From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ozlabs.org (ozlabs.org [IPv6:2401:3900:2:1::2]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 41Jw0y0vlHzF1D8 for ; Mon, 2 Jul 2018 14:55:14 +1000 (AEST) Date: Mon, 2 Jul 2018 14:52:43 +1000 From: David Gibson To: Alexey Kardashevskiy Cc: linuxppc-dev@lists.ozlabs.org, kvm-ppc@vger.kernel.org, Alex Williamson , Paul Mackerras Subject: Re: [PATCH kernel v2 2/2] KVM: PPC: Check if IOMMU page is contained in the pinned physical page Message-ID: <20180702045243.GX3422@umbus.fritz.box> References: <20180626055926.27703-1-aik@ozlabs.ru> <20180626055926.27703-3-aik@ozlabs.ru> <20180629041241.GC3422@umbus.fritz.box> <20180629145121.5d03e067@aik.ozlabs.ibm.com> <20180629045702.GI3422@umbus.fritz.box> <20180629151820.461ae112@aik.ozlabs.ibm.com> <20180629170747.471bea35@aik.ozlabs.ibm.com> <20180702040852.GW3422@umbus.fritz.box> <20180702143244.3a08ebe2@aik.ozlabs.ibm.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="Weqbz6OltVu8XOAh" In-Reply-To: <20180702143244.3a08ebe2@aik.ozlabs.ibm.com> List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , --Weqbz6OltVu8XOAh Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Jul 02, 2018 at 02:33:30PM +1000, Alexey Kardashevskiy wrote: > On Mon, 2 Jul 2018 14:08:52 +1000 > David Gibson wrote: >=20 > > On Fri, Jun 29, 2018 at 05:07:47PM +1000, Alexey Kardashevskiy wrote: > > > On Fri, 29 Jun 2018 15:18:20 +1000 > > > Alexey Kardashevskiy wrote: > > > =20 > > > > On Fri, 29 Jun 2018 14:57:02 +1000 > > > > David Gibson wrote: > > > > =20 > > > > > On Fri, Jun 29, 2018 at 02:51:21PM +1000, Alexey Kardashevskiy wr= ote: =20 > > > > > > On Fri, 29 Jun 2018 14:12:41 +1000 > > > > > > David Gibson wrote: > > > > > > =20 > > > > > > > On Tue, Jun 26, 2018 at 03:59:26PM +1000, Alexey Kardashevski= y wrote: =20 > > > > > > > > We already have a check in drivers/vfio/vfio_iommu_spapr_tc= e.c that > > > > > > > > an IOMMU page is contained in the physical page so the PCI = hardware won't > > > > > > > > get access to unassigned host memory. > > > > > > > >=20 > > > > > > > > However we do not have this check in KVM fastpath (H_PUT_TC= E accelerated > > > > > > > > code) so the user space can pin memory backed with 64k page= s and create > > > > > > > > a hardware TCE table with a bigger page size. We were lucky= so far and > > > > > > > > did not hit this yet as the very first time the mapping hap= pens > > > > > > > > we do not have tbl::it_userspace allocated yet and fall bac= k to > > > > > > > > the userspace which in turn calls VFIO IOMMU driver and tha= t fails > > > > > > > > because of the check in vfio_iommu_spapr_tce.c which is rea= lly > > > > > > > > sustainable solution. > > > > > > > >=20 > > > > > > > > This stores the smallest preregistered page size in the pre= registered > > > > > > > > region descriptor and changes the mm_iommu_xxx API to check= this against > > > > > > > > the IOMMU page size. > > > > > > > >=20 > > > > > > > > Signed-off-by: Alexey Kardashevskiy > > > > > > > > --- > > > > > > > > Changes: > > > > > > > > v2: > > > > > > > > * explicitly check for compound pages before calling compou= nd_order() > > > > > > > >=20 > > > > > > > > --- > > > > > > > > The bug is: run QEMU _without_ hugepages (no -mempath) and = tell it to > > > > > > > > advertise 16MB pages to the guest; a typical pseries guest = will use 16MB > > > > > > > > for IOMMU pages without checking the mmu pagesize and this = will fail > > > > > > > > at https://git.qemu.org/?p=3Dqemu.git;a=3Dblob;f=3Dhw/vfio/= common.c;h=3Dfb396cf00ac40eb35967a04c9cc798ca896eed57;hb=3Drefs/heads/maste= r#l256 > > > > > > > >=20 > > > > > > > > With the change, mapping will fail in KVM and the guest wil= l print: > > > > > > > >=20 > > > > > > > > mlx5_core 0000:00:00.0: ibm,create-pe-dma-window(2027) 0 80= 00000 20000000 18 1f returned 0 (liobn =3D 0x80000001 starting addr =3D 800= 0000 0) > > > > > > > > mlx5_core 0000:00:00.0: created tce table LIOBN 0x80000001 = for /pci@800000020000000/ethernet@0 > > > > > > > > mlx5_core 0000:00:00.0: failed to map direct window for > > > > > > > > /pci@800000020000000/ethernet@0: -1 =20 > > > > > > >=20 > > > > > > > [snip] =20 > > > > > > > > @@ -124,7 +125,7 @@ long mm_iommu_get(struct mm_struct *mm,= unsigned long ua, unsigned long entries, > > > > > > > > struct mm_iommu_table_group_mem_t **pmem) > > > > > > > > { > > > > > > > > struct