From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-yw1-f181.google.com (mail-yw1-f181.google.com [209.85.128.181]) (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 CB8EC3F86E7 for ; Thu, 16 Jul 2026 18:40:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.128.181 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784227247; cv=none; b=dG/rbAjiFt0Wc5axJlau8GVlsnV344jr7sju4+VicWY8Wa5qmezxOihCW1RroA7zCejeNcR07rNkX5P2/5fvSN2lyv2g36lgfAs7lRzHAQzapk9gkGCnfTvpvLtzjN5rZCiYnPhw7SRCGQOeqDClvx6ncDVfu8hleOh822shWIk= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784227247; c=relaxed/simple; bh=ODTMwkCkSLRy00cr58DdZi88uC4XC16VI0Hq3Qdwtww=; h=Message-ID:Subject:From:To:Cc:Date:In-Reply-To:References: Content-Type:MIME-Version; b=C4zruec8GRVJQ7a87qBRnR6di9xbLDjO/B2C42H84V4Vi9xR1Kv92SVDXWNIYsFeniSn7rXXq+em7/3FmhStuGPSyhZDy305FX+IwLGAr6nDwZS7Kfn4COJuwRQ2zsU31nL6q35IArumpaauVNplKWHNVy7mAimFEyzWPh4VIyI= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dubeyko.com; spf=pass smtp.mailfrom=dubeyko.com; dkim=pass (2048-bit key) header.d=dubeyko-com.20251104.gappssmtp.com header.i=@dubeyko-com.20251104.gappssmtp.com header.b=xigU0t0x; arc=none smtp.client-ip=209.85.128.181 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dubeyko.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=dubeyko.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=dubeyko-com.20251104.gappssmtp.com header.i=@dubeyko-com.20251104.gappssmtp.com header.b="xigU0t0x" Received: by mail-yw1-f181.google.com with SMTP id 00721157ae682-81e8fa1b8d6so63027037b3.1 for ; Thu, 16 Jul 2026 11:40:44 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=dubeyko-com.20251104.gappssmtp.com; s=20251104; t=1784227244; x=1784832044; darn=vger.kernel.org; h=mime-version:user-agent:content-transfer-encoding:content-type :autocrypt:references:in-reply-to:date:cc:to:from:subject:message-id :from:to:cc:subject:date:message-id:reply-to:content-type; bh=vdSy1R8iBM3oBFxb7L5pKTCR5nmCXrmLm7RioHMDBNw=; b=xigU0t0xoT5Mj0Rcpi56QlDdjCLkDlqrn1ArSFNp5+XfyQ+yT2YKscOrm8LTV3h1QZ +gyqtgW26Ii7nFV0D0FpGyv5M5b9KiB+mLr/YgISUGVZFC6xoz7tGf4LDLjSrsnjqLtX jQ9Wz5yqy6waK1KSid27n1gdotlgNy6SSgq0F9fEqm+/JLomTVZXhBAqNKkmnWhSmPO4 aRNyRKEpLFMqnUCkBcF5/a159CTDPjCmyAiscmmm8aYWBMOFVsTRHUp/69ij2vQvCKku 5C7ceaKzBm7Fob2QxytU3vyJaYPcsqP4/D1naK+WQk2gn2pQKDkRu/90gUzM/Kbpp23y zQOA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1784227244; x=1784832044; h=mime-version:user-agent:content-transfer-encoding:content-type :autocrypt:references:in-reply-to:date:cc:to:from:subject:message-id :x-gm-gg:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to:content-type; bh=vdSy1R8iBM3oBFxb7L5pKTCR5nmCXrmLm7RioHMDBNw=; b=JG5Q2IiSNA6jYn9r0TUBzC0kdfy6bRo70BMATOjLs6DFCwMxR230kUaWCj/fbdIUWS DO75PE0r8/fOUKUPGW06iH+FtnVx/aA7h7uybVXPHytLxUQS7cDAjPdgj3vNtyFrZFsl 5F+PvawmJyp2lqms0ZYyNyIy9Yhu8KT8qSL2rTWvLaqEwR64Zd7FLn1CgxqqPRqpf3Av GyJqWms2iiUHygHtjNoDiKAjz+5LqGZITcyVEhqZB+x753NB/fTwVXNf+QNcmOFqgyYG WWod2BHARe/Nu80ngYOFiwNQ9T8LgG4NHRn57eYKFMtRwFKqMMg3ZLxscSyp2aOljbwv