From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f170.google.com (mail-pf1-f170.google.com [209.85.210.170]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6EB6933436A for ; Wed, 10 Sep 2025 16:08:51 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.170 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520533; cv=none; b=GNu6kTtw8oc4LbEBqO2RS4ckObveD/oaK26wcQiD83j8Tlu6c9XO9sBQWjKMzQHxijo9wpxVh96dItqiQjpDQpTjzca75jfIAGZ9pOplGT/dPb82LJuxM5fT4YW0LftUUc5i8NfnJD4sa0UnEjE6iPWgv+eDkY+8RiH3e5bjXY8= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520533; c=relaxed/simple; bh=1RcgNooYVkUukgc0lvOVRhy9e13KiuRZjCpuMpTyTSk=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=dZ511r5bTacnNl7TWDWwksHyCBNAIGh6YTyUlfA4TfW0+/2gG7pC9Jy1+Srtzj9m7cpjcoGidZC3GygP8eBsvZSQ/Ul7w182J5XjXpLUmGf3KYYKFbHNauPzqfAn72jM6oy9mlP515VEz59VTwcEOecml+pURpr2KIsAlQ9zqmU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=I+aXVfId; arc=none smtp.client-ip=209.85.210.170 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="I+aXVfId" Received: by mail-pf1-f170.google.com with SMTP id d2e1a72fcca58-76e2ea933b7so868363b3a.1 for ; Wed, 10 Sep 2025 09:08:51 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520531; x=1758125331; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=+iG/DaIyXWsx1WWZmoqNKoMlpLoSTABw5yoCvRsMNiQ=; b=I+aXVfId8uJC7l7JJ2cI9hNv857p4ibTX7z9T5S1EDdTFWAvBtRUIopijY7L5qqr+H MVDlpjIJLoChbo6Nnadnb4hPG9CFIB0a/WOO77FcoZAef3tHqlNVv0x1waW7jHWuoGcj 40iL+gaRiN5dlY5pGVFCHOH+WoaMI1jqEYFfAMAU6zhZGXj7LXQnyawHbyj0AOCRdItl xUpxKkFs974e6+Xtt3K6dvtPSSb1kjXYdrmLm6t2KUvVVFPVzuVYHU226M4qurFsM4kk 22hnfGOI+i7XbyDOvRuZrgeE+DIX5RyabHkVpX5lRst+tm4pupP7Ta0Vo4TAf+toOTdF dFng== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520531; x=1758125331; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=+iG/DaIyXWsx1WWZmoqNKoMlpLoSTABw5yoCvRsMNiQ=; b=NvBTvqtJ83mVMKAG5Wl4d/9+s/hbuE/k11lGtixxlmDls74m7oNufjMcY1Aat7JJZx +8JxeoJ09QuIP2dvSifha1SJaYlXwfm1MdWE6LqFC9DGLOYSjBd4LnSqxbezgnMvVdHo 3OiQsFZ5cHCTI2slsYYOznTqbA95s5V21lBwzvptQHqs8gVYM9d2pFdpHUgOSduNOkOy qtUoNLvOeyhZ8VJNnn8Jf4PyJKjJBSHDTty1oiz6pOnC9vbGZ7juxKwJtgw096wkny1r DXHg784mkQRdfmMKj9Wva6cVjUBeIQDeR2GPTFCJiyvsDAEuAxfjZAxAywX2sRBlMu/A ykQg== X-Forwarded-Encrypted: i=1; AJvYcCWbXJf2HjkSy3DGHMWN1w5MOk7lZcz69Y/rlbwOZV1khXiSRqTVVY7ing9bdCSf8MI1Jl5wK1Uouua5BmA=@vger.kernel.org X-Gm-Message-State: AOJu0YznZDitteztuYqnTmr+kkxaAY9Hjki2WPbx9AHU47YLrDxMCe1x Llo/XkE1gdwrLyQhQqOMBfhxYtYQJw4TpYP0TNYhIerqgtdoOYvMP81J X-Gm-Gg: ASbGnctvUfwP9d/eBXLDIKDrTIse+6MG7qE0YR+RPB6xPDcT2ppXcPZhG//htycCBOR /6faayYWWqimv7ZzXE76X11Exu6NrNz1etrR4jq3cOMWVK9KCDV2VY8eI1orGUCQMIMwM8WxMCJ a56pZMj/TuRjMlNRhZawD+mZDwyGfLCfS4sCdeoIoSNwmGrYriT01aw4EBvHsWAA1ZzF6f7qHkA 1/NKyCumS3h+n32gP24d4SlD4Oyi8eYV5PEJdn7MVYO0CI/dc2Z4ZPELhZxlZA5h6nt4NkZ7qBF 6zTFGMNiUm9QEuy+sUSRq7yiHklxQWIqwmC36++47NCoAgoiRjhKW4mG4AwiUOqMkMC/Uj29GVV qDMR8QyK1tjfDRPYtcx6D08vsYaUTQX1izbcg06tLj87FmEE= X-Google-Smtp-Source: AGHT+IHGtESNtbI/xZsyuKB3DJLeJiycjaitn//SnhhsXY3KbjYMDyPR0k+gvTRPXGFd9JgFnAJd4g== X-Received: by 2002:a05:6a20:258e:b0:243:d3d3:61d4 with SMTP id adf61e73a8af0-25cf6c38869mr168609637.16.1757520530570; Wed, 10 Sep 2025 09:08:50 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.08.45 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:08:50 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 01/15] docs/mm: add document for swap table Date: Thu, 11 Sep 2025 00:08:19 +0800 Message-ID: <20250910160833.3464-2-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song From: Chris Li Swap table is the new swap cache. Signed-off-by: Chris Li Signed-off-by: Kairui Song Suggested-by: Chris Li --- Documentation/mm/index.rst | 1 + Documentation/mm/swap-table.rst | 72 +++++++++++++++++++++++++++++++++ MAINTAINERS | 1 + 3 files changed, 74 insertions(+) create mode 100644 Documentation/mm/swap-table.rst diff --git a/Documentation/mm/index.rst b/Documentation/mm/index.rst index fb45acba16ac..828ad9b019b3 100644 --- a/Documentation/mm/index.rst +++ b/Documentation/mm/index.rst @@ -57,6 +57,7 @@ documentation, or deleted if it has served its purpose. page_table_check remap_file_pages split_page_table_lock + swap-table transhuge unevictable-lru vmalloced-kernel-stacks diff --git a/Documentation/mm/swap-table.rst b/Documentation/mm/swap-table.= rst new file mode 100644 index 000000000000..acae6ceb4f7b --- /dev/null +++ b/Documentation/mm/swap-table.rst @@ -0,0 +1,72 @@ +.. SPDX-License-Identifier: GPL-2.0 + +:Author: Chris Li , Kairui Song + +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D +Swap Table +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D + +Swap table implements swap cache as a per-cluster swap cache value array. + +Swap Entry +---------- + +A swap entry contains the information required to serve the anonymous page +fault. + +Swap entry is encoded as two parts: swap type and swap offset. + +The swap type indicates which swap device to use. +The swap offset is the offset of the swap file to read the page data from. + +Swap Cache +---------- + +Swap cache is a map to look up folios using swap entry as the key. The res= ult +value can have three possible types depending on which stage of this swap = entry +was in. + +1. NULL: This swap entry is not used. + +2. folio: A folio has been allocated and bound to this swap entry. This is + the transient state of swap out or swap in. The folio data can be in + the folio or swap file, or both. + +3. shadow: The shadow contains the working set information of the swapped + out folio. This is the normal state for a swapped out page. + +Swap Table Internals +-------------------- + +The previous swap cache is implemented by XArray. The XArray is a tree +structure. Each lookup will go through multiple nodes. Can we do better? + +Notice that most of the time when we look up the swap cache, we are either +in a swap in or swap out path. We should already have the swap cluster, +which contains the swap entry. + +If we have a per-cluster array to store swap cache value in the cluster. +Swap cache lookup within the cluster can be a very simple array lookup. + +We give such a per-cluster swap cache value array a name: the swap table. + +Each swap cluster contains 512 entries, so a swap table stores one cluster +worth of swap cache values, which is exactly one page. This is not +coincidental because the cluster size is determined by the huge page size. +The swap table is holding an array of pointers. The pointer has the same +size as the PTE. The size of the swap table should match to the second +last level of the page table page, exactly one page. + +With swap table, swap cache lookup can achieve great locality, simpler, +and faster. + +Locking +------- + +Swap table modification requires taking the cluster lock. If a folio +is being added to or removed from the swap table, the folio must be +locked prior to the cluster lock. After adding or removing is done, the +folio shall be unlocked. + +Swap table lookup is protected by RCU and atomic read. If the lookup +returns a folio, the user must lock the folio before use. diff --git a/MAINTAINERS b/MAINTAINERS index 68d29f0220fc..3d113bfc3c82 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16225,6 +16225,7 @@ R: Barry Song R: Chris Li L: linux-mm@kvack.org S: Maintained +F: Documentation/mm/swap-table.rst F: include/linux/swap.h F: include/linux/swapfile.h F: include/linux/swapops.h --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f180.google.com (mail-pf1-f180.google.com [209.85.210.180]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8E5E6334720 for ; Wed, 10 Sep 2025 16:08:56 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.180 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520538; cv=none; b=nue0NzCbhEgcwCiPsNEuD+WZQdzA+LCJ58hmslqp3AEGx6vemtcWGzJBnQP04CMk6yDvK58EXouBShKHN3WVggiQ1mXh6Zhp05J7o29MNeSM6WiundFpfQY4FYTaW/HrTYIMChHuWLr3LVNPI2Fs8IBWlt1hfkmn1ab0wWPeEfo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520538; c=relaxed/simple; bh=xgBQSstXvBp8VeP+kZcAikb8t1kgokUoq5Ewfa+sYSk=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=gn32yK3G0PyMGIf6ruBGlv1szLzAyGJp+cgfbY/FZIzLOdC76qzyjGbvaiGZRWcleV9cY4HBp2vtNlV0zJfPk2b6DHzQaxjYSLrayaHhfAPjk85vLUOxTSIJ7IS1Anjfrh0w+/6EyJPlrFPrnrCcYuPudVmNihhJUyB9Tj1m/zc= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=YuX/t1Fu; arc=none smtp.client-ip=209.85.210.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="YuX/t1Fu" Received: by mail-pf1-f180.google.com with SMTP id d2e1a72fcca58-77246079bc9so8132134b3a.3 for ; Wed, 10 Sep 2025 09:08:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520536; x=1758125336; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=QBZjrSlGrCuoXEovfI0RqN6ge66DgK4hS3OmggQQcuM=; b=YuX/t1FuB7Ma+BaZ5vH6lZsY2NcnPceKgetzb1z0ZzNjKc1151mtqH26joT687m517 2lzcXn+Q+TDAU7gal7C2WQSoxkmYbsV3c6iCigePm8wz12ByvHRLKcc5NGe1gANMJZEd vW8792r6Wr4CcERjOhwcS0HG2OG59Z6DVt7HlVNWZR5ZluKG6GTG7Vms7HJ5Wc+3yKrK TwQH2XWvuv3j2Aak6FmqVROqw8qhbyckFbtY+uk3oIXF1Y+y2U35yx/Xd2DuwBbyYNGC gwiejiuvCs4aLVEurpv8jReMqtm/UTwW593m/aQ8NPTqBLkgxpkaUVDDxskwSVEMTfmb 3tRw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520536; x=1758125336; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=QBZjrSlGrCuoXEovfI0RqN6ge66DgK4hS3OmggQQcuM=; b=n9XE2JkQGsrvUtvu+dHy1VDd07sYABtsYy4IbOzsJGtYt71xu4M7tXwxhTGoGdfTSD QZr3PlJniYKQ0RTQ+qlgAbpUIosWfUU3l/jhBBLd6SHOaSUBOK6pIc5tqf1QwDS5lTAL gxY5JPDW8stWscoUzYOYkZ8NCrlPPtty3iIGbYUnJ6G1IfwlDfoZQt/jpz+ITzTglmFC bfyjE+6yPTSuyP2yOAtFjKO+sDEAuX5zBWNeAa5KkdC/P/b5Ul6vD0lAyRVRCTTk3sJ3 UMZAW7Sf0oMzShJyJw4+odJxfUCQVNAZfPHDiIdJcaY8d0U2IukUjyrIKvyvMYA8rSE6 6a3A== X-Forwarded-Encrypted: i=1; AJvYcCVY/8sxjXoJFBVICV+6b4s4RL6j/66fmInefYC0RbWiRtTWFXjIcIRlihy5eGyqpHwfY7FblANtiXjPlPc=@vger.kernel.org X-Gm-Message-State: AOJu0YyQ0hTSo/o/3TG78wOx7WoZyj7KkWd0IvNqbgL82poeikZWuDGD 1BB6nUQVrvxuxiDDpgI274qs6ccSnhFQWr1+csV8UmpFJB4VIAXb/nnc X-Gm-Gg: ASbGncvKzVObMkiXt9Z4qP93OV2E/5QZ5hU3aPRcOcDmi0KSMXaexPMhWifkOPF5FIt sArara9Sjj3OZAmHJ4JzwLT3nxaLaVMNyowPNnnE2eurBufEI9051X4PKO6m+YtOYlkpna4w7mz RHTPNDUSD5XFdZFj6gublcRHuOaXvxsT1LZKJwiNRebDPbaAQ16znTTJrX6/g2mPDHsvKcIYSK0 ZOFofqZdx4Aic2wOkVj6wSqkOEtvQqxTTX/G0RGbHEZwtWVr16PULbFNMcPJpJh+tbYFiCwGDWZ 6rOxXZIQ5D29KSBeAyzGAV+clCPLrqyuK2kzhgWvvP1WDpqNyR73ILAHxgYB7LdFi+LaTjFbdM1 nI8YQjv29rbkaeYQnFs52dV7Cj5L0NNBPCeW0+6qjo9PSZQ8LEym+MvhBCA== X-Google-Smtp-Source: AGHT+IGGmJewUeNY+nNFHeQ/nHWhKuHcuJNLNLUkFHUvOr8qFnybJf+g/nmOah+d5OTnLDQWDhYpXw== X-Received: by 2002:a05:6a20:158a:b0:24d:b11b:e8ff with SMTP id adf61e73a8af0-25344412ff4mr27412031637.32.1757520535592; Wed, 10 Sep 2025 09:08:55 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.08.50 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:08:55 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 02/15] mm, swap: use unified helper for swap cache look up Date: Thu, 11 Sep 2025 00:08:20 +0800 Message-ID: <20250910160833.3464-3-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song The swap cache lookup helper swap_cache_get_folio currently does readahead updates as well, so callers that are not doing swapin from any VMA or mapping are forced to reuse filemap helpers instead, and have to access the swap cache space directly. So decouple readahead update with swap cache lookup. Move the readahead update part into a standalone helper. Let the caller call the readahead update helper if they do readahead. And convert all swap cache lookups to use swap_cache_get_folio. After this commit, there are only three special cases for accessing swap cache space now: huge memory splitting, migration, and shmem replacing, because they need to lock the XArray. The following commits will wrap their accesses to the swap cache too, with special helpers. And worth noting, currently dropbehind is not supported for anon folio, and we will never see a dropbehind folio in swap cache. The unified helper can be updated later to handle that. While at it, add proper kernedoc for touched helpers. No functional change. Signed-off-by: Kairui Song Reviewed-by: Baolin Wang Reviewed-by: Barry Song Acked-by: David Hildenbrand Acked-by: Chris Li Acked-by: Nhat Pham Suggested-by: Chris Li --- mm/memory.c | 6 ++- mm/mincore.c | 3 +- mm/shmem.c | 4 +- mm/swap.h | 13 ++++-- mm/swap_state.c | 109 +++++++++++++++++++++++++---------------------- mm/swapfile.c | 11 +++-- mm/userfaultfd.c | 5 +-- 7 files changed, 81 insertions(+), 70 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index d9de6c056179..10ef528a5f44 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4660,9 +4660,11 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) if (unlikely(!si)) goto out; =20 - folio =3D swap_cache_get_folio(entry, vma, vmf->address); - if (folio) + folio =3D swap_cache_get_folio(entry); + if (folio) { + swap_update_readahead(folio, vma, vmf->address); page =3D folio_file_page(folio, swp_offset(entry)); + } swapcache =3D folio; =20 if (!folio) { diff --git a/mm/mincore.c b/mm/mincore.c index 2f3e1816a30d..8ec4719370e1 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -76,8 +76,7 @@ static unsigned char mincore_swap(swp_entry_t entry, bool= shmem) if (!si) return 0; } - folio =3D filemap_get_entry(swap_address_space(entry), - swap_cache_index(entry)); + folio =3D swap_cache_get_folio(entry); if (shmem) put_swap_device(si); /* The swap cache space contains either folio, shadow or NULL */ diff --git a/mm/shmem.c b/mm/shmem.c index 29e1eb690125..410f27bc4752 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2317,7 +2317,7 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, } =20 /* Look it up and read it in.. */ - folio =3D swap_cache_get_folio(swap, NULL, 0); + folio =3D swap_cache_get_folio(swap); if (!folio) { if (data_race(si->flags & SWP_SYNCHRONOUS_IO)) { /* Direct swapin skipping swap cache & readahead */ @@ -2342,6 +2342,8 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, count_vm_event(PGMAJFAULT); count_memcg_event_mm(fault_mm, PGMAJFAULT); } + } else { + swap_update_readahead(folio, NULL, 0); } =20 if (order > folio_order(folio)) { diff --git a/mm/swap.h b/mm/swap.h index 1ae44d4193b1..efb6d7ff9f30 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -62,8 +62,7 @@ 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, int n= r); -struct folio *swap_cache_get_folio(swp_entry_t entry, - struct vm_area_struct *vma, unsigned long addr); +struct folio *swap_cache_get_folio(swp_entry_t entry); struct folio *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, struct vm_area_struct *vma, unsigned long addr, struct swap_iocb **plug); @@ -74,6 +73,8 @@ struct folio *swap_cluster_readahead(swp_entry_t entry, g= fp_t flag, struct mempolicy *mpol, pgoff_t ilx); struct folio *swapin_readahead(swp_entry_t entry, gfp_t flag, struct vm_fault *vmf); +void swap_update_readahead(struct folio *folio, struct vm_area_struct *vma, + unsigned long addr); =20 static inline unsigned int folio_swap_flags(struct folio *folio) { @@ -159,6 +160,11 @@ static inline struct folio *swapin_readahead(swp_entry= _t swp, gfp_t gfp_mask, return NULL; } =20 +static inline void swap_update_readahead(struct folio *folio, + struct vm_area_struct *vma, unsigned long addr) +{ +} + static inline int swap_writeout(struct folio *folio, struct swap_iocb **swap_plug) { @@ -169,8 +175,7 @@ static inline void swapcache_clear(struct swap_info_str= uct *si, swp_entry_t entr { } =20 -static inline struct folio *swap_cache_get_folio(swp_entry_t entry, - struct vm_area_struct *vma, unsigned long addr) +static inline struct folio *swap_cache_get_folio(swp_entry_t entry) { return NULL; } diff --git a/mm/swap_state.c b/mm/swap_state.c index 99513b74b5d8..68ec531d0f2b 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -69,6 +69,27 @@ void show_swap_cache_info(void) printk("Total swap =3D %lukB\n", K(total_swap_pages)); } =20 +/** + * swap_cache_get_folio - Looks up a folio in the swap cache. + * @entry: swap entry used for the lookup. + * + * A found folio will be returned unlocked and with its refcount increased. + * + * Context: Caller must ensure @entry is valid and protect the swap device + * with reference count or locks. + * Return: Returns the found folio on success, NULL otherwise. The caller + * must lock and check if the folio still matches the swap entry before + * use. + */ +struct folio *swap_cache_get_folio(swp_entry_t entry) +{ + struct folio *folio =3D filemap_get_folio(swap_address_space(entry), + swap_cache_index(entry)); + if (IS_ERR(folio)) + return NULL; + return folio; +} + void *get_shadow_from_swap_cache(swp_entry_t entry) { struct address_space *address_space =3D swap_address_space(entry); @@ -272,55 +293,43 @@ static inline bool swap_use_vma_readahead(void) return READ_ONCE(enable_vma_readahead) && !atomic_read(&nr_rotate_swap); } =20 -/* - * Lookup a swap entry in the swap cache. A found folio will be returned - * unlocked and with its refcount incremented - we rely on the kernel - * lock getting page table operations atomic even if we drop the folio - * lock before returning. - * - * Caller must lock the swap device or hold a reference to keep it valid. +/** + * swap_update_readahead - Update the readahead statistics of VMA or globa= lly. + * @folio: the swap cache folio that just got hit. + * @vma: the VMA that should be updated, could be NULL for global update. + * @addr: the addr that triggered the swapin, ignored if @vma is NULL. */ -struct folio *swap_cache_get_folio(swp_entry_t entry, - struct vm_area_struct *vma, unsigned long addr) +void swap_update_readahead(struct folio *folio, struct vm_area_struct *vma, + unsigned long addr) { - struct folio *folio; - - folio =3D filemap_get_folio(swap_address_space(entry), swap_cache_index(e= ntry)); - if (!IS_ERR(folio)) { - bool vma_ra =3D swap_use_vma_readahead(); - bool readahead; + bool readahead, vma_ra =3D swap_use_vma_readahead(); =20 - /* - * At the moment, we don't support PG_readahead for anon THP - * so let's bail out rather than confusing the readahead stat. - */ - if (unlikely(folio_test_large(folio))) - return folio; - - readahead =3D folio_test_clear_readahead(folio); - if (vma && vma_ra) { - unsigned long ra_val; - int win, hits; - - ra_val =3D GET_SWAP_RA_VAL(vma); - win =3D SWAP_RA_WIN(ra_val); - hits =3D SWAP_RA_HITS(ra_val); - if (readahead) - hits =3D min_t(int, hits + 1, SWAP_RA_HITS_MAX); - atomic_long_set(&vma->swap_readahead_info, - SWAP_RA_VAL(addr, win, hits)); - } - - if (readahead) { - count_vm_event(SWAP_RA_HIT); - if (!vma || !vma_ra) - atomic_inc(&swapin_readahead_hits); - } - } else { - folio =3D NULL; + /* + * At the moment, we don't support PG_readahead for anon THP + * so let's bail out rather than confusing the readahead stat. + */ + if (unlikely(folio_test_large(folio))) + return; + + readahead =3D folio_test_clear_readahead(folio); + if (vma && vma_ra) { + unsigned long ra_val; + int win, hits; + + ra_val =3D GET_SWAP_RA_VAL(vma); + win =3D SWAP_RA_WIN(ra_val); + hits =3D SWAP_RA_HITS(ra_val); + if (readahead) + hits =3D min_t(int, hits + 1, SWAP_RA_HITS_MAX); + atomic_long_set(&vma->swap_readahead_info, + SWAP_RA_VAL(addr, win, hits)); } =20 - return folio; + if (readahead) { + count_vm_event(SWAP_RA_HIT); + if (!vma || !vma_ra) + atomic_inc(&swapin_readahead_hits); + } } =20 struct folio *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, @@ -336,14 +345,10 @@ struct folio *__read_swap_cache_async(swp_entry_t ent= ry, gfp_t gfp_mask, *new_page_allocated =3D false; for (;;) { int err; - /* - * First check the swap cache. Since this is normally - * called after swap_cache_get_folio() failed, re-calling - * that would confuse statistics. - */ - folio =3D filemap_get_folio(swap_address_space(entry), - swap_cache_index(entry)); - if (!IS_ERR(folio)) + + /* Check the swap cache in case the folio is already there */ + folio =3D swap_cache_get_folio(entry); + if (folio) goto got_folio; =20 /* diff --git a/mm/swapfile.c b/mm/swapfile.c index a7ffabbe65ef..4b8ab2cb49ca 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -213,15 +213,14 @@ static int __try_to_reclaim_swap(struct swap_info_str= uct *si, unsigned long offset, unsigned long flags) { swp_entry_t entry =3D swp_entry(si->type, offset); - struct address_space *address_space =3D swap_address_space(entry); struct swap_cluster_info *ci; struct folio *folio; int ret, nr_pages; bool need_reclaim; =20 again: - folio =3D filemap_get_folio(address_space, swap_cache_index(entry)); - if (IS_ERR(folio)) + folio =3D swap_cache_get_folio(entry); + if (!folio) return 0; =20 nr_pages =3D folio_nr_pages(folio); @@ -2131,7 +2130,7 @@ static int unuse_pte_range(struct vm_area_struct *vma= , pmd_t *pmd, pte_unmap(pte); pte =3D NULL; =20 - folio =3D swap_cache_get_folio(entry, vma, addr); + folio =3D swap_cache_get_folio(entry); if (!folio) { struct vm_fault vmf =3D { .vma =3D vma, @@ -2357,8 +2356,8 @@ static int try_to_unuse(unsigned int type) (i =3D find_next_to_unuse(si, i)) !=3D 0) { =20 entry =3D swp_entry(type, i); - folio =3D filemap_get_folio(swap_address_space(entry), swap_cache_index(= entry)); - if (IS_ERR(folio)) + folio =3D swap_cache_get_folio(entry); + if (!folio) continue; =20 /* diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 50aaa8dcd24c..af61b95c89e4 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -1489,9 +1489,8 @@ static long move_pages_ptes(struct mm_struct *mm, pmd= _t *dst_pmd, pmd_t *src_pmd * separately to allow proper handling. */ if (!src_folio) - folio =3D filemap_get_folio(swap_address_space(entry), - swap_cache_index(entry)); - if (!IS_ERR_OR_NULL(folio)) { + folio =3D swap_cache_get_folio(entry); + if (folio) { if (folio_test_large(folio)) { ret =3D -EBUSY; folio_put(folio); --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pj1-f42.google.com (mail-pj1-f42.google.com [209.85.216.42]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 351F5335BB7 for ; Wed, 10 Sep 2025 16:09:01 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.216.42 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520542; cv=none; b=pD5GPb4VQ9Zs0124Q/V5ubOpw6QqDkS+SQHh0tu0Mj/sf0PwcuUsM45NU7sYzjgPxNujK+bO40zaO+suwQw64eps6mlCl0cTarYBf9t82gMqgANq7XqPYtEl4XhQG0Pw8KheFuX8rsQd/ml3Cy81O5FoFy8YOJMHgjyGr9Q/rko= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520542; c=relaxed/simple; bh=Khh5987+1u6MVC/NKJ8ASpt3UuFIKNSR0pBiIhzVLUQ=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=atmuSu3o1mzqrEf05osasHOUL2A2XgNV1oCwmIfEL02ZyhP67iq+axPvUcSEV305vaGQ0lEM71pt4yUZWuXtnbLdKpyQKYwYs9FdoJcoEKp/48zkJhwZSnObWZxydAeKxGtUzyf8I3wK+mRjyLznjMGHyz5QCyXlPtdXaqpD6p4= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=KRC/dZij; arc=none smtp.client-ip=209.85.216.42 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="KRC/dZij" Received: by mail-pj1-f42.google.com with SMTP id 98e67ed59e1d1-32d3e17d927so4912418a91.2 for ; Wed, 10 Sep 2025 09:09:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520540; x=1758125340; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=xvXia67Cq9g+cEA2tdQsRO5nx9SORMf8wIBhDb/3W4o=; b=KRC/dZijvkM0r08bV52WxKoE+t94KIfQgDz5IElSkU7cM/RFUsF85hEfbp6vlx/c65 bDkaIE2cp2INehPaSye75JBZ6mTU/WZ7FJkFMBYJVpfHPPjKR9MNnBfVePUo5fn+rK7u E48J+ul7yIPMuWhFdcSA5g18bkhngAHhhECs+1vJWJSqUWxugJYVoxxzC8QEr4puFUI2 lTtyI6iKpTb+uPDxCdqtSgsB3aWX282UWpp3X9gaQsxlMa5CaC9TwwyiWrugB6EfWlNY a/g0pQS0IxJXRma7aPKTOkq7eXjemkEYPPDiQFimPEN3lumHllFFsOWnK/8OMfOL4Kex pCwA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520540; x=1758125340; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=xvXia67Cq9g+cEA2tdQsRO5nx9SORMf8wIBhDb/3W4o=; b=ei8OSdTP9gVj/NsfXIiqQYbcI3LgVWAt6qKmhIcyXDWYdNXjxo4AzwyL8Eaa63WvFj V5RJ0WReRE0e+EQ/GcXCboRGwKISA5VpyrbR2BNGQHIuCtTn1rIVSKhJyrQ9l7Fe2RKs HEGCfDQ7NWAmGUgR32ccyzHoO8QMxaySXzEAsubCYz0EPiIjsDnblsfqUsW32jThYPvi JipPSq2claF05GzHHf8PMjKFiyksviTP2arXSTIkr4jPrOW3ZInh2CVloxfR99EdhFXu yN+CXjluXM67vAeSIUizDThq0NnEXdHTDohMY0GVfCHVAnio0rJEetrmCkTcQaUYf84S 6cFA== X-Forwarded-Encrypted: i=1; AJvYcCXvYx115yVChNT76GtjFGAJ5/07a8hNJSZR0ohT9Zm81J4nNQ96c0/XJ2JXkHurLpCQT0u0aa8dfAgVk/I=@vger.kernel.org X-Gm-Message-State: AOJu0YwAA3jSC4uVHW3vghcsobj57YOKas8UQ/epuNGaknh3QtlYLdTG o5e1sOVZht9nHknhndFsLYp3uTrcJ3jIoVJzFm4MNizzxH+gJKNJPSuC X-Gm-Gg: ASbGncsltQAk11ymZXlQNhMDSPB1APNN7EM6hXbuRaiTz5ApKIy4xGOz06y9+JbSLe3 1aFkKXx7plMAWa9nyPa2KYbmveyu3bnAZU+82P6l7VF6Uen4fGHb6nwsC9I4/XfuXS1t/z13ko0 fHJs7hUae0mIDtoo8iNddbY6md3J17MEq25tOjjmJAruU5lOvwBthQp6nqdt31RR96jFhWHdXtX UI1HluZyLNbPoo9pnebwk0v5rNh462iZlTOujcYTDun2vIWKToEgYtRn36oIIaTvdf5dYHIuCFf 7aTXL+8TEchMLbD1nf3CILZO2MO5jacXKIcsBscuV0/12QeV8V2iRApX+g0rGRUjQAE6sq66nnj FtmJLPlOY+www1fj2LlgYSGVcJaCKmK4w605lVH+IPxL0rHmt/YHi2/2LiA== X-Google-Smtp-Source: AGHT+IEVwQm61s1dQOtymtJfnxETL8PNIjVzDNadoWh1/HbK11a27qV9e2wO2k/6lR4hXgcT7hmMCA== X-Received: by 2002:a17:90a:c2c7:b0:321:6e1a:1b70 with SMTP id 98e67ed59e1d1-32d43e11be3mr22608409a91.0.1757520540507; Wed, 10 Sep 2025 09:09:00 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.08.55 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:08:59 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 03/15] mm, swap: fix swap cache index error when retrying reclaim Date: Thu, 11 Sep 2025 00:08:21 +0800 Message-ID: <20250910160833.3464-4-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song The allocator will reclaim cached slots while scanning. Currently, it will try again if reclaim found a folio that is already removed from the swap cache due to a race. But the following lookup will be using the wrong index. It won't cause any OOB issue since the swap cache index is truncated upon lookup, but it may lead to reclaiming of an irrelevant folio. This should not cause a measurable issue, but we should fix it. Fixes: fae8595505313 ("mm, swap: avoid reclaiming irrelevant swap cache") Signed-off-by: Kairui Song Reviewed-by: Baolin Wang Acked-by: Nhat Pham Acked-by: Chris Li Acked-by: David Hildenbrand Suggested-by: Chris Li --- mm/swapfile.