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 kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 472CFC48BC3 for ; Tue, 20 Feb 2024 01:11:10 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 5D6E36B0071; Mon, 19 Feb 2024 20:11:09 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 5854F6B0072; Mon, 19 Feb 2024 20:11:09 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 44D446B0074; Mon, 19 Feb 2024 20:11:09 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id 368FB6B0071 for ; Mon, 19 Feb 2024 20:11:09 -0500 (EST) Received: from smtpin03.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay08.hostedemail.com (Postfix) with ESMTP id 0182A140397 for ; Tue, 20 Feb 2024 01:11:08 +0000 (UTC) X-FDA: 81810403458.03.034DD49 Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.12]) by imf03.hostedemail.com (Postfix) with ESMTP id C7F6C20018 for ; Tue, 20 Feb 2024 01:11:05 +0000 (UTC) Authentication-Results: imf03.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=ERQMak2u; spf=pass (imf03.hostedemail.com: domain of ying.huang@intel.com designates 192.198.163.12 as permitted sender) smtp.mailfrom=ying.huang@intel.com; dmarc=pass (policy=none) header.from=intel.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1708391466; a=rsa-sha256; cv=none; b=GSjZl+MSkFSYFBfIXeWfoJTdy89oIU8NV1nG5o55vEpQIu6m5Qlf158ZA8PBT8+BNK0+O8 YKvSSjNhyDfsjN4g+QTNN8zqVVNWLo8rGDHMnhR8o5K+yIITMpFtBEj4TsyRV4f6cMdeJ9 8BTEqBSqRij8EY7KKaSqgEwPluhkEhI= ARC-Authentication-Results: i=1; imf03.hostedemail.com; dkim=pass header.d=intel.com header.s=Intel header.b=ERQMak2u; spf=pass (imf03.hostedemail.com: domain of ying.huang@intel.com designates 192.198.163.12 as permitted sender) smtp.mailfrom=ying.huang@intel.com; dmarc=pass (policy=none) header.from=intel.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1708391466; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=s1NEbIDJUwAe++nZl4QO24+GpDnfSy3LX7NEzIBB/lI=; b=ObXeC5lG1/hwCb/mNBX4qwutRb+5xfuvBiPqJy2CY9KFPB160P8NlanQ5aUPFptv0FIwZh WJAuyj4V0Qch4FAi0Wv+qaUI49HvN3Nu8tjRlbMyUoPpKEzPQ3kdahpUcUBVoHuk0GIaVg 2qfL+QXwKQSTqizsFX8o3U4KEmXtLSo= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1708391466; x=1739927466; h=from:to:cc:subject:in-reply-to:references:date: message-id:mime-version:content-transfer-encoding; bh=zbwGqBLzRD3fZTXy/MlmyAoiH93kbhAQ+M07A+hLbNU=; b=ERQMak2usDbJ2NuXN1ZLzp+Yb+keZcU/4NnGnWukOa0VPifXMY8/3nip YdRLLItiIxI9OMXG1xsh1jxN4OwAqbVrqpz2FJLXf1AN1z1Cm0CR7tHpC lD/xYynjpV/XUVfvhRm4MfsgpuggVdtmNLU2uEljqcSMaP63Mpa3hcd/q ZFa6ZTduirlZFK6DPy5B6qUVmvErSGpwaXQhHyr1o7PTKkZ+Sr9+1OW0G SthNKL7qB7aWtwO48C3tYpRCg1+SALFLGlj2L2WpMJaRmmrcpph8t06As 4Du4hVMQdA5MrkrgcKMw74gKSIThJQWuSw0yOA2T+XYkRcaDD9UDNNFu5 Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10989"; a="6250895" X-IronPort-AV: E=Sophos;i="6.06,171,1705392000"; d="scan'208";a="6250895" Received: from fmviesa002.fm.intel.com ([10.60.135.142]) by fmvoesa106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 19 Feb 2024 17:11:04 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.06,171,1705392000"; d="scan'208";a="27788054" Received: from yhuang6-desk2.sh.intel.com (HELO yhuang6-desk2.ccr.corp.intel.com) ([10.238.208.55]) by fmviesa002-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 19 Feb 2024 17:10:59 -0800 From: "Huang, Ying" To: Barry Song <21cnbao@gmail.com> Cc: Kairui Song , linux-mm@kvack.org, Andrew Morton , Chris Li , Minchan Kim , Barry Song , Yu Zhao , SeongJae Park , David Hildenbrand , Hugh Dickins , Johannes Weiner , Matthew Wilcox , Michal Hocko , Yosry Ahmed , Aaron Lu , stable@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH v4] mm/swap: fix race when skipping swapcache In-Reply-To: (Barry Song's message of "Tue, 20 Feb 2024 11:10:03 +1300") References: <20240219082040.7495-1-ryncsn@gmail.com> Date: Tue, 20 Feb 2024 09:09:03 +0800 Message-ID: <87frxoq6vk.fsf@yhuang6-desk2.ccr.corp.