ybGQ== X-Forwarded-Encrypted: i=1; AHgh+RqvygiMRD9+RhAD3VEfhznFZDRUMNl4/thVE9JgcCg//Nk7N/mlwM1pu0iSJl5h3bDsjRA7RNeEiuPmd20=@vger.kernel.org X-Gm-Message-State: AOJu0YzJJ/fskiQafrVc39FTq9aWniLGdcBEihRvuBTfunNlHSg7vVel A1pOo6j/BMvlg36T1O8VNhw6U+l95ZZrfPxajf7nRdLzZVUprum68Xj7OpeHCLMS0lg= X-Gm-Gg: AfdE7clTeqXp+neFjD594xYUJ3YQMnHq6P5nmfGsphYYJB489cim/4/HjI3olbykpl/ KWHbUtiprHMDrebm9oIRG092jYMXzrCzYSg5KnZY5hR2kiRXj1hrHvrjGp6hZ58yzzOY4y0EqXC DXjpSIde+eegGu9MS5uSwurbwtxTMEfRlkkQboQ0Q4dKSFJJBTajleWmGIr7phpqLDaMKbWEAqy 8JRaO0KrqILisnTBKgTq85Qn4v+R6UO/NNBsEQxnhKwKI9sIg46VBNCBFp8ENWI7fu3JygJ6sdD Yj9cAHn6eCQkIXPOCjb9EW+jIciWF0mG3U9Gc00r0C9RaqsOirZutiXlQl+tvuEaj3V36h+EJb0 Y9UO2lE+Keqzg4KXrUZsFWHP0zZAMp7wntGg+31beni6X0QPT3xCmZLlt8nea+xKu9VAMEf3uHd OQ71v6iLNW4BGZc1x20qlycx+JOHa2NIw2KCmztb9z/4srGAdq+dTKGwqdzTDg3JJiE0mV1o0G+ mx38urIOfQI+DCuCF2fQ2IpkP3+cW1HMVRqMq0xIOv22b01jLrs10DHoWv42aV6wj7pCuY= X-Received: by 2002:a05:690c:e1f:b0:80d:5978:9730 with SMTP id 00721157ae682-81ebffea556mr86174907b3.43.1784227243634; Thu, 16 Jul 2026 11:40:43 -0700 (PDT) Received: from ?IPv6:2600:1700:6476:1430:1ec9:9caa:73bc:e646? ([2600:1700:6476:1430:1ec9:9caa:73bc:e646]) by smtp.gmail.com with ESMTPSA id 00721157ae682-81e6c1a32e2sm201582307b3.30.2026.07.16.11.40.42 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 16 Jul 2026 11:40:43 -0700 (PDT) Message-ID: Subject: Re: [PATCH v6] hfs: port HFS+ b-tree bitmap corruption check From: Viacheslav Dubeyko To: Aditya Prakash Srivastava Cc: glaubitz@physik.fu-berlin.de, frank.li@vivo.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org Date: Thu, 16 Jul 2026 11:40:39 -0700 In-Reply-To: <20260716074150.1660-1-aditya.ansh182@gmail.com> References: <20260716074150.1660-1-aditya.ansh182@gmail.com> Autocrypt: addr=slava@dubeyko.com; prefer-encrypt=mutual; keydata=mQINBGgaTLYBEADaJc/WqWTeunGetXyyGJ5Za7b23M/ozuDCWCp+yWUa2GqQKH40dxRIR zshgOmAue7t9RQJU9lxZ4ZHWbi1Hzz85+0omefEdAKFmxTO6+CYV0g/sapU0wPJws3sC2Pbda9/eJ ZcvScAX2n/PlhpTnzJKf3JkHh3nM1ACO3jzSe2/muSQJvqMLG2D71ccekr1RyUh8V+OZdrPtfkDam V6GOT6IvyE+d+55fzmo20nJKecvbyvdikWwZvjjCENsG9qOf3TcCJ9DDYwjyYe1To8b+mQM9nHcxp jUsUuH074BhISFwt99/htZdSgp4csiGeXr8f9BEotRB6+kjMBHaiJ6B7BIlDmlffyR4f3oR/5hxgy dvIxMocqyc03xVyM6tA4ZrshKkwDgZIFEKkx37ec22ZJczNwGywKQW2TGXUTZVbdooiG4tXbRBLxe ga/NTZ52ZdEkSxAUGw/l0y0InTtdDIWvfUT+WXtQcEPRBE6HHhoeFehLzWL/o7w5Hog+0hXhNjqte fzKpI2fWmYzoIb6ueNmE/8sP9fWXo6Av9m8B5hRvF/hVWfEysr/2LSqN+xjt9NEbg8WNRMLy/Y0MS p5fgf9pmGF78waFiBvgZIQNuQnHrM+0BmYOhR0JKoHjt7r5wLyNiKFc8b7xXndyCDYfniO3ljbr0j tXWRGxx4to6FwARAQABtCZWaWFjaGVzbGF2IER1YmV5a28gPHNsYXZhQGR1YmV5a28uY29tPokCVw QTAQoAQQIbAQUJA8JnAAULCQgHAgYVCgkICwIEFgIDAQIeAQIXgBYhBFXDC2tnzsoLQtrbBDlc2cL fhEB1BQJoGl5PAhkBAAoJEDlc2cLfhEB17DsP/jy/Dx19MtxWOniPqpQf2s65enkDZuMIQ94jSg7B F2qTKIbNR9SmsczjyjC+/J7m7WZRmcqnwFYMOyNfh12aF2WhjT7p5xEAbvfGVYwUpUrg/lcacdT0D Yk61GGc5ZB89OAWHLr0FJjI54bd7kn7E/JRQF4dqNsxU8qcPXQ0wLHxTHUPZu/w5Zu/cO+lQ3H0Pj pSEGaTAh+tBYGSvQ4YPYBcV8+qjTxzeNwkw4ARza8EjTwWKP2jWAfA/ay4VobRfqNQ2zLoo84qDtN