From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7E18EC7619A for ; Fri, 24 Mar 2023 01:56:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231127AbjCXB4u (ORCPT ); Thu, 23 Mar 2023 21:56:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43650 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229499AbjCXB4s (ORCPT ); Thu, 23 Mar 2023 21:56:48 -0400 Received: from mail-yb1-xb2d.google.com (mail-yb1-xb2d.google.com [IPv6:2607:f8b0:4864:20::b2d]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A533723C62 for ; Thu, 23 Mar 2023 18:56:47 -0700 (PDT) Received: by mail-yb1-xb2d.google.com with SMTP id r187so558624ybr.6 for ; Thu, 23 Mar 2023 18:56:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; t=1679623007; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:from:to:cc:subject:date:message-id:reply-to; bh=OsiMTy6N1RgTGOvTb4Xekz2TfgGYj4qNFGKLfiNrC5c=; b=i/OsODyRFwhKGCQVE5cOGTdtvHCCj1uWMzF2K26hcBSijC8Bs40+uJrOhLxFbYU5z6 6UAmtrLHZO3hGGsrgY0Qi6Wr24Solw5pnxAXpRqqejv1QKhj15Hnoe0y66GD5GogEMMJ uy/BlrCfRTtxAyb0tiLsLk8DeyDYABIs2p5N7H+u4ZnG5fOeBodeZYRDuFNa1XmGSSgF VVU+G15TSxO7Bj1pvQDJv+mWp7bhdosjZhCKqIThklR+vB6hKeCl17Jv47VDXzhfEL6e fkD4a/RjHjBwdn/9jK2eWZqlAwM7pVq8cLCzYln2IEk5vlmcC572oIqH5nT6eEGJ7Wjt q2Uw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1679623007; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=OsiMTy6N1RgTGOvTb4Xekz2TfgGYj4qNFGKLfiNrC5c=; b=wWinInuwrbNoFHtdu84etmgrA77GY760UURKnQhZoS5bY+WpsMs10J1vAycrfCC06V F9p4n2BmAo5sM1LzRlYTCbxFHcuD/5A5s4cn4/PRzCSzfn3JwUlFwzDxK8rIrvvha5X6 ZiCE/OY/SvkHayxgmv9vMkUgO39LDiheMDQGeVwltbdqS1IDdCx7cOfd9b3z/pd3ykU5 lDOsl9FFEGB3kI9s0LHt6l49qUubdr9WdfPqNdi7/H47bIOhhKcelGzkHZzHLYHhV+j3 qZZ45EhFM1riEiAwFSZPeJqJCntE8KlO0u/wjrVL9GZhQ0sKSy2vKLwJcV4e/E9rNfz0 tUtQ== X-Gm-Message-State: AAQBX9ctqmUFRAld5UpfO3gk2P+eE7KtREGovtSaiErm41X16MNli9rS gxj+LPSZvKPCLwSB2MSMVuhDppUi/i4+JngEfZ92aQ== X-Google-Smtp-Source: AKy350YpjYImsNOySKYjBd7nGIWFGk6giMK3LiQUNQ9zymyu1GFBcYRTvVx6oVVhLqLV/H1dUZLH/g== X-Received: by 2002:a25:ad86:0:b0:b26:91a2:61bc with SMTP id z6-20020a25ad86000000b00b2691a261bcmr734070ybi.37.1679623006756; Thu, 23 Mar 2023 18:56:46 -0700 (PDT) Received: from ripple.attlocal.net (172-10-233-147.lightspeed.sntcca.sbcglobal.net. [172.10.233.147]) by smtp.gmail.com with ESMTPSA id j83-20020a252356000000b00b7767ca7469sm178947ybj.6.2023.03.23.18.56.44 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 23 Mar 2023 18:56:45 -0700 (PDT) Date: Thu, 23 Mar 2023 18:56:34 -0700 (PDT) From: Hugh Dickins X-X-Sender: hugh@ripple.attlocal.net To: Song Liu cc: Matthew Wilcox , Hugh Dickins , Song Liu , Jiri Olsa , David Stevens , linux-mm@kvack.org, Andrew Morton , Peter Xu , "Kirill A . Shutemov" , Yang Shi , David Hildenbrand , Jiaqi Yan , linux-kernel@vger.kernel.org Subject: Re: [PATCH v5 3/3] mm/khugepaged: maintain page cache uptodate flag In-Reply-To: Message-ID: References: <20230307052036.1520708-1-stevensd@google.com> <20230307052036.1520708-4-stevensd@google.com> <866d1a75-d462-563-dfd7-1aa2971a285b@google.com> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="-1463760895-38525656-1679623005=:6497" Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---1463760895-38525656-1679623005=:6497 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE On Thu, 23 Mar 2023, Song Liu wrote: > On Thu, Mar 23, 2023 at 2:56=E2=80=AFPM Matthew Wilcox wrote: > > > > On Thu, Mar 23, 2023 at 12:07:46PM -0700, Hugh Dickins wrote: > > > On an earlier audit, for different reasons, I did also run across > > > lib/buildid.c build_id_parse() using find_get_page() without checking > > > PageUptodate() - looks as if it might do the wrong thing if it races > > > with khugepaged collapsing text to huge, and should probably have a > > > similar fix. > > > > That shouldn't be using find_get_page(). It should probably use > > read_cache_folio() which will actually read in the data if it's not > > present in the page cache, and return an ERR_PTR if the data couldn't > > be read. >=20 > build_id_parse() can be called from NMI, so I don't think we can let > read_cache_folio() read-in the data. Interesting. This being the same Layering_Violation_ID which is asking for a home in everyone's struct file? (Okay, I'm being disagreeable, no need to answer!) I think even the current find_get_page() is unsafe from NMI: imagine that NMI interrupting a sequence (maybe THP collapse or splitting, maybe page migration, maybe others) when the page refcount has been frozen to 0: you'll just have to reboot the machine? I guess the RCU-safety of find_get_page() implies that its XArray basics are safe in NMI; but you need a low-level variant (xas_find()?) which does none of the "goto retry"s, and fails immediately if anything is wrong - including !PageUptodate. Hugh ---1463760895-38525656-1679623005=:6497--