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index 4b8ab2cb49ca..4baebd8b48f4 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -212,7 +212,7 @@ static bool swap_is_last_map(struct swap_info_struct *s= i, static int __try_to_reclaim_swap(struct swap_info_struct *si, unsigned long offset, unsigned long flags) { - swp_entry_t entry =3D swp_entry(si->type, offset); + const swp_entry_t entry =3D swp_entry(si->type, offset); struct swap_cluster_info *ci; struct folio *folio; int ret, nr_pages; @@ -240,13 +240,13 @@ static int __try_to_reclaim_swap(struct swap_info_str= uct *si, * Offset could point to the middle of a large folio, or folio * may no longer point to the expected offset before it's locked. */ - entry =3D folio->swap; - if (offset < swp_offset(entry) || offset >=3D swp_offset(entry) + nr_page= s) { + if (offset < swp_offset(folio->swap) || + offset >=3D swp_offset(folio->swap) + nr_pages) { folio_unlock(folio); folio_put(folio); goto again; } - offset =3D swp_offset(entry); + offset =3D swp_offset(folio->swap); =20 need_reclaim =3D ((flags & TTRS_ANYWAY) || ((flags & TTRS_UNMAPPED) && !folio_mapped(folio)) || --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f181.google.com (mail-pf1-f181.google.com [209.85.210.181]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3228F337684 for ; Wed, 10 Sep 2025 16:09:06 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.181 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520547; cv=none; b=mmv5Jzk0uWpWX/eJhcgZdpxPUlTZT7KvH2FBOb5LKslqYKovFoVIBcw+rZhyKc5iFsmlzaKRoghOusmD6Y/fs6fLCpiEjErRf0MQA4jq+NCM9WGaH8I77TsicjIcJbrlFFWRuZ4kq9DOHygGPY26IvAGPKJqGkTzATEGO+/Gdns= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520547; c=relaxed/simple; bh=rapjB7hGajfLUCWCoU/6tBi3ZKfcxCBTKav4eRx7lV8=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=r0q+G2eVIo+shaJ++bXutw4/uBRtmqMsNIN5uobJbdHHSauxAPIiQMp5vhZJUabH2zjc1zrIQTVELmVGAlfCrbi86qEBdeDbSJOqpqOPH2f4K5ZIJLb0s50MH0xB5n6nMg/YyVwh77T0dXQO4aayYyQr6PuulAMicooEoOEkMss= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=WthYWnwc; arc=none smtp.client-ip=209.85.210.181 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="WthYWnwc" Received: by mail-pf1-f181.google.com with SMTP id d2e1a72fcca58-7725de6b57dso8158745b3a.0 for ; Wed, 10 Sep 2025 09:09:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520545; x=1758125345; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=BxLX4xyttOXeHBJj3J8i0UpcIMZfFqRb74t6SVPVOEc=; b=WthYWnwcmWR3DgsQQtq8s2N7hr+gFoZUX+3YcpvQNBSmkwHl81krK3eAh+aIO04Nsl iAGU/EdWuxSyGsrqHJZUINqiGTR2Eb3nBAZGAUueGoiXsC435aDk/ZO/MboU4JOR4sSI wEn5JKLntUqk64X5KuOCK1sQuBPCtv5ESyXaWehWPyMjapPFoxfVhLcBpDoDTaBWzORh 3YLAMZC1WxqogN1Q5PZ2B8fkM4ZycAZkzME7tznXMPjRUh3l1+yPcCDOeqQrnljxODtX Lrhec9m/MnUgK/zoSUpDqCIRvNiG3+xCDsZtbLxxWkG2PElV+OxJfpwFijHN3Lk4dSQ3 8/FA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520545; x=1758125345; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=BxLX4xyttOXeHBJj3J8i0UpcIMZfFqRb74t6SVPVOEc=; b=Z4nAYqNOP8n/8EQrk+7qluyFLiBL8StxPFHX6Ltv5zVl5RHqmtgIRsynnyIsHsx2w8 e0JYGizr16pPiSsjdzl+1c1FmIUrjghL4o/9jL6iYiS00RulqnC1DFHWt5CcE4yqI28/ +tyhNR0hqdUrG2KBJj1xbZT2STm6bazX+yT5nHGewPZVNE3oM2oO+0CUG3MRumY0P6ox arF+vvqH+CELTfrh4v0TpkHPgQ6Xy/7HAaPpEYTMxngSkht+PPhHpM2+2LF7jJtrJDYL 1a6G+TKRlBmfi1pp8WF0bS3o/RbtC0H8zUOv4dToqo85Kr3OTXqa8S1mWx4bv7AYh4KD tLgA== X-Forwarded-Encrypted: i=1; AJvYcCU/53Y52ZegQNhfIaOxSbX+VzY0+IBURXBdT05xHqHwKJdIsu/FqEYPCIOmCoGFZvc7k6J4aTv8kDCgU4k=@vger.kernel.org X-Gm-Message-State: AOJu0Yx0pHMuw1uarCBLED4hdtXOfIVCfghKEj9wly1QCKpyIoNi29hY yaCXlxYfVZ2iqhIgDhYKB4QjOdTEoxOX2yLkFUAdVbEgkmH6CneN5fXa X-Gm-Gg: ASbGncsUVNamkjk7RWZdx6Cwi6fnh5fz7Z6ctG49RxRT3iWkxZIqP7JXKyEQywF8SeJ N7owXNkvTWPQd8NHeDZKmYFJqFpIqMqtoPcQ25K//xxd5fOkRSo7rlU7+DgzHIrFNnofXCWEjut 7Ygj1j2uE5XpVDuxK+PD4eVZDDlebHHqO+NmjSnfWGYuA92MwylCmXUjlkQ2GTLDofHN5AVexyX rKDELCRGe7qwBz+72ddoFSDCum0fw/cfyrkX0lp4XPcutTQinqCapWvI0x6ghSvBH9dsijTErcX ksDmpEcrJAF5R6jOBYSISteS1MRDGInnnlbubOpGrT2hSufEZZ3hy/3ghdG7g5ADWt3Rl7xLwAK 9L80zl4wP09k9O0YyWWIieUCiXb/s3Kz1QH19HxtaQG2Y3s4vBBl/bn69Dg== X-Google-Smtp-Source: AGHT+IFGnnTS7ld9IQLIIRQsAt7maX5kJ+os/mQb2BSw6qlHk3VZHZQENlh+ewVjuD1hIx8qBenBkg== X-Received: by 2002:a05:6a20:1584:b0:24e:9d94:7b17 with SMTP id adf61e73a8af0-2533e8532f4mr19272123637.9.1757520545462; Wed, 10 Sep 2025 09:09:05 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.00 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:04 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 04/15] mm, swap: check page poison flag after locking it Date: Thu, 11 Sep 2025 00:08:22 +0800 Message-ID: <20250910160833.3464-5-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Instead of checking the poison flag only in the fast swap cache lookup path, always check the poison flags after locking a swap cache folio. There are two reasons to do so. The folio is unstable and could be removed from the swap cache anytime, so it's totally possible that the folio is no longer the backing folio of a swap entry, and could be an irrelevant poisoned folio. We might mistakenly kill a faulting process. And it's totally possible or even common for the slow swap in path (swapin_readahead) to bring in a cached folio. The cache folio could be poisoned, too. Only checking the poison flag in the fast path will miss such folios. The race window is tiny, so it's very unlikely to happen, though. While at it, also add a unlikely prefix. Signed-off-by: Kairui Song Acked-by: Chris Li Suggested-by: Chris Li --- mm/memory.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 10ef528a5f44..94a5928e8ace 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4661,10 +4661,8 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) goto out; =20 folio =3D swap_cache_get_folio(entry); - if (folio) { + if (folio) swap_update_readahead(folio, vma, vmf->address); - page =3D folio_file_page(folio, swp_offset(entry)); - } swapcache =3D folio; =20 if (!folio) { @@ -4735,20 +4733,13 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) ret =3D VM_FAULT_MAJOR; count_vm_event(PGMAJFAULT); count_memcg_event_mm(vma->vm_mm, PGMAJFAULT); - page =3D folio_file_page(folio, swp_offset(entry)); - } else if (PageHWPoison(page)) { - /* - * hwpoisoned dirty swapcache pages are kept for killing - * owner processes (which may be unknown at hwpoison time) - */ - ret =3D VM_FAULT_HWPOISON; - goto out_release; } =20 ret |=3D folio_lock_or_retry(folio, vmf); if (ret & VM_FAULT_RETRY) goto out_release; =20 + page =3D folio_file_page(folio, swp_offset(entry)); if (swapcache) { /* * Make sure folio_free_swap() or swapoff did not release the @@ -4761,6 +4752,15 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) page_swap_entry(page).val !=3D entry.val)) goto out_page; =20 + if (unlikely(PageHWPoison(page))) { + /* + * hwpoisoned dirty swapcache pages are kept for killing + * owner processes (which may be unknown at hwpoison time) + */ + ret =3D VM_FAULT_HWPOISON; + goto out_page; + } + /* * KSM sometimes has to copy on read faults, for example, if * folio->index of non-ksm folios would be nonlinear inside the --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f175.google.com (mail-pf1-f175.google.com [209.85.210.175]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4774C32ED31 for ; Wed, 10 Sep 2025 16:09:11 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.175 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520552; cv=none; b=tp+0CcQ8w2Jpmt3cVv25ANgpe6mTIXqISiagsF5JYsVeAavQ4mXeWx97RLd/MCzJ5yWrzhsVxaNUIghAIusQcPrwGE9WGrcB1+3NAp9NEWJXSiKx9D/wMXiZ+H1d28HITHfiqMM4lqYF+sIlurei4e8V01q5koXkM/HhtkqKGks= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520552; c=relaxed/simple; bh=X64UaZp8N953S724ZjvtTL010EKdcF41YX+AMjP9KzI=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=dRRAkFkx5aSCGb8PCSA3BjAzD0ZOi5oy5EcGQk2oxmGur4GDKKZCgqtmd7XP9qY3crY9A5u0RrjhyN4mSuMJ/3Obi21D29vI+ptU/Kf7raEomLybeeTbfzzIonejxBBdRHP2J+nSvIq71/D281YhZaP0jShoA6PVZQQgRSc+x6s= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=Bpo5w/td; arc=none smtp.client-ip=209.85.210.175 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="Bpo5w/td" Received: by mail-pf1-f175.google.com with SMTP id d2e1a72fcca58-7722bcb989aso5278901b3a.1 for ; Wed, 10 Sep 2025 09:09:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520550; x=1758125350; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=2/WGEOxJHrA21wqgG+5kS81ezpjU55LK2AmGw+vBNHk=; b=Bpo5w/td3nUiL4xnYOLW/eWUJK+Hp47zfDiZhXHWMotScHIYlDrP7XaKX9td48Jnqs L2zxTG7pHIII6D2w7Am1C2MrlpFKmUBVlZ3wuGU4KVjGGh+8xhrCUfkF5SVQDJggoMgR pyhSCOqjVT9prTxxGkt5OLHbzRjy7ZIXRRsyO+7iDTO67T9C9rApc87yFijHcIbXQ6yV i3s53YXyYWAIBq5mdDQKFpNOZbJv87WCsELDaicv8K82TmDclQRmfP210LyUF/TE3rDR OIUz38SeEzqouN11zpgZjaNcW4ZCtchgj3tNDNhsq1MFwfaskoz0lY6/m8Abth2Xkmql b3RQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520550; x=1758125350; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=2/WGEOxJHrA21wqgG+5kS81ezpjU55LK2AmGw+vBNHk=; b=FmIsNI4ZDlHN6UfcfMBoksstlTyZBSYiuttGcmeudL2sN9XHiF2n+yv0yh0amIks5E 3UuAWCAguKuztEXX9u3W8hPn0haIbriNIZ7Fxr/TqB/qtfcEBS5m4l438Qrej4q42JRh m8GdVmrBW9AGJe4B78p077+rbnD6328SBZCqAGwjuaYJo7ZsnwZPL04KlBMu7pKi9PdK HKfx1DU/WMkXCJT8Ehoe+jeShZQ7ozbqJX5FoNn9o3vSOG3vgdy5LmF/nYG/y8HtyobM 3M5pWHFWpsgQEoHV6frJRNpfJB7eL9dYx6RGr5aHVXv7B4JWJoN/S5d1RJyyNKw5bN+2 csiA== X-Forwarded-Encrypted: i=1; AJvYcCUhoDFLS0X0dsQIqyIv5QodaVskeHYGDpbDkQEks3NfBUMybHMURh33vwyu4B0JjV4JSTCLkNdR6dx/L4o=@vger.kernel.org X-Gm-Message-State: AOJu0Yxjon4MdBjg0ZtABR17YstqJVYpqOXVyE7ELasvZl3jC7rFJ6Gp UIwICrv6C6tsTJeBJuiY+gqFuxJiExDb9iJMglsmj+5FIyMrGV5iAU01 X-Gm-Gg: ASbGncuZVbOkdK5LvA4lNL9bqaFI1w6k82D0NOS+Y96qiooi17zLDPYre/g6rdqNUKK EWKqYDRM0pT1epdysmGkB90ju7m9M2KYPcmCn2ia0ajYBLhV6jP+8Ok+fiif3yOumPhQGAc9L5S 8pB+YREDMglHXfWGz5fz37ZirIirjvaN7w5vtM1qQwLfdYywmi6SdWILJMah0aEmxVb7q5KZxQE jIIv0sXc0oXrW4BQAqOr2Ra0FdTofe0fJKkMBrv4GOcQQGCz3OCJks36ms9fcpXBTLZQfrO87Fu GIMQG4D7KGiqTb8SJptF2SmY6c1/VWXR2r1rWb53uBtnjbF/3whAnS7qK8Jzj2KGG8yXv2VM1bR p7qoBqHoJM69/yqBhWSl37KwJy5B1OjURQoK2b0KhnwUzlNI= X-Google-Smtp-Source: AGHT+IGtx55UQI+upmfsPN7ALnjl37EWjQTi7JoiDM2lB8YgnmOo7UdwmjQczornIDM65bBjDgqC/Q== X-Received: by 2002:a05:6a20:6a10:b0:248:e0f7:1331 with SMTP id adf61e73a8af0-2533e94fa7fmr24671362637.3.1757520550410; Wed, 10 Sep 2025 09:09:10 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.05 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:09 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 05/15] mm, swap: always lock and check the swap cache folio before use Date: Thu, 11 Sep 2025 00:08:23 +0800 Message-ID: <20250910160833.3464-6-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Swap cache lookup only increases the reference count of the returned folio. That's not enough to ensure a folio is stable in the swap cache, so the folio could be removed from the swap cache at any time. The caller should always lock and check the folio before using it. We have just documented this in kerneldoc, now introduce a helper for swap cache folio verification with proper sanity checks. Also, sanitize a few current users to use this convention and the new helper for easier debugging. They were not having observable problems yet, only trivial issues like wasted CPU cycles on swapoff or reclaiming. They would fail in some other way, but it is still better to always follow this convention to make things robust and make later commits easier to do. Signed-off-by: Kairui Song Suggested-by: Chris Li --- mm/memory.c | 3 +-- mm/swap.h | 27 +++++++++++++++++++++++++++ mm/swap_state.c | 7 +++++-- mm/swapfile.c | 11 ++++++++--- 4 files changed, 41 insertions(+), 7 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 94a5928e8ace..5808c4ef21b3 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4748,8 +4748,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) * swapcache, we need to check that the page's swap has not * changed. */ - if (unlikely(!folio_test_swapcache(folio) || - page_swap_entry(page).val !=3D entry.val)) + if (unlikely(!folio_matches_swap_entry(folio, entry))) goto out_page; =20 if (unlikely(PageHWPoison(page))) { diff --git a/mm/swap.h b/mm/swap.h index efb6d7ff9f30..7d868f8de696 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -52,6 +52,28 @@ static inline pgoff_t swap_cache_index(swp_entry_t entry) return swp_offset(entry) & SWAP_ADDRESS_SPACE_MASK; } =20 +/** + * folio_matches_swap_entry - Check if a folio matches a given swap entry. + * @folio: The folio. + * @entry: The swap entry to check against. + * + * Context: The caller should have the folio locked to ensure it's stable + * and nothing will move it in or out of the swap cache. + * Return: true or false. + */ +static inline bool folio_matches_swap_entry(const struct folio *folio, + swp_entry_t entry) +{ + swp_entry_t folio_entry =3D folio->swap; + long nr_pages =3D folio_nr_pages(folio); + + VM_WARN_ON_ONCE_FOLIO(!folio_test_locked(folio), folio); + if (!folio_test_swapcache(folio)) + return false; + VM_WARN_ON_ONCE_FOLIO(!IS_ALIGNED(folio_entry.val, nr_pages), folio); + return folio_entry.val =3D=3D round_down(entry.val, nr_pages); +} + void show_swap_cache_info(void); void *get_shadow_from_swap_cache(swp_entry_t entry); int add_to_swap_cache(struct folio *folio, swp_entry_t entry, @@ -144,6 +166,11 @@ static inline pgoff_t swap_cache_index(swp_entry_t ent= ry) return 0; } =20 +static inline bool folio_matches_swap_entry(const struct folio *folio, swp= _entry_t entry) +{ + return false; +} + static inline void show_swap_cache_info(void) { } diff --git a/mm/swap_state.c b/mm/swap_state.c index 68ec531d0f2b..9225d6b695ad 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -79,7 +79,7 @@ void show_swap_cache_info(void) * with reference count or locks. * Return: Returns the found folio on success, NULL otherwise. The caller * must lock and check if the folio still matches the swap entry before - * use. + * use (e.g. with folio_matches_swap_entry). */ struct folio *swap_cache_get_folio(swp_entry_t entry) { @@ -346,7 +346,10 @@ struct folio *__read_swap_cache_async(swp_entry_t entr= y, gfp_t gfp_mask, for (;;) { int err; =20 - /* Check the swap cache in case the folio is already there */ + /* + * Check the swap cache first, if a cached folio is found, + * return it unlocked. The caller will lock and check it. + */ folio =3D swap_cache_get_folio(entry); if (folio) goto got_folio; diff --git a/mm/swapfile.c b/mm/swapfile.c index 4baebd8b48f4..f1a4d381d719 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -240,14 +240,12 @@ static int __try_to_reclaim_swap(struct swap_info_str= uct *si, * Offset could point to the middle of a large folio, or folio * may no longer point to the expected offset before it's locked. */ - if (offset < swp_offset(folio->swap) || - offset >=3D swp_offset(folio->swap) + nr_pages) { + if (!folio_matches_swap_entry(folio, entry)) { folio_unlock(folio); folio_put(folio); goto again; } offset =3D swp_offset(folio->swap); - need_reclaim =3D ((flags & TTRS_ANYWAY) || ((flags & TTRS_UNMAPPED) && !folio_mapped(folio)) || ((flags & TTRS_FULL) && mem_cgroup_swap_full(folio))); @@ -2004,6 +2002,13 @@ static int unuse_pte(struct vm_area_struct *vma, pmd= _t *pmd, bool hwpoisoned =3D false; int ret =3D 1; =20 + /* + * If the folio is removed from swap cache by others, continue to + * unuse other PTEs. try_to_unuse may try again if we missed this one. + */ + if (!folio_matches_swap_entry(folio, entry)) + return 0; + swapcache =3D folio; folio =3D ksm_might_need_to_copy(folio, vma, addr); if (unlikely(!folio)) --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f173.google.com (mail-pf1-f173.google.com [209.85.210.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 77A68338F3D for ; Wed, 10 Sep 2025 16:09:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520558; cv=none; b=n+4ucRXJ+PSdA+uIOJnIGz6y7rQs9UCtGTfKvKIMUGk3/NIgtSXDXkco2+EYClnTu9EciKrMxQp8jsvNU3pPDZk5f7vvuKzOFQrO2kmt3L7abdsWOsnKG7i3wpZxcSKdxqspYla0/SRz5EfYDS6UGb1tAJRNheR3DZ2L4LhH/Fk= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520558; c=relaxed/simple; bh=tp9jK+cTl4ybe0NoVzXfXV2KqwiGZy4ZtyO0uOEresg=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=bJPcN9BiKmfzy5N28m+rY72WA7uE2XTC+f4ONULUroyVTzg1faw3XakGUPilY+m+Fz3hdCo9RGwtDUb5L8EEgoMaT4uEmmFFNOtsHR2uwh6Tt26ZKHMpBB/5EJchC2MYjx4+ir74Wl76D1NHPPPWao3eby6E7c6dTa/0GMMbnLw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=gI0SvO3S; arc=none smtp.client-ip=209.85.210.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="gI0SvO3S" Received: by mail-pf1-f173.google.com with SMTP id d2e1a72fcca58-7728a8862ccso870502b3a.0 for ; Wed, 10 Sep 2025 09:09:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520556; x=1758125356; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=s5HcJVxJqYzxUZjEo3aLu6n8KHT0yS8P2J4twu7LXww=; b=gI0SvO3SJ68e4t+k/PPvLJtfMNHne6qWnbTnUOxzPJgdIpEnSdb2nRbf377KCJfjmj IwN2XLR2b5HOgU5ZyKlR4wBLb2SbfP2f+JraCyusnMRFmf6fJ7VLBztzqWi0GGRtIc76 4CQyOOn5WD2TYotdrs4h+2Uglwtxj+a6fJKk3lO+Zfx4c7Ed2+UCbD41Jsy3ebbL0frJ E2Pwt696vl/oOHUiBIRiNKtJr00WMAaMeVRAA/a+F6W9umZxokqRKJRT6RoO9pZrMz87 NhvOyh1V56CcyufrD4LQR6qHvnVqwY86NUFnsf+EqGmQtSHUx1mSebHEjcJ9/mB1QhYc jhfQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520556; x=1758125356; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=s5HcJVxJqYzxUZjEo3aLu6n8KHT0yS8P2J4twu7LXww=; b=Ma19kBRz4dlc8l7W+PeI31sePIXUt2DZx2Jw1VwLUUgBt6fd3JpmqhuordlpMSiuNi Wu8QZd/+w4PVojd5sCCn8eGent28VYjpag0CrFVHv/+ZnsM1oJgAlf2nUVx3d4cCw0WK fVYpATET3IPH9PyKY8t8Llqkv2I0osudGhyfod0jiiJ30rmby7sCAFC+fSB/bjJFA0XZ Nyyl9ofAYdiZV0Jq1DKpU19FRWUOSKg75Bj/0Pbd6H9cdz1NLzCjvemJ41gqZYpS6iah 9q8gvWSXlffKtKX6B8LQPCZTT67E5u4JOFP3ziuAxK+dRHl6DiCB8vL4ipVviDwRe81I i/8g== X-Forwarded-Encrypted: i=1; AJvYcCVOsHzS0xdN02gMB5h2jzURwft1mV6J5UrEdVsYu3xFCEexsfLNhJ2s1Kj2plAXxr6IzpCie0kgw5DInHM=@vger.kernel.org X-Gm-Message-State: AOJu0YyX/vFhQ07J1QWFBviYlPNasmZi/LvHApOCqR3XVWJQXsFmlwJE LDbQnU3g44yL+D6lF32r3NvI7rXvjVs4pREI8ufTiKhN93/+4dFyNjZh X-Gm-Gg: ASbGncspFI+ifqn9RR8SHkyAJfbFL9yTkzsnt3NsJCV6gCdYE53pl9PcZBmAr7utlOA ghPGmb/52VKtXnFNYnpOgvR5i02HDY0hyoFHXXDyys5x6b2BvJl9sdPPQEgy48pxnP/ueLDgu91 ngJ7SCkMXRf9zrDBS5RbmNlyL5VOsdtUhwVE/OkBQ3ozCIfXUlLIME7cJlipsSZpamVpPTarY5j O5kq19Q7rEKJYvCtWuya2vEPAClP0VlqxfG13TalTygEM6juekG+EH6rHtJdnMGn/HSsoASez4x 6TtiWDaDnNfQL5ZzjeV/p623U3UeTtsnoWdOZgDM1xbmoUY1jflYdiZTZ5Qcu9bYVDEMff4TjAO nJLMaiOLfHGw+UVo4tAxOXI5Dj17kaL6Xtuu6cdCMNJkIgRo= X-Google-Smtp-Source: AGHT+IGH4gi7IyV/95U3LFCHhW1okK8q/O9STR4M1KwOMk1ioCbsGLzlOVyMoaTZn7dFGHmQS+afqg== X-Received: by 2002:a05:6a20:4327:b0:243:b965:1bb7 with SMTP id adf61e73a8af0-25d0a5cd9c0mr198373637.19.1757520555556; Wed, 10 Sep 2025 09:09:15 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.10 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:14 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 06/15] mm, swap: rename and move some swap cluster definition and helpers Date: Thu, 11 Sep 2025 00:08:24 +0800 Message-ID: <20250910160833.3464-7-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song No feature change, move cluster related definitions and helpers to mm/swap.h, also tidy up and add a "swap_" prefix for cluster lock/unlock helpers, so they can be used outside of swap files. And while at it, add kerneldoc. Signed-off-by: Kairui Song Reviewed-by: Baolin Wang Reviewed-by: Barry Song Acked-by: Chris Li Acked-by: David Hildenbrand Suggested-by: Chris Li --- include/linux/swap.h | 34 ---------------- mm/swap.h | 70 ++++++++++++++++++++++++++++++++ mm/swapfile.c | 97 +++++++++++++------------------------------- 3 files changed, 99 insertions(+), 102 deletions(-) diff --git a/include/linux/swap.h b/include/linux/swap.h index a2bb20841616..78cc48a65512 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -235,40 +235,6 @@ enum { /* Special value in each swap_map continuation */ #define SWAP_CONT_MAX 0x7f /* Max count */ =20 -/* - * We use this to track usage of a cluster. A cluster is a block of swap d= isk - * space with SWAPFILE_CLUSTER pages long and naturally aligns in disk. All - * free clusters are organized into a list. We fetch an entry from the lis= t to - * get a free cluster. - * - * The flags field determines if a cluster is free. This is - * protected by cluster lock. - */ -struct swap_cluster_info { - spinlock_t lock; /* - * Protect swap_cluster_info fields - * other than list, and swap_info_struct->swap_map - * elements corresponding to the swap cluster. - */ - u16 count; - u8 flags; - u8 order; - struct list_head list; -}; - -/* All on-list cluster must have a non-zero flag. */ -enum swap_cluster_flags { - CLUSTER_FLAG_NONE =3D 0, /* For temporary off-list cluster */ - CLUSTER_FLAG_FREE, - CLUSTER_FLAG_NONFULL, - CLUSTER_FLAG_FRAG, - /* Clusters with flags above are allocatable */ - CLUSTER_FLAG_USABLE =3D CLUSTER_FLAG_FRAG, - CLUSTER_FLAG_FULL, - CLUSTER_FLAG_DISCARD, - CLUSTER_FLAG_MAX, -}; - /* * The first page in the swap file is the swap header, which is always mar= ked * bad to prevent it from being allocated as an entry. This also prevents = the diff --git a/mm/swap.h b/mm/swap.h index 7d868f8de696..138b5197c35e 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -7,10 +7,80 @@ struct swap_iocb; =20 extern int page_cluster; =20 +#ifdef CONFIG_THP_SWAP +#define SWAPFILE_CLUSTER HPAGE_PMD_NR +#define swap_entry_order(order) (order) +#else +#define SWAPFILE_CLUSTER 256 +#define swap_entry_order(order) 0 +#endif + +/* + * We use this to track usage of a cluster. A cluster is a block of swap d= isk + * space with SWAPFILE_CLUSTER pages long and naturally aligns in disk. All + * free clusters are organized into a list. We fetch an entry from the lis= t to + * get a free cluster. + * + * The flags field determines if a cluster is free. This is + * protected by cluster lock. + */ +struct swap_cluster_info { + spinlock_t lock; /* + * Protect swap_cluster_info fields + * other than list, and swap_info_struct->swap_map + * elements corresponding to the swap cluster. + */ + u16 count; + u8 flags; + u8 order; + struct list_head list; +}; + +/* All on-list cluster must have a non-zero flag. */ +enum swap_cluster_flags { + CLUSTER_FLAG_NONE =3D 0, /* For temporary off-list cluster */ + CLUSTER_FLAG_FREE, + CLUSTER_FLAG_NONFULL, + CLUSTER_FLAG_FRAG, + /* Clusters with flags above are allocatable */ + CLUSTER_FLAG_USABLE =3D CLUSTER_FLAG_FRAG, + CLUSTER_FLAG_FULL, + CLUSTER_FLAG_DISCARD, + CLUSTER_FLAG_MAX, +}; + #ifdef CONFIG_SWAP #include /* for swp_offset */ #include /* for bio_end_io_t */ =20 +static inline struct swap_cluster_info *swp_offset_cluster( + struct swap_info_struct *si, pgoff_t offset) +{ + return &si->cluster_info[offset / SWAPFILE_CLUSTER]; +} + +/** + * swap_cluster_lock - Lock and return the swap cluster of given offset. + * @si: swap device the cluster belongs to. + * @offset: the swap entry offset, pointing to a valid slot. + * + * Context: The caller must ensure the offset is in the valid range and + * protect the swap device with reference count or locks. + */ +static inline struct swap_cluster_info *swap_cluster_lock( + struct swap_info_struct *si, unsigned long offset) +{ + struct swap_cluster_info *ci =3D swp_offset_cluster(si, offset); + + spin_lock(&ci->lock); + return ci; +} + +static inline void swap_cluster_unlock(struct swap_cluster_info *ci) +{ + spin_unlock(&ci->lock); +} + /* linux/mm/page_io.c */ int sio_pool_init(void); struct swap_iocb; diff --git a/mm/swapfile.c b/mm/swapfile.c index f1a4d381d719..3bfe9cf06fd2 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -58,9 +58,6 @@ static void swap_entries_free(struct swap_info_struct *si, static void swap_range_alloc(struct swap_info_struct *si, unsigned int nr_entries); static bool folio_swapcache_freeable(struct folio *folio); -static struct swap_cluster_info *lock_cluster(struct swap_info_struct *si, - unsigned long offset); -static inline void unlock_cluster(struct swap_cluster_info *ci); =20 static DEFINE_SPINLOCK(swap_lock); static unsigned int nr_swapfiles; @@ -257,9 +254,9 @@ static int __try_to_reclaim_swap(struct swap_info_struc= t *si, * swap_map is HAS_CACHE only, which means the slots have no page table * reference or pending writeback, and can't be allocated to others. */ - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); need_reclaim =3D swap_only_has_cache(si, offset, nr_pages); - unlock_cluster(ci); + swap_cluster_unlock(ci); if (!need_reclaim) goto out_unlock; =20 @@ -384,19 +381,6 @@ static void discard_swap_cluster(struct swap_info_stru= ct *si, } } =20 -#ifdef CONFIG_THP_SWAP -#define SWAPFILE_CLUSTER HPAGE_PMD_NR - -#define swap_entry_order(order) (order) -#else -#define SWAPFILE_CLUSTER 256 - -/* - * Define swap_entry_order() as constant to let compiler to optimize - * out some code if !CONFIG_THP_SWAP - */ -#define swap_entry_order(order) 0 -#endif #define LATENCY_LIMIT 256 =20 static inline bool cluster_is_empty(struct swap_cluster_info *info) @@ -424,34 +408,12 @@ static inline unsigned int cluster_index(struct swap_= info_struct *si, return ci - si->cluster_info; } =20 -static inline struct swap_cluster_info *offset_to_cluster(struct swap_info= _struct *si, - unsigned long offset) -{ - return &si->cluster_info[offset / SWAPFILE_CLUSTER]; -} - static inline unsigned int cluster_offset(struct swap_info_struct *si, struct swap_cluster_info *ci) { return cluster_index(si, ci) * SWAPFILE_CLUSTER; } =20 -static inline struct swap_cluster_info *lock_cluster(struct swap_info_stru= ct *si, - unsigned long offset) -{ - struct swap_cluster_info *ci; - - ci =3D offset_to_cluster(si, offset); - spin_lock(&ci->lock); - - return ci; -} - -static inline void unlock_cluster(struct swap_cluster_info *ci) -{ - spin_unlock(&ci->lock); -} - static void move_cluster(struct swap_info_struct *si, struct swap_cluster_info *ci, struct list_head *list, enum swap_cluster_flags new_flags) @@ -807,7 +769,7 @@ static unsigned int alloc_swap_scan_cluster(struct swap= _info_struct *si, } out: relocate_cluster(si, ci); - unlock_cluster(ci); + swap_cluster_unlock(ci); if (si->flags & SWP_SOLIDSTATE) { this_cpu_write(percpu_swap_cluster.offset[order], next); this_cpu_write(percpu_swap_cluster.si[order], si); @@ -874,7 +836,7 @@ static void swap_reclaim_full_clusters(struct swap_info= _struct *si, bool force) if (ci->flags =3D=3D CLUSTER_FLAG_NONE) relocate_cluster(si, ci); =20 - unlock_cluster(ci); + swap_cluster_unlock(ci); if (to_scan <=3D 0) break; } @@ -913,7 +875,7 @@ static unsigned long cluster_alloc_swap_entry(struct sw= ap_info_struct *si, int o if (offset =3D=3D SWAP_ENTRY_INVALID) goto new_cluster; =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); /* Cluster could have been used by another order */ if (cluster_is_usable(ci, order)) { if (cluster_is_empty(ci)) @@ -921,7 +883,7 @@ static unsigned long cluster_alloc_swap_entry(struct sw= ap_info_struct *si, int o found =3D alloc_swap_scan_cluster(si, ci, offset, order, usage); } else { - unlock_cluster(ci); + swap_cluster_unlock(ci); } if (found) goto done; @@ -1202,7 +1164,7 @@ static bool swap_alloc_fast(swp_entry_t *entry, if (!si || !offset || !get_swap_device_info(si)) return false; =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); if (cluster_is_usable(ci, order)) { if (cluster_is_empty(ci)) offset =3D cluster_offset(si, ci); @@ -1210,7 +1172,7 @@ static bool swap_alloc_fast(swp_entry_t *entry, if (found) *entry =3D swp_entry(si->type, found); } else { - unlock_cluster(ci); + swap_cluster_unlock(ci); } =20 put_swap_device(si); @@ -1478,14 +1440,14 @@ static void swap_entries_put_cache(struct swap_info= _struct *si, unsigned long offset =3D swp_offset(entry); struct swap_cluster_info *ci; =20 - ci =3D lock_cluster(si, offset); - if (swap_only_has_cache(si, offset, nr)) + ci =3D swap_cluster_lock(si, offset); + if (swap_only_has_cache(si, offset, nr)) { swap_entries_free(si, ci, entry, nr); - else { + } else { for (int i =3D 0; i < nr; i++, entry.val++) swap_entry_put_locked(si, ci, entry, SWAP_HAS_CACHE); } - unlock_cluster(ci); + swap_cluster_unlock(ci); } =20 static bool swap_entries_put_map(struct swap_info_struct *si, @@ -1503,7 +1465,7 @@ static bool swap_entries_put_map(struct swap_info_str= uct *si, if (count !=3D 1 && count !=3D SWAP_MAP_SHMEM) goto fallback; =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); if (!swap_is_last_map(si, offset, nr, &has_cache)) { goto locked_fallback; } @@ -1512,21 +1474,20 @@ static bool swap_entries_put_map(struct swap_info_s= truct *si, else for (i =3D 0; i < nr; i++) WRITE_ONCE(si->swap_map[offset + i], SWAP_HAS_CACHE); - unlock_cluster(ci); + swap_cluster_unlock(ci); =20 return has_cache; =20 fallback: - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); locked_fallback: for (i =3D 0; i < nr; i++, entry.val++) { count =3D swap_entry_put_locked(si, ci, entry, 1); if (count =3D=3D SWAP_HAS_CACHE) has_cache =3D true; } - unlock_cluster(ci); + swap_cluster_unlock(ci); return has_cache; - } =20 /* @@ -1576,7 +1537,7 @@ static void swap_entries_free(struct swap_info_struct= *si, unsigned char *map_end =3D map + nr_pages; =20 /* It should never free entries across different clusters */ - VM_BUG_ON(ci !=3D offset_to_cluster(si, offset + nr_pages - 1)); + VM_BUG_ON(ci !=3D swp_offset_cluster(si, offset + nr_pages - 1)); VM_BUG_ON(cluster_is_empty(ci)); VM_BUG_ON(ci->count < nr_pages); =20 @@ -1651,9 +1612,9 @@ bool swap_entry_swapped(struct swap_info_struct *si, = swp_entry_t entry) struct swap_cluster_info *ci; int count; =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); count =3D swap_count(si->swap_map[offset]); - unlock_cluster(ci); + swap_cluster_unlock(ci); return !!count; } =20 @@ -1676,7 +1637,7 @@ int swp_swapcount(swp_entry_t entry) =20 offset =3D swp_offset(entry); =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); =20 count =3D swap_count(si->swap_map[offset]); if (!(count & COUNT_CONTINUED)) @@ -1699,7 +1660,7 @@ int swp_swapcount(swp_entry_t entry) n *=3D (SWAP_CONT_MAX + 1); } while (tmp_count & COUNT_CONTINUED); out: - unlock_cluster(ci); + swap_cluster_unlock(ci); return count; } =20 @@ -1714,7 +1675,7 @@ static bool swap_page_trans_huge_swapped(struct swap_= info_struct *si, int i; bool ret =3D false; =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); if (nr_pages =3D=3D 1) { if (swap_count(map[roffset])) ret =3D true; @@ -1727,7 +1688,7 @@ static bool swap_page_trans_huge_swapped(struct swap_= info_struct *si, } } unlock_out: - unlock_cluster(ci); + swap_cluster_unlock(ci); return ret; } =20 @@ -2661,8 +2622,8 @@ static void wait_for_allocation(struct swap_info_stru= ct *si) BUG_ON(si->flags & SWP_WRITEOK); =20 for (offset =3D 0; offset < end; offset +=3D SWAPFILE_CLUSTER) { - ci =3D lock_cluster(si, offset); - unlock_cluster(ci); + ci =3D swap_cluster_lock(si, offset); + swap_cluster_unlock(ci); } } =20 @@ -3578,7 +3539,7 @@ static int __swap_duplicate(swp_entry_t entry, unsign= ed char usage, int nr) offset =3D swp_offset(entry); VM_WARN_ON(nr > SWAPFILE_CLUSTER - offset % SWAPFILE_CLUSTER); VM_WARN_ON(usage =3D=3D 1 && nr > 1); - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); =20 err =3D 0; for (i =3D 0; i < nr; i++) { @@ -3633,7 +3594,7 @@ static int __swap_duplicate(swp_entry_t entry, unsign= ed char usage, int nr) } =20 unlock_out: - unlock_cluster(ci); + swap_cluster_unlock(ci); return err; } =20 @@ -3732,7 +3693,7 @@ int add_swap_count_continuation(swp_entry_t entry, gf= p_t gfp_mask) =20 offset =3D swp_offset(entry); =20 - ci =3D lock_cluster(si, offset); + ci =3D swap_cluster_lock(si, offset); =20 count =3D swap_count(si->swap_map[offset]); =20 @@ -3792,7 +3753,7 @@ int add_swap_count_continuation(swp_entry_t entry, gf= p_t gfp_mask) out_unlock_cont: spin_unlock(&si->cont_lock); out: - unlock_cluster(ci); + swap_cluster_unlock(ci); put_swap_device(si); outer: if (page) --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f180.google.com (mail-pf1-f180.google.com [209.85.210.180]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C288B32F774 for ; Wed, 10 Sep 2025 16:09:21 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.180 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520563; cv=none; b=IhgIzogGHaXVVogLwAFIJBvGhhYOt3qDWoBEEZiJTfL5bEU7/CgnT+lfTcH2sEoejciPDQDMm8Np1Fl1XGx1JneVeadaMQ3reiNGsGUTCipUXBkSWq7aYACnXjp5oMO/P9XFkHxH6v+6JpCbv4EljDSWtYQ1CAJ0rCiig5uoDRw= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520563; c=relaxed/simple; bh=NdekbX19Sr3DSBdAq8Pu05RLNECHsFUhFrPbW8yuf68=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=qnDzNyujiYKv7dIYJvcKztJx53ElsiJ3Q1AMmzQfckewjbX3b1kyQE49wN2DVLHkkuYbOkrDX7MvgXOz+AfZLHHF9xAR3r4h2pgG5btguMJRk/ZLlRFaS1RGh4GHUjFOoOHXUIYdvjTCvy8V2S6kYS0zqwZgRIeHr3uTm6u0CEs= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=Ldb8UjjY; arc=none smtp.client-ip=209.85.210.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="Ldb8UjjY" Received: by mail-pf1-f180.google.com with SMTP id d2e1a72fcca58-77256e75eacso6346527b3a.0 for ; Wed, 10 Sep 2025 09:09:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520561; x=1758125361; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=r30iRUUJzVW6YjOo4TAiZlOld85R21y5sr96AWgus+8=; b=Ldb8UjjYNZ2whpVgyZ/UN4FK1/bnOy+80R/Z4/UaIX5idV2JFP+h0rhW3Gf2mRyRtI 55ecOu8TDxhd8KEEGyx/LFD/tI7Y3At6UJo73NL1jZ3W5Aap+ukTwUNI3q2TGqLue+r7 22TJXLYPXfNHY3M7QOiYdF/XYZBv1ttqZGhbudV5UJeAUX/qgkikijXShLoQHVyrgkDk SLnsNJjC6ZXoVNLRzCTQHt/d7+8y5wEv1dhZeku8pSHUwOKcA8U0/emQshfofWCF6maA 6r8SAHtdyfcMlWG5uqRraMao15ouQVWGRPz/O1l5C8SrSdvXNPulnIBwCO6o0grBCsGf FIQQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520561; x=1758125361; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=r30iRUUJzVW6YjOo4TAiZlOld85R21y5sr96AWgus+8=; b=N5+XlL/JtJMSnqYZAKHSYsDZcjRK6dmdinA/EeqlC99sBmnwQ29L9ogX9+gPHxwgcj B58Tdx0uQN/btfGSQTHO4nMUmPddH2KXVK6CovbF5dnw+LqfADkCDc6YUBvUfNNK4+qw bk1C2OhmYZBfei87pVnOJsDBJX3U95ARRZYPnO86InWGyYZCkBw4W/w0PHMZKs/NDIgY S70OpmwGCg6T93W6qzHTXpqDwCKdQDnZQTyaNAnASoAAEglkxLEk/UlfBLXU5WQV2RP5 bBU2TN+rsYHbsvirRoV990cxiZfCRqoeHu8SEpk1+p472y1+K01aeo1SgRHxJ2Nfs2bA 0FNg== X-Forwarded-Encrypted: i=1; AJvYcCVF1bAAvi3rf5UqTMWIWH/Zl+JdHmujOr9wUxjrbi+4sLKMBQfri9CQbTA5T7AVFllqZeD/fEQ6um3PKl4=@vger.kernel.org X-Gm-Message-State: AOJu0Yyx1QtIkpayEZXEEqAFk8yoxLyBfEDYTJB2zUfMtphlHaWyIhCx tpzNUEho82SskyYvC5ghMmcgusqxOtSnhBBFSPfVhx7jyXQgLghryalF X-Gm-Gg: ASbGncsBwdXGrQ6QXUOgctYi+i/rKjoN0lC5gok9RAACGrhkklVZeasj3QLIfNyiS9R TYcaqd7Wo6rABSamcV69jxo5nTFjJeZyTFDi/4EO2tRx9wLAQEETPzwjNUi7RYhBGeQwmwSTcl7 XDmuaGAT6r6YzH77FwGsr5Kn2SdMUDM18w6lDhz02CD34NHQKqGQ6E87t/B+RBIqbMzxdl0/E/J 9h0v4lz0DZoANjLUx5oU70a1IV/jQfOd6vkOEkeGPwxT7LPHBpLAnJktpXOehxmv87aN1g10IZT TRBIytcL6ItT27obvG4OqxtjuOKMycqOWJoamiTTa6QFliVa3qo6+OEAS8slGmbXU8BUYRgq521 G1osbLiEZdIjJQNrV2gqlKOGS9AuidI4bPggIZNIVZQDtmjg= X-Google-Smtp-Source: AGHT+IG4rgicLdld9JX3OJJkcVKNbRft/Ys3FKlmoalIK9jozRWnjhDtssS7p6xY441+nTXDOOTp4g== X-Received: by 2002:a05:6a20:7d8a:b0:24f:f79d:669c with SMTP id adf61e73a8af0-2534557a96fmr23074527637.59.1757520560630; Wed, 10 Sep 2025 09:09:20 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.15 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:19 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 07/15] mm, swap: tidy up swap device and cluster info helpers Date: Thu, 11 Sep 2025 00:08:25 +0800 Message-ID: <20250910160833.3464-8-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song swp_swap_info is the most commonly used helper for retrieving swap info. It has an internal check that may lead to a NULL return value, but almost none of its caller checks the return value, making the internal check pointless. In fact, most of these callers already ensured the entry is valid and never expect a NULL value. Tidy this up and improve the function names. If the caller can make sure the swap entry/type is valid and the device is pinned, use the new introduced __swap_entry_to_info/__swap_type_to_info instead. They have more debug sanity checks and lower overhead as they are inlined. Callers that may expect a NULL value should use swap_entry_to_info/swap_type_to_info instead. No feature change. The rearranged codes should have had no effect, or they should have been hitting NULL de-ref bugs already. Only some new sanity checks are added so potential issues may show up in debug build. The new helpers will be frequently used with swap table later when working with swap cache folios. A locked swap cache folio ensures the entries are valid and stable so these helpers are very helpful. Signed-off-by: Kairui Song Acked-by: Chris Li Reviewed-by: Barry Song Acked-by: David Hildenbrand Suggested-by: Chris Li --- include/linux/swap.h | 6 ------ mm/page_io.c | 12 ++++++------ mm/swap.h | 38 +++++++++++++++++++++++++++++++++----- mm/swap_state.c | 4 ++-- mm/swapfile.c | 37 +++++++++++++++++++------------------ 5 files changed, 60 insertions(+), 37 deletions(-) diff --git a/include/linux/swap.h b/include/linux/swap.h index 78cc48a65512..762f8db0e811 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -479,7 +479,6 @@ extern sector_t swapdev_block(int, pgoff_t); extern int __swap_count(swp_entry_t entry); extern bool swap_entry_swapped(struct swap_info_struct *si, swp_entry_t en= try); extern int swp_swapcount(swp_entry_t entry); -struct swap_info_struct *swp_swap_info(swp_entry_t entry); struct backing_dev_info; extern int init_swap_address_space(unsigned int type, unsigned long nr_pag= es); extern void exit_swap_address_space(unsigned int type); @@ -492,11 +491,6 @@ static inline void put_swap_device(struct swap_info_st= ruct *si) } =20 #else /* CONFIG_SWAP */ -static inline struct swap_info_struct *swp_swap_info(swp_entry_t entry) -{ - return NULL; -} - static inline struct swap_info_struct *get_swap_device(swp_entry_t entry) { return NULL; diff --git a/mm/page_io.c b/mm/page_io.c index a2056a5ecb13..3c342db77ce3 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -204,7 +204,7 @@ static bool is_folio_zero_filled(struct folio *folio) static void swap_zeromap_folio_set(struct folio *folio) { struct obj_cgroup *objcg =3D get_obj_cgroup_from_folio(folio); - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); int nr_pages =3D folio_nr_pages(folio); swp_entry_t entry; unsigned int i; @@ -223,7 +223,7 @@ static void swap_zeromap_folio_set(struct folio *folio) =20 static void swap_zeromap_folio_clear(struct folio *folio) { - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); swp_entry_t entry; unsigned int i; =20 @@ -374,7 +374,7 @@ static void sio_write_complete(struct kiocb *iocb, long= ret) static void swap_writepage_fs(struct folio *folio, struct swap_iocb **swap= _plug) { struct swap_iocb *sio =3D swap_plug ? *swap_plug : NULL; - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); struct file *swap_file =3D sis->swap_file; loff_t pos =3D swap_dev_pos(folio->swap); =20 @@ -446,7 +446,7 @@ static void swap_writepage_bdev_async(struct folio *fol= io, =20 void __swap_writepage(struct folio *folio, struct swap_iocb **swap_plug) { - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); =20 VM_BUG_ON_FOLIO(!folio_test_swapcache(folio), folio); /* @@ -537,7 +537,7 @@ static bool swap_read_folio_zeromap(struct folio *folio) =20 static void swap_read_folio_fs(struct folio *folio, struct swap_iocb **plu= g) { - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); struct swap_iocb *sio =3D NULL; loff_t pos =3D swap_dev_pos(folio->swap); =20 @@ -608,7 +608,7 @@ static void swap_read_folio_bdev_async(struct folio *fo= lio, =20 void swap_read_folio(struct folio *folio, struct swap_iocb **plug) { - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); bool synchronous =3D sis->flags & SWP_SYNCHRONOUS_IO; bool workingset =3D folio_test_workingset(folio); unsigned long pflags; diff --git a/mm/swap.h b/mm/swap.h index 138b5197c35e..30b1039c27fe 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -15,6 +15,8 @@ extern int page_cluster; #define swap_entry_order(order) 0 #endif =20 +extern struct swap_info_struct *swap_info[]; + /* * We use this to track usage of a cluster. A cluster is a block of swap d= isk * space with SWAPFILE_CLUSTER pages long and naturally aligns in disk. All @@ -53,9 +55,29 @@ enum swap_cluster_flags { #include /* for swp_offset */ #include /* for bio_end_io_t */ =20 -static inline struct swap_cluster_info *swp_offset_cluster( +/* + * Callers of all helpers below must ensure the entry, type, or offset is + * valid, and protect the swap device with reference count or locks. + */ +static inline struct swap_info_struct *__swap_type_to_info(int type) +{ + struct swap_info_struct *si; + + si =3D READ_ONCE(swap_info[type]); /* rcu_dereference() */ + VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ + return si; +} + +static inline struct swap_info_struct *__swap_entry_to_info(swp_entry_t en= try) +{ + return __swap_type_to_info(swp_type(entry)); +} + +static inline struct swap_cluster_info *__swap_offset_to_cluster( struct swap_info_struct *si, pgoff_t offset) { + VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ + VM_WARN_ON_ONCE(offset >=3D si->max); return &si->cluster_info[offset / SWAPFILE_CLUSTER]; } =20 @@ -70,8 +92,9 @@ static inline struct swap_cluster_info *swp_offset_cluste= r( static inline struct swap_cluster_info *swap_cluster_lock( struct swap_info_struct *si, unsigned long offset) { - struct swap_cluster_info *ci =3D swp_offset_cluster(si, offset); + struct swap_cluster_info *ci =3D __swap_offset_to_cluster(si, offset); =20 + VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ spin_lock(&ci->lock); return ci; } @@ -170,7 +193,7 @@ void swap_update_readahead(struct folio *folio, struct = vm_area_struct *vma, =20 static inline unsigned int folio_swap_flags(struct folio *folio) { - return swp_swap_info(folio->swap)->flags; + return __swap_entry_to_info(folio->swap)->flags; } =20 /* @@ -181,7 +204,7 @@ static inline unsigned int folio_swap_flags(struct foli= o *folio) static inline int swap_zeromap_batch(swp_entry_t entry, int max_nr, bool *is_zeromap) { - struct swap_info_struct *sis =3D swp_swap_info(entry); + struct swap_info_struct *sis =3D __swap_entry_to_info(entry); unsigned long start =3D swp_offset(entry); unsigned long end =3D start + max_nr; bool first_bit; @@ -200,7 +223,7 @@ static inline int swap_zeromap_batch(swp_entry_t entry,= int max_nr, =20 static inline int non_swapcache_batch(swp_entry_t entry, int max_nr) { - struct swap_info_struct *si =3D swp_swap_info(entry); + struct swap_info_struct *si =3D __swap_entry_to_info(entry); pgoff_t offset =3D swp_offset(entry); int i; =20 @@ -219,6 +242,11 @@ static inline int non_swapcache_batch(swp_entry_t entr= y, int max_nr) =20 #else /* CONFIG_SWAP */ struct swap_iocb; +static inline struct swap_info_struct *__swap_entry_to_info(swp_entry_t en= try) +{ + return NULL; +} + static inline void swap_read_folio(struct folio *folio, struct swap_iocb *= *plug) { } diff --git a/mm/swap_state.c b/mm/swap_state.c index 9225d6b695ad..0ad4f3b41f1b 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -336,7 +336,7 @@ struct folio *__read_swap_cache_async(swp_entry_t entry= , gfp_t gfp_mask, struct mempolicy *mpol, pgoff_t ilx, bool *new_page_allocated, bool skip_if_exists) { - struct swap_info_struct *si =3D swp_swap_info(entry); + struct swap_info_struct *si =3D __swap_entry_to_info(entry); struct folio *folio; struct folio *new_folio =3D NULL; struct folio *result =3D NULL; @@ -560,7 +560,7 @@ struct folio *swap_cluster_readahead(swp_entry_t entry,= gfp_t gfp_mask, unsigned long offset =3D entry_offset; unsigned long start_offset, end_offset; unsigned long mask; - struct swap_info_struct *si =3D swp_swap_info(entry); + struct swap_info_struct *si =3D __swap_entry_to_info(entry); struct blk_plug plug; struct swap_iocb *splug =3D NULL; bool page_allocated; diff --git a/mm/swapfile.c b/mm/swapfile.c index 3bfe9cf06fd2..51c159aa646c 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -102,7 +102,7 @@ static PLIST_HEAD(swap_active_head); static struct plist_head *swap_avail_heads; static DEFINE_SPINLOCK(swap_avail_lock); =20 -static struct swap_info_struct *swap_info[MAX_SWAPFILES]; +struct swap_info_struct *swap_info[MAX_SWAPFILES]; =20 static DEFINE_MUTEX(swapon_mutex); =20 @@ -124,14 +124,20 @@ static DEFINE_PER_CPU(struct percpu_swap_cluster, per= cpu_swap_cluster) =3D { .lock =3D INIT_LOCAL_LOCK(), }; =20 -static struct swap_info_struct *swap_type_to_swap_info(int type) +/* May return NULL on invalid type, caller must check for NULL return */ +static struct swap_info_struct *swap_type_to_info(int type) { if (type >=3D MAX_SWAPFILES) return NULL; - return READ_ONCE(swap_info[type]); /* rcu_dereference() */ } =20 +/* May return NULL on invalid entry, caller must check for NULL return */ +static struct swap_info_struct *swap_entry_to_info(swp_entry_t entry) +{ + return swap_type_to_info(swp_type(entry)); +} + static inline unsigned char swap_count(unsigned char ent) { return ent & ~SWAP_HAS_CACHE; /* may include COUNT_CONTINUED flag */ @@ -341,7 +347,7 @@ offset_to_swap_extent(struct swap_info_struct *sis, uns= igned long offset) =20 sector_t swap_folio_sector(struct folio *folio) { - struct swap_info_struct *sis =3D swp_swap_info(folio->swap); + struct swap_info_struct *sis =3D __swap_entry_to_info(folio->swap); struct swap_extent *se; sector_t sector; pgoff_t offset; @@ -1299,7 +1305,7 @@ static struct swap_info_struct *_swap_info_get(swp_en= try_t entry) =20 if (!entry.val) goto out; - si =3D swp_swap_info(entry); + si =3D swap_entry_to_info(entry); if (!si) goto bad_nofile; if (data_race(!(si->flags & SWP_USED))) @@ -1414,7 +1420,7 @@ struct swap_info_struct *get_swap_device(swp_entry_t = entry) =20 if (!entry.val) goto out; - si =3D swp_swap_info(entry); + si =3D swap_entry_to_info(entry); if (!si) goto bad_nofile; if (!get_swap_device_info(si)) @@ -1537,7 +1543,7 @@ static void swap_entries_free(struct swap_info_struct= *si, unsigned char *map_end =3D map + nr_pages; =20 /* It should never free entries across different clusters */ - VM_BUG_ON(ci !=3D swp_offset_cluster(si, offset + nr_pages - 1)); + VM_BUG_ON(ci !=3D __swap_offset_to_cluster(si, offset + nr_pages - 1)); VM_BUG_ON(cluster_is_empty(ci)); VM_BUG_ON(ci->count < nr_pages); =20 @@ -1595,7 +1601,7 @@ void put_swap_folio(struct folio *folio, swp_entry_t = entry) =20 int __swap_count(swp_entry_t entry) { - struct swap_info_struct *si =3D swp_swap_info(entry); + struct swap_info_struct *si =3D __swap_entry_to_info(entry); pgoff_t offset =3D swp_offset(entry); =20 return swap_count(si->swap_map[offset]); @@ -1826,7 +1832,7 @@ void free_swap_and_cache_nr(swp_entry_t entry, int nr) =20 swp_entry_t get_swap_page_of_type(int type) { - struct swap_info_struct *si =3D swap_type_to_swap_info(type); + struct swap_info_struct *si =3D swap_type_to_info(type); unsigned long offset; swp_entry_t entry =3D {0}; =20 @@ -1907,7 +1913,7 @@ int find_first_swap(dev_t *device) */ sector_t swapdev_block(int type, pgoff_t offset) { - struct swap_info_struct *si =3D swap_type_to_swap_info(type); + struct swap_info_struct *si =3D swap_type_to_info(type); struct swap_extent *se; =20 if (!si || !