From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: david@fromorbit.com, darrick.wong@oracle.com
Cc: linux-xfs@vger.kernel.org
Subject: [PATCH 01/55] xfs: forbid AG btrees with level == 0
Date: Fri, 02 Dec 2016 17:35:21 -0800 [thread overview]
Message-ID: <148072892114.12995.3258531215811110809.stgit@birch.djwong.org> (raw)
In-Reply-To: <148072891404.12995.15510849192837089093.stgit@birch.djwong.org>
There is no such thing as a zero-level AG btree since even a single-node
zero-records btree has one level. Btree cursor constructors read
cur_nlevels straight from disk and then access things like
cur_bufs[cur_nlevels - 1] which is /really/ bad if cur_nlevels is zero!
Therefore, strengthen the verifiers to prevent this possibility.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
fs/xfs/libxfs/xfs_alloc.c | 10 +++++++---
fs/xfs/libxfs/xfs_ialloc.c | 9 ++++++++-
2 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index effb64c..5050056 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2455,12 +2455,15 @@ xfs_agf_verify(
be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp)))
return false;
- if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > XFS_BTREE_MAXLEVELS ||
+ if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) < 1 ||
+ be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) < 1 ||
+ be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > XFS_BTREE_MAXLEVELS ||
be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) > XFS_BTREE_MAXLEVELS)
return false;
if (xfs_sb_version_hasrmapbt(&mp->m_sb) &&
- be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) > XFS_BTREE_MAXLEVELS)
+ (be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) < 1 ||
+ be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]) > XFS_BTREE_MAXLEVELS))
return false;
/*
@@ -2477,7 +2480,8 @@ xfs_agf_verify(
return false;
if (xfs_sb_version_hasreflink(&mp->m_sb) &&
- be32_to_cpu(agf->agf_refcount_level) > XFS_BTREE_MAXLEVELS)
+ (be32_to_cpu(agf->agf_refcount_level) < 1 ||
+ be32_to_cpu(agf->agf_refcount_level) > XFS_BTREE_MAXLEVELS))
return false;
return true;;
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index c507c1b..30967c8 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -2513,8 +2513,15 @@ xfs_agi_verify(
if (!XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)))
return false;
- if (be32_to_cpu(agi->agi_level) > XFS_BTREE_MAXLEVELS)
+ if (be32_to_cpu(agi->agi_level) < 1 ||
+ be32_to_cpu(agi->agi_level) > XFS_BTREE_MAXLEVELS)
return false;
+
+ if (xfs_sb_version_hasfinobt(&mp->m_sb) &&
+ (be32_to_cpu(agi->agi_free_level) < 1 ||
+ be32_to_cpu(agi->agi_free_level) > XFS_BTREE_MAXLEVELS))
+ return false;
+
/*
* during growfs operations, the perag is not fully initialised,
* so we can't use it for any useful checking. growfs ensures we can't
next prev parent reply other threads:[~2016-12-03 1:35 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-03 1:35 [PATCH v3 00/55] xfs: online scrub/repair support Darrick J. Wong
2016-12-03 1:35 ` Darrick J. Wong [this message]
2016-12-03 1:35 ` [PATCH 02/55] xfs: check for bogus values in btree block headers Darrick J. Wong
2016-12-03 1:35 ` [PATCH 03/55] xfs: complain if we don't get nextents bmap records Darrick J. Wong
2016-12-05 1:35 ` Dave Chinner
2016-12-03 1:35 ` [PATCH 04/55] xfs: don't crash if reading a directory results in an unexpected hole Darrick J. Wong
2016-12-03 1:35 ` [PATCH 05/55] xfs: error out if trying to add attrs and anextents > 0 Darrick J. Wong
2016-12-03 1:35 ` [PATCH 06/55] xfs: don't allow di_size with high bit set Darrick J. Wong
2016-12-03 1:35 ` [PATCH 07/55] xfs: don't cap maximum dedupe request length Darrick J. Wong
2016-12-03 1:36 ` [PATCH 08/55] xfs: plumb in needed functions for range querying of the freespace btrees Darrick J. Wong
2016-12-03 1:36 ` [PATCH 09/55] xfs: provide a query_range function for " Darrick J. Wong
2016-12-03 1:36 ` [PATCH 10/55] xfs: create a function to query all records in a btree Darrick J. Wong
2016-12-03 1:36 ` [PATCH 11/55] xfs: introduce the XFS_IOC_GETFSMAP ioctl Darrick J. Wong
