From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wanpeng Li Subject: Re: [PATCH] KVM: use slowpath for cross page cached accesses Date: Thu, 9 Apr 2015 08:18:07 +0800 Message-ID: <20150409001807.GA5146@kernel> References: <1428438897-22206-1-git-send-email-rkrcmar@redhat.com> <5524EBB7.3080906@redhat.com> <20150408092611.GA2164@potion.brq.redhat.com> <55250643.3090402@redhat.com> <20150408121648.GA3519@potion.brq.redhat.com> Reply-To: Wanpeng Li Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Paolo Bonzini , linux-kernel@vger.kernel.org, kvm@vger.kernel.org To: Radim =?utf-8?B?S3LEjW3DocWZ?= Return-path: Content-Disposition: inline In-Reply-To: <20150408121648.GA3519@potion.brq.redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On Wed, Apr 08, 2015 at 02:16:48PM +0200, Radim Kr=C4=8Dm=C3=A1=C5=99 w= rote: >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 = more. >>>>> 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 w= e >> should just use ghc->memslot =3D NULL for cross page writes. This w= ould >> bypass the bug you are fixing here, and avoid worries about partial = writes. > >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 whic= h >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 an= ywhere >in the page. (nr_pages_needed is 0 on page boundary.) > >Fixes: 8f964525a121 ("KVM: Allow cross page reads and writes from cach= ed translations.") >Signed-off-by: Radim Kr=C4=8Dm=C3=A1=C5=99 Reviewed-by: Wanpeng Li >--- > 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, s= truct 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_avai= l); >- if (!kvm_is_error_hva(ghc->hva) && nr_pages_avail >=3D nr_pages_need= ed) { >+ 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 { > /* >-- >To unsubscribe from this list: send the line "unsubscribe linux-kernel= " in >the body of a message to majordomo@vger.kernel.org >More majordomo info at http://vger.kernel.org/majordomo-info.html >Please read the FAQ at http://www.tux.org/lkml/