(si->flags & SWP_WRITEOK)) @@ -2836,7 +2842,7 @@ static void *swap_start(struct seq_file *swap, loff_t= *pos) if (!l) return SEQ_START_TOKEN; =20 - for (type =3D 0; (si =3D swap_type_to_swap_info(type)); type++) { + for (type =3D 0; (si =3D swap_type_to_info(type)); type++) { if (!(si->flags & SWP_USED) || !si->swap_map) continue; if (!--l) @@ -2857,7 +2863,7 @@ static void *swap_next(struct seq_file *swap, void *v= , loff_t *pos) type =3D si->type + 1; =20 ++(*pos); - for (; (si =3D swap_type_to_swap_info(type)); type++) { + for (; (si =3D swap_type_to_info(type)); type++) { if (!(si->flags & SWP_USED) || !si->swap_map) continue; return si; @@ -3530,7 +3536,7 @@ static int __swap_duplicate(swp_entry_t entry, unsign= ed char usage, int nr) unsigned char has_cache; int err, i; =20 - si =3D swp_swap_info(entry); + si =3D swap_entry_to_info(entry); if (WARN_ON_ONCE(!si)) { pr_err("%s%08lx\n", Bad_file, entry.val); return -EINVAL; @@ -3645,11 +3651,6 @@ void swapcache_clear(struct swap_info_struct *si, sw= p_entry_t entry, int nr) swap_entries_put_cache(si, entry, nr); } =20 -struct swap_info_struct *swp_swap_info(swp_entry_t entry) -{ - return swap_type_to_swap_info(swp_type(entry)); -} - /* * add_swap_count_continuation - called when a swap count is duplicated * beyond SWAP_MAP_MAX, it allocates a new page and links that to the entr= y's --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f171.google.com (mail-pf1-f171.google.com [209.85.210.171]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id AE7E4334733 for ; Wed, 10 Sep 2025 16:09:26 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.171 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520568; cv=none; b=Z6KepTx3TGVxlSXMmf8XMq2xcfuYZWHx6IIjbTHHocINhsGf/7AlajLYvZiz4lznS/EWd17ETdRQWBjkjYVIu2qqN1MG73iJJiWS6mPmdZJUqRvgm5NL6iUUBZWzOe986SMKCeXU+hnFdcLCFcLTv3UJXcFM7pcbHadpl6e422E= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520568; c=relaxed/simple; bh=LOotFfbZVx0rWQf/JEDe/ZG6D+kArOuTSBSkBrqVv9s=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=uidb+93VzikIc3k9nBoyAyGLuDXj0Iu12ll+RmDWs3JwdORvDfBDkN7t2Fe8Ngqi6eTtYJTJQDwUHvXdlI59YRmYECV+wXOCu4d54IaTNfq1GzWRk5fW9qTA7t27awh1KQder3fKfWkFCk9UOrdzTtt04tIAqhFkQbIcFdIBS/M= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=Ayx9600/; arc=none smtp.client-ip=209.85.210.171 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="Ayx9600/" Received: by mail-pf1-f171.google.com with SMTP id d2e1a72fcca58-77256e75eacso6346656b3a.0 for ; Wed, 10 Sep 2025 09:09:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520566; x=1758125366; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=41IrheQrU1uNONBC2D87eHUDc3ROCWh6baPpZqwFKeI=; b=Ayx9600/wC14DhdS8GRdYU64qP2ApX6jUkqYW+akMkIW/ZTcMdwDAofRv0pO1celg/ iOEe+iQl+JAJFKkhXuUH3u0lct4LFq0SWJgOO6ekC4DkSf2zipmErRbDekpKsbH0zF6V f2RyGjRFojE1ImXNxys+zUdpJgjPqeMoI6JFCJhvcVVV3k+WnirEyxFYhte4vlbs97Fp C3NNLqPu/OwFYstLvhV1baS6fVZuW9Thmd0TuW3ilGkcz+nMd4hbr51u8lZr74HrFBeN Z8+KrHcTh8nXMLO1S+wHwi57o80P+1lg9K7hDimE9mYGCnMtTFweRiOWU0dPoh+eudF+ voWQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520566; x=1758125366; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=41IrheQrU1uNONBC2D87eHUDc3ROCWh6baPpZqwFKeI=; b=gLAfoRr2wScRNiSkUXeTkz5xLxTfzOmWKmDcOf80tN5n0t8HB3f6XI+BwzVa0LZV3h OGfmsSQ+hALSVR4N4US32PW8Peyuq2HPVUkrLNEHJHJutOMFc1o9oDt4u2xQPdzcJeYI ZYJWtUHdTjk92K2RLok92ymuFeWxAI+lAdEYpHHjtTgXZdPWYUPnglqJJQR1boWsotA8 EtwetT3gPWs2/p7uCt4z/hLCm5p2K5zmsG7T4r78kdEgv6k7kt+lEodw9TElpFyZDRCl /YmDtfa1vU1Xp5HiX6ax0y920NFUtAxHIwuuepME+zFbep5yRcYQgH/wzjTD9GVgAU+3 pV3A== X-Forwarded-Encrypted: i=1; AJvYcCV1x1QnEmwMYaBy/D6wlry/bQqnz2d1uw+1Jxi7F5b+JOziwnob8qUJuSzBRVX6fULNMQXchtkQHB7suXs=@vger.kernel.org X-Gm-Message-State: AOJu0YwzkLONRYWDSv8z4/9n1w7i23ZQv3HNVBSOrbq2LPSXUXvChUoy LS/juap/ggYLdnDtoPZOo93wOzQs6xPo5h9zDOd97xQ8TPzGtQJ6T5KI X-Gm-Gg: ASbGncu0crGVvTtWwAGWmfYyRjvqjXxMZn6nVxsraamkOaYuIRkzj1MwhvHRqofnSDy RWPypvsoY7mDsuJoz3rOjJVPJxCnNWiMJNIN9cZT95Y3mSjWzdBKMqOfxDPhyasfbbV3IeCB9Nh oqE8wjYGbCU243aIFChMGO2vciIH1EWDYqJ/QxHml5FCTv06OSfDtzq68j/wS9itzMSM48raikN HNJil9RATuSLfqt+uL6WKPCOkzzi8NTKzyG2VcOyaEBm/NOw9ub+9+MzKAkGp8dKMOThGUThpA6 x06kCPNSR8KT5BUJN5K2oyHRfGzH1Ql7UZsLpSeFnMLwTAC9sRPQijxrXE0WMYFza81eIsQugqY XsAHhLC5lMJ5R35LhM5epbStu7ZBi5ctCW2zlfm1wBiOF2OSQIRROQft2Jg== X-Google-Smtp-Source: AGHT+IGX1OhgXhkIxofCcoD4B7nQIR9FtlH0Q7qIyxMOhhjXQBkvBpqTM/Oc2XmvflNOYzzMsOTu9A== X-Received: by 2002:a05:6a20:2591:b0:24e:c235:d7e6 with SMTP id adf61e73a8af0-2533d5fed90mr25408969637.5.1757520565690; Wed, 10 Sep 2025 09:09:25 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.20 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:25 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 08/15] mm, swap: cleanup swap cache API and add kerneldoc Date: Thu, 11 Sep 2025 00:08:26 +0800 Message-ID: <20250910160833.3464-9-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song In preparation for replacing the swap cache backend with the swap table, clean up and add proper kernel doc for all swap cache APIs. Now all swap cache APIs are well-defined with consistent names. No feature change, only renaming and documenting. Signed-off-by: Kairui Song Acked-by: Chris Li Reviewed-by: Barry Song Reviewed-by: Baolin Wang Acked-by: David Hildenbrand Suggested-by: Chris Li --- mm/filemap.c | 2 +- mm/memory-failure.c | 2 +- mm/memory.c | 2 +- mm/swap.h | 48 ++++++++++++++----------- mm/swap_state.c | 86 ++++++++++++++++++++++++++++++++------------- mm/swapfile.c | 8 ++--- mm/vmscan.c | 2 +- mm/zswap.c | 2 +- 8 files changed, 98 insertions(+), 54 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 8d078aa2738a..2a05b1fdd445 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -4525,7 +4525,7 @@ static void filemap_cachestat(struct address_space *m= apping, * invalidation, so there might not be * a shadow in the swapcache (yet). */ - shadow =3D get_shadow_from_swap_cache(swp); + shadow =3D swap_cache_get_shadow(swp); if (!shadow) goto resched; } diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 6d9134e3d115..3edebb0cda30 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1127,7 +1127,7 @@ static int me_swapcache_clean(struct page_state *ps, = struct page *p) struct folio *folio =3D page_folio(p); int ret; =20 - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); =20 ret =3D delete_from_lru_cache(folio) ? MF_FAILED : MF_RECOVERED; folio_unlock(folio); diff --git a/mm/memory.c b/mm/memory.c index 5808c4ef21b3..41e641823558 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -4699,7 +4699,7 @@ vm_fault_t do_swap_page(struct vm_fault *vmf) =20 memcg1_swapin(entry, nr_pages); =20 - shadow =3D get_shadow_from_swap_cache(entry); + shadow =3D swap_cache_get_shadow(entry); if (shadow) workingset_refault(folio, shadow); =20 diff --git a/mm/swap.h b/mm/swap.h index 30b1039c27fe..6c4acb549bec 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -167,17 +167,29 @@ static inline bool folio_matches_swap_entry(const str= uct folio *folio, return folio_entry.val =3D=3D round_down(entry.val, nr_pages); } =20 +/* + * All swap cache helpers below require the caller to ensure the swap entr= ies + * used are valid and stablize the device by any of the following ways: + * - Hold a reference by get_swap_device(): this ensures a single entry is + * valid and increases the swap device's refcount. + * - Locking a folio in the swap cache: this ensures the folio's swap entr= ies + * are valid and pinned, also implies reference to the device. + * - Locking anything referencing the swap entry: e.g. PTL that protects + * swap entries in the page table, similar to locking swap cache folio. + * - See the comment of get_swap_device() for more complex usage. + */ +struct folio *swap_cache_get_folio(swp_entry_t entry); +void *swap_cache_get_shadow(swp_entry_t entry); +int swap_cache_add_folio(struct folio *folio, swp_entry_t entry, + gfp_t gfp, void **shadow); +void swap_cache_del_folio(struct folio *folio); +void __swap_cache_del_folio(struct folio *folio, + swp_entry_t entry, void *shadow); +void swap_cache_clear_shadow(int type, unsigned long begin, + unsigned long end); + void show_swap_cache_info(void); -void *get_shadow_from_swap_cache(swp_entry_t entry); -int add_to_swap_cache(struct folio *folio, swp_entry_t entry, - gfp_t gfp, void **shadowp); -void __delete_from_swap_cache(struct folio *folio, - swp_entry_t entry, void *shadow); -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, int n= r); -struct folio *swap_cache_get_folio(swp_entry_t entry); struct folio *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, struct vm_area_struct *vma, unsigned long addr, struct swap_iocb **plug); @@ -305,28 +317,22 @@ static inline struct folio *swap_cache_get_folio(swp_= entry_t entry) return NULL; } =20 -static inline void *get_shadow_from_swap_cache(swp_entry_t entry) +static inline void *swap_cache_get_shadow(swp_entry_t entry) { return NULL; } =20 -static inline int add_to_swap_cache(struct folio *folio, swp_entry_t entry, - gfp_t gfp_mask, void **shadowp) -{ - return -1; -} - -static inline void __delete_from_swap_cache(struct folio *folio, - swp_entry_t entry, void *shadow) +static inline int swap_cache_add_folio(swp_entry_t entry, struct folio *fo= lio, + gfp_t gfp, void **shadow) { + return -EINVAL; } =20 -static inline void delete_from_swap_cache(struct folio *folio) +static inline void swap_cache_del_folio(struct folio *folio) { } =20 -static inline void clear_shadow_from_swap_cache(int type, unsigned long be= gin, - unsigned long end) +static inline void __swap_cache_del_folio(struct folio *folio, swp_entry_t= entry, void *shadow) { } =20 diff --git a/mm/swap_state.c b/mm/swap_state.c index 0ad4f3b41f1b..f3a32a06a950 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -78,8 +78,8 @@ void show_swap_cache_info(void) * Context: Caller must ensure @entry is valid and protect the swap device * with reference count or locks. * Return: Returns the found folio on success, NULL otherwise. The caller - * must lock and check if the folio still matches the swap entry before - * use (e.g. with folio_matches_swap_entry). + * must lock nd check if the folio still matches the swap entry before + * use (e.g., folio_matches_swap_entry). */ struct folio *swap_cache_get_folio(swp_entry_t entry) { @@ -90,7 +90,15 @@ struct folio *swap_cache_get_folio(swp_entry_t entry) return folio; } =20 -void *get_shadow_from_swap_cache(swp_entry_t entry) +/** + * swap_cache_get_shadow - Looks up a shadow in the swap cache. + * @entry: swap entry used for the lookup. + * + * Context: Caller must ensure @entry is valid and protect the swap device + * with reference count or locks. + * Return: Returns either NULL or an XA_VALUE (shadow). + */ +void *swap_cache_get_shadow(swp_entry_t entry) { struct address_space *address_space =3D swap_address_space(entry); pgoff_t idx =3D swap_cache_index(entry); @@ -102,12 +110,21 @@ void *get_shadow_from_swap_cache(swp_entry_t entry) return NULL; } =20 -/* - * add_to_swap_cache resembles filemap_add_folio on swapper_space, - * but sets SwapCache flag and 'swap' instead of mapping and index. +/** + * swap_cache_add_folio - Add a folio into the swap cache. + * @folio: The folio to be added. + * @entry: The swap entry corresponding to the folio. + * @gfp: gfp_mask for XArray node allocation. + * @shadowp: If a shadow is found, return the shadow. + * + * Context: Caller must ensure @entry is valid and protect the swap device + * with reference count or locks. + * The caller also needs to mark the corresponding swap_map slots with + * SWAP_HAS_CACHE to avoid race or conflict. + * Return: Returns 0 on success, error code otherwise. */ -int add_to_swap_cache(struct folio *folio, swp_entry_t entry, - gfp_t gfp, void **shadowp) +int swap_cache_add_folio(struct folio *folio, swp_entry_t entry, + gfp_t gfp, void **shadowp) { struct address_space *address_space =3D swap_address_space(entry); pgoff_t idx =3D swap_cache_index(entry); @@ -155,12 +172,20 @@ int add_to_swap_cache(struct folio *folio, swp_entry_= t entry, return xas_error(&xas); } =20 -/* - * This must be called only on folios that have - * been verified to be in the swap cache. +/** + * __swap_cache_del_folio - Removes a folio from the swap cache. + * @folio: The folio. + * @entry: The first swap entry that the folio corresponds to. + * @shadow: shadow value to be filled in the swap cache. + * + * Removes a folio from the swap cache and fills a shadow in place. + * This won't put the folio's refcount. The caller has to do that. + * + * Context: Caller must hold the xa_lock, ensure the folio is + * locked and in the swap cache, using the index of @entry. */ -void __delete_from_swap_cache(struct folio *folio, - swp_entry_t entry, void *shadow) +void __swap_cache_del_folio(struct folio *folio, + swp_entry_t entry, void *shadow) { struct address_space *address_space =3D swap_address_space(entry); int i; @@ -186,27 +211,40 @@ void __delete_from_swap_cache(struct folio *folio, __lruvec_stat_mod_folio(folio, NR_SWAPCACHE, -nr); } =20 -/* - * This must be called only on folios that have - * been verified to be in the swap cache and locked. - * It will never put the folio into the free list, - * the caller has a reference on the folio. +/** + * swap_cache_del_folio - Removes a folio from the swap cache. + * @folio: The folio. + * + * Same as __swap_cache_del_folio, but handles lock and refcount. The + * caller must ensure the folio is either clean or has a swap count + * equal to zero, or it may cause data loss. + * + * Context: Caller must ensure the folio is locked and in the swap cache. */ -void delete_from_swap_cache(struct folio *folio) +void swap_cache_del_folio(struct folio *folio) { swp_entry_t entry =3D folio->swap; struct address_space *address_space =3D swap_address_space(entry); =20 xa_lock_irq(&address_space->i_pages); - __delete_from_swap_cache(folio, entry, NULL); + __swap_cache_del_folio(folio, entry, NULL); xa_unlock_irq(&address_space->i_pages); =20 put_swap_folio(folio, entry); folio_ref_sub(folio, folio_nr_pages(folio)); } =20 -void clear_shadow_from_swap_cache(int type, unsigned long begin, - unsigned long end) +/** + * swap_cache_clear_shadow - Clears a set of shadows in the swap cache. + * @type: Indicates the swap device. + * @begin: Beginning offset of the range. + * @end: Ending offset of the range. + * + * Context: Caller must ensure the range is valid and hold a reference to + * the swap device. + */ +void swap_cache_clear_shadow(int type, unsigned long begin, + unsigned long end) { unsigned long curr =3D begin; void *old; @@ -393,7 +431,7 @@ struct folio *__read_swap_cache_async(swp_entry_t entry= , gfp_t gfp_mask, goto put_and_return; =20 /* - * We might race against __delete_from_swap_cache(), and + * We might race against __swap_cache_del_folio(), and * stumble across a swap_map entry whose SWAP_HAS_CACHE * has not yet been cleared. Or race against another * __read_swap_cache_async(), which has set SWAP_HAS_CACHE @@ -412,7 +450,7 @@ struct folio *__read_swap_cache_async(swp_entry_t entry= , gfp_t gfp_mask, goto fail_unlock; =20 /* May fail (-ENOMEM) if XArray node allocation failed. */ - if (add_to_swap_cache(new_folio, entry, gfp_mask & GFP_RECLAIM_MASK, &sha= dow)) + if (swap_cache_add_folio(new_folio, entry, gfp_mask & GFP_RECLAIM_MASK, &= shadow)) goto fail_unlock; =20 memcg1_swapin(entry, 1); diff --git a/mm/swapfile.c b/mm/swapfile.c index 51c159aa646c..6f5206255789 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -266,7 +266,7 @@ static int __try_to_reclaim_swap(struct swap_info_struc= t *si, if (!need_reclaim) goto out_unlock; =20 - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); folio_set_dirty(folio); ret =3D nr_pages; out_unlock: @@ -1123,7 +1123,7 @@ static void swap_range_free(struct swap_info_struct *= si, unsigned long offset, swap_slot_free_notify(si->bdev, offset); offset++; } - clear_shadow_from_swap_cache(si->type, begin, end); + swap_cache_clear_shadow(si->type, begin, end); =20 /* * Make sure that try_to_unuse() observes si->inuse_pages reaching 0 @@ -1288,7 +1288,7 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp) * TODO: this could cause a theoretical memory reclaim * deadlock in the swap out path. */ - if (add_to_swap_cache(folio, entry, gfp | __GFP_NOMEMALLOC, NULL)) + if (swap_cache_add_folio(folio, entry, gfp | __GFP_NOMEMALLOC, NULL)) goto out_free; =20 return 0; @@ -1758,7 +1758,7 @@ bool folio_free_swap(struct folio *folio) if (folio_swapped(folio)) return false; =20 - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); folio_set_dirty(folio); return true; } diff --git a/mm/vmscan.c b/mm/vmscan.c index ca9e1cd3cd68..c79c6806560b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -776,7 +776,7 @@ static int __remove_mapping(struct address_space *mappi= ng, struct folio *folio, =20 if (reclaimed && !mapping_exiting(mapping)) shadow =3D workingset_eviction(folio, target_memcg); - __delete_from_swap_cache(folio, swap, shadow); + __swap_cache_del_folio(folio, swap, shadow); memcg1_swapout(folio, swap); xa_unlock_irq(&mapping->i_pages); put_swap_folio(folio, swap); diff --git a/mm/zswap.c b/mm/zswap.c index c88ad61b232c..3dda4310099e 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -1069,7 +1069,7 @@ static int zswap_writeback_entry(struct zswap_entry *= entry, =20 out: if (ret && ret !=3D -EEXIST) { - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); folio_unlock(folio); } folio_put(folio); --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f175.google.com (mail-pf1-f175.google.com [209.85.210.175]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7259632F774 for ; Wed, 10 Sep 2025 16:09:31 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.175 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520573; cv=none; b=tzGZBu/78Dh84VP9lF7PKlfqMu2Xooh/XzqDz3ArI60unRFc5Fr7wLQDlZbMVDrjrPKkEzYRo539g9+y3IHM2m7DU0SlYHcI0Kh70bjBoucZ+yfC8V7ZtC8CnSCIUaihKIBzUcm9QHoQ5oos4FXKDSm5IglKBMXr4RgOz8h0w3w= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520573; c=relaxed/simple; bh=89lTN+lYG5EkPRgJ5HkI886n9BOFunertfvU1h3jT8I=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=FN8Rzcf75s3ExLX0DQVlq1qJ6snt549n8N8ojUmpJGqFY2eXtxyiwAIxQQ2y+0s9swNRXEEJwFZxgMWVwaIHRxzS2iSgjttp2+Sl99CsMQ1202fnHjpbAOp8WaD2yhve63Nu39K0yBmhWHtxyZhnj+q4m5miGZTm3XsEJtygtg4= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=UirNdaie; arc=none smtp.client-ip=209.85.210.175 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="UirNdaie" Received: by mail-pf1-f175.google.com with SMTP id d2e1a72fcca58-7724cacc32bso5634629b3a.0 for ; Wed, 10 Sep 2025 09:09:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520571; x=1758125371; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=griGxIh9GEE3HF6pE5uHZRdQ9DQCWTXYvIepBVp/qTo=; b=UirNdaieFTuSuNpGd80X4zDZ6YrTKRenJQietkMGjOWkaV8oXWRkBvtxJiFjs9DiwO c6ZTOyamudGPe5YbiqhcVzlsBh2k51qjkhTh53s1B3TAmdQg0no7vgXK3L2t/AEdhgjk y4Foubef244pFM5srpAuREmxM8lW+IfWlVq54nvid3nkuEizuKUiEOV7cu0oz1LnQjvY 0X8XFYKCgeBUWOTlQixIYdquZ8ll1UKaDJljXJHz0noGgqx0yeGlbosab8eXf6Rj073A 4ztXXC8RvJw6NB3A8bVpK9O0yWzc76NpcVROF7h5v4IsY8PJZaC4lmZ9gY8BZsrGjqOH 1XCA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520571; x=1758125371; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=griGxIh9GEE3HF6pE5uHZRdQ9DQCWTXYvIepBVp/qTo=; b=uQM2irV93lHsgTJ4fBG1/x3jeqiAKV36jh0JIC7FWBGvAPenb8UZaOx9645wxt/uOP zzBz4wjPg0E4ypeNCogTsJzk6pMzgrBVKJKzEo2Z4J8m6Wn1xJLCHz9pBKvzO+Tj3vHB tOn9JoHpMOT8p0HIHfIdi4ua2vOsqxWOFcstFqZu9OOiH11fHJ4CEJKRvPjTYx6O95cb V3b67EBeeJUkVXuJyH6mFpLIBusoTTdq32tfsjGTzI0lynjtuQSebU66soqrrAZwKt/6 ej30cUlkEozVJ3c6siVAg/Sz0abvYFiLO38Np9J2ln9cdYESK2EgfpY7xsFO8Y/AKYS6 ISxQ== X-Forwarded-Encrypted: i=1; AJvYcCXN/089L60zDC/o+NN6NRxs5Z83HCGJwStrZjzMDKEYmHE4i4L3/a7vPXDzkYGJGLtbDrvOQgAOXHg80wE=@vger.kernel.org X-Gm-Message-State: AOJu0Yy0W4RHm7qO/IDREyEvAIsWEnSdVJiCbQ8TMw1/kKYqCGvHm0hS jtptc90o85MPZyWtzafiEcKxnmslvrTvnkI3kwDr/Kh/s5+3u7DZS1hO X-Gm-Gg: ASbGncuStsBH1u0+HwbzARSKewk80gYXCEUBHM2IwG7JOj3yZjl7Mf/rKaYIsoJNu+F jHEyE3ZnhE7ZsrJ0rcAbfmaWktYKm7nUplJ6xW0tYmMN2/mhY3FpE6t0qGVsIAsvRy/O2cGNZ9z QtfiFBS1uB2ot1isF4CjgT7XPp58YxxRzpsA5mB3IgJonMSJh9Qh4+YUlTwzYuu8JLDlmNlJ8d4 qjybExTZjO+75ywM1xSOh8Uf14xkpT6BQEwO3wLR6lxv3lVAA5oxftzK2E6HycNkT4OwkXFY46t pfmZQti5BAnZaGQK+YUMiP5hOjhE8kBl7sGLFi4b0xjFdWWXQ9CaEDkCdTFBEE5USVVvX/SeK3q flod1nX6D/FPyNB8aRdF8HgaDnHgsDcASmnMsY+hVDy7A0RM= X-Google-Smtp-Source: AGHT+IF79Trm+wcbzKOddKHrBqAVQp9c3H0wicsNS7W5K+jd07nVhIvIvzw66BfjwtQW3jqIqjmWZQ== X-Received: by 2002:a05:6a20:7484:b0:250:9175:96df with SMTP id adf61e73a8af0-25344415c9bmr25513067637.48.1757520570590; Wed, 10 Sep 2025 09:09:30 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.26 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:30 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 09/15] mm/shmem, swap: remove redundant error handling for replacing folio Date: Thu, 11 Sep 2025 00:08:27 +0800 Message-ID: <20250910160833.3464-10-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Shmem may replace a folio in the swap cache if the cached one doesn't fit the swapin's GFP zone. When doing so, shmem has already double checked that the swap cache folio is locked, still has the swap cache flag set, and contains the wanted swap entry. So it is impossible to fail due to an XArray mismatch. There is even a comment for that. Delete the defensive error handling path, and add a WARN_ON instead: if that happened, something has broken the basic principle of how the swap cache works, we should catch and fix that. Signed-off-by: Kairui Song Reviewed-by: David Hildenbrand Suggested-by: Chris Li --- mm/shmem.c | 42 ++++++++++++------------------------------ 1 file changed, 12 insertions(+), 30 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index 410f27bc4752..5f395fab489c 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1661,13 +1661,13 @@ int shmem_writeout(struct folio *folio, struct swap= _iocb **plug, } =20 /* - * The delete_from_swap_cache() below could be left for + * The swap_cache_del_folio() below could be left for * shrink_folio_list()'s folio_free_swap() to dispose of; * but I'm a little nervous about letting this folio out of * shmem_writeout() in a hybrid half-tmpfs-half-swap state * e.g. folio_mapping(folio) might give an unexpected answer. */ - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); goto redirty; } if (nr_pages > 1) @@ -2045,7 +2045,7 @@ static struct folio *shmem_swap_alloc_folio(struct in= ode *inode, new->swap =3D entry; =20 memcg1_swapin(entry, nr_pages); - shadow =3D get_shadow_from_swap_cache(entry); + shadow =3D swap_cache_get_shadow(entry); if (shadow) workingset_refault(new, shadow); folio_add_lru(new); @@ -2121,35 +2121,17 @@ static int shmem_replace_folio(struct folio **folio= p, gfp_t gfp, /* Swap cache still stores N entries instead of a high-order entry */ xa_lock_irq(&swap_mapping->i_pages); for (i =3D 0; i < nr_pages; i++) { - void *item =3D xas_load(&xas); - - if (item !