From mboxrd@z Thu Jan 1 00:00:00 1970 From: Radim =?utf-8?B?S3LEjW3DocWZ?= Subject: [PATCH] KVM: use slowpath for cross page cached accesses Date: Wed, 8 Apr 2015 14:16:48 +0200 Message-ID: <20150408121648.GA3519@potion.brq.redhat.com> References: <1428438897-22206-1-git-send-email-rkrcmar@redhat.com> <5524EBB7.3080906@redhat.com> <20150408092611.GA2164@potion.brq.redhat.com> <55250643.3090402@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org To: Paolo Bonzini Return-path: Content-Disposition: inline In-Reply-To: <55250643.3090402@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org 2015-04-08 12:43+0200, Paolo Bonzini: > On 08/04/2015 11:26, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: >> 2015-04-08 10:49+0200, Paolo Bonzini: >>> On 07/04/2015 22:34, Radim Kr=C4=8Dm=C3=A1=C5=99 wrote: >>>> We dirtied only one page because writes originally couldn't span m= ore. >>>> Use improved syntax for '>> PAGE_SHIFT' while at it. >>>> >>>> Fixes: 8f964525a121 ("KVM: Allow cross page reads and writes from = cached translations.") >>>> Signed-off-by: Radim Kr=C4=8Dm=C3=A1=C5=99 >>> >>> Cross-page reads and writes should never get here; they have >>> ghc->memslot set to NULL and go through the slow path in kvm_write_= guest. >>=20 >> Only cross-memslot writes have NULL memslot. >=20 > The power of wrong comments... >=20 > Considering how kvm_gfn_to_hva_cache_init is used (one 1-byte field, = two > 4-byte fields, one 28-bytes struct that is 32-bytes aligned, one > 32-bytes field that is in practice cacheline-aligned), I wonder if we > should just use ghc->memslot =3D NULL for cross page writes. This wo= uld > bypass the bug you are fixing here, and avoid worries about partial w= rites. Good idea, and it could make those comments right :) (Though in general, I prefer less constraints on APIs ...) Partial writes would be a pain; copy_to_user API does not define which bytes were not written. I think the write can't fail mid-page, which makes our implementation ok, but I still worry a bit about it. Anyway, here's the patch: ---8<--- kvm_write_guest_cached() does not mark all written pages as dirty and code comments in kvm_gfn_to_hva_cache_init() talk about NULL memslot with cross page accesses. Fix all the easy way. The check is '<=3D 1' to have the same result for 'len =3D 0' cache any= where in the page. (nr_pages_needed is 0 on page boundary.) =46ixes: 8f964525a121 ("KVM: Allow cross page reads and writes from cac= hed translations.") Signed-off-by: Radim Kr=C4=8Dm=C3=A1=C5=99 --- virt/kvm/kvm_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index aadef264bed1..f3dc641f9640 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1637,8 +1637,8 @@ int kvm_gfn_to_hva_cache_init(struct kvm *kvm, st= ruct gfn_to_hva_cache *ghc, ghc->generation =3D slots->generation; ghc->len =3D len; ghc->memslot =3D gfn_to_memslot(kvm, start_gfn); - ghc->hva =3D gfn_to_hva_many(ghc->memslot, start_gfn, &nr_pages_avail= ); - if (!kvm_is_error_hva(ghc->hva) && nr_pages_avail >=3D nr_pages_neede= d) { + ghc->hva =3D gfn_to_hva_many(ghc->memslot, start_gfn, NULL); + if (!kvm_is_error_hva(ghc->hva) && nr_pages_needed <=3D 1) { ghc->hva +=3D offset; } else { /*