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Rspamd-Server: rspam08 X-Rspamd-Queue-Id: C7F6C20018 X-Stat-Signature: mxf5yug1ohijoe3g63pqkqijyyogtaxk X-Rspam-User: X-HE-Tag: 1708391465-574274 X-HE-Meta: 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 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: Barry Song <21cnbao@gmail.com> writes: > On Mon, Feb 19, 2024 at 9:21=E2=80=AFPM Kairui Song wr= ote: >> >> From: Kairui Song >> >> When skipping swapcache for SWP_SYNCHRONOUS_IO, if two or more threads >> swapin the same entry at the same time, they get different pages (A, B). >> Before one thread (T0) finishes the swapin and installs page (A) >> to the PTE, another thread (T1) could finish swapin of page (B), >> swap_free the entry, then swap out the possibly modified page >> reusing the same entry. It breaks the pte_same check in (T0) because >> PTE value is unchanged, causing ABA problem. Thread (T0) will >> install a stalled page (A) into the PTE and cause data corruption. >> >> One possible callstack is like this: >> >> CPU0 CPU1 >> ---- ---- >> do_swap_page() do_swap_page() with same entry >> >> >> swap_read_folio() <- read to page A swap_read_folio() <- read to page B >> >> .. set_pte_at() >> swap_free() <- entry is free >> >> >> pte_same() <- Check pass, PTE seems >> unchanged, but page A >> is stalled! >> swap_free() <- page B content lost! >> set_pte_at() <- staled page A installed! >> >> And besides, for ZRAM, swap_free() allows the swap device to discard >> the entry content, so even if page (B) is not modified, if >> swap_read_folio() on CPU0 happens later than swap_free() on CPU1, >> it may also cause data loss. >> >> To fix this, reuse swapcache_prepare which will pin the swap entry using >> the cache flag, and allow only one thread to swap it in, also prevent >> any parallel code from putting the entry in the cache. Release the pin >> after PT unlocked. >> >> Racers just loop and wait since it's a rare and very short event. >> A schedule_timeout_uninterruptible(1) call is added to avoid repeated >> page faults wasting too much CPU, causing livelock or adding too much >> noise to perf statistics. A similar livelock issue was described in >> commit 029c4628b2eb ("mm: swap: get rid of livelock in swapin readahead") >> >> Reproducer: >> >> This race issue can be triggered easily using a well constructed >> reproducer and patched brd (with a delay in read path) [1]: >> >> With latest 6.8 mainline, race caused data loss can be observed easily: >> $ gcc -g -lpthread test-thread-swap-race.c && ./a.out >> Polulating 32MB of memory region... >> Keep swapping out... >> Starting round 0... >> Spawning 65536 workers... >> 32746 workers spawned, wait for done... >> Round 0: Error on 0x5aa00, expected 32746, got 32743, 3 data loss! >> Round 0: Error on 0x395200, expected 32746, got 32743, 3 data loss! >> Round 0: Error on 0x3fd000, expected 32746, got 32737, 9 data loss! >> Round 0 Failed, 15 data loss! >> >> This reproducer spawns multiple threads sharing the same memory region >> using a small swap device. Every two threads updates mapped pages one by >> one in opposite direction trying to create a race, with one dedicated >> thread keep swapping out the data out using madvise. >> >> The reproducer created a reproduce rate of about once every 5 minutes, >> so the race should