2016-12-03 1:36 ` [PATCH 12/55] xfs: report shared extents in getfsmapx Darrick J. Wong
2016-12-03 1:36 ` [PATCH 13/55] xfs: have getfsmap fall back to the freesp btrees when rmap is not present Darrick J. Wong
2016-12-03 1:36 ` [PATCH 14/55] xfs: getfsmap should fall back to rtbitmap when rtrmapbt " Darrick J. Wong
2016-12-03 1:36 ` [PATCH 15/55] xfs: use GPF_NOFS when allocating btree cursors Darrick J. Wong
2016-12-03 1:36 ` [PATCH 16/55] xfs: add scrub tracepoints Darrick J. Wong
2016-12-03 1:37 ` [PATCH 17/55] xfs: create an ioctl to scrub AG metadata Darrick J. Wong
2016-12-03 1:37 ` [PATCH 18/55] xfs: generic functions to scrub metadata and btrees Darrick J. Wong
2016-12-03 1:37 ` [PATCH 19/55] xfs: scrub the backup superblocks Darrick J. Wong
2016-12-03 1:37 ` [PATCH 20/55] xfs: scrub AGF and AGFL Darrick J. Wong
2016-12-03 1:37 ` [PATCH 21/55] xfs: scrub the AGI Darrick J. Wong
2016-12-03 1:37 ` [PATCH 22/55] xfs: support scrubbing free space btrees Darrick J. Wong
2016-12-03 1:37 ` [PATCH 23/55] xfs: support scrubbing inode btrees Darrick J. Wong
2016-12-03 1:37 ` [PATCH 24/55] xfs: support scrubbing rmap btree Darrick J. Wong
2016-12-03 1:37 ` [PATCH 25/55] xfs: support scrubbing refcount btree Darrick J. Wong
2016-12-03 1:38 ` [PATCH 26/55] xfs: scrub inodes Darrick J. Wong
2016-12-03 1:38 ` [PATCH 27/55] xfs: scrub inode block mappings Darrick J. Wong
2016-12-03 1:38 ` [PATCH 28/55] xfs: scrub directory/attribute btrees Darrick J. Wong
2016-12-03 1:38 ` [PATCH 29/55] xfs: scrub directory metadata Darrick J. Wong
2016-12-03 1:38 ` [PATCH 30/55] xfs: scrub extended attributes Darrick J. Wong
2016-12-03 1:38 ` [PATCH 31/55] xfs: scrub symbolic links Darrick J. Wong
2016-12-03 1:38 ` [PATCH 32/55] xfs: scrub realtime bitmap/summary Darrick J. Wong
2016-12-03 1:38 ` [PATCH 33/55] xfs: scrub should cross-reference with the bnobt Darrick J. Wong
2016-12-03 1:38 ` [PATCH 34/55] xfs: cross-reference bnobt records with cntbt Darrick J. Wong
2016-12-03 1:39 ` [PATCH 35/55] xfs: cross-reference extents with AG header Darrick J. Wong
2016-12-03 1:39 ` [PATCH 36/55] xfs: cross-reference inode btrees during scrub Darrick J. Wong
2016-12-03 1:39 ` [PATCH 37/55] xfs: cross-reference reverse-mapping btree Darrick J. Wong
2016-12-03 1:39 ` [PATCH 38/55] xfs: cross-reference refcount btree during scrub Darrick J. Wong
2016-12-03 1:39 ` [PATCH 39/55] xfs: scrub should cross-reference the realtime bitmap Darrick J. Wong
2016-12-03 1:39 ` [PATCH 40/55] xfs: cross-reference the block mappings when possible Darrick J. Wong
2016-12-03 1:39 ` [PATCH 41/55] xfs: create tracepoints for online repair Darrick J. Wong
2016-12-03 1:39 ` [PATCH 42/55] xfs: implement the metadata repair ioctl flag Darrick J. Wong
2016-12-03 1:40 ` [PATCH 43/55] xfs: add helper routines for the repair code Darrick J. Wong
2016-12-03 1:40 ` [PATCH 44/55] xfs: repair superblocks Darrick J. Wong
2016-12-03 1:40 ` [PATCH 45/55] xfs: repair the AGF and AGFL Darrick J. Wong
2016-12-03 1:40 ` [PATCH 46/55] xfs: rebuild the AGI Darrick J. Wong
2016-12-03 1:40 ` [PATCH 47/55] xfs: repair free space btrees Darrick J. Wong
2016-12-03 1:40 ` [PATCH 48/55] xfs: repair inode btrees Darrick J. Wong
2016-12-03 1:40 ` [PATCH 49/55] xfs: rebuild the rmapbt Darrick J. Wong
2016-12-03 1:40 ` [PATCH 50/55] xfs: repair refcount btrees Darrick J. Wong
2016-12-03 1:40 ` [PATCH 51/55] xfs: online repair of inodes Darrick J. Wong
2016-12-03 1:40 ` [PATCH 52/55] xfs: repair inode block maps Darrick J. Wong
2016-12-03 1:41 ` [PATCH 53/55] xfs: repair damaged symlinks Darrick J. Wong
2016-12-03 1:41 ` [PATCH 54/55] xfs: query the per-AG reservation counters Darrick J. Wong
2016-12-03 1:41 ` [PATCH 55/55] xfs: avoid mount-time deadlock in CoW extent recovery Darrick J. Wong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=148072892114.12995.3258531215811110809.stgit@birch.djwong.org \
--to=darrick.wong@oracle.com \
--cc=david@fromorbit.com \
--cc=linux-xfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).