From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Gibson Subject: Re: [PATCH V4 05/18] iommu/ioasid: Redefine IOASID set and allocation APIs Date: Wed, 28 Apr 2021 10:49:03 +1000 Message-ID: References: <20210421230301.GP1370958@nvidia.com> <20210422111337.6ac3624d@redhat.com> <20210422175715.GA1370958@nvidia.com> <20210422133747.23322269@redhat.com> <20210422200024.GC1370958@nvidia.com> <20210422163808.2d173225@redhat.com> <20210422233950.GD1370958@nvidia.com> <20210427163954.GC1370958@nvidia.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="JQRVQfh5FhbBFvT+" Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gibson.dropbear.id.au; s=201602; t=1619577710; bh=gE3cwxeK8l52V3mXQ8Xznwj1fLwrrrjLu8KWJT/DX9U=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=SflghnlghzACtBYRzPNbsCJ+dAV56XMiDSEKz9RA3l5nSe+zufSUmI3DT5EaR/A3T BaKjdbudVH0aFttmCtWqp5/X40LE0tVirFJvWdJVJNqXw6uDi46p2SUUxuf/6kS/HU f/ygGsxHlo7TvpMhoX7/NxFIL1i96S+MJgESlHkE= Content-Disposition: inline In-Reply-To: <20210427163954.GC1370958@nvidia.com> List-ID: To: Jason Gunthorpe Cc: "Tian, Kevin" , Alex Williamson , "Liu, Yi L" , Jacob Pan , Auger Eric , Jean-Philippe Brucker , LKML , Joerg Roedel , Lu Baolu , David Woodhouse , "iommu@lists.linux-foundation.org" , "cgroups@vger.kernel.org" , Tejun Heo , Li Zefan , Johannes Weiner , Jean-Philippe Brucker , Jonathan Corbet , "Raj, Ashok" , "Wu, Hao" --JQRVQfh5FhbBFvT+ Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Apr 27, 2021 at 01:39:54PM -0300, Jason Gunthorpe wrote: > On Tue, Apr 27, 2021 at 03:11:25PM +1000, David Gibson wrote: >=20 > > > So your proposal sort of moves the entire container/group/domain=20 > > > managment into /dev/ioasid and then leaves vfio only provide device > > > specific uAPI. An ioasid represents a page table (address space), thu= s=20 > > > is equivalent to the scope of VFIO container. > >=20 > > Right. I don't really know how /dev/iosasid is supposed to work, and > > so far I don't see how it conceptually differs from a container. What > > is it adding? >=20 > There are three motivating topics: > 1) /dev/vfio/vfio is only usable by VFIO and we have many interesting > use cases now where we need the same thing usable outside VFIO > 2) /dev/vfio/vfio does not support modern stuff like PASID and > updating to support that is going to be a big change, like adding > multiple IOASIDs so they can be modeled as as a tree inside a > single FD > 3) I understand there is some desire to revise the uAPI here a bit, > ie Alex mentioned the poor mapping performance. >=20 > I would say it is not conceptually different from what VFIO calls a > container, it is just a different uAPI with the goal to be cross > subsystem. Ok, that makes sense. --=20 David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson --JQRVQfh5FhbBFvT+ Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEdfRlhq5hpmzETofcbDjKyiDZs5IFAmCIsP8ACgkQbDjKyiDZ s5LRgBAAnUYT4mzHS9z12bEDDjxIJlNISL9eC8ubKY8SIUm+mchxLYgEY3wAkoJL eFglIjrPvJD6R2vnpBVT9z8SNbHHULxTcVcLbhb1M0Qf5TrzniLGaBfdJbrgxomz lMTL3xTJ/djQcAeE4ulfQheA5Rh5OMmEmrFgzUUkI0y2aHK04YCyjxnxFYMSmIje ECTKFFl4sCwH6d7mDfOvz0CcelTIuI4OvpaFPw8jdKCv8rhF56C0En63sSdrGDIm iU2d/im/d4d7cnQfUHtcdw9neI4ia2NNYf2tIzLz2Ow9TBavv3VaeHMjUKQueKTE Adk6pdb9yZxeKurPYFoU9Cm1mhi4zbd+K48TGAexE57xCDWSEZcr44nM4ALv8YjC 6ZCU67IEM+J/uM0XYyv99G/ctqQeyQRRw9VtxT2sM/qBiI3S+t+S80LcFXV63La9 uP/5VvEe+C5+yMatJxFkdMQ92djhlFqFFnu84vhJnQX0U9zJvEx2XvSpoc/cxOUj uORn3Pex3YVOoh8SSIM8eJOEZxYy/Zk0EMZgP1Ub0zgD2PaS1DUqYgkVNslUZ1qs qhkZttZsxvwsPKBLXoSSarYj6h25ic6a8vGr/GMPxZJ6EDHuivzkqsJY0owcdIus AlCuTBNJvmY/5OUSao/MZ/XrbOXP/hCgYagOjMsTsudi3e/WYp8= =hSc/ -----END PGP SIGNATURE----- --JQRVQfh5FhbBFvT+--