linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <djwong@us.ibm.com>
To: Andreas Dilger <adilger.kernel@dilger.ca>,
	Theodore Tso <tytso@mit.edu>,
	"Darrick J. Wong" <djwong@us.ibm.com>
Cc: Sunil Mushran <sunil.mushran@oracle.com>,
	Amir Goldstein <amir73il@gmail.com>,
	Andi Kleen <andi@firstfloor.org>, Mingming Cao <cmm@us.ibm.com>,
	Joel Becker <jlbec@evilplan.org>,
	linux-ext4@vger.kernel.org, Coly Li <colyli@gmail.com>
Subject: [PATCH 44/47] e2fsck: Check revoke block checksum during recovery
Date: Sat, 08 Oct 2011 00:37:57 -0700	[thread overview]
Message-ID: <20111008073757.17888.62465.stgit@elm3c44.beaverton.ibm.com> (raw)
In-Reply-To: <20111008073315.17888.22132.stgit@elm3c44.beaverton.ibm.com>

Verify the revoke block checksum when recovering the journal.

Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
---
 e2fsck/recovery.c       |   37 +++++++++++++++++++++++++++++++++++--
 lib/ext2fs/kernel-jbd.h |    4 ++++
 2 files changed, 39 insertions(+), 2 deletions(-)


diff --git a/e2fsck/recovery.c b/e2fsck/recovery.c
index b669941..8979193 100644
--- a/e2fsck/recovery.c
+++ b/e2fsck/recovery.c
@@ -674,8 +674,17 @@ static int do_one_pass(journal_t *journal,
 			err = scan_revoke_records(journal, bh,
 						  next_commit_ID, info);
 			brelse(bh);
-			if (err)
-				goto failed;
+			if (err) {
+				if (err != -EINVAL)
+					goto failed;
+				/*
+				 * Ignoring corrupt revoke blocks is safe
+				 * because at worst it results in unnecessary
+				 * writes during recovery.
+				 */
+				jbd_debug(3, "Skipping corrupt revoke "
+					  "block.\n");
+			}
 			continue;
 
 		default:
@@ -715,6 +724,27 @@ static int do_one_pass(journal_t *journal,
 	return err;
 }
 
+static int jbd2_revoke_block_csum_verify(journal_t *j,
+					 void *buf)
+{
+	struct journal_revoke_tail *tail;
+	__u32 provided, calculated;
+
+	if (!JFS_HAS_INCOMPAT_FEATURE(j, JFS_FEATURE_INCOMPAT_CSUM_V2))
+		return 1;
+
+	tail = (struct journal_revoke_tail *)(buf + j->j_blocksize -
+			sizeof(struct journal_revoke_tail));
+	provided = tail->r_checksum;
+	tail->r_checksum = 0;
+	calculated = ext2fs_crc32c_le(~0, j->j_superblock->s_uuid,
+				      sizeof(j->j_superblock->s_uuid));
+	calculated = ext2fs_crc32c_le(calculated, buf, j->j_blocksize);
+	tail->r_checksum = provided;
+
+	provided = ext2fs_be32_to_cpu(provided);
+	return provided == calculated;
+}
 
 /* Scan a revoke record, marking all blocks mentioned as revoked. */
 
@@ -729,6 +759,9 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
 	offset = sizeof(journal_revoke_header_t);
 	max = be32_to_cpu(header->r_count);
 
+	if (!jbd2_revoke_block_csum_verify(journal, header))
+		return -EINVAL;
+
 	if (JFS_HAS_INCOMPAT_FEATURE(journal, JFS_FEATURE_INCOMPAT_64BIT))
 		record_len = 8;
 
diff --git a/lib/ext2fs/kernel-jbd.h b/lib/ext2fs/kernel-jbd.h
index 08479d3..187cb38 100644
--- a/lib/ext2fs/kernel-jbd.h
+++ b/lib/ext2fs/kernel-jbd.h
@@ -157,6 +157,10 @@ typedef struct journal_revoke_header_s
 	int		 r_count;	/* Count of bytes used in the block */
 } journal_revoke_header_t;
 
+/* Tail of revoke block, for checksumming */
+struct journal_revoke_tail {
+	__u32		r_checksum;
+};
 
 /* Definitions for the journal tag flags word: */
 #define JFS_FLAG_ESCAPE		1	/* on-disk block is escaped */