Uxe0zPE2wobIXELWkbuW/6hoQFPpMlJWz+mbvVms57NAA1HO8F5c1SLFaJ6dN0AQbxrHi45/cQXla 9hSEOJjxcEnJG/ZmcomYHFneM9K1p1K6HcGajiY2BFWkVet9vuHygkLWXVYZ0lr1paLFR52S7T+cf 6dkxOqu1ZiRegvFoyzBUzlLh/elgp3tWUfG2VmJD3lGpB3m5ZhwQ3rFpK8A7cKzgKjwPp61Me0o9z HX53THoG+QG+o0nnIKK7M8+coToTSyznYoq9C3eKeM/J97x9+h9tbizaeUQvWzQOgG8myUJ5u5Dr4 6tv9KXrOJy0iy/dcyreMYV5lwODaFfOeA4Lbnn5vRn9OjuMg1PFhCi3yMI4lA4umXFw0V2/OI5rgW BQELhfvW6mxkihkl6KLZX8m1zcHitCpWaWFjaGVzbGF2IER1YmV5a28gPFNsYXZhLkR1YmV5a29Aa WJtLmNvbT6JAlQEEwEKAD4WIQRVwwtrZ87KC0La2wQ5XNnC34RAdQUCaBpd7AIbAQUJA8JnAAULCQ gHAgYVCgkICwIEFgIDAQIeAQIXgAAKCRA5XNnC34RAdYjFEACiWBEybMt1xjRbEgaZ3UP5i2bSway DwYDvgWW5EbRP7JcqOcZ2vkJwrK3gsqC3FKpjOPh7ecE0I4vrabH1Qobe2N8B2Y396z24mGnkTBbb 16Uz3PC93nFN1BA0wuOjlr1/oOTy5gBY563vybhnXPfSEUcXRd28jI7z8tRyzXh2tL8ZLdv1u4vQ8 E0O7lVJ55p9yGxbwgb5vXU4T2irqRKLxRvU80rZIXoEM7zLf5r7RaRxgwjTKdu6rYMUOfoyEQQZTD 4Xg9YE/X8pZzcbYFs4IlscyK6cXU0pjwr2ssjearOLLDJ7ygvfOiOuCZL+6zHRunLwq2JH/RmwuLV mWWSbgosZD6c5+wu6DxV15y7zZaR3NFPOR5ErpCFUorKzBO1nA4dwOAbNym9OGkhRgLAyxwpea0V0 ZlStfp0kfVaSZYo7PXd8Bbtyjali0niBjPpEVZdgtVUpBlPr97jBYZ+L5GF3hd6WJFbEYgj+5Af7C UjbX9DHweGQ/tdXWRnJHRzorxzjOS3003ddRnPtQDDN3Z/XzdAZwQAs0RqqXrTeeJrLppFUbAP+HZ TyOLVJcAAlVQROoq8PbM3ZKIaOygjj6Yw0emJi1D9OsN2UKjoe4W185vamFWX4Ba41jmCPrYJWAWH fAMjjkInIPg7RLGs8FiwxfcpkILP0YbVWHiNAabQoVmlhY2hlc2xhdiBEdWJleWtvIDx2ZHViZXlr b0BrZXJuZWwub3JnPokCVAQTAQoAPhYhBFXDC2tnzsoLQtrbBDlc2cLfhEB1BQJoVemuAhsBBQkDw mcABQsJCAcCBhUKCQgLAgQWAgMBAh4BAheAAAoJEDlc2cLfhEB1GRwP/1scX5HO9Sk7dRicLD/fxo ipwEs+UbeA0/TM8OQfdRI4C/tFBYbQCR7lD05dfq8VsYLEyrgeLqP/iRhabLky8LTaEdwoAqPDc/O 9HRffx/faJZqkKc1dZryjqS6b8NExhKOVWmDqN357+Cl/H4hT9wnvjCj1YEqXIxSd/2Pc8+yw/KRC AP7jtRzXHcc/49Lpz/NU5irScusxy2GLKa5o/13jFK3F1fWX1wsOJF8NlTx3rLtBy4GWHITwkBmu8 zI4qcJGp7eudI0l4xmIKKQWanEhVdzBm5UnfyLIa7gQ2T48UbxJlWnMhLxMPrxgtC4Kos1G3zovEy Ep+fJN7D1pwN9aR36jVKvRsX7V4leIDWGzCdfw1FGWkMUfrRwgIl6i3wgqcCP6r9YSWVQYXdmwdMu 1RFLC44iF9340S0hw9+30yGP8TWwd1mm8V/+zsdDAFAoAwisi5QLLkQnEsJSgLzJ9daAsE8KjMthv hUWHdpiUSjyCpigT+KPl9YunZhyrC1jZXERCDPCQVYgaPt+Xbhdjcem/ykv8UVIDAGVXjuk4OW8la nf8SP+uxkTTDKcPHOa5rYRaeNj7T/NClRSd4z6aV3F6pKEJnEGvv/DFMXtSHlbylhyiGKN2Amd0b4 9jg+DW85oNN7q2UYzYuPwkHsFFq5iyF1QggiwYYTpoVXsw Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable User-Agent: Evolution 3.60.1 (by Flathub.org) Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 On Thu, 2026-07-16 at 07:41 +0000, Aditya Prakash Srivastava wrote: > In HFS+ filesystems, during b-tree open (hfs_btree_open()), the code > verifies that the allocation map bit for the tree header (node 0) is > set. If not, it indicates a corrupted map record/bitmap and mounts > the volume as read-only (SB_RDONLY) to prevent further damage. >=20 > HFS filesystems share the same b-tree structure but currently lack > this corruption detection check. >=20 > Port this check to HFS, aligning its implementation with HFS+ to > maintain consistent b-tree logic across both filesystems: > 1. Define struct hfs_bmap_ctx, and define HFS_TREE_HEAD and the > relevant > =C2=A0=C2=A0 map record indices in include/linux/hfs_common.h. > 