, which this machine type does not provide > > > $ qemu-system-x86_64 -device vmgenid -machine pc-i440fx-2.6 -glob= al fw_cfg.dma_enabled=3Don > > > [boots normally] > > >=20 > > > Suggested-by: Eduardo Habkost > > > Signed-off-by: Marc-Andr=E9 Lureau > >=20 > > It's a nice cleanup, but I suspect we need to first implement > > a framework for initialization ordering. I don't much like it > > that we are adding more dependencies to the current bag of hacks. >=20 > I agree we should address this, but in this case there's no need to > introduce new mechanisms for initialization ordering if we just check > the dependencies on machine_done notifier or acpi_setup() (which is > called by machine_done). I guess what should fail is attempt to register a writeable blob. This sounds reasonable. > --=20 > Eduardo