be totally possible in production. >> >> After this patch, I ran the reproducer for over a few hundred rounds >> and no data loss observed. >> >> Performance overhead is minimal, microbenchmark swapin 10G from 32G >> zram: >> >> Before: 10934698 us >> After: 11157121 us >> Cached: 13155355 us (Dropping SWP_SYNCHRONOUS_IO flag) >> >> Fixes: 0bcac06f27d7 ("mm, swap: skip swapcache for swapin of synchronous= device") >> Link: https://github.com/ryncsn/emm-test-project/tree/master/swap-stress= -race [1] >> Reported-by: "Huang, Ying" >> Closes: https://lore.kernel.org/lkml/87bk92gqpx.fsf_-_@yhuang6-desk2.ccr= .corp.intel.com/ >> Signed-off-by: Kairui Song >> Cc: stable@vger.kernel.org >> >> --- >> V3: https://lore.kernel.org/all/20240216095105.14502-1-ryncsn@gmail.com/ >> Update from V3: >> - Use schedule_timeout_uninterruptible(1) for now instead of schedule() = to >> prevent the busy faulting task holds CPU and livelocks [Huang, Ying] >> >> V2: https://lore.kernel.org/all/20240206182559.32264-1-ryncsn@gmail.com/ >> Update from V2: >> - Add a schedule() if raced to prevent repeated page faults wasting CPU >> and add noise to perf statistics. >> - Use a bool to state the special case instead of reusing existing >> variables fixing error handling [Minchan Kim]. >> >> V1: https://lore.kernel.org/all/20240205110959.4021-1-ryncsn@gmail.com/ >> Update from V1: >> - Add some words on ZRAM case, it will discard swap content on swap_free >> so the race window is a bit different but cure is the same. [Barry Son= g] >> - Update comments make it cleaner [Huang, Ying] >> - Add a function place holder to fix CONFIG_SWAP=3Dn built [SeongJae Par= k] >> - Update the commit message and summary, refer to SWP_SYNCHRONOUS_IO >> instead of "direct swapin path" [Yu Zhao] >> - Update commit message. >> - Collect Review and Acks. >> >> include/linux/swap.h | 5 +++++ >> mm/memory.c | 20 ++++++++++++++++++++ >> mm/swap.h | 5 +++++ >> mm/swapfile.c | 13 +++++++++++++ >> 4 files changed, 43 insertions(+) >> >> diff --git a/include/linux/swap.h b/include/linux/swap.h >> index 4db00ddad261..8d28f6091a32 100644 >> --- a/include/linux/swap.h >> +++ b/include/linux/swap.h >> @@ -549,6 +549,11 @@ static inline int swap_duplicate(swp_entry_t swp) >> return 0; >> } >> >> +static inline int swapcache_prepare(swp_entry_t swp) >> +{ >> + return 0; >> +} >> + >> static inline void swap_free(swp_entry_t swp) >> { >> } >> diff --git a/mm/memory.c b/mm/memory.c >> index 7e1f4849463a..a99f5e7be9a5 100644 >> --- a/mm/memory.c >> +++ b/mm/memory.c >> @@ -3799,6 +3799,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) >> struct page *page; >> struct swap_info_struct *si =3D NULL; >> rmap_t rmap_flags =3D RMAP_NONE; >> + bool need_clear_cache =3D false; >> bool exclusive =3D false; >> swp_entry_t entry; >> pte_t pte; >> @@ -3867,6 +3868,20 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) >> if (!folio) { >> if (data_race(si->flags & SWP_SYNCHRONOUS_IO) && >> __swap_count(entry) =3D=3D 1) { >> + /* >> + * Prevent parallel swapin from proceeding with >> + * the cache flag. Otherwise, another thread may >> + * finish swapin first, free the entry, and swap= out >> + * reusing the same entry. It's undetectable as >> + * pte_same() returns true due to entry reuse. >> + */ >> + if (swapcache_prepare(entry)) { >> + /* Relax a bit to prevent rapid repeated= page faults */ >> + schedule_timeout_uninterruptible(1); > > Not a ideal model, imaging two tasks, > > task A - low priority running on a LITTLE core > task B - high priority and have real-time requirements such as audio, > graphics running on a big