2. Port the necessary offset and length validation helpers to > =C2=A0=C2=A0 fs/hfs/btree.h as static inline functions with robust null p= ointer > =C2=A0=C2=A0 checks. > 3. Implement static hfs_bmap_get_map_page() in fs/hfs/btree.c. > 4. Implement static hfs_bmap_test_bit() in fs/hfs/btree.c to inspect > =C2=A0=C2=A0 the B-tree bitmap using the get_map_page helper. > 5. Implement static hfs_bmap_clear_bit() in fs/hfs/btree.c. > 6. Rewrite hfs_bmap_alloc() in fs/hfs/btree.c to use the unified > =C2=A0=C2=A0 hfs_bmap_get_map_page() helper, eliminating redundant page a= nd > offset > =C2=A0=C2=A0 calculation code. > 7. Refactor hfs_bmap_free() in fs/hfs/btree.c to use the new > =C2=A0=C2=A0 hfs_bmap_clear_bit() helper. > 8. In hfs_btree_open(), retrieve the header node via > hfs_bnode_find(), > =C2=A0=C2=A0 test its allocation bit with hfs_bmap_test_bit(), and releas= e it > =C2=A0=C2=A0 using hfs_bnode_put(). >=20 > Suggested-by: Viacheslav Dubeyko > Link: > https://lore.kernel.org/all/6a36101b.be22b350.2a3e9.0001.GAE@google.com/T= /#r446d0fed2a2900bd805534bbcb799d86619ae2ea > Signed-off-by: Aditya Prakash Srivastava > --- > =C2=A0fs/hfs/bnode.c=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 |=C2=A0 42 -------- > =C2=A0fs/hfs/btree.c=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 | 210 +++++++++++++++++++++++++++++------ > -- > =C2=A0fs/hfs/btree.h=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 |=C2=A0 40 +++++++ > =C2=A0include/linux/hfs_common.h |=C2=A0 13 ++- > =C2=A04 files changed, 218 insertions(+), 87 deletions(-) >=20 > diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c > index da8e5342c91c..1b331108d9c0 100644 > --- a/fs/hfs/bnode.c > +++ b/fs/hfs/bnode.c > @@ -15,48 +15,6 @@ > =C2=A0 > =C2=A0#include "btree.h" > =C2=A0 > -static inline > -bool is_bnode_offset_valid(struct hfs_bnode *node, u32 off) > -{ > - bool is_valid =3D off < node->tree->node_size; > - > - if (!is_valid) { > - pr_err("requested invalid offset: " > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 "NODE: id %u, type %#x, height %u= , " > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 "node_size %u, offset %u\n", > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->this, node->type, node->hei= ght, > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->tree->node_size, off); > - } > - > - return is_valid; > -} > - > -static inline > -u32 check_and_correct_requested_length(struct hfs_bnode *node, u32 > off, u32 len) > -{ > - unsigned int node_size; > - > - if (!is_bnode_offset_valid(node, off)) > - return 0; > - > - node_size =3D node->tree->node_size; > - > - if ((u64)off + len > node_size) { > - u32 new_len =3D node_size - off; > - > - pr_err("requested length has been corrected: " > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 "NODE: id %u, type %#x, height %u= , " > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 "node_size %u, offset %u, " > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 "requested_len %u, corrected_len = %u\n", > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->this, node->type, node->hei= ght, > - =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->tree->node_size, off, len, = new_len); > - > - return new_len; > - } > - > - return len; > -} > - > =C2=A0void hfs_bnode_read(struct hfs_bnode *node, void *buf, u32 off, u32 > len) > =C2=A0{ > =C2=A0 struct page *page; > diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c > index 2eb37a2f64e8..41b4e8fc9463 100644 > --- a/fs/hfs/btree.c > +++ b/fs/hfs/btree.c > @@ -15,6 +15,121 @@ > =C2=A0 > =C2=A0#include "btree.h" > =C2=A0 > +/* Context for iterating b-tree map pages > + * @page_idx: The index of the page within the b-node's page array > + * @off: The byte offset within the mapped page > + * @len: The remaining length of the map record > + */ > +struct hfs_bmap_ctx { > + unsigned int page_idx; > + unsigned int off; > + u16 len; > +}; > + > +/* > + * Finds the specific page containing the requested byte offset > within the map > + * record. Automatically handles the difference between header and > map nodes. > + * Returns the struct page pointer, or an ERR_PTR on failure. > + * Note: The caller is responsible for mapping/unmapping the > returned page. > + */ > +static struct page *hfs_bmap_get_map_page(struct hfs_bnode *node, > + =C2=A0 struct hfs_bmap_ctx *ctx, > + =C2=A0 u32 byte_offset) > +{ > + u16 rec_idx, off16; > + unsigned int page_off; > + > + if (node->this =3D=3D HFS_TREE_HEAD) { > + if (node->type !=3D HFS_NODE_HEADER) { > + pr_err("hfs: invalid btree header node\n"); > + return ERR_PTR(-EIO); > + } > + rec_idx =3D HFS_BTREE_HDR_MAP_REC_INDEX; > + } else { > + if (node->type !=3D HFS_NODE_MAP) { > + pr_err("hfs: invalid btree map node\n"); > + return ERR_PTR(-EIO); > + } > + rec_idx =3D HFS_BTREE_MAP_NODE_REC_INDEX; > + } > + > + ctx->len =3D hfs_brec_lenoff(node, rec_idx, &off16); > + if (!ctx->len) > + return ERR_PTR(-ENOENT); > + > + if (!is_bnode_offset_valid(node, off16)) > + return ERR_PTR(-EIO); > + > + ctx->len =3D check_and_correct_requested_length(node, off16, > ctx->len); > + > + if (byte_offset >=3D ctx->len) > + return ERR_PTR(-EINVAL); > + > + page_off =3D (u32)off16 + node->page_offset + byte_offset; > + ctx->page_idx =3D page_off >> PAGE_SHIFT; > + ctx->off =3D page_off & ~PAGE_MASK; > + > + return node->page[ctx->page_idx]; > +} > + > +/** > + * hfs_bmap_test_bit - test a bit in the b-tree map > + * @node: the b-tree node containing the map record > + * @node_bit_idx: the relative bit index within the node's map > record > + * > + * Returns true if set, false if clear or on failure. > + */ > +static bool hfs_bmap_test_bit(struct hfs_bnode *node, u32 > node_bit_idx) > +{ > + struct hfs_bmap_ctx ctx; > + struct page *page; > + u8 *bmap, byte, mask; > + > + page =3D hfs_bmap_get_map_page(node, &ctx, node_bit_idx / > BITS_PER_BYTE); > + if (IS_ERR(page)) > + return false; > + > + bmap =3D kmap_local_page(page); > + byte =3D bmap[ctx.off]; > + kunmap_local(bmap); > + > + mask =3D 1 << (7 - (node_bit_idx % BITS_PER_BYTE)); > + return (byte & mask) !