From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wanpeng Li Subject: Re: [PATCH v4] kvm: Fix page ageing bugs Date: Wed, 24 Sep 2014 15:20:26 +0800 Message-ID: <542270BA.4090708@gmail.com> References: <1411410865-3603-1-git-send-email-andreslc@google.com> <1411422882-16245-1-git-send-email-andreslc@google.com> <20140924022729.GA2889@kernel> <54226CEB.9080504@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable Cc: Gleb Natapov , Radim Krcmar , Rik van Riel , Andrew Morton , Andrea Arcangeli , Peter Feiner , kvm@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org To: Paolo Bonzini , Wanpeng Li , Andres Lagar-Cavilla Return-path: In-Reply-To: <54226CEB.9080504@redhat.com> Sender: owner-linux-mm@kvack.org List-Id: kvm.vger.kernel.org Hi Paolo, =E4=BA=8E 9/24/14, 3:04 PM, Paolo Bonzini =E5=86=99=E9=81=93: > Il 24/09/2014 04:27, Wanpeng Li ha scritto: >> Hi Andres, >> On Mon, Sep 22, 2014 at 02:54:42PM -0700, Andres Lagar-Cavilla wrote: >>> 1. We were calling clear_flush_young_notify in unmap_one, but we are >>> within an mmu notifier invalidate range scope. The spte exists no mor= e >>> (due to range_start) and the accessed bit info has already been >>> propagated (due to kvm_pfn_set_accessed). Simply call >>> clear_flush_young. >>> >>> 2. We clear_flush_young on a primary MMU PMD, but this may be mapped >>> as a collection of PTEs by the secondary MMU (e.g. during log-dirty). >>> This required expanding the interface of the clear_flush_young mmu >>> notifier, so a lot of code has been trivially touched. >>> >>> 3. In the absence of shadow_accessed_mask (e.g. EPT A bit), we emulat= e >>> the access bit by blowing the spte. This requires proper synchronizin= g >>> with MMU notifier consumers, like every other removal of spte's does. >>> >> [...] >>> --- >>> + BUG_ON(!shadow_accessed_mask); >>> >>> for (sptep =3D rmap_get_first(*rmapp, &iter); sptep; >>> sptep =3D rmap_get_next(&iter)) { >>> + struct kvm_mmu_page *sp; >>> + gfn_t gfn; >>> BUG_ON(!is_shadow_present_pte(*sptep)); >>> + /* From spte to gfn. */ >>> + sp =3D page_header(__pa(sptep)); >>> + gfn =3D kvm_mmu_page_get_gfn(sp, sptep - sp->spt); >>> >>> if (*sptep & shadow_accessed_mask) { >>> young =3D 1; >>> clear_bit((ffs(shadow_accessed_mask) - 1), >>> (unsigned long *)sptep); >>> } >>> + trace_kvm_age_page(gfn, slot, young); >> IIUC, all the rmapps in this for loop are against the same gfn which >> results in the above trace point dump the message duplicated. > You're right; Andres's patch "[PATCH] kvm/x86/mmu: Pass gfn and level t= o > rmapp callback" helps avoiding that. From Andres's patch "[PATCH] kvm/x86/mmu: Pass gfn and level to rmapp=20 callback" @@ -1410,25 +1421,20 @@ static int kvm_age_rmapp(struct kvm *kvm,=20 unsigned long *rmapp, for (sptep =3D rmap_get_first(*rmapp, &iter); sptep; sptep =3D rmap_get_next(&iter)) { - struct kvm_mmu_page *sp; - gfn_t gfn; BUG_ON(!is_shadow_present_pte(*sptep)); - /* From spte to gfn. */ - sp =3D page_header(__pa(sptep)); - gfn =3D kvm_mmu_page_get_gfn(sp, sptep - sp->spt); - if (*sptep & shadow_accessed_mask) { young =3D 1; clear_bit((ffs(shadow_accessed_mask) - 1), (unsigned long *)sptep); } - trace_kvm_age_page(gfn, slot, young); + trace_kvm_age_page(gfn, level, slot, young); } return young; } This trace point still dup duplicated message for the same gfn in the=20 for loop. Regards, Wanpeng Li > > Paolo > -- > 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/ -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org