All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: sandeen@redhat.com
Cc: linux-xfs@vger.kernel.org
Subject: [PATCH v2 3/5] xfs_repair: strengthen geometry checks
Date: Mon, 23 Jan 2017 16:55:56 -0800	[thread overview]
Message-ID: <20170124005556.GH31202@birch.djwong.org> (raw)
In-Reply-To: <148494393485.5256.16822617236354659766.stgit@birch.djwong.org>

In xfs_repair, the inodelog, sectlog, and dirblklog values are read
directly into the xfs_mount structure without any sanity checking by the
verifier.  This results in xfs_repair segfaulting when those fields have
ridiculously high values because the pointer arithmetic runs us off the
end of the metadata buffers.  Therefore, reject the superblock if these
values are garbage and try to find one of the other ones.  Clean up the
dblocks checking to use the relevant macros.

The superblock field fuzzer (xfs/1301) triggers all these segfaults.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
v2: make the inode geometry more consistent with the kernel sb verifier,
and prevent an ASSERT if the fs is !dalign but sunit != 0.
v3: add a new failure message for directory geometry inconsistency and
complain if sunit/swidth are set and dalign is not.
---
 repair/globals.h    |    3 ++-
 repair/sb.c         |   27 +++++++++++++++++----------
 repair/xfs_repair.c |    2 ++
 3 files changed, 21 insertions(+), 11 deletions(-)

diff --git a/repair/globals.h b/repair/globals.h
index efd1d03..4085ccc 100644
--- a/repair/globals.h
+++ b/repair/globals.h
@@ -50,7 +50,8 @@
 #define XR_BAD_SB_WIDTH		18	/* bad stripe width */
 #define XR_BAD_SVN		19	/* bad shared version number */
 #define XR_BAD_CRC		20	/* Bad CRC */
-#define XR_BAD_ERR_CODE		21	/* Bad error code */
+#define XR_BAD_DIR_SIZE_DATA	21	/* Bad directory geometry */
+#define XR_BAD_ERR_CODE		22	/* Bad error code */
 
 /* XFS filesystem (il)legal values */
 
diff --git a/repair/sb.c b/repair/sb.c
index 004702c..77e5154 100644
--- a/repair/sb.c
+++ b/repair/sb.c
@@ -395,20 +395,22 @@ verify_sb(char *sb_buf, xfs_sb_t *sb, int is_primary_sb)
 	/* sanity check ag count, size fields against data size field */
 
 	if (sb->sb_dblocks == 0 ||
-		sb->sb_dblocks >
-			((__uint64_t)sb->sb_agcount * sb->sb_agblocks) ||
-		sb->sb_dblocks <
-			((__uint64_t)(sb->sb_agcount - 1) * sb->sb_agblocks
-			+ XFS_MIN_AG_BLOCKS))
+		sb->sb_dblocks > XFS_MAX_DBLOCKS(sb) ||
+		sb->sb_dblocks < XFS_MIN_DBLOCKS(sb))
 		return(XR_BAD_FS_SIZE_DATA);
 
 	if (sb->sb_agblklog != (__uint8_t)libxfs_log2_roundup(sb->sb_agblocks))
 		return(XR_BAD_FS_SIZE_DATA);
 
-	if (sb->sb_inodesize < XFS_DINODE_MIN_SIZE ||
-		sb->sb_inodesize > XFS_DINODE_MAX_SIZE ||
-		sb->sb_inopblock != howmany(sb->sb_blocksize,sb->sb_inodesize))
-		return(XR_BAD_INO_SIZE_DATA);
+	if (sb->sb_inodesize < XFS_DINODE_MIN_SIZE                     ||
+	    sb->sb_inodesize > XFS_DINODE_MAX_SIZE                     ||
+	    sb->sb_inodelog < XFS_DINODE_MIN_LOG                       ||
+	    sb->sb_inodelog > XFS_DINODE_MAX_LOG                       ||
+	    sb->sb_inodesize != (1 << sb->sb_inodelog)                 ||
+	    sb->sb_logsunit > XLOG_MAX_RECORD_BSIZE                    ||
+	    sb->sb_inopblock != howmany(sb->sb_blocksize, sb->sb_inodesize) ||
+	    (sb->sb_blocklog - sb->sb_inodelog != sb->sb_inopblog))
+		return XR_BAD_INO_SIZE_DATA;
 