  parent reply	other threads:[~2011-10-08  7:39 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-08  7:33 [PATCH v2 00/47] e2fsprogs: Add metadata checksumming Darrick J. Wong
2011-10-08  7:33 ` [PATCH 01/47] libext2fs: Read and write full size inodes Darrick J. Wong
2011-10-08  7:33 ` [PATCH 02/47] libext2fs: Add metadata checksum flag Darrick J. Wong
2011-10-08  7:33 ` [PATCH 03/47] debugfs: Optionally ignore bad checksums Darrick J. Wong
2011-10-08  7:33 ` [PATCH 04/47] libext2fs: Add inode checksum support Darrick J. Wong
2011-10-08  7:33 ` [PATCH 05/47] debugfs: Dump inode checksum when appropriate Darrick J. Wong
2011-10-08  7:33 ` [PATCH 06/47] tune2fs: Add inode checksum support Darrick J. Wong
2011-10-08  7:34 ` [PATCH 07/47] e2fsck: Verify and correct inode checksums Darrick J. Wong
2011-10-08  7:34 ` [PATCH 08/47] mke2fs: Allow metadata checksums to be turned on at mkfs time Darrick J. Wong
2011-10-08  7:34 ` [PATCH 09/47] libext2fs: Create the inode bitmap checksum Darrick J. Wong
2011-10-08  7:34 ` [PATCH 10/47] tune2fs: Rewrite inode bitmap checksums Darrick J. Wong
2011-10-08  7:34 ` [PATCH 11/47] dumpe2fs: Display inode bitmap checksum Darrick J. Wong
2011-10-08  7:34 ` [PATCH 12/47] e2fsck: Verify " Darrick J. Wong
2011-10-08  7:34 ` [PATCH 13/47] libext2fs: Create the block " Darrick J. Wong
2011-10-08  7:34 ` [PATCH 14/47] dumpe2fs: Display " Darrick J. Wong
2011-10-08  7:34 ` [PATCH 15/47] e2fsck: Verify " Darrick J. Wong
2011-10-08  7:34 ` [PATCH 16/47] e2fsck: Don't verify bitmap checksums Darrick J. Wong
2011-10-08  7:35 ` [PATCH 17/47] tune2fs: Rewrite block " Darrick J. Wong
2011-10-08  7:35 ` [PATCH 18/47] libext2fs: Verify and calculate extent tree block checksums Darrick J. Wong
2011-10-08  7:35 ` [PATCH 19/47] tune2fs: Enable extent tree checksums Darrick J. Wong
2011-10-08  7:35 ` [PATCH 20/47] libext2fs: Introduce dx_tail and dir_entry_tail Darrick J. Wong
2011-10-08  7:35 ` [PATCH 21/47] debugfs: Print htree internal node checksums Darrick J. Wong
2011-10-08  7:35 ` [PATCH 22/47] libext2fs: Add dx_root/dx_node checksum calculation and verification helpers Darrick J. Wong
2011-10-08  7:35 ` [PATCH 23/47] e2fsck: Verify htree root/node checksums Darrick J. Wong
2011-10-08  7:35 ` [PATCH 24/47] libext2fs: Introduce dir_entry_tail to provide checksums for directory leaf nodes Darrick J. Wong
2011-10-08  7:35 ` [PATCH 25/47] e2fsck: Check directory leaf block checksums Darrick J. Wong
2011-10-08  7:36 ` [PATCH 26/47] tune2fs: Rebuild and checksum directories when toggling metadata_csum or changing UUID Darrick J. Wong
2011-10-08  7:36 ` [PATCH 27/47] libext2fs: Verify and calculate extended attribute block checksums Darrick J. Wong
2011-10-08  7:36 ` [PATCH 28/47] e2fsck: Check " Darrick J. Wong
2011-10-08  7:36 ` [PATCH 29/47] tune2fs: Rewrite " Darrick J. Wong
2011-10-08  7:36 ` [PATCH 30/47] libext2fs: Calculate and verify superblock checksums Darrick J. Wong
2011-10-08  7:36 ` [PATCH 31/47] e2fsck: Handle superblock checksum errors gracefully Darrick J. Wong
2011-10-08  7:36 ` [PATCH 32/47] libext2fs: Use i_generation in inode-related metadata checksums Darrick J. Wong
2011-10-08  7:36 ` [PATCH 33/47] libext2fs: Record the checksum algorithm in use in the superblock Darrick J. Wong
2011-10-08  7:36 ` [PATCH 34/47] tune2fs: Store checksum algorithm type in superblock Darrick J. Wong
2011-10-08  7:37 ` [PATCH 35/47] mke2fs: Record the checksum algorithm in use in the superblock Darrick J. Wong
2011-10-08  7:37 ` [PATCH 36/47] libext2fs: Block group checksum should use metadata_csum algorithm (if feature flag set) Darrick J. Wong
2011-10-08  7:37 ` [PATCH 37/47] tune2fs: Rewrite block group checksums when changing bg_use_meta_csum feature Darrick J. Wong
2011-10-08  7:37 ` [PATCH 38/47] mke2fs: Warn if not enabling all the features that metadata_csum wants Darrick J. Wong
2011-10-08  7:37 ` [PATCH 39/47] libext2fs: Add checksum to MMP block Darrick J. Wong
2011-10-08  7:37 ` [PATCH 40/47] e2fsck: Verify and correct MMP checksum problems Darrick J. Wong
2011-10-08  7:37 ` [PATCH 41/47] tune2fs: Force MMP update when changing metadata_csum flag Darrick J. Wong
2011-10-08  7:37 ` [PATCH 42/47] libext2fs: Add feature flags for jbd2 v2 checksums Darrick J. Wong
2011-10-08  7:37 ` [PATCH 43/47] e2fsck: Check journal superblock checksum prior to recovery Darrick J. Wong
2011-10-08  7:37 ` Darrick J. Wong [this message]
2011-10-08  7:38 ` [PATCH 45/47] e2fsck: Check descriptor block checksum when recovering journal Darrick J. Wong
2011-10-08  7:38 ` [PATCH 46/47] e2fsck: Check commit block checksum during recovery Darrick J. Wong
2011-10-08  7:38 ` [PATCH 47/47] e2fsck: Verify data block checksums when recovering journal 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=20111008073757.17888.62465.stgit@elm3c44.beaverton.ibm.com \
    --to=djwong@us.ibm.com \
    --cc=adilger.kernel@dilger.ca \
    --cc=amir73il@gmail.com \
    --cc=andi@firstfloor.org \
    --cc=cmm@us.ibm.com \
    --cc=colyli@gmail.com \
    --cc=jlbec@evilplan.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=sunil.mushran@oracle.com \
    --cc=tytso@mit.edu \
    /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).