mm_iommu_table_group_mem_t *mem; > > > > > > > > - long i, j, ret =3D 0, locked_entries =3D 0; > > > > > > > > + long i, j, ret =3D 0, locked_entries =3D 0, pageshift; > > > > > > > > struct page *page =3D NULL; > > > > > > > > =20 > > > > > > > > mutex_lock(&mem_list_mutex); > > > > > > > > @@ -166,6 +167,8 @@ long mm_iommu_get(struct mm_struct *mm,= unsigned long ua, unsigned long entries, > > > > > > > > goto unlock_exit; > > > > > > > > } > > > > > > > > =20 > > > > > > > > + mem->pageshift =3D 30; /* start from 1G pages - the bigg= est we have */ =20 > > > > > > >=20 > > > > > > > What about 16G pages on an HPT system? =20 > > > > > >=20 > > > > > >=20 > > > > > > Below in the loop mem->pageshift will reduce to the biggest act= ual size > > > > > > which will be 16mb/64k/4k. Or remain 1GB if no memory is actual= ly > > > > > > pinned, no loss there. =20 > > > > >=20 > > > > > Are you saying that 16G IOMMU pages aren't supported? Or that th= ere's > > > > > some reason a guest can never use them? =20 > > > >=20 > > > >=20 > > > > ah, 16_G_, not _M_. My bad. I just never tried such huge pages, I w= ill > > > > lift the limit up to 64 then, easier this way. =20 > > >=20 > > >=20 > > > Ah, no, rather this as the upper limit: > > >=20 > > > mem->pageshift =3D ilog2(entries) + PAGE_SHIFT; =20 > >=20 > > I can't make sense of this comment in context. I see how you're > > computing the minimum page size in the reserved region. > >=20 > > My question is about what the "maximum minimum" is - the starting > > value from which you calculate. Currently it's 1G, but I can't > > immediately see a reason that 16G is impossible here. >=20 >=20 > 16GB is impossible if the chunk we are preregistering here is smaller > than that, for example, the entire guest ram is 4GB. Of course. Just like it was for 1GiB if you had a 512MiB guest, for example. I'm talking about a case where you have a guest that's >=3D16GiB and you *have* allocated 16GiB hugepages to back it. > If that is the > case and we try mapping a 16GB IOMMU page, this should fail as I do not > really know what happens to the memory between 4GB..16GB. >=20 > imho if not that, than 1<<64 would make a good upper limit. >=20 >=20 >=20 > >=20 > > > @entries here is a number of system pages being pinned in that > > > function. > > >=20 > > >=20 > > > =20 > > > > =20 > > > > > =20 > > > > > > > > for (i =3D 0; i < entries; ++i) { > > > > > > > > if (1 !=3D get_user_pages_fast(ua + (i << PAGE_SHIFT), > > > > > > > > 1/* pages */, 1/* iswrite */, &page)) { > > > > > > > > @@ -199,6 +202,11 @@ long mm_iommu_get(struct mm_struct *mm= , unsigned long ua, unsigned long entries, > > > > > > > > } > > > > > > > > } > > > > > > > > populate: > > > > > > > > + pageshift =3D PAGE_SHIFT; > > > > > > > > + if (PageCompound(page)) > > > > > > > > + pageshift +=3D compound_order(compound_head(page)); > > > > > > > > + mem->pageshift =3D min_t(unsigned int, mem->pageshift, p= ageshift); =20 > > > > > > >=20 > > > > > > > Why not make mem->pageshift and pageshift local the same type= to avoid > > > > > > > the min_t() ? =20 > > > > > >=20 > > > > > > I was under impression min() is deprecated (misinterpret checkp= atch.pl > > > > > > may be) and therefore did not pay attention to it. I can fix th= is and > > > > > > repost if there is no other question. =20 > > > > >=20 > > > > > Hm, it's possible. =20 > > > >=20 > > > > Nah, tried min(), compiles fine. =20 > > >=20 > > >=20 > > > =20 > >=20 > >=20 > >=20 >=20 >=20 >=20 --=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 --Weqbz6OltVu8XOAh Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEdfRlhq5hpmzETofcbDjKyiDZs5IFAls5r5kACgkQbDjKyiDZ s5I5cg/+N4hro1z18LyIkTCl+Fx++L7WB0aWhrk6fnTJidlmGVUotIq1NFKyXG5h UwMKpEKSTnkSLZsJT2ulRsw26dYk68FKghIf7sqPt3d4i7YoGby4S6ADIu1NLAuD fF1BsjmqhuQxuBwHP+ALIncvEE0a0p4XwBVa170kgcuUniOr/hxz1DEM7jgNiCEW enqLKDOm6fBPaO6SPvAqo1+JVt+MuaCFEEvFPVIJz0FrrJas2de7mIX0E8DKhuCm boBeFtcdKqQF82dDw3SA3U7qKORn0uX6xTy+UT+ZH7agBcUDeHrS3wZzE8l+qK9y /46YjbfLJCe0CvF5NTZ1nkw2atOYSgtOCdskCIxEZOO/Yo/hAyX33VJVqyNwSrre 5Vjh30ietQmw3vvFDS1OR5i7Lh9FmsUVNyuReWnmmLWe1yZ4zmAqFY9pb96fnjmh Mnt6RGT3WkNSX8AqtKvmDusxAv1KmRvUIPAC4AnJM/I8Y2EcG3dGS128dGQVRUiV DEJ5nk5s0TVtGOzHwmskZSMw87Jzd8YP2k/buigFBor7jQUFivsBtujElH6mhZcI iCIzkVZ0/FYVXzIk837+Zl2/2Q7hPihrNPwgP2kFZ/o5hguk8mtOn2fVwYldGfYC 4j89jv48R9191iRuDOkoa6PcdvzCDpXHxwJIp9m6jSnr/tIpjwk= =ofm0 -----END PGP SIGNATURE----- --Weqbz6OltVu8XOAh--