=3D old) { - error =3D -ENOENT; - break; - } - - xas_store(&xas, new); + WARN_ON_ONCE(xas_store(&xas, new)); xas_next(&xas); } - if (!error) { - mem_cgroup_replace_folio(old, new); - shmem_update_stats(new, nr_pages); - shmem_update_stats(old, -nr_pages); - } xa_unlock_irq(&swap_mapping->i_pages); =20 - if (unlikely(error)) { - /* - * Is this possible? I think not, now that our callers - * check both the swapcache flag and folio->private - * after getting the folio lock; but be defensive. - * Reverse old to newpage for clear and free. - */ - old =3D new; - } else { - folio_add_lru(new); - *foliop =3D new; - } + mem_cgroup_replace_folio(old, new); + shmem_update_stats(new, nr_pages); + shmem_update_stats(old, -nr_pages); + + folio_add_lru(new); + *foliop =3D new; =20 folio_clear_swapcache(old); old->private =3D NULL; @@ -2183,7 +2165,7 @@ static void shmem_set_folio_swapin_error(struct inode= *inode, pgoff_t index, nr_pages =3D folio_nr_pages(folio); folio_wait_writeback(folio); if (!skip_swapcache) - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); /* * Don't treat swapin error folio as alloced. Otherwise inode->i_blocks * won't be 0 when inode is released and thus trigger WARN_ON(i_blocks) @@ -2422,7 +2404,7 @@ static int shmem_swapin_folio(struct inode *inode, pg= off_t index, folio->swap.val =3D 0; swapcache_clear(si, swap, nr_pages); } else { - delete_from_swap_cache(folio); + swap_cache_del_folio(folio); } folio_mark_dirty(folio); swap_free_nr(swap, nr_pages); --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f171.google.com (mail-pf1-f171.google.com [209.85.210.171]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 42C2333CEBA for ; Wed, 10 Sep 2025 16:09:36 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.171 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520578; cv=none; b=Vk/sXXYKV6eOXa9tp3aYYbRJ8mdGm9qJbc1xeW4ty1Y9HlZEMvfqtwgLlA24ND5uKrblX2l7Hh5AIXx7VTTG+xhkY66zYyb4iFNfGQp+n/XuAV4+HW7ROtilTTq6GpFWRnEh5QjWtIzvnVKtMMjHCoJpDj3Nh80zkQpx89Xlkpg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520578; c=relaxed/simple; bh=wezNe+McLtBEyF9B2AQrUKduS6WWFRf6QeCr+ValA6A=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=NhbFysAauzJB8dLiDB+UVG4qFn6iiJBLf0/hjM9HrvVvlFtegmIyYI/GkxMOxRl3f8/XcRIHvtcfTcTLZlH28ovM5AOtftTXdf3gcBUqabdBvKllnlpyv8DNv3zshK3eZV6fwy166BMB2cdffEfEcPEndX44Uwvt96ruQcH+DMY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=NhqYjrI3; arc=none smtp.client-ip=209.85.210.171 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="NhqYjrI3" Received: by mail-pf1-f171.google.com with SMTP id d2e1a72fcca58-7724cacc32bso5634755b3a.0 for ; Wed, 10 Sep 2025 09:09:36 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520575; x=1758125375; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=TXr0Gr+eyBH1J2ZOC4B1nLxnwl36iHb/5ARHAVX+lWM=; b=NhqYjrI3efKGWa2VjQLoN+h/VLJxY/3ZfgJpmNryCHTSTw71gVVyCJkiMyBV9/cwUe 3EMigPg41ACLsKi5NYfuuIrKkC+mlnbACONb3///vCKVMn62oqppXlaX6ZDBJqODfOF4 A3W9VZV5wSDfOfzREQykNZGNWe27e2mQrIIh6smBYc1hGJ872VcGikX5veSU/3/lpaA5 wFFTCR5VhWbGhd6AhtWnIzBS5K5Ccpo/tg6CN4h4U1uQZePjtZJLhEHcuFXFhy8NLww/ 6nJkEP0gAgk5mn4m7R41VVtPx3Gd/N3RalP2r/fgyOSTAF0yruiweKEBRDdB76jZcfWZ VoxQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520575; x=1758125375; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=TXr0Gr+eyBH1J2ZOC4B1nLxnwl36iHb/5ARHAVX+lWM=; b=qtsREnUDL+8btQxvA7rfQMcDtnUCbXeLAyBlzRd/YnoWsR3JvBNyNAw0EjmShsNqlh bGSRhjYLKtXOHx42MTnCtLrh17op5gIMtj1/fD9WWxKgStBTjVaIGoEEi/YMz8RXk9Rr 1fyYg+wvB/z2aTyCaK1T7s4xyFwDd/Fuig8Wbh2q654fJH1ILVJ0ga7M7+g6YCuqbC9O oYknuLdICzjcdJKHBbcDCBgaenyuhbkcKIxGVZnbgW25mJ9yDHHxnDYbxJttX3ddF7/8 sCdVfP6O7cIb1UIA0j0X+afQgflFpTUrf3c+Og9J+YL8gHPg4xCK+F1DKs2SHbiWwA2S EDjA== X-Forwarded-Encrypted: i=1; AJvYcCWPYHCEbZbjZuLNyxPfp2fIwL/9pS4UENsRMfHzhC508apaWbDcQEQFUsZVFrqNnZA+AEpKaKe+TqgKN5E=@vger.kernel.org X-Gm-Message-State: AOJu0YyvoQEfHvq6pv9MuxOSI6OObZVhZCpGdQdMRJVFsLA22NHgtM7R k7pnGdV9VfQp7HxCuF1/NREuP8VuOGNnMvy9YPIPqhenE2F8+6f2Nt6B X-Gm-Gg: ASbGncv2uz4XINcxDNHaCNKT3hkzv6nDBZbGNlmPDpV7MNlzwZYPj3CGzoLtL03Eya9 Bw+N69Oy/QzPJNcqkkL4s+zpoeB9jcW2ZcpqKX0GVLEuXliSZKoW2maG2prjh/9dLpBm4kkuRp3 sBAvRGQ1QQYYDPKJfkaNvYaifQX1Nfo3GdvYqo3IcqdMmJpedm29UpdiAQbZ/7mfnVniRAv4Avj lw/nkvorRAPxmiXYx8pjUtk3T784ve191BBNxQbbNKtDp4mgBuxdf/F3G15Vk4rBD0Ch4BMALY6 dH6bRpouvSWYGSQW3IXA3jnNwS/pW+J/LS7pofMDNwPgsXggpN0YwHG/m3EZdDNhBgCUC6UKF0Z DLgG6ZC4IHnqFrjoJyWDT3e4PnvsjEhamCG9R0z7y+ifdMVAvB+JaB5b0Mw== X-Google-Smtp-Source: AGHT+IGd1fs3d/xLWFIH8x0QD2Wv+0XsIfWldgkL360dAU+YTjAIy2jV9nMZXESeItmlntKEcr6Daw== X-Received: by 2002:a05:6a20:3d19:b0:24d:301a:79b1 with SMTP id adf61e73a8af0-2533e94d169mr26623227637.15.1757520575417; Wed, 10 Sep 2025 09:09:35 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.30 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:34 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 10/15] mm, swap: wrap swap cache replacement with a helper Date: Thu, 11 Sep 2025 00:08:28 +0800 Message-ID: <20250910160833.3464-11-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song There are currently three swap cache users that are trying to replace an existing folio with a new one: huge memory splitting, migration, and shmem replacement. What they are doing is quite similar. Introduce a common helper for this. In later commits, this can be easily switched to use the swap table by updating this helper. The newly added helper also makes the swap cache API better defined, and make debugging easier by adding a few more debug checks. Migration and shmem replace are meant to clone the folio, including content, swap entry value, and flags. And splitting will adjust each sub folio's swap entry according to order, which could be non-uniform in the future. So document it clearly that it's the caller's responsibility to set up the new folio's swap entries and flags before calling the helper. The helper will just follow the new folio's entry value. This also prepares for replacing high-order folios in the swap cache. Currently, only splitting to order 0 is allowed for swap cache folios. Using the new helper, we can handle high-order folio splitting better. Signed-off-by: Kairui Song Reviewed-by: Baolin Wang Suggested-by: Chris Li --- mm/huge_memory.c | 4 +--- mm/migrate.c | 11 +++-------- mm/shmem.c | 10 ++-------- mm/swap.h | 5 +++++ mm/swap_state.c | 33 +++++++++++++++++++++++++++++++++ 5 files changed, 44 insertions(+), 19 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 26cedfcd7418..4c66e358685b 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3798,9 +3798,7 @@ static int __folio_split(struct folio *folio, unsigne= d int new_order, * NOTE: shmem in swap cache is not supported yet. */ if (swap_cache) { - __xa_store(&swap_cache->i_pages, - swap_cache_index(new_folio->swap), - new_folio, 0); + __swap_cache_replace_folio(folio, new_folio); continue; } =20 diff --git a/mm/migrate.c b/mm/migrate.c index 8e435a078fc3..c69cc13db692 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -566,7 +566,6 @@ static int __folio_migrate_mapping(struct address_space= *mapping, struct zone *oldzone, *newzone; int dirty; long nr =3D folio_nr_pages(folio); - long entries, i; =20 if (!mapping) { /* Take off deferred split queue while frozen and memcg set */ @@ -615,9 +614,6 @@ static int __folio_migrate_mapping(struct address_space= *mapping, if (folio_test_swapcache(folio)) { folio_set_swapcache(newfolio); newfolio->private =3D folio_get_private(folio); - entries =3D nr; - } else { - entries =3D 1; } =20 /* Move dirty while folio refs frozen and newfolio not yet exposed */ @@ -627,11 +623,10 @@ static int __folio_migrate_mapping(struct address_spa= ce *mapping, folio_set_dirty(newfolio); } =20 - /* Swap cache still stores N entries instead of a high-order entry */ - for (i =3D 0; i < entries; i++) { + if (folio_test_swapcache(folio)) + __swap_cache_replace_folio(folio, newfolio); + else xas_store(&xas, newfolio); - xas_next(&xas); - } =20 /* * Drop cache reference from old folio by unfreezing diff --git a/mm/shmem.c b/mm/shmem.c index 5f395fab489c..8930780325da 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2086,10 +2086,8 @@ static int shmem_replace_folio(struct folio **foliop= , gfp_t gfp, struct folio *new, *old =3D *foliop; swp_entry_t entry =3D old->swap; struct address_space *swap_mapping =3D swap_address_space(entry); - pgoff_t swap_index =3D swap_cache_index(entry); - XA_STATE(xas, &swap_mapping->i_pages, swap_index); int nr_pages =3D folio_nr_pages(old); - int error =3D 0, i; + int error =3D 0; =20 /* * We have arrived here because our zones are constrained, so don't @@ -2118,12 +2116,8 @@ static int shmem_replace_folio(struct folio **foliop= , gfp_t gfp, new->swap =3D entry; folio_set_swapcache(new); =20 - /* Swap cache still stores N entries instead of a high-order entry */ xa_lock_irq(&swap_mapping->i_pages); - for (i =3D 0; i < nr_pages; i++) { - WARN_ON_ONCE(xas_store(&xas, new)); - xas_next(&xas); - } + __swap_cache_replace_folio(old, new); xa_unlock_irq(&swap_mapping->i_pages); =20 mem_cgroup_replace_folio(old, new); diff --git a/mm/swap.h b/mm/swap.h index 6c4acb549bec..fe579c81c6c4 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -185,6 +185,7 @@ int swap_cache_add_folio(struct folio *folio, swp_entry= _t entry, void swap_cache_del_folio(struct folio *folio); void __swap_cache_del_folio(struct folio *folio, swp_entry_t entry, void *shadow); +void __swap_cache_replace_folio(struct folio *old, struct folio *new); void swap_cache_clear_shadow(int type, unsigned long begin, unsigned long end); =20 @@ -336,6 +337,10 @@ static inline void __swap_cache_del_folio(struct folio= *folio, swp_entry_t entry { } =20 +static inline void __swap_cache_replace_folio(struct folio *old, struct fo= lio *new) +{ +} + static inline unsigned int folio_swap_flags(struct folio *folio) { return 0; diff --git a/mm/swap_state.c b/mm/swap_state.c index f3a32a06a950..d1f5b8fa52fc 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -234,6 +234,39 @@ void swap_cache_del_folio(struct folio *folio) folio_ref_sub(folio, folio_nr_pages(folio)); } =20 +/** + * __swap_cache_replace_folio - Replace a folio in the swap cache. + * @old: The old folio to be replaced. + * @new: The new folio. + * + * Replace an existing folio in the swap cache with a new folio. The + * caller is responsible for setting up the new folio's flag and swap + * entries. Replacement will take the new folio's swap entry value as + * the starting offset to override all slots covered by the new folio. + * + * Context: Caller must ensure both folios are locked, also lock the + * swap address_space that holds the old folio to avoid races. + */ +void __swap_cache_replace_folio(struct folio *old, struct folio *new) +{ + swp_entry_t entry =3D new->swap; + unsigned long nr_pages =3D folio_nr_pages(new); + unsigned long offset =3D swap_cache_index(entry); + unsigned long end =3D offset + nr_pages; + + XA_STATE(xas, &swap_address_space(entry)->i_pages, offset); + + VM_WARN_ON_ONCE(!folio_test_swapcache(old) || !folio_test_swapcache(new)); + VM_WARN_ON_ONCE(!folio_test_locked(old) || !folio_test_locked(new)); + VM_WARN_ON_ONCE(!entry.val); + + /* Swap cache still stores N entries instead of a high-order entry */ + do { + WARN_ON_ONCE(xas_store(&xas, new) !=3D old); + xas_next(&xas); + } while (++offset < end); +} + /** * swap_cache_clear_shadow - Clears a set of shadows in the swap cache. * @type: Indicates the swap device. --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f176.google.com (mail-pf1-f176.google.com [209.85.210.176]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8E68C33EAE6 for ; Wed, 10 Sep 2025 16:09:41 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.176 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520584; cv=none; b=p9+2bTrpUoVa41CHQXKyXpPD/KmhNGRZvWIlZwSIyYOvDlgvtjiAnhn+XfnlUB+Uozl/LYcFT9OKlNLQ0CDRS1iel6pg+A4a5TlF9WfWW+0iq0DtCc4m5d0h9IMo5LWsLfTkZke3PN7pvhLvnmYFurSOlQnqo1yYebDzuO8T6to= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520584; c=relaxed/simple; bh=aeZ+y0q2aSQE8aNO1ajCkRy1xI+YWzxbLPzYGeb4V/c=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=Ry/sDWEY1Z/NwwLXCAXCv20LL2hTFausBb8Xw747AbV6XONqDcSKQmW4+pufYsWCxUm73yMCXp+ivtReWYcsNzEr5KlsfujcMSkNuh46p/nfNkg/fvir1xmxIlv6JVQGvXQBpTxbaI+EwYtcgY8ShioiEHPlXVRuLfwGU9Zx8kI= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=LoZpny/m; arc=none smtp.client-ip=209.85.210.176 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="LoZpny/m" Received: by mail-pf1-f176.google.com with SMTP id d2e1a72fcca58-772679eb358so6640976b3a.1 for ; Wed, 10 Sep 2025 09:09:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520581; x=1758125381; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=1JSpL7ZCq9Q4IZEbmZCVrVHvMcatBTvNzr50xCRGm9o=; b=LoZpny/mI3ZwF2AtdIzR/VTYCFWOhwIdm/6js9U9Fk5p5RFfyM4LQyEc5GiF3igJjq hy9otbLQVSOkgUJj/TtxnuOs6n3OKUrv0HJ5J3N6+SGlNofWz9ZAkcmg6CFx/HtHNBUk qd04WRXeW/XNB/9JvBiidSPQGfrSx92ZOqPiXEgmoWSFpO4p2IMNCL/UjAvb5jm7g4wX 1OlO9R2Vc/62io+VUWGIwWAZt4c60kdQWi9gpj2inTIEl8pLehzCmvAybybxLRF4y2H8 Q7XyyphiOYVoFgVASPEeEE055Z84zrzoGzOOmVfjIjLTUcSJ/rjJdAHznUDvHFrQStvI +e2g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520581; x=1758125381; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=1JSpL7ZCq9Q4IZEbmZCVrVHvMcatBTvNzr50xCRGm9o=; b=nl0Xl3FghItR48A8cLNUCljdiOCfVltx7ybeEVlAu/YvY7vSPEXjXNh7hmWrnUzVzL Ac6TduBKVVHcusrmXvEr1wrxEaR1O8cwHoGhHvcobvgDBlMHTb1i1/Gm21QtF0UVcNQB rJgNo2wWJiROVAbDnuVz2Fzymc9p2qsv8aykn5vyLWvCKf+567slNCd4zlJruBapspRA DDjme79+0rK1LcDbayVIwk5H9iyF/NPQqzutdyAaUgxjFI9Mu2OqAtlueLWBAosqepRo vAS1KeWfhVoBZJU2HDDWhgBD1ZMJ9uZyTGdD9DGE50pJqlSr0b1Cyqukd7MrlBurDB4V FpYw== X-Forwarded-Encrypted: i=1; AJvYcCV0Fyz6gcY6O61LWyu+B3JaF1zPSm1Zp4+15sxCIQBigxnJsBokThFSPCJhzWlDs6VlnYmQUlh/6NnNX2g=@vger.kernel.org X-Gm-Message-State: AOJu0YxXE1OA0u3lKg6PiefAscyI0yr3kTcxS0i3A7egSYJQH5SFnfod 0bkLp6WPXPVf83ELYMNplWTZLoQj2bA3gPycfXRp5U9cNI5ssNg/c3KM X-Gm-Gg: ASbGncsQ2Fwjc0nuIGZeyfXn2PTlqliJu6G8rQSI7GVtZKu3Ki2izxQbFLaxaG7OmRu 9yxZrloBqjHfpSjtPPxjA0NdFhN+oTnu1BedtlzwM+/DBh5ieJg0hG9jfKj+Usy4zAQq2HyrdTP 37IPsBds9RqFDa+w8l+nOyAMrJVoGx5uCMtoFgtvwBPxMLMWjwNdFQAYQ0XyhF/NBFkvWjKINaY Nns29TDvNVykMTpp86uHAnK3WAVU88oNO3jREAe4B07egGftbYzGtA0UI232RVbF5sfC8M+jf5o 3TF2GCxaNiIeoCNPoS9HI+D5BFpzHEGuxweXT8kd5vwpJx1WB8pOKdybaqw9oKa0c795PFZdrMF vJ//Ga34B78/gVBf5eLBZ1xGU1UlXkoqrsgb+LytUDRBG0gIqBVNZ1I9JFw== X-Google-Smtp-Source: AGHT+IGzdfVdQOR0+tFkF7v1qxuYTr8o+dH9tTF5cvN+EsZ096h2fv+en+7YHy9W1aYlsUBZZis87A== X-Received: by 2002:a05:6a21:32aa:b0:24e:84c9:e999 with SMTP id adf61e73a8af0-2533f4e683cmr23413568637.22.1757520580633; Wed, 10 Sep 2025 09:09:40 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.35 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:40 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 11/15] mm, swap: use the swap table for the swap cache and switch API Date: Thu, 11 Sep 2025 00:08:29 +0800 Message-ID: <20250910160833.3464-12-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Introduce basic swap table infrastructures, which are now just a fixed-sized flat array inside each swap cluster, with access wrappers. Each cluster contains a swap table of 512 entries. Each table entry is an opaque atomic long. It could be in 3 types: a shadow type (XA_VALUE), a folio type (pointer), or NULL. In this first step, it only supports storing a folio or shadow, and it is a drop-in replacement for the current swap cache. Convert all swap cache users to use the new sets of APIs. Chris Li has been suggesting using a new infrastructure for swap cache for better performance, and that idea combined well with the swap table as the new backing structure. Now the lock contention range is reduced to 2M clusters, which is much smaller than the 64M address_space. And we can also drop the multiple address_space design. All the internal works are done with swap_cache_get_* helpers. Swap cache lookup is still lock-less like before, and the helper's contexts are same with original swap cache helpers. They still require a pin on the swap device to prevent the backing data from being freed. Swap cache updates are now protected by the swap cluster lock instead of the Xarray lock. This is mostly handled internally, but new __swap_cache_* helpers require the caller to lock the cluster. So, a few new cluster access and locking helpers are also introduced. A fully cluster-based unified swap table can be implemented on top of this to take care of all count tracking and synchronization work, with dynamic allocation. It should reduce the memory usage while making the performance even better. Co-developed-by: Chris Li Signed-off-by: Chris Li Signed-off-by: Kairui Song Acked-by: Chris Li Suggested-by: Chris Li --- MAINTAINERS | 1 + include/linux/swap.h | 2 - mm/huge_memory.c | 13 +- mm/migrate.c | 19 ++- mm/shmem.c | 8 +- mm/swap.h | 154 +++++++++++++++++------ mm/swap_state.c | 293 +++++++++++++++++++------------------------ mm/swap_table.h | 97 ++++++++++++++ mm/swapfile.c | 100 +++++++++++---- mm/vmscan.c | 20 ++- 10 files changed, 459 insertions(+), 248 deletions(-) create mode 100644 mm/swap_table.h diff --git a/MAINTAINERS b/MAINTAINERS index 3d113bfc3c82..4c8bbf70a3c7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16232,6 +16232,7 @@ F: include/linux/swapops.h F: mm/page_io.c F: mm/swap.c F: mm/swap.h +F: mm/swap_table.h F: mm/swap_state.c F: mm/swapfile.c =20 diff --git a/include/linux/swap.h b/include/linux/swap.h index 762f8db0e811..e818fbade1e2 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -480,8 +480,6 @@ extern int __swap_count(swp_entry_t entry); extern bool swap_entry_swapped(struct swap_info_struct *si, swp_entry_t en= try); extern int swp_swapcount(swp_entry_t entry); struct backing_dev_info; -extern int init_swap_address_space(unsigned int type, unsigned long nr_pag= es); -extern void exit_swap_address_space(unsigned int type); extern struct swap_info_struct *get_swap_device(swp_entry_t entry); sector_t swap_folio_sector(struct folio *folio); =20 diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 4c66e358685b..a9fc7a09167a 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3720,7 +3720,7 @@ static int __folio_split(struct folio *folio, unsigne= d int new_order, /* Prevent deferred_split_scan() touching ->_refcount */ spin_lock(&ds_queue->split_queue_lock); if (folio_ref_freeze(folio, 1 + extra_pins)) { - struct address_space *swap_cache =3D NULL; + struct swap_cluster_info *ci =3D NULL; struct lruvec *lruvec; int expected_refs; =20 @@ -3764,8 +3764,7 @@ static int __folio_split(struct folio *folio, unsigne= d int new_order, goto fail; } =20 - swap_cache =3D swap_address_space(folio->swap); - xa_lock(&swap_cache->i_pages); + ci =3D swap_cluster_get_and_lock(folio); } =20 /* lock lru list/PageCompound, ref frozen by page_ref_freeze */ @@ -3797,8 +3796,8 @@ static int __folio_split(struct folio *folio, unsigne= d int new_order, * Anonymous folio with swap cache. * NOTE: shmem in swap cache is not supported yet. */ - if (swap_cache) { - __swap_cache_replace_folio(folio, new_folio); + if (ci) { + __swap_cache_replace_folio(ci, folio, new_folio); continue; } =20 @@ -3833,8 +3832,8 @@ static int __folio_split(struct folio *folio, unsigne= d int new_order, =20 unlock_page_lruvec(lruvec); =20 - if (swap_cache) - xa_unlock(&swap_cache->i_pages); + if (ci) + swap_cluster_unlock(ci); } else { spin_unlock(&ds_queue->split_queue_lock); ret =3D -EAGAIN; diff --git a/mm/migrate.c b/mm/migrate.c index c69cc13db692..946253c39807 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -563,6 +563,7 @@ static int __folio_migrate_mapping(struct address_space= *mapping, struct folio *newfolio, struct folio *folio, int expected_count) { XA_STATE(xas, &mapping->i_pages, folio_index(folio)); + struct swap_cluster_info *ci =3D NULL; struct zone *oldzone, *newzone; int dirty; long nr =3D folio_nr_pages(folio); @@ -591,9 +592,16 @@ static int __folio_migrate_mapping(struct address_spac= e *mapping, oldzone =3D folio_zone(folio); newzone =3D folio_zone(newfolio); =20 - xas_lock_irq(&xas); + if (folio_test_swapcache(folio)) + ci =3D swap_cluster_get_and_lock_irq(folio); + else + xas_lock_irq(&xas); + if (!folio_ref_freeze(folio, expected_count)) { - xas_unlock_irq(&xas); + if (ci) + swap_cluster_unlock(ci); + else + xas_unlock_irq(&xas); return -EAGAIN; } =20 @@ -624,7 +632,7 @@ static int __folio_migrate_mapping(struct address_space= *mapping, } =20 if (folio_test_swapcache(folio)) - __swap_cache_replace_folio(folio, newfolio); + __swap_cache_replace_folio(ci, folio, newfolio); else xas_store(&xas, newfolio); =20 @@ -635,8 +643,11 @@ static int __folio_migrate_mapping(struct address_spac= e *mapping, */ folio_ref_unfreeze(folio, expected_count - nr); =20 - xas_unlock(&xas); /* Leave irq disabled to prevent preemption while updating stats */ + if (ci) + swap_cluster_unlock(ci); + else + xas_unlock(&xas); =20 /* * If moved to a different zone then also account diff --git a/mm/shmem.c b/mm/shmem.c index 8930780325da..8147a99a4b07 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2083,9 +2083,9 @@ static int shmem_replace_folio(struct folio **foliop,= gfp_t gfp, struct shmem_inode_info *info, pgoff_t index, struct vm_area_struct *vma) { + struct swap_cluster_info *ci; struct folio *new, *old =3D *foliop; swp_entry_t entry =3D old->swap; - struct address_space *swap_mapping =3D swap_address_space(entry); int nr_pages =3D folio_nr_pages(old); int error =3D 0; =20 @@ -2116,9 +2116,9 @@ static int shmem_replace_folio(struct folio **foliop,= gfp_t gfp, new->swap =3D entry; folio_set_swapcache(new); =20 - xa_lock_irq(&swap_mapping->i_pages); - __swap_cache_replace_folio(old, new); - xa_unlock_irq(&swap_mapping->i_pages); + ci =3D swap_cluster_get_and_lock_irq(old); + __swap_cache_replace_folio(ci, old, new); + swap_cluster_unlock(ci); =20 mem_cgroup_replace_folio(old, new); shmem_update_stats(new, nr_pages); diff --git a/mm/swap.h b/mm/swap.h index fe579c81c6c4..742db4d46d23 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -2,6 +2,7 @@ #ifndef _MM_SWAP_H #define _MM_SWAP_H =20 +#include /* for atomic_long_t */ struct mempolicy; struct swap_iocb; =20 @@ -35,6 +36,7 @@ struct swap_cluster_info { u16 count; u8 flags; u8 order; + atomic_long_t *table; /* Swap table entries, see mm/swap_table.h */ struct list_head list; }; =20 @@ -55,6 +57,11 @@ enum swap_cluster_flags { #include /* for swp_offset */ #include /* for bio_end_io_t */ =20 +static inline unsigned int swp_cluster_offset(swp_entry_t entry) +{ + return swp_offset(entry) % SWAPFILE_CLUSTER; +} + /* * Callers of all helpers below must ensure the entry, type, or offset is * valid, and protect the swap device with reference count or locks. @@ -81,6 +88,25 @@ static inline struct swap_cluster_info *__swap_offset_to= _cluster( return &si->cluster_info[offset / SWAPFILE_CLUSTER]; } =20 +static inline struct swap_cluster_info *__swap_entry_to_cluster(swp_entry_= t entry) +{ + return __swap_offset_to_cluster(__swap_entry_to_info(entry), + swp_offset(entry)); +} + +static __always_inline struct swap_cluster_info *__swap_cluster_lock( + struct swap_info_struct *si, unsigned long offset, bool irq) +{ + struct swap_cluster_info *ci =3D __swap_offset_to_cluster(si, offset); + + VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ + if (irq) + spin_lock_irq(&ci->lock); + else + spin_lock(&ci->lock); + return ci; +} + /** * swap_cluster_lock - Lock and return the swap cluster of given offset. * @si: swap device the cluster belongs to. @@ -92,11 +118,49 @@ static inline struct swap_cluster_info *__swap_offset_= to_cluster( static inline struct swap_cluster_info *swap_cluster_lock( struct swap_info_struct *si, unsigned long offset) { - struct swap_cluster_info *ci =3D __swap_offset_to_cluster(si, offset); + return __swap_cluster_lock(si, offset, false); +} =20 - VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ - spin_lock(&ci->lock); - return ci; +static inline struct swap_cluster_info *__swap_cluster_get_and_lock( + const struct folio *folio, bool irq) +{ + VM_WARN_ON_ONCE_FOLIO(!folio_test_locked(folio), folio); + VM_WARN_ON_ONCE_FOLIO(!folio_test_swapcache(folio), folio); + return __swap_cluster_lock(__swap_entry_to_info(folio->swap), + swp_offset(folio->swap), irq); +} + +/* + * swap_cluster_get_and_lock - Locks the cluster that holds a folio's entr= ies. + * @folio: The folio. + * + * This locks and returns the swap cluster that contains a folio's swap + * entries. The swap entries of a folio are always in one single cluster. + * The folio has to be locked so its swap entries won't change and the + * cluster won't be freed. + * + * Context: Caller must ensure the folio is locked and in the swap cache. + * Return: Pointer to the swap cluster. + */ +static inline struct swap_cluster_info *swap_cluster_get_and_lock( + const struct folio *folio) +{ + return __swap_cluster_get_and_lock(folio, false); +} + +/* + * swap_cluster_get_and_lock_irq - Locks the cluster that holds a folio's = entries. + * @folio: The folio. + * + * Same as swap_cluster_get_and_lock but also disable IRQ. + * + * Context: Caller must ensure the folio is locked and in the swap cache. + * Return: Pointer to the swap cluster. + */ +static inline struct swap_cluster_info *swap_cluster_get_and_lock_irq( + const struct folio *folio) +{ + return __swap_cluster_get_and_lock(folio, true); } =20 static inline void swap_cluster_unlock(struct swap_cluster_info *ci) @@ -104,6 +168,11 @@ static inline void swap_cluster_unlock(struct swap_clu= ster_info *ci) spin_unlock(&ci->lock); } =20 +static inline void swap_cluster_unlock_irq(struct swap_cluster_info *ci) +{ + spin_unlock_irq(&ci->lock); +} + /* linux/mm/page_io.c */ int sio_pool_init(void); struct swap_iocb; @@ -123,10 +192,11 @@ void __swap_writepage(struct folio *folio, struct swa= p_iocb **swap_plug); #define SWAP_ADDRESS_SPACE_SHIFT 14 #define SWAP_ADDRESS_SPACE_PAGES (1 << SWAP_ADDRESS_SPACE_SHIFT) #define SWAP_ADDRESS_SPACE_MASK (SWAP_ADDRESS_SPACE_PAGES - 1) -extern struct address_space *swapper_spaces[]; -#define swap_address_space(entry) \ - (&swapper_spaces[swp_type(entry)][swp_offset(entry) \ - >> SWAP_ADDRESS_SPACE_SHIFT]) +extern struct address_space swap_space; +static inline struct address_space *swap_address_space(swp_entry_t entry) +{ + return &swap_space; +} =20 /* * Return the swap device position of the swap entry. @@ -136,15 +206,6 @@ static inline loff_t swap_dev_pos(swp_entry_t entry) return ((loff_t)swp_offset(entry)) << PAGE_SHIFT; } =20 -/* - * Return the swap cache index of the swap entry. - */ -static inline pgoff_t swap_cache_index(swp_entry_t entry) -{ - BUILD_BUG_ON((SWP_OFFSET_MASK | SWAP_ADDRESS_SPACE_MASK) !