 	if (xfs_sb_version_hassector(sb))  {
 
@@ -492,7 +494,12 @@ verify_sb(char *sb_buf, xfs_sb_t *sb, int is_primary_sb)
 		if ((sb->sb_unit && !sb->sb_width) ||
 		    (sb->sb_width && sb->sb_unit && sb->sb_width % sb->sb_unit))
 			return(XR_BAD_SB_WIDTH);
-	}
+	} else if (sb->sb_unit || sb->sb_width)
+		return XR_BAD_SB_WIDTH;
+
+	/* Directory block log */
+	if (sb->sb_blocklog + sb->sb_dirblklog > XFS_MAX_BLOCKSIZE_LOG)
+		return XR_BAD_DIR_SIZE_DATA;
 
 	return(XR_OK);
 }
diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
index 5c79fd9..b07567b 100644
--- a/repair/xfs_repair.c
+++ b/repair/xfs_repair.c
@@ -143,6 +143,8 @@ err_string(int err_code)
 			_("bad shared version number in superblock");
 		err_message[XR_BAD_CRC] =
 			_("bad CRC in superblock");
+		err_message[XR_BAD_DIR_SIZE_DATA] =
+			_("inconsistent directory geometry information");
 		done = 1;
 	}
 

  parent reply	other threads:[~2017-01-24  0:56 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-20 20:25 [PATCH 0/5] xfsprogs: miscellaneous cleanups Darrick J. Wong
2017-01-20 20:25 ` [PATCH 1/5] xfs_db: sanitize geometry on load Darrick J. Wong
2017-01-20 23:33   ` Eric Sandeen
2017-01-21  0:15   ` [PATCH v5 " Darrick J. Wong
2017-01-23 20:02     ` Eric Sandeen
2017-01-23 20:35       ` Darrick J. Wong
2017-01-23 21:30     ` Darrick J. Wong
2017-01-23 21:31   ` [PATCH v6 " Darrick J. Wong
2017-01-24 22:38     ` Eric Sandeen
2017-01-24 22:52     ` [PATCH v7 1/5] xfs_db: sanitize agcount " Eric Sandeen
2017-01-25  0:21       ` Darrick J. Wong
2017-01-25  0:55         ` Eric Sandeen
2017-01-25  3:09       ` [PATCH v8 " Eric Sandeen
2017-01-25  4:48         ` Darrick J. Wong
2017-01-26  1:05         ` [PATCH v9 " Eric Sandeen
2017-01-26  1:17           ` [PATCH v10 " Eric Sandeen
2017-01-26  1:27             ` Darrick J. Wong
2017-01-20 20:25 ` [PATCH 2/5] xfs_db: fix the 'source' command when passed as a -c option Darrick J. Wong
2017-01-23 22:29   ` Eric Sandeen
2017-01-23 23:39     ` Darrick J. Wong
2017-01-23 23:41   ` [PATCH v2 " Darrick J. Wong
2017-01-20 20:25 ` [PATCH 3/5] xfs_repair: strengthen geometry checks Darrick J. Wong
2017-01-23 23:47   ` Eric Sandeen
2017-01-24  0:13     ` Darrick J. Wong
2017-01-24  0:29       ` Eric Sandeen
2017-01-24  0:55   ` Darrick J. Wong [this message]
2017-01-20 20:25 ` [PATCH 4/5] xfs_repair: zero shared_vn Darrick J. Wong
2017-01-20 22:20   ` Eric Sandeen
2017-01-20 22:51     ` Darrick J. Wong
2017-01-20 22:52   ` [PATCH v2 " Darrick J. Wong
2017-01-20 23:08     ` Eric Sandeen
2017-01-21  0:08       ` Darrick J. Wong
2017-01-21  0:09   ` [PATCH v3 " Darrick J. Wong
2017-01-24  2:38     ` Eric Sandeen
2017-01-20 20:25 ` [PATCH 5/5] xfs_repair: trash dirattr btrees that cycle to the root Darrick J. Wong
2017-01-24  3:03   ` Eric Sandeen

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=20170124005556.GH31202@birch.djwong.org \
    --to=darrick.wong@oracle.com \
    --cc=linux-xfs@vger.kernel.org \
    --cc=sandeen@redhat.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.