core. > > The original code will make B win even if it is a bit later than A as its= CPU is > much faster to finish swap_read_folio for example from zRAM. task B's > swap-in can finish very soon. > > With the patch, B will wait a tick and its real-time performance will be > negatively affected from time to time once low priority and high priority > tasks fault in the same PTE and high priority tasks are a bit later than > low priority tasks. This is a kind of priority inversion. > > When we support large folio swap-in, things can get worse. For example, > to swap-in 16 or even more pages in one do_swap_page, the chance for > task A and task B located in the same range of 16 PTEs will increase > though they are not located in the same PTE. > > Please consider this is not a blocker for this patch. But I will put the = problem > in my list and run some real tests on Android phones later. Yes. This may hurt performance. But this is necessary to solve a livelock problem similar as commit 029c4628b2eb ("mm: swap: get rid of livelock in swapin readahead"). Please consider that too. -- Best Regards, Huang, Ying >> + goto out; >> + } >> + need_clear_cache =3D true; >> + >> /* skip swapcache */ >> folio =3D vma_alloc_folio(GFP_HIGHUSER_MOVABLE, = 0, >> vma, vmf->address, false= ); >> @@ -4117,6 +4132,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) >> if (vmf->pte) >> pte_unmap_unlock(vmf->pte, vmf->ptl); >> out: >> + /* Clear the swap cache pin for direct swapin after PTL unlock */ >> + if (need_clear_cache) >> + swapcache_clear(si, entry); >> if (si) >> put_swap_device(si); >> return ret; >> @@ -4131,6 +4149,8 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) >> folio_unlock(swapcache); >> folio_put(swapcache); >> } >> + if (need_clear_cache) >> + swapcache_clear(si, entry); >> if (si) >> put_swap_device(si); >> return ret; >> diff --git a/mm/swap.h b/mm/swap.h >> index 758c46ca671e..fc2f6ade7f80 100644 >> --- a/mm/swap.h >> +++ b/mm/swap.h >> @@ -41,6 +41,7 @@ void __delete_from_swap_cache(struct folio *folio, >> void delete_from_swap_cache(struct folio *folio); >> void clear_shadow_from_swap_cache(int type, unsigned long begin, >> unsigned long end); >> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry); >> struct folio *swap_cache_get_folio(swp_entry_t entry, >> struct vm_area_struct *vma, unsigned long addr); >> struct folio *filemap_get_incore_folio(struct address_space *mapping, >> @@ -97,6 +98,10 @@ static inline int swap_writepage(struct page *p, stru= ct writeback_control *wbc) >> return 0; >> } >> >> +static inline void swapcache_clear(struct swap_info_struct *si, swp_ent= ry_t entry) >> +{ >> +} >> + >> static inline struct folio *swap_cache_get_folio(swp_entry_t entry, >> struct vm_area_struct *vma, unsigned long addr) >> { >> diff --git a/mm/swapfile.c b/mm/swapfile.c >> index 556ff7347d5f..746aa9da5302 100644 >> --- a/mm/swapfile.c >> +++ b/mm/swapfile.c >> @@ -3365,6 +3365,19 @@ int swapcache_prepare(swp_entry_t entry) >> return __swap_duplicate(entry, SWAP_HAS_CACHE); >> } >> >> +void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry) >> +{ >> + struct swap_cluster_info *ci; >> + unsigned long offset =3D swp_offset(entry); >> + unsigned char usage; >> + >> + ci =3D lock_cluster_or_swap_info(si, offset); >> + usage =3D __swap_entry_free_locked(si, offset, SWAP_HAS_CACHE); >> + unlock_cluster_or_swap_info(si, ci); >> + if (!usage) >> + free_swap_slot(entry); >> +} >> + >> struct swap_info_struct *swp_swap_info(swp_entry_t entry) >> { >> return swap_type_to_swap_info(swp_type(entry)); >> -- >> 2.43.0 >> >> > > Thanks > Barry