=3D SWP_OFFSET_= MASK); - return swp_offset(entry) & SWAP_ADDRESS_SPACE_MASK; -} - /** * folio_matches_swap_entry - Check if a folio matches a given swap entry. * @folio: The folio. @@ -180,14 +241,14 @@ static inline bool folio_matches_swap_entry(const str= uct folio *folio, */ struct folio *swap_cache_get_folio(swp_entry_t entry); void *swap_cache_get_shadow(swp_entry_t entry); -int swap_cache_add_folio(struct folio *folio, swp_entry_t entry, - gfp_t gfp, void **shadow); +void swap_cache_add_folio(struct folio *folio, swp_entry_t entry, void **s= hadow); void swap_cache_del_folio(struct folio *folio); -void __swap_cache_del_folio(struct folio *folio, - swp_entry_t entry, void *shadow); -void __swap_cache_replace_folio(struct folio *old, struct folio *new); -void swap_cache_clear_shadow(int type, unsigned long begin, - unsigned long end); +/* Below helpers require the caller to lock and pass in the swap cluster. = */ +void __swap_cache_del_folio(struct swap_cluster_info *ci, + struct folio *folio, swp_entry_t entry, void *shadow); +void __swap_cache_replace_folio(struct swap_cluster_info *ci, + struct folio *old, struct folio *new); +void __swap_cache_clear_shadow(swp_entry_t entry, int nr_ents); =20 void show_swap_cache_info(void); void swapcache_clear(struct swap_info_struct *si, swp_entry_t entry, int n= r); @@ -255,6 +316,32 @@ static inline int non_swapcache_batch(swp_entry_t entr= y, int max_nr) =20 #else /* CONFIG_SWAP */ struct swap_iocb; +static inline struct swap_cluster_info *swap_cluster_lock( + struct swap_info_struct *si, pgoff_t offset, bool irq) +{ + return NULL; +} + +static inline struct swap_cluster_info *swap_cluster_get_and_lock( + struct folio *folio) +{ + return NULL; +} + +static inline struct swap_cluster_info *swap_cluster_get_and_lock_irq( + struct folio *folio) +{ + return NULL; +} + +static inline void swap_cluster_unlock(struct swap_cluster_info *ci) +{ +} + +static inline void swap_cluster_unlock_irq(struct swap_cluster_info *ci) +{ +} + static inline struct swap_info_struct *__swap_entry_to_info(swp_entry_t en= try) { return NULL; @@ -272,11 +359,6 @@ static inline struct address_space *swap_address_space= (swp_entry_t entry) return NULL; } =20 -static inline pgoff_t swap_cache_index(swp_entry_t entry) -{ - return 0; -} - static inline bool folio_matches_swap_entry(const struct folio *folio, swp= _entry_t entry) { return false; @@ -323,21 +405,21 @@ static inline void *swap_cache_get_shadow(swp_entry_t= entry) return NULL; } =20 -static inline int swap_cache_add_folio(swp_entry_t entry, struct folio *fo= lio, - gfp_t gfp, void **shadow) +static inline void swap_cache_add_folio(struct folio *folio, swp_entry_t e= ntry, void **shadow) { - return -EINVAL; } =20 static inline void swap_cache_del_folio(struct folio *folio) { } =20 -static inline void __swap_cache_del_folio(struct folio *folio, swp_entry_t= entry, void *shadow) +static inline void __swap_cache_del_folio(struct swap_cluster_info *ci, + struct folio *folio, swp_entry_t entry, void *shadow) { } =20 -static inline void __swap_cache_replace_folio(struct folio *old, struct fo= lio *new) +static inline void __swap_cache_replace_folio(struct swap_cluster_info *ci, + struct folio *old, struct folio *new) { } =20 @@ -371,8 +453,10 @@ static inline int non_swapcache_batch(swp_entry_t entr= y, int max_nr) */ static inline pgoff_t folio_index(struct folio *folio) { +#ifdef CONFIG_SWAP if (unlikely(folio_test_swapcache(folio))) - return swap_cache_index(folio->swap); + return swp_offset(folio->swap); +#endif return folio->index; } =20 diff --git a/mm/swap_state.c b/mm/swap_state.c index d1f5b8fa52fc..f4a579c23087 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -23,6 +23,7 @@ #include #include #include "internal.h" +#include "swap_table.h" #include "swap.h" =20 /* @@ -36,8 +37,10 @@ static const struct address_space_operations swap_aops = =3D { #endif }; =20 -struct address_space *swapper_spaces[MAX_SWAPFILES] __read_mostly; -static unsigned int nr_swapper_spaces[MAX_SWAPFILES] __read_mostly; +struct address_space swap_space __read_mostly =3D { + .a_ops =3D &swap_aops, +}; + static bool enable_vma_readahead __read_mostly =3D true; =20 #define SWAP_RA_ORDER_CEILING 5 @@ -83,11 +86,20 @@ void show_swap_cache_info(void) */ struct folio *swap_cache_get_folio(swp_entry_t entry) { - struct folio *folio =3D filemap_get_folio(swap_address_space(entry), - swap_cache_index(entry)); - if (IS_ERR(folio)) - return NULL; - return folio; + unsigned long swp_tb; + struct folio *folio; + + for (;;) { + swp_tb =3D __swap_table_get(__swap_entry_to_cluster(entry), + swp_cluster_offset(entry)); + if (!swp_tb_is_folio(swp_tb)) + return NULL; + folio =3D swp_tb_to_folio(swp_tb); + if (likely(folio_try_get(folio))) + return folio; + } + + return NULL; } =20 /** @@ -100,13 +112,13 @@ struct folio *swap_cache_get_folio(swp_entry_t entry) */ void *swap_cache_get_shadow(swp_entry_t entry) { - struct address_space *address_space =3D swap_address_space(entry); - pgoff_t idx =3D swap_cache_index(entry); - void *shadow; + unsigned long swp_tb; + + swp_tb =3D __swap_table_get(__swap_entry_to_cluster(entry), + swp_cluster_offset(entry)); + if (swp_tb_is_shadow(swp_tb)) + return swp_tb_to_shadow(swp_tb); =20 - shadow =3D xa_load(&address_space->i_pages, idx); - if (xa_is_value(shadow)) - return shadow; return NULL; } =20 @@ -119,61 +131,48 @@ void *swap_cache_get_shadow(swp_entry_t entry) * * Context: Caller must ensure @entry is valid and protect the swap device * with reference count or locks. - * The caller also needs to mark the corresponding swap_map slots with - * SWAP_HAS_CACHE to avoid race or conflict. - * Return: Returns 0 on success, error code otherwise. + * The caller also needs to update the corresponding swap_map slots with + * SWAP_HAS_CACHE bit to avoid race or conflict. */ -int swap_cache_add_folio(struct folio *folio, swp_entry_t entry, - gfp_t gfp, void **shadowp) +void swap_cache_add_folio(struct folio *folio, swp_entry_t entry, void **s= hadowp) { - struct address_space *address_space =3D swap_address_space(entry); - pgoff_t idx =3D swap_cache_index(entry); - XA_STATE_ORDER(xas, &address_space->i_pages, idx, folio_order(folio)); - unsigned long i, nr =3D folio_nr_pages(folio); - void *old; - - xas_set_update(&xas, workingset_update_node); - - VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); - VM_BUG_ON_FOLIO(folio_test_swapcache(folio), folio); - VM_BUG_ON_FOLIO(!folio_test_swapbacked(folio), folio); + void *shadow =3D NULL; + unsigned long old_tb, new_tb; + struct swap_cluster_info *ci; + unsigned int ci_start, ci_off, ci_end; + unsigned long nr_pages =3D folio_nr_pages(folio); + + VM_WARN_ON_ONCE_FOLIO(!folio_test_locked(folio), folio); + VM_WARN_ON_ONCE_FOLIO(folio_test_swapcache(folio), folio); + VM_WARN_ON_ONCE_FOLIO(!folio_test_swapbacked(folio), folio); + + new_tb =3D folio_to_swp_tb(folio); + ci_start =3D swp_cluster_offset(entry); + ci_end =3D ci_start + nr_pages; + ci_off =3D ci_start; + ci =3D swap_cluster_lock(__swap_entry_to_info(entry), swp_offset(entry)); + do { + old_tb =3D __swap_table_xchg(ci, ci_off, new_tb); + WARN_ON_ONCE(swp_tb_is_folio(old_tb)); + if (swp_tb_is_shadow(old_tb)) + shadow =3D swp_tb_to_shadow(old_tb); + } while (++ci_off < ci_end); =20 - folio_ref_add(folio, nr); + folio_ref_add(folio, nr_pages); folio_set_swapcache(folio); folio->swap =3D entry; + swap_cluster_unlock(ci); =20 - do { - xas_lock_irq(&xas); - xas_create_range(&xas); - if (xas_error(&xas)) - goto unlock; - for (i =3D 0; i < nr; i++) { - VM_BUG_ON_FOLIO(xas.xa_index !=3D idx + i, folio); - if (shadowp) { - old =3D xas_load(&xas); - if (xa_is_value(old)) - *shadowp =3D old; - } - xas_store(&xas, folio); - xas_next(&xas); - } - address_space->nrpages +=3D nr; - __node_stat_mod_folio(folio, NR_FILE_PAGES, nr); - __lruvec_stat_mod_folio(folio, NR_SWAPCACHE, nr); -unlock: - xas_unlock_irq(&xas); - } while (xas_nomem(&xas, gfp)); - - if (!xas_error(&xas)) - return 0; + node_stat_mod_folio(folio, NR_FILE_PAGES, nr_pages); + lruvec_stat_mod_folio(folio, NR_SWAPCACHE, nr_pages); =20 - folio_clear_swapcache(folio); - folio_ref_sub(folio, nr); - return xas_error(&xas); + if (shadowp) + *shadowp =3D shadow; } =20 /** * __swap_cache_del_folio - Removes a folio from the swap cache. + * @ci: The locked swap cluster. * @folio: The folio. * @entry: The first swap entry that the folio corresponds to. * @shadow: shadow value to be filled in the swap cache. @@ -181,34 +180,36 @@ int swap_cache_add_folio(struct folio *folio, swp_ent= ry_t entry, * Removes a folio from the swap cache and fills a shadow in place. * This won't put the folio's refcount. The caller has to do that. * - * Context: Caller must hold the xa_lock, ensure the folio is - * locked and in the swap cache, using the index of @entry. + * Context: Caller must ensure the folio is locked and in the swap cache + * using the index of @entry, and lock the cluster that holds the entries. */ -void __swap_cache_del_folio(struct folio *folio, +void __swap_cache_del_folio(struct swap_cluster_info *ci, struct folio *fo= lio, swp_entry_t entry, void *shadow) { - struct address_space *address_space =3D swap_address_space(entry); - int i; - long nr =3D folio_nr_pages(folio); - pgoff_t idx =3D swap_cache_index(entry); - XA_STATE(xas, &address_space->i_pages, idx); - - xas_set_update(&xas, workingset_update_node); - - VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio); - VM_BUG_ON_FOLIO(!folio_test_swapcache(folio), folio); - VM_BUG_ON_FOLIO(folio_test_writeback(folio), folio); - - for (i =3D 0; i < nr; i++) { - void *entry =3D xas_store(&xas, shadow); - VM_BUG_ON_PAGE(entry !=3D folio, entry); - xas_next(&xas); - } + unsigned long old_tb, new_tb; + unsigned int ci_start, ci_off, ci_end; + unsigned long nr_pages =3D folio_nr_pages(folio); + + VM_WARN_ON_ONCE(__swap_entry_to_cluster(entry) !=3D ci); + VM_WARN_ON_ONCE_FOLIO(!folio_test_locked(folio), folio); + VM_WARN_ON_ONCE_FOLIO(!folio_test_swapcache(folio), folio); + VM_WARN_ON_ONCE_FOLIO(folio_test_writeback(folio), folio); + + new_tb =3D shadow_swp_to_tb(shadow); + ci_start =3D swp_cluster_offset(entry); + ci_end =3D ci_start + nr_pages; + ci_off =3D ci_start; + do { + /* If shadow is NULL, we sets an empty shadow */ + old_tb =3D __swap_table_xchg(ci, ci_off, new_tb); + WARN_ON_ONCE(!swp_tb_is_folio(old_tb) || + swp_tb_to_folio(old_tb) !=3D folio); + } while (++ci_off < ci_end); + folio->swap.val =3D 0; folio_clear_swapcache(folio); - address_space->nrpages -=3D nr; - __node_stat_mod_folio(folio, NR_FILE_PAGES, -nr); - __lruvec_stat_mod_folio(folio, NR_SWAPCACHE, -nr); + node_stat_mod_folio(folio, NR_FILE_PAGES, -nr_pages); + lruvec_stat_mod_folio(folio, NR_SWAPCACHE, -nr_pages); } =20 /** @@ -223,12 +224,12 @@ void __swap_cache_del_folio(struct folio *folio, */ void swap_cache_del_folio(struct folio *folio) { + struct swap_cluster_info *ci; swp_entry_t entry =3D folio->swap; - struct address_space *address_space =3D swap_address_space(entry); =20 - xa_lock_irq(&address_space->i_pages); - __swap_cache_del_folio(folio, entry, NULL); - xa_unlock_irq(&address_space->i_pages); + ci =3D swap_cluster_lock(__swap_entry_to_info(entry), swp_offset(entry)); + __swap_cache_del_folio(ci, folio, entry, NULL); + swap_cluster_unlock(ci); =20 put_swap_folio(folio, entry); folio_ref_sub(folio, folio_nr_pages(folio)); @@ -236,6 +237,7 @@ void swap_cache_del_folio(struct folio *folio) =20 /** * __swap_cache_replace_folio - Replace a folio in the swap cache. + * @ci: The locked swap cluster. * @old: The old folio to be replaced. * @new: The new folio. * @@ -244,65 +246,62 @@ void swap_cache_del_folio(struct folio *folio) * entries. Replacement will take the new folio's swap entry value as * the starting offset to override all slots covered by the new folio. * - * Context: Caller must ensure both folios are locked, also lock the - * swap address_space that holds the old folio to avoid races. + * Context: Caller must ensure both folios are locked, and lock the + * cluster that holds the old folio to be replaced. */ -void __swap_cache_replace_folio(struct folio *old, struct folio *new) +void __swap_cache_replace_folio(struct swap_cluster_info *ci, + struct folio *old, struct folio *new) { swp_entry_t entry =3D new->swap; unsigned long nr_pages =3D folio_nr_pages(new); - unsigned long offset =3D swap_cache_index(entry); - unsigned long end =3D offset + nr_pages; - - XA_STATE(xas, &swap_address_space(entry)->i_pages, offset); + unsigned int ci_off =3D swp_cluster_offset(entry); + unsigned int ci_end =3D ci_off + nr_pages; + unsigned long old_tb, new_tb; =20 VM_WARN_ON_ONCE(!folio_test_swapcache(old) || !folio_test_swapcache(new)); VM_WARN_ON_ONCE(!folio_test_locked(old) || !folio_test_locked(new)); VM_WARN_ON_ONCE(!entry.val); =20 /* Swap cache still stores N entries instead of a high-order entry */ + new_tb =3D folio_to_swp_tb(new); do { - WARN_ON_ONCE(xas_store(&xas, new) !=3D old); - xas_next(&xas); - } while (++offset < end); + old_tb =3D __swap_table_xchg(ci, ci_off, new_tb); + WARN_ON_ONCE(!swp_tb_is_folio(old_tb) || swp_tb_to_folio(old_tb) !=3D ol= d); + } while (++ci_off < ci_end); + + /* + * If the old folio is partially replaced (e.g., splitting a large + * folio, the old folio is shrunk, and new split sub folios replace + * the shrunk part), ensure the new folio doesn't overlap it. + */ + if (IS_ENABLED(CONFIG_DEBUG_VM) && + folio_order(old) !=3D folio_order(new)) { + ci_off =3D swp_cluster_offset(old->swap); + ci_end =3D ci_off + folio_nr_pages(old); + while (ci_off++ < ci_end) + WARN_ON_ONCE(swp_tb_to_folio(__swap_table_get(ci, ci_off)) !=3D old); + } } =20 /** * swap_cache_clear_shadow - Clears a set of shadows in the swap cache. - * @type: Indicates the swap device. - * @begin: Beginning offset of the range. - * @end: Ending offset of the range. + * @entry: The starting index entry. + * @nr_ents: How many slots need to be cleared. * - * Context: Caller must ensure the range is valid and hold a reference to - * the swap device. + * Context: Caller must ensure the range is valid, not occupied by, + * any folio and protect the swap device with reference count or locks. */ -void swap_cache_clear_shadow(int type, unsigned long begin, - unsigned long end) +void __swap_cache_clear_shadow(swp_entry_t entry, int nr_ents) { - unsigned long curr =3D begin; - void *old; - - for (;;) { - swp_entry_t entry =3D swp_entry(type, curr); - unsigned long index =3D curr & SWAP_ADDRESS_SPACE_MASK; - struct address_space *address_space =3D swap_address_space(entry); - XA_STATE(xas, &address_space->i_pages, index); - - xas_set_update(&xas, workingset_update_node); - - xa_lock_irq(&address_space->i_pages); - xas_for_each(&xas, old, min(index + (end - curr), SWAP_ADDRESS_SPACE_PAG= ES)) { - if (!xa_is_value(old)) - continue; - xas_store(&xas, NULL); - } - xa_unlock_irq(&address_space->i_pages); + struct swap_cluster_info *ci =3D __swap_entry_to_cluster(entry); + unsigned int ci_off =3D swp_cluster_offset(entry), ci_end; + unsigned long old; =20 - /* search the next swapcache until we meet end */ - curr =3D ALIGN((curr + 1), SWAP_ADDRESS_SPACE_PAGES); - if (curr > end) - break; - } + ci_end =3D ci_off + nr_ents; + do { + old =3D __swap_table_xchg(ci, ci_off, null_to_swp_tb()); + WARN_ON_ONCE(swp_tb_is_folio(old)); + } while (++ci_off < ci_end); } =20 /* @@ -482,10 +481,7 @@ struct folio *__read_swap_cache_async(swp_entry_t entr= y, gfp_t gfp_mask, if (mem_cgroup_swapin_charge_folio(new_folio, NULL, gfp_mask, entry)) goto fail_unlock; =20 - /* May fail (-ENOMEM) if XArray node allocation failed. */ - if (swap_cache_add_folio(new_folio, entry, gfp_mask & GFP_RECLAIM_MASK, &= shadow)) - goto fail_unlock; - + swap_cache_add_folio(new_folio, entry, &shadow); memcg1_swapin(entry, 1); =20 if (shadow) @@ -677,41 +673,6 @@ struct folio *swap_cluster_readahead(swp_entry_t entry= , gfp_t gfp_mask, return folio; } =20 -int init_swap_address_space(unsigned int type, unsigned long nr_pages) -{ - struct address_space *spaces, *space; - unsigned int i, nr; - - nr =3D DIV_ROUND_UP(nr_pages, SWAP_ADDRESS_SPACE_PAGES); - spaces =3D kvcalloc(nr, sizeof(struct address_space), GFP_KERNEL); - if (!spaces) - return -ENOMEM; - for (i =3D 0; i < nr; i++) { - space =3D spaces + i; - xa_init_flags(&space->i_pages, XA_FLAGS_LOCK_IRQ); - atomic_set(&space->i_mmap_writable, 0); - space->a_ops =3D &swap_aops; - /* swap cache doesn't use writeback related tags */ - mapping_set_no_writeback_tags(space); - } - nr_swapper_spaces[type] =3D nr; - swapper_spaces[type] =3D spaces; - - return 0; -} - -void exit_swap_address_space(unsigned int type) -{ - int i; - struct address_space *spaces =3D swapper_spaces[type]; - - for (i =3D 0; i < nr_swapper_spaces[type]; i++) - VM_WARN_ON_ONCE(!mapping_empty(&spaces[i])); - kvfree(spaces); - nr_swapper_spaces[type] =3D 0; - swapper_spaces[type] =3D NULL; -} - static int swap_vma_ra_win(struct vm_fault *vmf, unsigned long *start, unsigned long *end) { @@ -884,7 +845,7 @@ static const struct attribute_group swap_attr_group =3D= { .attrs =3D swap_attrs, }; =20 -static int __init swap_init_sysfs(void) +static int __init swap_init(void) { int err; struct kobject *swap_kobj; @@ -899,11 +860,13 @@ static int __init swap_init_sysfs(void) pr_err("failed to register swap group\n"); goto delete_obj; } + /* Swap cache writeback is LRU based, no tags for it */ + mapping_set_no_writeback_tags(&swap_space); return 0; =20 delete_obj: kobject_put(swap_kobj); return err; } -subsys_initcall(swap_init_sysfs); +subsys_initcall(swap_init); #endif diff --git a/mm/swap_table.h b/mm/swap_table.h new file mode 100644 index 000000000000..e1f7cc009701 --- /dev/null +++ b/mm/swap_table.h @@ -0,0 +1,97 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef _MM_SWAP_TABLE_H +#define _MM_SWAP_TABLE_H + +#include "swap.h" + +/* + * A swap table entry represents the status of a swap slot on a swap + * (physical or virtual) device. The swap table in each cluster is a + * 1:1 map of the swap slots in this cluster. + * + * Each swap table entry could be a pointer (folio), a XA_VALUE + * (shadow), or NULL. + */ + +/* + * Helpers for casting one type of info into a swap table entry. + */ +static inline unsigned long null_to_swp_tb(void) +{ + BUILD_BUG_ON(sizeof(unsigned long) !=3D sizeof(atomic_long_t)); + return 0; +} + +static inline unsigned long folio_to_swp_tb(struct folio *folio) +{ + BUILD_BUG_ON(sizeof(unsigned long) !=3D sizeof(void *)); + return (unsigned long)folio; +} + +static inline unsigned long shadow_swp_to_tb(void *shadow) +{ + BUILD_BUG_ON((BITS_PER_XA_VALUE + 1) !=3D + BITS_PER_BYTE * sizeof(unsigned long)); + VM_WARN_ON_ONCE(shadow && !xa_is_value(shadow)); + return (unsigned long)shadow; +} + +/* + * Helpers for swap table entry type checking. + */ +static inline bool swp_tb_is_null(unsigned long swp_tb) +{ + return !swp_tb; +} + +static inline bool swp_tb_is_folio(unsigned long swp_tb) +{ + return !xa_is_value((void *)swp_tb) && !swp_tb_is_null(swp_tb); +} + +static inline bool swp_tb_is_shadow(unsigned long swp_tb) +{ + return xa_is_value((void *)swp_tb); +} + +/* + * Helpers for retrieving info from swap table. + */ +static inline struct folio *swp_tb_to_folio(unsigned long swp_tb) +{ + VM_WARN_ON(!swp_tb_is_folio(swp_tb)); + return (void *)swp_tb; +} + +static inline void *swp_tb_to_shadow(unsigned long swp_tb) +{ + VM_WARN_ON(!swp_tb_is_shadow(swp_tb)); + return (void *)swp_tb; +} + +/* + * Helpers for accessing or modifying the swap table of a cluster, + * the swap cluster must be locked. + */ +static inline void __swap_table_set(struct swap_cluster_info *ci, + unsigned int off, unsigned long swp_tb) +{ + VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); + atomic_long_set(&ci->table[off], swp_tb); +} + +static inline unsigned long __swap_table_xchg(struct swap_cluster_info *ci, + unsigned int off, unsigned long swp_tb) +{ + VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); + /* Ordering is guaranteed by cluster lock, relax */ + return atomic_long_xchg_relaxed(&ci->table[off], swp_tb); +} + +static inline unsigned long __swap_table_get(struct swap_cluster_info *ci, + unsigned int off) +{ + VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); + return atomic_long_read(&ci->table[off]); +} +#endif diff --git a/mm/swapfile.c b/mm/swapfile.c index 6f5206255789..a81ada4a361d 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -46,6 +46,7 @@ #include #include #include +#include "swap_table.h" #include "internal.h" #include "swap.h" =20 @@ -420,6 +421,34 @@ static inline unsigned int cluster_offset(struct swap_= info_struct *si, return cluster_index(si, ci) * SWAPFILE_CLUSTER; } =20 +static int swap_cluster_alloc_table(struct swap_cluster_info *ci) +{ + WARN_ON(ci->table); + ci->table =3D kzalloc(sizeof(unsigned long) * SWAPFILE_CLUSTER, GFP_KERNE= L); + if (!ci->table) + return -ENOMEM; + return 0; +} + +static void swap_cluster_free_table(struct swap_cluster_info *ci) +{ + unsigned int ci_off; + unsigned long swp_tb; + + if (!ci->table) + return; + + for (ci_off =3D 0; ci_off < SWAPFILE_CLUSTER; ci_off++) { + swp_tb =3D __swap_table_get(ci, ci_off); + if (!swp_tb_is_null(swp_tb)) + pr_err_once("swap: unclean swap space on swapoff: 0x%lx", + swp_tb); + } + + kfree(ci->table); + ci->table =3D NULL; +} + static void move_cluster(struct swap_info_struct *si, struct swap_cluster_info *ci, struct list_head *list, enum swap_cluster_flags new_flags) @@ -702,6 +731,26 @@ static bool cluster_scan_range(struct swap_info_struct= *si, return true; } =20 +/* + * Currently, the swap table is not used for count tracking, just + * do a sanity check here to ensure nothing leaked, so the swap + * table should be empty upon freeing. + */ +static void swap_cluster_assert_table_empty(struct swap_cluster_info *ci, + unsigned int start, unsigned int nr) +{ + unsigned int ci_off =3D start % SWAPFILE_CLUSTER; + unsigned int ci_end =3D ci_off + nr; + unsigned long swp_tb; + + if (IS_ENABLED(CONFIG_DEBUG_VM)) { + do { + swp_tb =3D __swap_table_get(ci, ci_off); + VM_WARN_ON_ONCE(!swp_tb_is_null(swp_tb)); + } while (++ci_off < ci_end); + } +} + static bool cluster_alloc_range(struct swap_info_struct *si, struct swap_c= luster_info *ci, unsigned int start, unsigned char usage, unsigned int order) @@ -721,6 +770,7 @@ static bool cluster_alloc_range(struct swap_info_struct= *si, struct swap_cluster ci->order =3D order; =20 memset(si->swap_map + start, usage, nr_pages); + swap_cluster_assert_table_empty(ci, start, nr_pages); swap_range_alloc(si, nr_pages); ci->count +=3D nr_pages; =20 @@ -1123,7 +1173,7 @@ static void swap_range_free(struct swap_info_struct *= si, unsigned long offset, swap_slot_free_notify(si->bdev, offset); offset++; } - swap_cache_clear_shadow(si->type, begin, end); + __swap_cache_clear_shadow(swp_entry(si->type, begin), nr_entries); =20 /* * Make sure that try_to_unuse() observes si->inuse_pages reaching 0 @@ -1280,16 +1330,7 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp) if (!entry.val) return -ENOMEM; =20 - /* - * XArray node allocations from PF_MEMALLOC contexts could - * completely exhaust the page allocator. __GFP_NOMEMALLOC - * stops emergency reserves from being allocated. - * - * TODO: this could cause a theoretical memory reclaim - * deadlock in the swap out path. - */ - if (swap_cache_add_folio(folio, entry, gfp | __GFP_NOMEMALLOC, NULL)) - goto out_free; + swap_cache_add_folio(folio, entry, NULL); =20 return 0; =20 @@ -1555,6 +1596,7 @@ static void swap_entries_free(struct swap_info_struct= *si, =20 mem_cgroup_uncharge_swap(entry, nr_pages); swap_range_free(si, offset, nr_pages); + swap_cluster_assert_table_empty(ci, offset, nr_pages); =20 if (!ci->count) free_cluster(si, ci); @@ -2633,6 +2675,18 @@ static void wait_for_allocation(struct swap_info_str= uct *si) } } =20 +static void free_cluster_info(struct swap_cluster_info *cluster_info, + unsigned long maxpages) +{ + int i, nr_clusters =3D DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); + + if (!cluster_info) + return; + for (i =3D 0; i < nr_clusters; i++) + swap_cluster_free_table(&cluster_info[i]); + kvfree(cluster_info); +} + /* * Called after swap device's reference count is dead, so * neither scan nor allocation will use it. @@ -2767,12 +2821,13 @@ SYSCALL_DEFINE1(swapoff, const char __user *, speci= alfile) =20 swap_file =3D p->swap_file; p->swap_file =3D NULL; - p->max =3D 0; swap_map =3D p->swap_map; p->swap_map =3D NULL; zeromap =3D p->zeromap; p->zeromap =3D NULL; cluster_info =3D p->cluster_info; + free_cluster_info(cluster_info, p->max); + p->max =3D 0; p->cluster_info =3D NULL; spin_unlock(&p->lock); spin_unlock(&swap_lock); @@ -2783,10 +2838,8 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specia= lfile) p->global_cluster =3D NULL; vfree(swap_map); kvfree(zeromap); - kvfree(cluster_info); /* Destroy swap account information */ swap_cgroup_swapoff(p->type); - exit_swap_address_space(p->type); =20 inode =3D mapping->host; =20 @@ -3170,8 +3223,11 @@ static struct swap_cluster_info *setup_clusters(stru= ct swap_info_struct *si, if (!cluster_info) goto err; =20 - for (i =3D 0; i < nr_clusters; i++) + for (i =3D 0; i < nr_clusters; i++) { spin_lock_init(&cluster_info[i].lock); + if (swap_cluster_alloc_table(&cluster_info[i])) + goto err_free; + } =20 if (!