From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrea Arcangeli Subject: Re: [RFC][PATCH v3 1/6] mm: Don't unmap gup()ed page Date: Wed, 15 Apr 2009 13:41:54 +0200 Message-ID: <20090415114154.GI9809@random.random> References: <20090414143252.GE28265@random.random> <200904150042.15653.nickpiggin@yahoo.com.au> <20090415165431.AC4C.A69D9226@jp.fujitsu.com> <20090415104615.GG9809@random.random> <2f11576a0904150439k6e828307ja97b6729650bcb94@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Cc: Nick Piggin , LKML , Linus Torvalds , Andrew Morton , Jeff Moyer , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, Hugh Dickins To: KOSAKI Motohiro Return-path: Content-Disposition: inline In-Reply-To: <2f11576a0904150439k6e828307ja97b6729650bcb94@mail.gmail.com> Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Wed, Apr 15, 2009 at 08:39:04PM +0900, KOSAKI Motohiro wrote: > >> + =A0 =A0 if (!migration) { > >> + =A0 =A0 =A0 =A0 =A0 =A0 /* re-check */ > >> + =A0 =A0 =A0 =A0 =A0 =A0 if (PageSwapCache(page) && > >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 page_count(page) !=3D page_mapcoun= t(page) + 2) { > >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 /* We lose race against ge= t_user_pages_fast() */ > >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 set_pte_at(mm, address, pt= e, pteval); > >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 ret =3D SWAP_FAIL; > >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 goto out_unmap; > >> + =A0 =A0 =A0 =A0 =A0 =A0 } > >> + =A0 =A0 } > >> + =A0 =A0 mmu_notifier_invalidate_page(vma->vm_mm, address); > > > > With regard to mmu notifier, this is the opposite of the right > > ordering. One mmu_notifier_invalidate_page must run _before_ the firs= t > > check. The ptep_clear_flush_notify will then stay and there's no need > > of a further mmu_notifier_invalidate_page after the second check. >=20 > OK. but I have one question. >=20 > Can we assume mmu_notifier is only used by kvm now? > if not, we need to make new notifier. KVM is no fundamentally different from other users in this respect, so I don't see why need a new notifier. If it works for others it'll work for KVM and the other way around is true too. mmu notifier users can or cannot take a page pin. KVM does. GRU doesn't. XPMEM does. All of them releases any pin after mmu_notifier_invalidate_page. All that is important is to run mmu_notifier_invalidate_page _after_ the ptep_clear_young_notify, so that we don't nuke secondary mappings on the pages unless we really go to nuke the pte. -- 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