From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755950AbZDNNkO (ORCPT ); Tue, 14 Apr 2009 09:40:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753731AbZDNNj4 (ORCPT ); Tue, 14 Apr 2009 09:39:56 -0400 Received: from wf-out-1314.google.com ([209.85.200.173]:32912 "EHLO wf-out-1314.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752260AbZDNNjz convert rfc822-to-8bit (ORCPT ); Tue, 14 Apr 2009 09:39:55 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; b=F5bm4q8BO2zuFyF8XmEg5gTqqDd/SDHLWE+MPigsx0lveO2EjJUTXGdMpIBuXI8OiU lJ/xuaM9DHFk1CRIg+nQQHmobwmOXR8K8TAj+9Mm96srXYkJpNiHl3T4tUankbknV1bv 67A0BFylbIOfaChPSfROlhod34oUVmMhdHZns= MIME-Version: 1.0 In-Reply-To: <200904142225.10788.nickpiggin@yahoo.com.au> References: <20090414151204.C647.A69D9226@jp.fujitsu.com> <200904141925.46012.nickpiggin@yahoo.com.au> <2f11576a0904140502h295faf33qcea9a39ff7f230a5@mail.gmail.com> <200904142225.10788.nickpiggin@yahoo.com.au> Date: Tue, 14 Apr 2009 22:39:54 +0900 X-Google-Sender-Auth: ae0a55b4fab9f09f Message-ID: <2f11576a0904140639l426e137ewdc46296cdb377dd@mail.gmail.com> Subject: Re: [RFC][PATCH v3 1/6] mm: Don't unmap gup()ed page From: KOSAKI Motohiro To: Nick Piggin Cc: LKML , Linus Torvalds , Andrew Morton , Andrea Arcangeli , Jeff Moyer , linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, Hugh Dickins Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org >> >> @@ -547,7 +549,13 @@ int reuse_swap_page(struct page *page) >> >>                       SetPageDirty(page); >> >>               } >> >>       } >> >> -     return count == 1; >> >> + >> >> +     /* >> >> +      * If we can re-use the swap page _and_ the end >> >> +      * result has only one user (the mapping), then >> >> +      * we reuse the whole page >> >> +      */ >> >> +     return count + page_count(page) == 2; >> >>  } >> > >> > I guess this patch does work to close the read-side race, but I slightly don't >> > like using page_count for things like this. page_count can be temporarily >> > raised for reasons other than access through their user mapping. Swapcache, >> > page reclaim, LRU pagevecs, concurrent do_wp_page, etc. >> >> Yes, that's trade-off. >> your early decow also can misjudge and make unnecessary copy. > > Yes indeed it can. Although it would only ever do so in case of pages > that have had get_user_pages run against them previously, and not from > random interactions from any other parts of the kernel. Agreed. > I would be interested, using an anon vma field as you say for keeping > a gup count... it could potentially be used to avoid the extra copy. > But hmm, I don't have much time to go down that path so long as the > basic concept of my proposal is in question. ok, I try to make it. thanks. > +     if (PageSwapCache(page) && > +         page_count(page) != page_mapcount(page) + 2) { > +             ret = SWAP_FAIL; > +             goto out_unmap; > +     } > > Now if another thread does a get_user_pages_fast after it passes this > check, it can take a gup reference to the page which is now about to > be unmapped. Then after it is unmapped, if a wp fault is caused on the > page, then it will not be reused and thus you lose data as explained > in your big comment. Grrr, I lose. I've misunderstood get_user_pages_fast() also grab pte_lock. I must think it again. I guess you dislike get_user_page_fast() grab pte_lock too, right?