(si->flags & SWP_SOLIDSTATE)) { si->global_cluster =3D kmalloc(sizeof(*si->global_cluster), @@ -3232,9 +3288,8 @@ static struct swap_cluster_info *setup_clusters(struc= t swap_info_struct *si, } =20 return cluster_info; - err_free: - kvfree(cluster_info); + free_cluster_info(cluster_info, maxpages); err: return ERR_PTR(err); } @@ -3428,13 +3483,9 @@ SYSCALL_DEFINE2(swapon, const char __user *, special= file, int, swap_flags) } } =20 - error =3D init_swap_address_space(si->type, maxpages); - if (error) - goto bad_swap_unlock_inode; - error =3D zswap_swapon(si->type, maxpages); if (error) - goto free_swap_address_space; + goto bad_swap_unlock_inode; =20 /* * Flush any pending IO and dirty mappings before we start using this @@ -3469,8 +3520,6 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialf= ile, int, swap_flags) goto out; free_swap_zswap: zswap_swapoff(si->type); -free_swap_address_space: - exit_swap_address_space(si->type); bad_swap_unlock_inode: inode_unlock(inode); bad_swap: @@ -3485,7 +3534,8 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialf= ile, int, swap_flags) spin_unlock(&swap_lock); vfree(swap_map); kvfree(zeromap); - kvfree(cluster_info); + if (cluster_info) + free_cluster_info(cluster_info, maxpages); if (inced_nr_rotate_swap) atomic_dec(&nr_rotate_swap); if (swap_file) diff --git a/mm/vmscan.c b/mm/vmscan.c index c79c6806560b..e170c12e2065 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -730,13 +730,18 @@ static int __remove_mapping(struct address_space *map= ping, struct folio *folio, { int refcount; void *shadow =3D NULL; + struct swap_cluster_info *ci; =20 BUG_ON(!folio_test_locked(folio)); BUG_ON(mapping !=3D folio_mapping(folio)); =20 - if (!folio_test_swapcache(folio)) + if (folio_test_swapcache(folio)) { + ci =3D swap_cluster_get_and_lock_irq(folio); + } else { spin_lock(&mapping->host->i_lock); - xa_lock_irq(&mapping->i_pages); + xa_lock_irq(&mapping->i_pages); + } + /* * The non racy check for a busy folio. * @@ -776,9 +781,9 @@ static int __remove_mapping(struct address_space *mappi= ng, struct folio *folio, =20 if (reclaimed && !mapping_exiting(mapping)) shadow =3D workingset_eviction(folio, target_memcg); - __swap_cache_del_folio(folio, swap, shadow); + __swap_cache_del_folio(ci, folio, swap, shadow); memcg1_swapout(folio, swap); - xa_unlock_irq(&mapping->i_pages); + swap_cluster_unlock_irq(ci); put_swap_folio(folio, swap); } else { void (*free_folio)(struct folio *); @@ -816,9 +821,12 @@ static int __remove_mapping(struct address_space *mapp= ing, struct folio *folio, return 1; =20 cannot_free: - xa_unlock_irq(&mapping->i_pages); - if (!folio_test_swapcache(folio)) + if (folio_test_swapcache(folio)) { + swap_cluster_unlock_irq(ci); + } else { + xa_unlock_irq(&mapping->i_pages); spin_unlock(&mapping->host->i_lock); + } return 0; } =20 --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f175.google.com (mail-pf1-f175.google.com [209.85.210.175]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 32C623375D0 for ; Wed, 10 Sep 2025 16:09:46 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.175 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520587; cv=none; b=a3H20wLA+HUWb2Xpz7FPTAN+ukK6RQ087GGt1Au4qhI9e7m891x5Gy9zSSejE75YZFOWbEQzxjhVOR6chaJfSJtLEsqqg7m+vYiTSaUq+HvN/Ajxe0Xt4+N3LgfZLVHp6GA+KzUzX1Tk7ySZN31vWhnOCT0OAA0F4hYetuG0VvQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520587; c=relaxed/simple; bh=J0AulNAsnT8Ymo6KtZvstHEj395E13+pa9k9CH/G17A=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=dQwHKtW6rcmn+p7DhQUPsOmv5KlQPQksMEuAF5F10a4RLvr3642283YsZKJz5ix9clj40hgFT2yBYxu3ycMZtZshYguBAaHk94H1urobYYVhSRK8qPM5rm6ijvNv/XcmYvnre+Wn99+TmlwYUf919zmdfEsGdw6BnUxiUk2kO3U= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=X7ze4DPH; arc=none smtp.client-ip=209.85.210.175 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="X7ze4DPH" Received: by mail-pf1-f175.google.com with SMTP id d2e1a72fcca58-77256200f1bso6152005b3a.3 for ; Wed, 10 Sep 2025 09:09:46 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520585; x=1758125385; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=mUMwPFkC6/4gEbz67OObm145GGunBjCLIUgwlHXltYw=; b=X7ze4DPHIL9G2/u0D2EduOXBFsDg5u9+SrMhfRt67fIb9vSfrWCRT2ADvD1xlUMMk5 i2X75MT/qd31NZUfkyfFhDG+nWLqSCoJ5rsxmDt4alPYP9DUi+WhABW+V4ngoI3Pjy4A rj8alK9uXwzWeKq3k70BxCSuqkBsEeeZm1mJpsSxwCevzupJd5sIPaBNWM0lvFD04DuW DmdFc1h9lImeQyoW5gYDAHJKEOgSlxlc+i9SXYNWASgJstIo/D6LWNjEDJRaDEJ+Q04v wiKPKD1JGayTxxBBVv/tk3/jISg3sgO9TMwb60IH1cNCGSkqia1nLXGEbCvvypYqQ8ZB fpsA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520585; x=1758125385; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=mUMwPFkC6/4gEbz67OObm145GGunBjCLIUgwlHXltYw=; b=StscRIiRo27eGwodOA8XlS/TUF1mCrU/LAY4JdFh4g2VMmLtD7/5ac0SMFjhR7a1SF 9sUc1xGW41JhPUdO2dQvVNfKQZhWdjVOr76P6ruFX7skgEiDcql5jh/6N+lIA9Sonleg AtqE52HLee2W1d5eeFez+U0zFtUVW2m+/Z6ZewbeLdE5VIkNXQKSrmhUt4wyqstp0Lyc aVDguRTBUCPHn2zc67P9o0UdBXXkAH35Ti5lsJ+CZdI/HW4SdqI3Sz/ImVI5wq2I6zK1 qItZC7Djec+ofNYSKShp+ys/JwxQoiZsi/ZAjU4vdiHlRM+4zPNbEtVnwhdvt4RJAXIu 5OcA== X-Forwarded-Encrypted: i=1; AJvYcCUAVCXFJnUH1URt0Zkm80cZoKrrlkpFqBcudNdSQmJNOWxV4wEoW6IFtCoZEWfnXmuTu9sW10fll6NyvGs=@vger.kernel.org X-Gm-Message-State: AOJu0Yw6jq2gJHwxUaJMix/hNjwhY4afyGTCwsjYD0Z+Kver4JdUrb5M ztXPdQNmJfqxTn8qPrxRToPOz7dbIK4VyRy84eq1/N+MnF1FBROQG7Lc X-Gm-Gg: ASbGncsxcqo+If/RCQe3D3jUleNlW6CySFm/DTe25aHtaiDR69BZMpxLS/20yiRh8Fy N1OROmcuEvODxkzkGedXlDmwqDH/ixiuRUjuWsREvsjiatDH5ttlS3CyBQGFqWQud2LF4FpzWWD /dxm+fokVeLy7UlUuC5Qsrz60tex8Pa8xDRTg7/fdZsnAT3UDHsBhm1MIGTEv1XrPnEdg1eYqRV q4QhCDQ8FxBaMVeuodZbHPXe3xCXnIeO8LXnuPRXoDKMJMZd4KthLVqXziA/mDmyxayXZTGn1tm +eltWVQprV1Z1qMDGJ3b6oEBQzYgklUva3y3recjEeDPJME5LH3t3hvezpfXWofmY73LzNSv74z 5slS7hUXSm282zDXz+oXrNYJdoYBatSkjMQv3Rk8/ltpm9F4= X-Google-Smtp-Source: AGHT+IF4aEJFdwdzqld0Qv3VuyHOHub7zXw74OiqWKQuRHCc9dAp9UkKKUX0sq6k06ILw38fGnOA9Q== X-Received: by 2002:a05:6a20:2584:b0:246:3a6:3e45 with SMTP id adf61e73a8af0-2534557cb54mr23558299637.46.1757520585439; Wed, 10 Sep 2025 09:09:45 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.40 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:45 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 12/15] mm, swap: mark swap address space ro and add context debug check Date: Thu, 11 Sep 2025 00:08:30 +0800 Message-ID: <20250910160833.3464-13-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Swap cache is now backed by swap table, and the address space is not holding any mutable data anymore. And swap cache is now protected by the swap cluster lock, instead of the XArray lock. All access to swap cache are wrapped by swap cache helpers. Locking is mostly handled internally by swap cache helpers, only a few __swap_cache_* helpers require the caller to lock the cluster by themselves. Worth noting that, unlike XArray, the cluster lock is not IRQ safe. The swap cache was very different compared to filemap, and now it's completely separated from filemap. Nothing wants to mark or change anything or do a writeback callback in IRQ. So explicitly document this and add a debug check to avoid further potential misuse. And mark the swap cache space as read-only to avoid any user wrongly mixing unexpected filemap helpers with swap cache. Signed-off-by: Kairui Song Acked-by: Chris Li Acked-by: David Hildenbrand Suggested-by: Chris Li --- mm/swap.h | 12 +++++++++++- mm/swap_state.c | 3 ++- 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/mm/swap.h b/mm/swap.h index 742db4d46d23..adcd85fa8538 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -99,6 +99,16 @@ static __always_inline struct swap_cluster_info *__swap_= cluster_lock( { struct swap_cluster_info *ci =3D __swap_offset_to_cluster(si, offset); =20 + /* + * Nothing modifies swap cache in an IRQ context. All access to + * swap cache is wrapped by swap_cache_* helpers, and swap cache + * writeback is handled outside of IRQs. Swapin or swapout never + * occurs in IRQ, and neither does in-place split or replace. + * + * Besides, modifying swap cache requires synchronization with + * swap_map, which was never IRQ safe. + */ + VM_WARN_ON_ONCE(!in_task()); VM_WARN_ON_ONCE(percpu_ref_is_zero(&si->users)); /* race with swapoff */ if (irq) spin_lock_irq(&ci->lock); @@ -192,7 +202,7 @@ void __swap_writepage(struct folio *folio, struct swap_= iocb **swap_plug); #define SWAP_ADDRESS_SPACE_SHIFT 14 #define SWAP_ADDRESS_SPACE_PAGES (1 << SWAP_ADDRESS_SPACE_SHIFT) #define SWAP_ADDRESS_SPACE_MASK (SWAP_ADDRESS_SPACE_PAGES - 1) -extern struct address_space swap_space; +extern struct address_space swap_space __ro_after_init; static inline struct address_space *swap_address_space(swp_entry_t entry) { return &swap_space; diff --git a/mm/swap_state.c b/mm/swap_state.c index f4a579c23087..97372539a575 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -37,7 +37,8 @@ static const struct address_space_operations swap_aops = =3D { #endif }; =20 -struct address_space swap_space __read_mostly =3D { +/* Set swap_space as read only as swap cache is handled by swap table */ +struct address_space swap_space __ro_after_init =3D { .a_ops =3D &swap_aops, }; =20 --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pf1-f172.google.com (mail-pf1-f172.google.com [209.85.210.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6582033EB18 for ; Wed, 10 Sep 2025 16:09:51 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520593; cv=none; b=o162a6/u/qC+1ArfWHkvqBGYkHiMEczlIF3vne14+H9FQxvrK3QVx8NpV5KPrUrDcx1OxaeeMpa2XR0/o4zytSXCqChKFQdZaiWJz0FqJO1hEJUOgJkvBAUQVKYh8Zd0ch7WTVAbMGeaJUNMSqimRYqj08EbCvvZW1EEkho//hg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520593; c=relaxed/simple; bh=gzjZTwxGbEP21GLGwFbxoi6pwA4ybOJR49Hf3hcPCNY=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=LgUFGtIbmIyssVbMH9NMn1svCQmu/wwsFSVByBMl/OGjy03KppklYkA7izV0+wvl6pkq8oVMA4ATSUBOXj3xFv6v6Pw0hxC9jwDZ3Bwry5qASD20wSFUmm2sbkBOsrY0qfhQ3wLwODXihrFf/Fb+Neg9htQ3mV9VlSQU5Bm8y94= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=KcfMHYcQ; arc=none smtp.client-ip=209.85.210.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="KcfMHYcQ" Received: by mail-pf1-f172.google.com with SMTP id d2e1a72fcca58-77264a94031so4816611b3a.2 for ; Wed, 10 Sep 2025 09:09:51 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520591; x=1758125391; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=su4PnDYf8OUfLgI2Lrokhalmhp7VgFbGjYZtLVCSA2o=; b=KcfMHYcQjpEWSJWWgpKWlMTwqn6z1Ie1NeaNNiKB9G6X2xugN/+0SMfBCh8L0twPGY FQ0MCLEOEH+VTCdaHdUVDlNwV3/0pQ6UkBshQiKCfPGtsSHLo9HgPzpGL09nSgkW6Xmu nedX2qsrwDaF4dBa7fKckTDdejqZv9SzolB0Wc3pNZbk9WY/hw+GFn2RyHJyMT2j5Jn+ JkXUUm8I/J41H+qRhNQTMgTjB+J4ouJ1XMzxKrCJWpgFBkLpHWvxJ5EnKmgqKtmXPRQg y+dAODlkWSZLg7IOplLCzhBdCaHVH7s8FX7sfM6Rkv6PqYdp6ymGUnwM6Vd6sAQAe9Eu 2R7A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520591; x=1758125391; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=su4PnDYf8OUfLgI2Lrokhalmhp7VgFbGjYZtLVCSA2o=; b=KjU7mjDEhXfYM6c2gxUq+H05tMTbBeJ4PCBC58JoKqWwkwTaBZtklBp7Ub/0bX4JpQ dGqz/8ujc3I3GXX8jXeSBcSly5uG1rGE6BhyHlnIvHJ8C2Pz1NQRcrGxmPVROWMY5udC REukL2hqnEsFOT0vosMp9rxdQSco64CcU5FjwzGXtR4Uj5cCQksUUuoEYjF/eaxoBrjE zZORWBpmuArnOHVLnv/G9HSuqaSEk2AZwfaPxcCSbLXxnghyyk+oeleyH3G0+kO+u8VA WM+i97E3ZDYCq7LhRw9iXHMjafuUK0cAbCm+2FHOHkLN1kN8CO9JOQM7ZfuUXfD01gWs RINQ== X-Forwarded-Encrypted: i=1; AJvYcCU+PatyGQAPfkBQqNfh+t1Zz61kUJihZ8kegG98QSaIttIWFLf72yTeT31+OrZIVfFY8BwPOo/WQ2erwjg=@vger.kernel.org X-Gm-Message-State: AOJu0YyTU9020zfieYpPFkpEk9X4HfOkoqpe0jPNCAHlJKJc/S1Ab8mE EfZmsscsfYrOC5JGKC9lrIjOjFBngbxg6HaPZjCoPUwGD7dyMaUrfnxA X-Gm-Gg: ASbGncv8cf0LnGxULReqq1r3OgxdWNPhCp0Nw0JcrYVg2+j9mU7mPDoyN3P1GQXaPIR VOVi3s2bvM6P2RH2YaxK0yw6ApR6Ji9//zh8wMgGxRSeXXdfOAMqo5mlTqfwgAHyQscVI8Bl3gS OIguQn5NLz2zH8NIG3HnOJG5jbc2p99VdEiwvGhNMuzUiZtQt+cWxX5RaBLepEliNcMBdAaTqJa q/jyD4Wn/61TgYYPfJOYps1lbkwIeeLEq9R7qExX6olOFH7mJC0AajbJLXQdnS/GN7rGbXZl7Jj O5+s+PFsdj2O652K+qmgR5PptrYw9t4qkWEWGQ0t74T+Ir3UO4dK/StFugYAXAqjSOsV4t+X5dO 0xeKYx9BTYyxZnY00/wUsa1RkjJRNhPA/ll3kDXI/cIldXYU= X-Google-Smtp-Source: AGHT+IEXDaG6nULakGwHLuDrhEcX+0zJ4/JSzZr8GFi2sn5vxnMvwdyigmF9NkrFHeceSC2DScxt2A== X-Received: by 2002:a05:6a20:1584:b0:246:1e3:1f6f with SMTP id adf61e73a8af0-2533dcc0596mr25473628637.12.1757520590541; Wed, 10 Sep 2025 09:09:50 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.45 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:50 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song , kernel test robot Subject: [PATCH v3 13/15] mm, swap: remove contention workaround for swap cache Date: Thu, 11 Sep 2025 00:08:31 +0800 Message-ID: <20250910160833.3464-14-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Swap cluster setup will try to shuffle the clusters on initialization. It was helpful to avoid contention for the swap cache space. The cluster size (2M) was much smaller than each swap cache space (64M), so shuffling the cluster means the allocator will try to allocate swap slots that are in different swap cache spaces for each CPU, reducing the chance of two CPUs using the same swap cache space, and hence reducing the contention. Now, swap cache is managed by swap clusters, this shuffle is pointless. Just remove it, and clean up related macros. This also improves the HDD swap performance as shuffling IO is a bad idea for HDD, and now the shuffling is gone. Test have shown a ~40% performance gain for HDD [1]: Doing sequential swap in of 8G data using 8 processes with usemem, average of 3 test runs: Before: 1270.91 KB/s per process After: 1849.54 KB/s per process Link: https://lore.kernel.org/linux-mm/CAMgjq7AdauQ8=3DX0zeih2r21QoV=3D-WWj= 1hyBxLWRzq74n-C=3D-Ng@mail.gmail.com/ [1] Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202504241621.f27743ec-lkp@intel.com Signed-off-by: Kairui Song Acked-by: Chris Li Reviewed-by: Barry Song Acked-by: David Hildenbrand Suggested-by: Chris Li --- mm/swap.h | 4 ---- mm/swapfile.c | 32 ++++++++------------------------ mm/zswap.c | 7 +++++-- 3 files changed, 13 insertions(+), 30 deletions(-) diff --git a/mm/swap.h b/mm/swap.h index adcd85fa8538..fe5c20922082 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -198,10 +198,6 @@ int swap_writeout(struct folio *folio, struct swap_ioc= b **swap_plug); void __swap_writepage(struct folio *folio, struct swap_iocb **swap_plug); =20 /* linux/mm/swap_state.c */ -/* One swap address space for each 64M swap space */ -#define SWAP_ADDRESS_SPACE_SHIFT 14 -#define SWAP_ADDRESS_SPACE_PAGES (1 << SWAP_ADDRESS_SPACE_SHIFT) -#define SWAP_ADDRESS_SPACE_MASK (SWAP_ADDRESS_SPACE_PAGES - 1) extern struct address_space swap_space __ro_after_init; static inline struct address_space *swap_address_space(swp_entry_t entry) { diff --git a/mm/swapfile.c b/mm/swapfile.c index a81ada4a361d..89659928465e 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -3203,21 +3203,14 @@ static int setup_swap_map(struct swap_info_struct *= si, return 0; } =20 -#define SWAP_CLUSTER_INFO_COLS \ - DIV_ROUND_UP(L1_CACHE_BYTES, sizeof(struct swap_cluster_info)) -#define SWAP_CLUSTER_SPACE_COLS \ - DIV_ROUND_UP(SWAP_ADDRESS_SPACE_PAGES, SWAPFILE_CLUSTER) -#define SWAP_CLUSTER_COLS \ - max_t(unsigned int, SWAP_CLUSTER_INFO_COLS, SWAP_CLUSTER_SPACE_COLS) - static struct swap_cluster_info *setup_clusters(struct swap_info_struct *s= i, union swap_header *swap_header, unsigned long maxpages) { unsigned long nr_clusters =3D DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); struct swap_cluster_info *cluster_info; - unsigned long i, j, idx; int err =3D -ENOMEM; + unsigned long i; =20 cluster_info =3D kvcalloc(nr_clusters, sizeof(*cluster_info), GFP_KERNEL); if (!cluster_info) @@ -3266,22 +3259,13 @@ static struct swap_cluster_info *setup_clusters(str= uct swap_info_struct *si, INIT_LIST_HEAD(&si->frag_clusters[i]); } =20 - /* - * Reduce false cache line sharing between cluster_info and - * sharing same address space. - */ - for (j =3D 0; j < SWAP_CLUSTER_COLS; j++) { - for (i =3D 0; i < DIV_ROUND_UP(nr_clusters, SWAP_CLUSTER_COLS); i++) { - struct swap_cluster_info *ci; - idx =3D i * SWAP_CLUSTER_COLS + j; - ci =3D cluster_info + idx; - if (idx >=3D nr_clusters) - continue; - if (ci->count) { - ci->flags =3D CLUSTER_FLAG_NONFULL; - list_add_tail(&ci->list, &si->nonfull_clusters[0]); - continue; - } + for (i =3D 0; i < nr_clusters; i++) { + struct swap_cluster_info *ci =3D &cluster_info[i]; + + if (ci->count) { + ci->flags =3D CLUSTER_FLAG_NONFULL; + list_add_tail(&ci->list, &si->nonfull_clusters[0]); + } else { ci->flags =3D CLUSTER_FLAG_FREE; list_add_tail(&ci->list, &si->free_clusters); } diff --git a/mm/zswap.c b/mm/zswap.c index 3dda4310099e..cba7077fda40 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -225,10 +225,13 @@ static bool zswap_has_pool; * helpers and fwd declarations **********************************/ =20 +/* One swap address space for each 64M swap space */ +#define ZSWAP_ADDRESS_SPACE_SHIFT 14 +#define ZSWAP_ADDRESS_SPACE_PAGES (1 << ZSWAP_ADDRESS_SPACE_SHIFT) static inline struct xarray *swap_zswap_tree(swp_entry_t swp) { return &zswap_trees[swp_type(swp)][swp_offset(swp) - >> SWAP_ADDRESS_SPACE_SHIFT]; + >> ZSWAP_ADDRESS_SPACE_SHIFT]; } =20 #define zswap_pool_debug(msg, p) \ @@ -1674,7 +1677,7 @@ int zswap_swapon(int type, unsigned long nr_pages) struct xarray *trees, *tree; unsigned int nr, i; =20 - nr =3D DIV_ROUND_UP(nr_pages, SWAP_ADDRESS_SPACE_PAGES); + nr =3D DIV_ROUND_UP(nr_pages, ZSWAP_ADDRESS_SPACE_PAGES); trees =3D kvcalloc(nr, sizeof(*tree), GFP_KERNEL); if (!trees) { pr_err("alloc failed, zswap disabled for swap type %d\n", type); --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pg1-f172.google.com (mail-pg1-f172.google.com [209.85.215.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7E133340DA6 for ; Wed, 10 Sep 2025 16:09:56 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.172 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520598; cv=none; b=q3+0mscO+wyI7RRixHcUNjDkpsLR9dFaM9tXWNt3q1IzXeXev56o//zw6k+WRCsu/ZqKtKQuW5y6eCQbiMcaQAB2uSSkLrRTycWt6pUSbqbHbrJPtKEErBcAq3N0ZhYtYcDY4UR98q3cHnTf/5COlET2rGVwImNszbVKBTcUWnA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520598; c=relaxed/simple; bh=Jd6aPP8P7OyaDhFSAzivqWPYpoDwtzJxgjnfvz85MaY=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=qOT3BzdcefHUSCtg9ugnbN0rvhF6GZon95qjMWu9mYuy5NgtrXzQ2wUG0GGXuDV6fi5PnkUqypvbpY1GZzyMTKxCtF1XUS4jVbp6GiDWXMLrajXj+ztrVNqGeigTivsdXpyeq5T1QvZ5pMg6wRZAeJ3PKYq5CeQI9NMAWvEA9wY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=nG9/OLl3; arc=none smtp.client-ip=209.85.215.172 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="nG9/OLl3" Received: by mail-pg1-f172.google.com with SMTP id 41be03b00d2f7-b4c8bee055cso5947574a12.2 for ; Wed, 10 Sep 2025 09:09:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520596; x=1758125396; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=gqqOh+GurbkmfIODxteOTLsF6NzKFWqzUJ203IGMWLQ=; b=nG9/OLl3xknbBeVba5h1iju1dtHRmn/4ENDeQovAHUH5h12RPJ32uYvPgxQdWiEq1x ovY/EjPtrZTX05K6lp/pLQAnT1IcamD2EERDL/E3+Frd8c8YE5RdSRht/l8eQMF5Tqsp 8tfbZKLU+7vq7drbN+Lj7khN+zIZ3risKFmeFFE0498pP+RKCABRukvB1zv03TJZqO9J Xg0v5J4/Iqc3yHEZKNwJNiHt86VSBvbdu7YBePd2CqJTD6P3xBhhMGn1p7dK9C1DLHaC mFvoKly0uNVbcDR5ILMffB5foq85S6avWJoqGblYSDU094Z+4dazJySOhTZBDkeJDAXe O6Vw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520596; x=1758125396; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=gqqOh+GurbkmfIODxteOTLsF6NzKFWqzUJ203IGMWLQ=; b=pki196aWLfhshcO8vcazN+05UApYanrSNDMuCu+9Hiz1GZsq7OTMHfTD+5Wgc4VaHd lT3dkHCnUsTTatgvYK6rcZhmFlyTDrO25ClDqFMNfVL3luGAKKBiMljL+LTsEBWKiyhZ pQsEn+w6jAhzJj5YEwPzBjgmzF+0kdUPSBurQDDyp7THyk/n1Grv50cAOam31LlUZPgk cgFzlatJDR/I9zbGq9PmKrwWEmLAF/X2gEoldpl6Vgp173mRYQxXN/jSggqeEWeUdBRU o+x+K04R+rckn7R/0Igy4hNPye0gy0VkbiLRIXbEhOv68oVIVUZnbTtHkfL0zx2oGUmu FN9g== X-Forwarded-Encrypted: i=1; AJvYcCUNettcneJ5xBlO13VpePNZvQJnmp8vclavVW6IlY39kzcLAOyGpsSOSbSMeG8/oe2V3ZSQ5tnU7sMxS0E=@vger.kernel.org X-Gm-Message-State: AOJu0Yw+fejRc7xOMc+B9JX00PI7rRJ3Zq7VvzV/3Z0o8GSFMlIu7wHV WEOhrO2C8jvz4/rI181fJUK5MhAo6mhA1vEznnE17ACUXtGrjjt+ha9l X-Gm-Gg: ASbGnctKcCFjlgiuTAuW/S4xOBh19Zwp7glpvP6hQn/xjGYSS39f79koZP7EIrsoOgm pad87vQOq0uQXaAgIrsLpAOve/+1DvDWDKGCxeWjdyRQXMwihoLx4SKwdAZqlemCUQRnwwq5HLY 6eZ9PvagT+Urk8i1glV51iWPUIzdSfSbf4dQD2j5W0M58/KIYY2gK0HM2IK21K79W6rmmF30pXz mTDRw+9qSh2fu5NshM+4OMGnK5vm4XNYtLYrsLF+T3HA3t2d9/4LaRKSbVAjJsLCVFO3TpZ0fBr kvTDHIhCRkcvfhShqQ7hSPc4z6zO8JXN8uuuuJ9SiJo9VhOW2x7ucFYdtOcd6UIoxwXgdqKCtuZ PhYoBxEJxfyf6dhHAce5Eg8Jjrp5GmoUMm6euOOVEPOnAQG9fFgviR+ljCw== X-Google-Smtp-Source: AGHT+IG2o7GH+yYlWdqcHRNV6LJZ6go0DoQV2Ggmh0ko7bTKjKCSNDKVLaOYql/uI8eSn27eyHKBeQ== X-Received: by 2002:a17:90b:1802:b0:32b:ca6f:1231 with SMTP id 98e67ed59e1d1-32d43f05005mr19821868a91.1.1757520595521; Wed, 10 Sep 2025 09:09:55 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.50 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:55 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 14/15] mm, swap: implement dynamic allocation of swap table Date: Thu, 11 Sep 2025 00:08:32 +0800 Message-ID: <20250910160833.3464-15-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song Now swap table is cluster based, which means free clusters can free its table since no one should modify it. There could be speculative readers, like swap cache look up, protect them by making them RCU protected. All swap table should be filled with null entries before free, so such readers will either see a NULL pointer or a null filled table being lazy freed. On allocation, allocate the table when a cluster is used by any order. This way, we can reduce the memory usage of large swap device significantly. This idea to