=3D 0; > +} > + > +/** > + * hfs_bmap_clear_bit - clear a bit in the b-tree map > + * @node: the b-tree node containing the map record > + * @node_bit_idx: the relative bit index within the node's map > record > + * > + * Returns 0 on success, -EINVAL if already clear, or negative error > code. > + */ > +static int hfs_bmap_clear_bit(struct hfs_bnode *node, u32 > node_bit_idx) > +{ > + struct hfs_bmap_ctx ctx; > + struct page *page; > + u8 *bmap, mask; > + > + page =3D hfs_bmap_get_map_page(node, &ctx, node_bit_idx / > BITS_PER_BYTE); > + if (IS_ERR(page)) > + return PTR_ERR(page); > + > + bmap =3D kmap_local_page(page); > + > + mask =3D 1 << (7 - (node_bit_idx % BITS_PER_BYTE)); > + > + if (!(bmap[ctx.off] & mask)) { > + kunmap_local(bmap); > + return -EINVAL; > + } > + > + bmap[ctx.off] &=3D ~mask; > + set_page_dirty(page); > + kunmap_local(bmap); > + > + return 0; > +} > + > =C2=A0/* Get a reference to a B*Tree and do some initial checks */ > =C2=A0struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id, > btree_keycmp keycmp) > =C2=A0{ > @@ -23,6 +138,7 @@ struct hfs_btree *hfs_btree_open(struct > super_block *sb, u32 id, btree_keycmp ke > =C2=A0 struct address_space *mapping; > =C2=A0 struct folio *folio; > =C2=A0 struct buffer_head *bh; > + struct hfs_bnode *node; > =C2=A0 unsigned int size; > =C2=A0 u16 dblock; > =C2=A0 sector_t start_block; > @@ -155,6 +271,20 @@ struct hfs_btree *hfs_btree_open(struct > super_block *sb, u32 id, btree_keycmp ke > =C2=A0 kunmap_local(head); > =C2=A0 folio_unlock(folio); > =C2=A0 folio_put(folio); > + > + node =3D hfs_bnode_find(tree, HFS_TREE_HEAD); > + if (IS_ERR(node)) > + goto free_inode; > + > + if (!hfs_bmap_test_bit(node, HFS_TREE_HEAD)) { > + pr_warn("(%s): %s (cnid 0x%x) bitmap corrupted, > forcing rdonly\n", > + sb->s_id, id =3D=3D HFS_EXT_CNID ? "extents" : > "catalog", id); > + pr_warn("Run fsck.hfs to repair.\n"); > + sb->s_flags |=3D SB_RDONLY; > + } > + > + hfs_bnode_put(node); > + > =C2=A0 return tree; > =C2=A0 > =C2=A0fail_folio: > @@ -285,11 +415,9 @@ int hfs_bmap_reserve(struct hfs_btree *tree, u32 > rsvd_nodes) > =C2=A0struct hfs_bnode *hfs_bmap_alloc(struct hfs_btree *tree) > =C2=A0{ > =C2=A0 struct hfs_bnode *node, *next_node; > - struct page **pagep; > + struct hfs_bmap_ctx ctx; > + struct page *page; > =C2=A0 u32 nidx, idx; > - unsigned off; > - u16 off16; > - u16 len; > =C2=A0 u8 *data, byte, m; > =C2=A0 int i, res; > =C2=A0 > @@ -301,24 +429,26 @@ struct hfs_bnode *hfs_bmap_alloc(struct > hfs_btree *tree) > =C2=A0 node =3D hfs_bnode_find(tree, nidx); > =C2=A0 if (IS_ERR(node)) > =C2=A0 return node; > - len =3D hfs_brec_lenoff(node, 2, &off16); > - off =3D off16; > =C2=A0 > - off +=3D node->page_offset; > - pagep =3D node->page + (off >> PAGE_SHIFT); > - data =3D kmap_local_page(*pagep); > - off &=3D ~PAGE_MASK; > + page =3D hfs_bmap_get_map_page(node, &ctx, 0); > + if (IS_ERR(page)) { > + res =3D PTR_ERR(page); > + hfs_bnode_put(node); > + return ERR_PTR(res); > + } > + > + data =3D kmap_local_page(page); > =C2=A0 idx =3D 0; > =C2=A0 > =C2=A0 for (;;) { > - while (len) { > - byte =3D data[off]; > + while (ctx.len) { > + byte =3D data[ctx.off]; > =C2=A0 if (byte !