From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Rutland Subject: Re: [PATCH 2/3] Initial skeleton of VFIO support for Device Tree based devices Date: Mon, 5 Aug 2013 14:50:09 +0100 Message-ID: <20130805135009.GG9910@e106331-lin.cambridge.arm.com> References: <1375708632-26356-1-git-send-email-a.motakis@virtualopensystems.com> <1375708632-26356-3-git-send-email-a.motakis@virtualopensystems.com> <20130805133757.GF9910@e106331-lin.cambridge.arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "linux-arm-kernel@lists.infradead.org" , "alex.williamson@redhat.com" , "linux-samsung-soc@vger.kernel.org" , "kvm@vger.kernel.org" , "agraf@suse.de" , "B08248@freescale.com" , "iommu@lists.linux-foundation.org" , "kvmarm@lists.cs.columbia.edu" , Pawel Moll , "swarren@wwwdotorg.org" , "ian.campbell@citrix.com" , "grant.likely@linaro.org" , "tomasz.figa@gmail.com" , "rob.herring@calxeda.com" To: Antonios Motakis Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-samsung-soc-owner@vger.kernel.org List-Id: kvm.vger.kernel.org > > +static const struct of_device_id vfio_dt_match[] =3D { > > + =EF=BF=BD =EF=BF=BD /* In the future, we can implement a bet= ter mechanism to bind > the > > + =EF=BF=BD =EF=BF=BD =EF=BF=BD* module to any device. For now= add the compatible property to > the > > + =EF=BF=BD =EF=BF=BD =EF=BF=BD* dtb of the devices we want to= use. =EF=BF=BD */ > > + =EF=BF=BD =EF=BF=BD { > > + =EF=BF=BD =EF=BF=BD =EF=BF=BD =EF=BF=BD =EF=BF=BD =EF=BF=BD = =2Ecompatible =3D "vfio-dt", > > + =EF=BF=BD =EF=BF=BD }, > > + =EF=BF=BD =EF=BF=BD {}, > > +}; >=20 > This definitely doesn't belong in the dt. It's purely a Linux > abstraction and does not represent a piece of hardware or common > interface. >=20 > We need to think of a better mechanism for binding the module to= these > devices now. >=20 > I already make this remark in the cover letter; thanks for confirm= ing it. > =EF=BF=BD > Antonios Sorry, I found the cover letter a little unclear in that regard. Thanks for the clarification :) Mark.