dynamically release unused swap cluster data was initially suggested by Chris Li while proposing the cluster swap allocator and it suits the swap table idea very well. Co-developed-by: Chris Li Signed-off-by: Chris Li Signed-off-by: Kairui Song Acked-by: Chris Li Suggested-by: Chris Li --- mm/swap.h | 2 +- mm/swap_state.c | 9 +-- mm/swap_table.h | 37 ++++++++- mm/swapfile.c | 202 ++++++++++++++++++++++++++++++++++++++---------- 4 files changed, 199 insertions(+), 51 deletions(-) diff --git a/mm/swap.h b/mm/swap.h index fe5c20922082..8d8efdf1297a 100644 --- a/mm/swap.h +++ b/mm/swap.h @@ -36,7 +36,7 @@ struct swap_cluster_info { u16 count; u8 flags; u8 order; - atomic_long_t *table; /* Swap table entries, see mm/swap_table.h */ + atomic_long_t __rcu *table; /* Swap table entries, see mm/swap_table.h */ struct list_head list; }; =20 diff --git a/mm/swap_state.c b/mm/swap_state.c index 97372539a575..1fc3a9eff8f2 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -91,8 +91,8 @@ struct folio *swap_cache_get_folio(swp_entry_t entry) struct folio *folio; =20 for (;;) { - swp_tb =3D __swap_table_get(__swap_entry_to_cluster(entry), - swp_cluster_offset(entry)); + swp_tb =3D swap_table_get(__swap_entry_to_cluster(entry), + swp_cluster_offset(entry)); if (!swp_tb_is_folio(swp_tb)) return NULL; folio =3D swp_tb_to_folio(swp_tb); @@ -115,11 +115,10 @@ void *swap_cache_get_shadow(swp_entry_t entry) { unsigned long swp_tb; =20 - swp_tb =3D __swap_table_get(__swap_entry_to_cluster(entry), - swp_cluster_offset(entry)); + swp_tb =3D swap_table_get(__swap_entry_to_cluster(entry), + swp_cluster_offset(entry)); if (swp_tb_is_shadow(swp_tb)) return swp_tb_to_shadow(swp_tb); - return NULL; } =20 diff --git a/mm/swap_table.h b/mm/swap_table.h index e1f7cc009701..52254e455304 100644 --- a/mm/swap_table.h +++ b/mm/swap_table.h @@ -2,8 +2,15 @@ #ifndef _MM_SWAP_TABLE_H #define _MM_SWAP_TABLE_H =20 +#include +#include #include "swap.h" =20 +/* A typical flat array in each cluster as swap table */ +struct swap_table { + atomic_long_t entries[SWAPFILE_CLUSTER]; +}; + /* * A swap table entry represents the status of a swap slot on a swap * (physical or virtual) device. The swap table in each cluster is a @@ -76,22 +83,46 @@ static inline void *swp_tb_to_shadow(unsigned long swp_= tb) static inline void __swap_table_set(struct swap_cluster_info *ci, unsigned int off, unsigned long swp_tb) { + atomic_long_t *table =3D rcu_dereference_protected(ci->table, true); + + lockdep_assert_held(&ci->lock); VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); - atomic_long_set(&ci->table[off], swp_tb); + atomic_long_set(&table[off], swp_tb); } =20 static inline unsigned long __swap_table_xchg(struct swap_cluster_info *ci, unsigned int off, unsigned long swp_tb) { + atomic_long_t *table =3D rcu_dereference_protected(ci->table, true); + + lockdep_assert_held(&ci->lock); VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); /* Ordering is guaranteed by cluster lock, relax */ - return atomic_long_xchg_relaxed(&ci->table[off], swp_tb); + return atomic_long_xchg_relaxed(&table[off], swp_tb); } =20 static inline unsigned long __swap_table_get(struct swap_cluster_info *ci, unsigned int off) { + atomic_long_t *table; + VM_WARN_ON_ONCE(off >=3D SWAPFILE_CLUSTER); - return atomic_long_read(&ci->table[off]); + table =3D rcu_dereference_check(ci->table, lockdep_is_held(&ci->lock)); + + return atomic_long_read(&table[off]); +} + +static inline unsigned long swap_table_get(struct swap_cluster_info *ci, + unsigned int off) +{ + atomic_long_t *table; + unsigned long swp_tb; + + rcu_read_lock(); + table =3D rcu_dereference(ci->table); + swp_tb =3D table ? atomic_long_read(&table[off]) : null_to_swp_tb(); + rcu_read_unlock(); + + return swp_tb; } #endif diff --git a/mm/swapfile.c b/mm/swapfile.c index 89659928465e..faf867a6c5c1 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -105,6 +105,8 @@ static DEFINE_SPINLOCK(swap_avail_lock); =20 struct swap_info_struct *swap_info[MAX_SWAPFILES]; =20 +static struct kmem_cache *swap_table_cachep; + static DEFINE_MUTEX(swapon_mutex); =20 static DECLARE_WAIT_QUEUE_HEAD(proc_poll_wait); @@ -400,10 +402,17 @@ static inline bool cluster_is_discard(struct swap_clu= ster_info *info) return info->flags =3D=3D CLUSTER_FLAG_DISCARD; } =20 +static inline bool cluster_table_is_alloced(struct swap_cluster_info *ci) +{ + return rcu_dereference_protected(ci->table, lockdep_is_held(&ci->lock)); +} + static inline bool cluster_is_usable(struct swap_cluster_info *ci, int ord= er) { if (unlikely(ci->flags > CLUSTER_FLAG_USABLE)) return false; + if (!cluster_table_is_alloced(ci)) + return false; if (!order) return true; return cluster_is_empty(ci) || order =3D=3D ci->order; @@ -421,32 +430,98 @@ static inline unsigned int cluster_offset(struct swap= _info_struct *si, return cluster_index(si, ci) * SWAPFILE_CLUSTER; } =20 -static int swap_cluster_alloc_table(struct swap_cluster_info *ci) +static void swap_cluster_free_table(struct swap_cluster_info *ci) { - WARN_ON(ci->table); - ci->table =3D kzalloc(sizeof(unsigned long) * SWAPFILE_CLUSTER, GFP_KERNE= L); - if (!ci->table) - return -ENOMEM; - return 0; + unsigned int ci_off; + struct swap_table *table; + + /* Only empty cluster's table is allow to be freed */ + lockdep_assert_held(&ci->lock); + VM_WARN_ON_ONCE(!cluster_is_empty(ci)); + for (ci_off =3D 0; ci_off < SWAPFILE_CLUSTER; ci_off++) + VM_WARN_ON_ONCE(!swp_tb_is_null(__swap_table_get(ci, ci_off))); + table =3D (void *)rcu_dereference_protected(ci->table, true); + rcu_assign_pointer(ci->table, NULL); + + kmem_cache_free(swap_table_cachep, table); } =20 -static void swap_cluster_free_table(struct swap_cluster_info *ci) +/* + * Allocate a swap table may need to sleep, which leads to migration, + * so attempt an atomic allocation first then fallback and handle + * potential race. + */ +static struct swap_cluster_info * +swap_cluster_alloc_table(struct swap_info_struct *si, + struct swap_cluster_info *ci, + int order) { - unsigned int ci_off; - unsigned long swp_tb; + struct swap_cluster_info *pcp_ci; + struct swap_table *table; + unsigned long offset; =20 - if (!ci->table) - return; + /* + * Only cluster isolation from the allocator does table allocation. + * Swap allocator uses a percpu cluster and holds the local lock. + */ + lockdep_assert_held(&ci->lock); + lockdep_assert_held(&this_cpu_ptr(&percpu_swap_cluster)->lock); + + table =3D kmem_cache_zalloc(swap_table_cachep, + __GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN); + if (table) { + rcu_assign_pointer(ci->table, table); + return ci; + } + + /* + * Try a sleep allocation. Each isolated free cluster may cause + * a sleep allocation, but there is a limited number of them, so + * the potential recursive allocation should be limited. + */ + spin_unlock(&ci->lock); + if (!(si->flags & SWP_SOLIDSTATE)) + spin_unlock(&si->global_cluster_lock); + local_unlock(&percpu_swap_cluster.lock); + table =3D kmem_cache_zalloc(swap_table_cachep, __GFP_HIGH | GFP_KERNEL); =20 - for (ci_off =3D 0; ci_off < SWAPFILE_CLUSTER; ci_off++) { - swp_tb =3D __swap_table_get(ci, ci_off); - if (!swp_tb_is_null(swp_tb)) - pr_err_once("swap: unclean swap space on swapoff: 0x%lx", - swp_tb); + local_lock(&percpu_swap_cluster.lock); + if (!(si->flags & SWP_SOLIDSTATE)) + spin_lock(&si->global_cluster_lock); + /* + * Back to atomic context. First, check if we migrated to a new + * CPU with a usable percpu cluster. If so, try using that instead. + * No need to check it for the spinning device, as swap is + * serialized by the global lock on them. + * + * The is_usable check is a bit rough, but ensures order 0 success. + */ + offset =3D this_cpu_read(percpu_swap_cluster.offset[order]); + if ((si->flags & SWP_SOLIDSTATE) && offset) { + pcp_ci =3D swap_cluster_lock(si, offset); + if (cluster_is_usable(pcp_ci, order) && + pcp_ci->count < SWAPFILE_CLUSTER) { + ci =3D pcp_ci; + goto free_table; + } + swap_cluster_unlock(pcp_ci); } =20 - kfree(ci->table); - ci->table =3D NULL; + if (!table) + return NULL; + + spin_lock(&ci->lock); + /* Nothing should have touched the dangling empty cluster. */ + if (WARN_ON_ONCE(cluster_table_is_alloced(ci))) + goto free_table; + + rcu_assign_pointer(ci->table, table); + return ci; + +free_table: + if (table) + kmem_cache_free(swap_table_cachep, table); + return ci; } =20 static void move_cluster(struct swap_info_struct *si, @@ -478,7 +553,7 @@ static void swap_cluster_schedule_discard(struct swap_i= nfo_struct *si, =20 static void __free_cluster(struct swap_info_struct *si, struct swap_cluste= r_info *ci) { - lockdep_assert_held(&ci->lock); + swap_cluster_free_table(ci); move_cluster(si, ci, &si->free_clusters, CLUSTER_FLAG_FREE); ci->order =3D 0; } @@ -493,15 +568,11 @@ static void __free_cluster(struct swap_info_struct *s= i, struct swap_cluster_info * this returns NULL for an non-empty list. */ static struct swap_cluster_info *isolate_lock_cluster( - struct swap_info_struct *si, struct list_head *list) + struct swap_info_struct *si, struct list_head *list, int order) { - struct swap_cluster_info *ci, *ret =3D NULL; + struct swap_cluster_info *ci, *found =3D NULL; =20 spin_lock(&si->lock); - - if (unlikely(!(si->flags & SWP_WRITEOK))) - goto out; - list_for_each_entry(ci, list, list) { if (!spin_trylock(&ci->lock)) continue; @@ -513,13 +584,19 @@ static struct swap_cluster_info *isolate_lock_cluster( =20 list_del(&ci->list); ci->flags =3D CLUSTER_FLAG_NONE; - ret =3D ci; + found =3D ci; break; } -out: spin_unlock(&si->lock); =20 - return ret; + if (found && !cluster_table_is_alloced(found)) { + /* Only an empty free cluster's swap table can be freed. */ + VM_WARN_ON_ONCE(list !=3D &si->free_clusters); + VM_WARN_ON_ONCE(!cluster_is_empty(found)); + return swap_cluster_alloc_table(si, found, order); + } + + return found; } =20 /* @@ -652,17 +729,27 @@ static void relocate_cluster(struct swap_info_struct = *si, * added to free cluster list and its usage counter will be increased by 1. * Only used for initialization. */ -static void inc_cluster_info_page(struct swap_info_struct *si, +static int inc_cluster_info_page(struct swap_info_struct *si, struct swap_cluster_info *cluster_info, unsigned long page_nr) { unsigned long idx =3D page_nr / SWAPFILE_CLUSTER; + struct swap_table *table; struct swap_cluster_info *ci; =20 ci =3D cluster_info + idx; + if (!ci->table) { + table =3D kmem_cache_zalloc(swap_table_cachep, GFP_KERNEL); + if (!table) + return -ENOMEM; + rcu_assign_pointer(ci->table, table); + } + ci->count++; =20 VM_BUG_ON(ci->count > SWAPFILE_CLUSTER); VM_BUG_ON(ci->flags); + + return 0; } =20 static bool cluster_reclaim_range(struct swap_info_struct *si, @@ -844,7 +931,7 @@ static unsigned int alloc_swap_scan_list(struct swap_in= fo_struct *si, unsigned int found =3D SWAP_ENTRY_INVALID; =20 do { - struct swap_cluster_info *ci =3D isolate_lock_cluster(si, list); + struct swap_cluster_info *ci =3D isolate_lock_cluster(si, list, order); unsigned long offset; =20 if (!ci) @@ -869,7 +956,7 @@ static void swap_reclaim_full_clusters(struct swap_info= _struct *si, bool force) if (force) to_scan =3D swap_usage_in_pages(si) / SWAPFILE_CLUSTER; =20 - while ((ci =3D isolate_lock_cluster(si, &si->full_clusters))) { + while ((ci =3D isolate_lock_cluster(si, &si->full_clusters, 0))) { offset =3D cluster_offset(si, ci); end =3D min(si->max, offset + SWAPFILE_CLUSTER); to_scan--; @@ -1017,6 +1104,7 @@ static unsigned long cluster_alloc_swap_entry(struct = swap_info_struct *si, int o done: if (!(si->flags & SWP_SOLIDSTATE)) spin_unlock(&si->global_cluster_lock); + return found; } =20 @@ -1884,7 +1972,13 @@ swp_entry_t get_swap_page_of_type(int type) /* This is called for allocating swap entry, not cache */ if (get_swap_device_info(si)) { if (si->flags & SWP_WRITEOK) { + /* + * Grab the local lock to be complaint + * with swap table allocation. + */ + local_lock(&percpu_swap_cluster.lock); offset =3D cluster_alloc_swap_entry(si, 0, 1); + local_unlock(&percpu_swap_cluster.lock); if (offset) { entry =3D swp_entry(si->type, offset); atomic_long_dec(&nr_swap_pages); @@ -2678,12 +2772,21 @@ static void wait_for_allocation(struct swap_info_st= ruct *si) static void free_cluster_info(struct swap_cluster_info *cluster_info, unsigned long maxpages) { + struct swap_cluster_info *ci; int i, nr_clusters =3D DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER); =20 if (!cluster_info) return; - for (i =3D 0; i < nr_clusters; i++) - swap_cluster_free_table(&cluster_info[i]); + for (i =3D 0; i < nr_clusters; i++) { + ci =3D cluster_info + i; + /* Cluster with bad marks count will have a remaining table */ + spin_lock(&ci->lock); + if (rcu_dereference_protected(ci->table, true)) { + ci->count =3D 0; + swap_cluster_free_table(ci); + } + spin_unlock(&ci->lock); + } kvfree(cluster_info); } =20 @@ -2719,6 +2822,7 @@ SYSCALL_DEFINE1(swapoff, const char __user *, special= file) struct address_space *mapping; struct inode *inode; struct filename *pathname; + unsigned int maxpages; int err, found =3D 0; =20 if (!capable(CAP_SYS_ADMIN)) @@ -2825,8 +2929,8 @@ SYSCALL_DEFINE1(swapoff, const char __user *, special= file) p->swap_map =3D NULL; zeromap =3D p->zeromap; p->zeromap =3D NULL; + maxpages =3D p->max; cluster_info =3D p->cluster_info; - free_cluster_info(cluster_info, p->max); p->max =3D 0; p->cluster_info =3D NULL; spin_unlock(&p->lock); @@ -2838,6 +2942,7 @@ SYSCALL_DEFINE1(swapoff, const char __user *, special= file) p->global_cluster =3D NULL; vfree(swap_map); kvfree(zeromap); + free_cluster_info(cluster_info, maxpages); /* Destroy swap account information */ swap_cgroup_swapoff(p->type); =20 @@ -3216,11 +3321,8 @@ static struct swap_cluster_info *setup_clusters(stru= ct swap_info_struct *si, if (!cluster_info) goto err; =20 - for (i =3D 0; i < nr_clusters; i++) { + for (i =3D 0; i < nr_clusters; i++) spin_lock_init(&cluster_info[i].lock); - if (swap_cluster_alloc_table(&cluster_info[i])) - goto err_free; - } =20 if (!(si->flags & SWP_SOLIDSTATE)) { si->global_cluster =3D kmalloc(sizeof(*si->global_cluster), @@ -3239,16 +3341,23 @@ static struct swap_cluster_info *setup_clusters(str= uct swap_info_struct *si, * See setup_swap_map(): header page, bad pages, * and the EOF part of the last cluster. */ - inc_cluster_info_page(si, cluster_info, 0); + err =3D inc_cluster_info_page(si, cluster_info, 0); + if (err) + goto err; for (i =3D 0; i < swap_header->info.nr_badpages; i++) { unsigned int page_nr =3D swap_header->info.badpages[i]; =20 if (page_nr >=3D maxpages) continue; - inc_cluster_info_page(si, cluster_info, page_nr); + err =3D inc_cluster_info_page(si, cluster_info, page_nr); + if (err) + goto err; + } + for (i =3D maxpages; i < round_up(maxpages, SWAPFILE_CLUSTER); i++) { + err =3D inc_cluster_info_page(si, cluster_info, i); + if (err) + goto err; } - for (i =3D maxpages; i < round_up(maxpages, SWAPFILE_CLUSTER); i++) - inc_cluster_info_page(si, cluster_info, i); =20 INIT_LIST_HEAD(&si->free_clusters); INIT_LIST_HEAD(&si->full_clusters); @@ -3962,6 +4071,15 @@ static int __init swapfile_init(void) =20 swapfile_maximum_size =3D arch_max_swapfile_size(); =20 + /* + * Once a cluster is freed, it's swap table content is read + * only, and all swap cache readers (swap_cache_*) verifies + * the content before use. So it's safe to use RCU slab here. + */ + swap_table_cachep =3D kmem_cache_create("swap_table", + sizeof(struct swap_table), + 0, SLAB_PANIC | SLAB_TYPESAFE_BY_RCU, NULL); + #ifdef CONFIG_MIGRATION if (swapfile_maximum_size >=3D (1UL << SWP_MIG_TOTAL_BITS)) swap_migration_ad_supported =3D true; --=20 2.51.0 From nobody Wed Sep 10 23:46:17 2025 Received: from mail-pg1-f176.google.com (mail-pg1-f176.google.com [209.85.215.176]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 643973376B7 for ; Wed, 10 Sep 2025 16:10:01 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.215.176 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520603; cv=none; b=a5ZHWSIWpVN6+P5J4aFzEPr9TgaaEAIkJ61ZbGUA0wpdUsmtxU+xtK9LgIFpq53XZb6RX4WCCuv+X2JjWPjhyBbKQJVLtShVIKTsX/RvBzIVbDapegpt3A+43xZ79bXaNX1N9kozuOBANLFEX4ArRNTEULyuIzYDG9yfdRZbt/s= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1757520603; c=relaxed/simple; bh=QL36JjOd6NS2P9etFIDhKrt8BasNgA9JZNK8noTcJxE=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=FJAYZGoXdVuIesZ3Iynk5kyJt88XZ73mWhy5wcsM/aVOIMTv5Ae4h7CDLQ/BsARC6E6mKS+sKzgNS5HuQjWod+IcNQ64FEjGFsO6IyYFo/NG34nDO27dd5i3NaVM2OLLN5DW0RSVD8z2SKyGkJVf7PIqcDMTw00NKNuFm22xMGw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=YhPYkQLv; arc=none smtp.client-ip=209.85.215.176 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="YhPYkQLv" Received: by mail-pg1-f176.google.com with SMTP id 41be03b00d2f7-b4f9d61e7deso4496242a12.2 for ; Wed, 10 Sep 2025 09:10:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1757520601; x=1758125401; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:from:to:cc:subject :date:message-id:reply-to; bh=9X8IezR3SoWMo8wYuh3xUkeHhs7bAkncRRB5ILkQmNw=; b=YhPYkQLv1OSmGQm0fAL3bdtLtE2aye6wSAf4fIbp3cYSF5A6Gzu+EXfj3/vMKWpkfC aTluOPt11XKbEtW9qGUdFZWFScjDDJpH/7ffow0M98cPf43+yl4lP9LUvI8r/68E6D31 GN0Ix88Ti6wE9KskQt3ZLsa3LaF7NIUGdtvwhFk45jtkpnR/XbrMmYgpULAeRC3z5NG8 DsNAKz3abXUv/DVydKUJxEzAYqO1DKZ5x183HwZMle8fEgL81c8slOFTnDqgWRjOg3wn g+SZfV2aftMyUVWQdHFiy+tEcNbWM5eVZdV5COVa35ZQjVDmdfu3bPj8ofkmFyhmFqkO xp0A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757520601; x=1758125401; h=content-transfer-encoding:mime-version:reply-to:references :in-reply-to:message-id:date:subject:cc:to:from:x-gm-message-state :from:to:cc:subject:date:message-id:reply-to; bh=9X8IezR3SoWMo8wYuh3xUkeHhs7bAkncRRB5ILkQmNw=; b=PsnKJEx2ETxEtD1u9EozNLJf9xdTvuHy15aNRYiqa3isnDUqd98ahCn2iKzeB9WEYR xzM1/k2bfZMEfZETelYoJaoUJRRqeQO0GfM+e4c52NORyCT062IcTK4P43XZ1B+Y5Srw zsanCFZNqN43MSusZCkNPxLDqAEMtsIhDTMQhKOyoAhQgDwPpAl4u+NQWwXtefLUshBJ sbvmLivLYKBhL0x7+/xyEvLm4UNw3pvWWMJXlEUSyBULhGdcCH05mwlDqPLHWBiRYxbU 6V8aNYb5kww0ZN76UZVriWZCwDlYKG88PMp4QQN+B8Z3LAdjIj8QnPzKQdd+1s3m9Ce3 PL7A== X-Forwarded-Encrypted: i=1; AJvYcCVOpziJOjEI66TfmHfZ0Ebq28HdMVxAHQG5uhlA1IeC0CC9Y7QUsnT/pqC5kfFn4xpjf3nvaxsTlWCKspk=@vger.kernel.org X-Gm-Message-State: AOJu0YyvwStBRpnJwlyy+J83kOZX8sTE4Bly6acNYGw0+LggXDdwMQ03 9zJe2tOBMdK0IJ+KsZMDbTTXHjH3YOCYuS+RaSn0N7B2Iexzi7+ufzGG X-Gm-Gg: ASbGncsibfEmwaHBmcznDzg1v6IgHS1zFlCnkrZQTTxT2e/VsiRev2IyHrlAw0xy9dC jYAez+TDtHNeBjFLgDMJH5mivSD3taX81KruoHtLNr7I4eV/rdOpAxP5ISDutWEtMVXHed9IzDn 8n3z4D2e1SttjrGFUbYqdFRMA2JAGGbo4E1Y3+qyH4Ij1luAVkLuZOMiqg0398rsh/96ck1qMAH Txv0D1z9vc0K7CbA12OtcjKWNZ346QKmQh76X5if9j1mRc4f5evDzUG4meFINieWCoFRLEYOcDN I1J8N1ofGLaoK3IwaILXpJuhvMrq+KDwXblMRgPdctS1GZ0iZjyVO4ZOAXZcSMMhq+rlEC+c+mm VQajze2S8zr04ciAIMSTASyGPvcRu76OP4EKrUbaiT3lygObofux2eGbnbw== X-Google-Smtp-Source: AGHT+IGH73IxV7jUuJYAM9aSrRqAGj9iR4ECuxZxMfuieRYNWITXMqhhRKZss7t1YGKHwvp/sLs/2Q== X-Received: by 2002:a17:90b:3d4e:b0:32d:cab5:a51b with SMTP id 98e67ed59e1d1-32dcab5a716mr1672236a91.29.1757520600531; Wed, 10 Sep 2025 09:10:00 -0700 (PDT) Received: from KASONG-MC4.tencent.com ([101.32.222.185]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-b54998a1068sm1000142a12.31.2025.09.10.09.09.55 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Wed, 10 Sep 2025 09:09:59 -0700 (PDT) From: Kairui Song To: linux-mm@kvack.org Cc: Kairui Song , Andrew Morton , Matthew Wilcox , Hugh Dickins , Chris Li , Barry Song , Baoquan He , Nhat Pham , Kemeng Shi , Baolin Wang , Ying Huang , Johannes Weiner , David Hildenbrand , Yosry Ahmed , Lorenzo Stoakes , Zi Yan , linux-kernel@vger.kernel.org, Kairui Song Subject: [PATCH v3 15/15] mm, swap: use a single page for swap table when the size fits Date: Thu, 11 Sep 2025 00:08:33 +0800 Message-ID: <20250910160833.3464-16-ryncsn@gmail.com> X-Mailer: git-send-email 2.51.0 In-Reply-To: <20250910160833.3464-1-ryncsn@gmail.com> References: <20250910160833.3464-1-ryncsn@gmail.com> Reply-To: Kairui Song Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Kairui Song We have a cluster size of 512 slots. Each slot consumes 8 bytes in swap table so the swap table size of each cluster is exactly one page (4K). If that condition is true, allocate one page direct and disable the slab cache to reduce the memory usage of swap table and avoid fragmentation. Co-developed-by: Chris Li Signed-off-by: Chris Li Signed-off-by: Kairui Song Acked-by: Chris Li Suggested-by: Chris Li --- mm/swap_table.h | 2 ++ mm/swapfile.c | 50 ++++++++++++++++++++++++++++++++++++++++--------- 2 files changed, 43 insertions(+), 9 deletions(-) diff --git a/mm/swap_table.h b/mm/swap_table.h index 52254e455304..ea244a57a5b7 100644 --- a/mm/swap_table.h +++ b/mm/swap_table.h @@ -11,6 +11,8 @@ struct swap_table { atomic_long_t entries[SWAPFILE_CLUSTER]; }; =20 +#define SWP_TABLE_USE_PAGE (sizeof(struct swap_table) =3D=3D PAGE_SIZE) + /* * A swap table entry represents the status of a swap slot on a swap * (physical or virtual) device. The swap table in each cluster is a diff --git a/mm/swapfile.c b/mm/swapfile.c index faf867a6c5c1..cebf2dcfaab0 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -430,6 +430,38 @@ static inline unsigned int cluster_offset(struct swap_= info_struct *si, return cluster_index(si, ci) * SWAPFILE_CLUSTER; } =20 +static struct swap_table *swap_table_alloc(gfp_t gfp) +{ + struct folio *folio; + + if (!SWP_TABLE_USE_PAGE) + return kmem_cache_zalloc(swap_table_cachep, gfp); + + folio =3D folio_alloc(gfp | __GFP_ZERO, 0); + if (folio) + return folio_address(folio); + return NULL; +} + +static void swap_table_free_folio_rcu_cb(struct rcu_head *head) +{ + struct folio *folio; + + folio =3D page_folio(container_of(head, struct page, rcu_head)); + folio_put(folio); +} + +static void swap_table_free(struct swap_table *table) +{ + if (!SWP_TABLE_USE_PAGE) { + kmem_cache_free(swap_table_cachep, table); + return; + } + + call_rcu(&(folio_page(virt_to_folio(table), 0)->rcu_head), + swap_table_free_folio_rcu_cb); +} + static void swap_cluster_free_table(struct swap_cluster_info *ci) { unsigned int ci_off; @@ -443,7 +475,7 @@ static void swap_cluster_free_table(struct swap_cluster= _info *ci) table =3D (void *)rcu_dereference_protected(ci->table, true); rcu_assign_pointer(ci->table, NULL); =20 - kmem_cache_free(swap_table_cachep, table); + swap_table_free(table); } =20 /* @@ -467,8 +499,7 @@ swap_cluster_alloc_table(struct swap_info_struct *si, lockdep_assert_held(&ci->lock); lockdep_assert_held(&this_cpu_ptr(&percpu_swap_cluster)->lock); =20 - table =3D kmem_cache_zalloc(swap_table_cachep, - __GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN); + table =3D swap_table_alloc(__GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN); if (table) { rcu_assign_pointer(ci->table, table); return ci; @@ -483,7 +514,7 @@ swap_cluster_alloc_table(struct swap_info_struct *si, if (!(si->flags & SWP_SOLIDSTATE)) spin_unlock(&si->global_cluster_lock); local_unlock(&percpu_swap_cluster.lock); - table =3D kmem_cache_zalloc(swap_table_cachep, __GFP_HIGH | GFP_KERNEL); + table =3D swap_table_alloc(__GFP_HIGH | GFP_KERNEL); =20 local_lock(&percpu_swap_cluster.lock); if (!(si->flags & SWP_SOLIDSTATE)) @@ -520,7 +551,7 @@ swap_cluster_alloc_table(struct swap_info_struct *si, =20 free_table: if (table) - kmem_cache_free(swap_table_cachep, table); + swap_table_free(table); return ci; } =20 @@ -738,7 +769,7 @@ static int inc_cluster_info_page(struct swap_info_struc= t *si, =20 ci =3D cluster_info + idx; if (!ci->table) { - table =3D kmem_cache_zalloc(swap_table_cachep, GFP_KERNEL); + table =3D swap_table_alloc(GFP_KERNEL); if (!table) return -ENOMEM; rcu_assign_pointer(ci->table, table); @@ -4076,9 +4107,10 @@ static int __init swapfile_init(void) * only, and all swap cache readers (swap_cache_*) verifies * the content before use. So it's safe to use RCU slab here. */ - swap_table_cachep =3D kmem_cache_create("swap_table", - sizeof(struct swap_table), - 0, SLAB_PANIC | SLAB_TYPESAFE_BY_RCU, NULL); + if (!SWP_TABLE_USE_PAGE) + swap_table_cachep =3D kmem_cache_create("swap_table", + sizeof(struct swap_table), + 0, SLAB_PANIC | SLAB_TYPESAFE_BY_RCU, NULL); =20 #ifdef CONFIG_MIGRATION if (swapfile_maximum_size >=3D (1UL << SWP_MIG_TOTAL_BITS)) --=20 2.51.0