=3D 0xff) { > =C2=A0 for (m =3D 0x80, i =3D 0; i < 8; m >>=3D > 1, i++) { > =C2=A0 if (!(byte & m)) { > =C2=A0 idx +=3D i; > - data[off] |=3D m; > - > set_page_dirty(*pagep); > + data[ctx.off] |=3D m; > + set_page_dirty(page) > ; > =C2=A0 kunmap_local(data); > =C2=A0 tree->free_nodes--; > =C2=A0 mark_inode_dirty(tre > e->inode); > @@ -327,13 +457,14 @@ struct hfs_bnode *hfs_bmap_alloc(struct > hfs_btree *tree) > =C2=A0 } > =C2=A0 } > =C2=A0 } > - if (++off >=3D PAGE_SIZE) { > + if (++ctx.off >=3D PAGE_SIZE) { > =C2=A0 kunmap_local(data); > - data =3D kmap_local_page(*++pagep); > - off =3D 0; > + page =3D node->page[++ctx.page_idx]; > + data =3D kmap_local_page(page); > + ctx.off =3D 0; > =C2=A0 } > =C2=A0 idx +=3D 8; > - len--; > + ctx.len--; > =C2=A0 } > =C2=A0 kunmap_local(data); > =C2=A0 nidx =3D node->next; > @@ -347,22 +478,22 @@ struct hfs_bnode *hfs_bmap_alloc(struct > hfs_btree *tree) > =C2=A0 return next_node; > =C2=A0 node =3D next_node; > =C2=A0 > - len =3D hfs_brec_lenoff(node, 0, &off16); > - off =3D off16; > - off +=3D node->page_offset; > - pagep =3D node->page + (off >> PAGE_SHIFT); > - data =3D kmap_local_page(*pagep); > - off &=3D ~PAGE_MASK; > + page =3D hfs_bmap_get_map_page(node, &ctx, 0); > + if (IS_ERR(page)) { > + res =3D PTR_ERR(page); > + hfs_bnode_put(node); > + return ERR_PTR(res); > + } > + data =3D kmap_local_page(page); > =C2=A0 } > =C2=A0} > =C2=A0 > =C2=A0void hfs_bmap_free(struct hfs_bnode *node) > =C2=A0{ > =C2=A0 struct hfs_btree *tree; > - struct page *page; > =C2=A0 u16 off, len; > =C2=A0 u32 nidx; > - u8 *data, byte, m; > + int res; > =C2=A0 > =C2=A0 hfs_dbg("node %u\n", node->this); > =C2=A0 tree =3D node->tree; > @@ -396,23 +527,18 @@ void hfs_bmap_free(struct hfs_bnode *node) > =C2=A0 } > =C2=A0 len =3D hfs_brec_lenoff(node, 0, &off); > =C2=A0 } > - off +=3D node->page_offset + nidx / 8; > - page =3D node->page[off >> PAGE_SHIFT]; > - data =3D kmap_local_page(page); > - off &=3D ~PAGE_MASK; > - m =3D 1 << (~nidx & 7); > - byte =3D data[off]; > - if (!(byte & m)) { > + > + res =3D hfs_bmap_clear_bit(node, nidx); > + if (res =3D=3D -EINVAL) { > =C2=A0 pr_crit("trying to free free bnode %u(%d)\n", > - node->this, node->type); > - kunmap_local(data); > - hfs_bnode_put(node); > - return; > + nidx, node->type); > + } else if (res) { > + pr_crit("fail to free bnode %u(%d)\n", > + nidx, node->type); > + } else { > + tree->free_nodes++; > + mark_inode_dirty(tree->inode); > =C2=A0 } > - data[off] =3D byte & ~m; > - set_page_dirty(page); > - kunmap_local(data); > + > =C2=A0 hfs_bnode_put(node); > - tree->free_nodes++; > - mark_inode_dirty(tree->inode); > =C2=A0} > diff --git a/fs/hfs/btree.h b/fs/hfs/btree.h > index 99be858b2446..b4c3f2a31471 100644 > --- a/fs/hfs/btree.h > +++ b/fs/hfs/btree.h > @@ -129,3 +129,43 @@ extern int __hfs_brec_find(struct hfs_bnode > *bnode, struct hfs_find_data *fd); > =C2=A0extern int hfs_brec_find(struct hfs_find_data *fd); > =C2=A0extern int hfs_brec_read(struct hfs_find_data *fd, void *rec, u32 > rec_len); > =C2=A0extern int hfs_brec_goto(struct hfs_find_data *fd, int cnt); > + > +static inline bool is_bnode_offset_valid(struct hfs_bnode *node, u32 > off) > +{ > + bool is_valid; > + > + if (!node || !node->tree) > + return false; > + > + is_valid =3D off < node->tree->node_size; > + > + if (!is_valid) { > + pr_err("invalid offset: id %u, type %#x, h %u, sz > %u, off %u\n", > + =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->this, node->type, node->hei= ght, > + =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->tree->node_size, off); > + } > + > + return is_valid; > +} > + > +static inline u32 check_and_correct_requested_length(struct > hfs_bnode *node, u32 off, u32 len) > +{ > + unsigned int node_size; > + > + if (!is_bnode_offset_valid(node, off)) > + return 0; > + > + node_size =3D node->tree->node_size; > + > + if ((u64)off + len > node_size) { > + u32 new_len =3D node_size - off; > + > + pr_err("corrected len: id %u, type %#x, h %u, sz %u, > off %u, len %u->%u\n", > + =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node->this, node->type, node->hei= ght, > + =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 node_size, off, len, new_len); > + > + return new_len; > + } > + > + return len; > +} > diff --git a/include/linux/hfs_common.h b/include/linux/hfs_common.h > index 45fb4c9ff9f5..d6a615e74b26 100644 > --- a/include/linux/hfs_common.h > +++ b/include/linux/hfs_common.h > @@ -510,14 +510,21 @@ struct hfs_btree_header_rec { > =C2=A0 =C2=A0=C2=A0 b-tree but not in > extents > =C2=A0 =C2=A0=C2=A0 b-tree (hfsplus). > */ > =C2=A0 > +/* HFS BTree misc info */ > +#define HFS_TREE_HEAD 0 > +#define HFS_BTREE_HDR_MAP_REC_INDEX 2 /* Map > (bitmap) record in Header node */ > +#define HFS_BTREE_MAP_NODE_REC_INDEX 0 /* Map > record in Map Node */ > + > =C2=A0/* HFS+ BTree misc info */ > -#define HFSPLUS_TREE_HEAD 0 > +#define HFSPLUS_TREE_HEAD HFS_TREE_HEAD > =C2=A0#define HFSPLUS_NODE_MXSZ 32768 > =C2=A0#define HFSPLUS_NODE_MINSZ 512 > =C2=A0#define HFSPLUS_ATTR_TREE_NODE_SIZE 8192 > =C2=A0#define HFSPLUS_BTREE_HDR_NODE_RECS_COUNT 3 > -#define HFSPLUS_BTREE_HDR_MAP_REC_INDEX 2 /* > Map (bitmap) record in Header node */ > -#define HFSPLUS_BTREE_MAP_NODE_REC_INDEX 0 /* Map > record in Map Node */ > +/* Map (bitmap) record in Header node */ > +#define > HFSPLUS_BTREE_HDR_MAP_REC_INDEX HFS_BTREE_HDR_MAP_REC_INDEX > +/* Map record in Map Node */ > +#define > HFSPLUS_BTREE_MAP_NODE_REC_INDEX HFS_BTREE_MAP_NODE_REC_INDEX > =C2=A0#define HFSPLUS_BTREE_HDR_USER_BYTES 128 > =C2=A0#define HFSPLUS_BTREE_MAP_NODE_RECS_COUNT 2 > =C2=A0#define HFSPLUS_BTREE_MAP_NODE_RESERVED_BYTES 2 Looks good. Thanks a lot for the patch. Reviewed-by: Viacheslav Dubeyko Thanks, Slava.