Updated patches for journal checksums.

Updated patches for journal checksums.

[Girish Shilamkar]

[-- Attachment #1: Type: text/plain, Size: 613 bytes --]

Hi,
	The journal checksums patches for ext4 and e2fsprogs sent on 30th May
had run into problems on big-endian machines. I have made the required
changes. 
The major updates are:
1. Checksum was written in little endian format in the commit header,
changed it to big-endian.
2. The crc32 code ported to userspace (which is used by e2fsprogs) had
problem with #ifdef, fixed that.
3. The crc32 user space code used i386 assembly code for swabbing
changed it to generic one.
4. Removed various #includes, which caused compilation errors on ppc
machine.

Any comments/suggestions are always welcome.

Thanks,
Girish.

[-- Attachment #2: Type: text/x-patch, Size: 19252 bytes --]

Index: linux-2.6.20/fs/jbd2/commit.c
===================================================================
--- linux-2.6.20.orig/fs/jbd2/commit.c
+++ linux-2.6.20/fs/jbd2/commit.c
@@ -21,6 +21,7 @@
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/smp_lock.h>
+#include <linux/crc32.h>
 
 /*
  * Default IO end handler for temporary BJ_IO buffer_heads.
@@ -93,15 +94,18 @@ static int inverted_lock(journal_t *jour
 	return 1;
 }
 
-/* Done it all: now write the commit record.  We should have
+/*
+ * Done it all: now submit the commit record.  We should have
  * cleaned up our previous buffers by now, so if we are in abort
  * mode we can now just skip the rest of the journal write
  * entirely.
  *
  * Returns 1 if the journal needs to be aborted or 0 on success
  */
-static int journal_write_commit_record(journal_t *journal,
-					transaction_t *commit_transaction)
+static int journal_submit_commit_record(journal_t *journal,
+					transaction_t *commit_transaction,
+					struct buffer_head **cbh,
+					__u32 crc32_sum)
 {
 	struct journal_head *descriptor;
 	struct buffer_head *bh;
@@ -117,21 +121,36 @@ static int journal_write_commit_record(j
 
 	bh = jh2bh(descriptor);
 
-	/* AKPM: buglet - add `i' to tmp! */
 	for (i = 0; i < bh->b_size; i += 512) {
-		journal_header_t *tmp = (journal_header_t*)bh->b_data;
+ 		struct commit_header *tmp = (struct commit_header*)bh->b_data +
+					    i;
 		tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER);
 		tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK);
 		tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid);
-	}
+		 		
+ 		if (JBD2_HAS_COMPAT_FEATURE(journal,
+					    JBD2_FEATURE_COMPAT_CHECKSUM)) {
+ 			tmp->h_chksum_type 	= JBD2_CRC32_CHKSUM;
+ 			tmp->h_chksum_size 	= JBD2_CRC32_CHKSUM_SIZE;
+ 			tmp->h_chksum[0] 	= crc32_sum;
+ 		}
+  	}
+
+ 	JBUFFER_TRACE(descriptor, "submit commit block");
+ 	lock_buffer(bh);
 
-	JBUFFER_TRACE(descriptor, "write commit block");
 	set_buffer_dirty(bh);
-	if (journal->j_flags & JBD2_BARRIER) {
+ 	set_buffer_uptodate(bh);
+ 	bh->b_end_io = journal_end_buffer_io_sync;
+
+ 	if (journal->j_flags & JBD2_BARRIER &&
+ 		!JBD2_HAS_COMPAT_FEATURE(journal,
+					 JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
 		set_buffer_ordered(bh);
 		barrier_done = 1;
-	}
-	ret = sync_dirty_buffer(bh);
+ 	}	
+ 	ret = submit_bh(WRITE, bh);
+
 	/* is it possible for another commit to fail at roughly
 	 * the same time as this one?  If so, we don't want to
 	 * trust the barrier flag in the super, but instead want
@@ -152,14 +171,72 @@ static int journal_write_commit_record(j
 		clear_buffer_ordered(bh);
 		set_buffer_uptodate(bh);
 		set_buffer_dirty(bh);
-		ret = sync_dirty_buffer(bh);
+		ret = submit_bh(WRITE, bh);
 	}
-	put_bh(bh);		/* One for getblk() */
-	jbd2_journal_put_journal_head(descriptor);
+	*cbh = bh;
+	return ret;
+}
+
+/*
+ * This function along with journal_submit_commit_record
+ * allows to write the commit record asynchronously.
+ */
+static int journal_wait_on_commit_record(struct buffer_head *bh)
+{
+	int ret = 0;
 
-	return (ret == -EIO);
+	if (buffer_locked(bh))
+		wait_on_buffer(bh);
+	
+	if (unlikely(!buffer_uptodate(bh)))
+		ret = -EIO;
+	put_bh(bh);            /* One for getblk() */
+	jbd2_journal_put_journal_head(bh2jh(bh));
+	
+	return ret;
 }
 
+/*
+ * Wait for all submitted IO to complete.
+ */
+static int journal_wait_on_locked_list(journal_t *journal,
+				       transaction_t *commit_transaction)
+{
+	int ret = 0;
+	struct journal_head *jh;
+
+	while (commit_transaction->t_locked_list) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_locked_list->b_tprev;
+		bh = jh2bh(jh);
+		get_bh(bh);
+		if (buffer_locked(bh)) {
+			spin_unlock(&journal->j_list_lock);
+			wait_on_buffer(bh);
+			if (unlikely(!buffer_uptodate(bh)))
+				ret = -EIO;
+			spin_lock(&journal->j_list_lock);
+		}
+		if (!inverted_lock(journal, bh)) {
+			put_bh(bh);
+			spin_lock(&journal->j_list_lock);
+			continue;
+		}
+		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
+			__jbd2_journal_unfile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+			jbd2_journal_remove_journal_head(bh);
+			put_bh(bh);
+		} else {
+			jbd_unlock_bh_state(bh);
+		}
+		put_bh(bh);
+		cond_resched_lock(&journal->j_list_lock);
+	}
+	return ret;
+  }
+
 static void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
 {
 	int i;
@@ -306,6 +383,8 @@ void jbd2_journal_commit_transaction(jou
 	int tag_flag;
 	int i;
 	int tag_bytes = journal_tag_bytes(journal);
+	struct buffer_head *cbh = NULL; /* For transactional checksums */
+	__u32 crc32_sum = ~0;
 
 	/*
 	 * First job: lock down the current transaction and wait for
@@ -441,38 +520,15 @@ void jbd2_journal_commit_transaction(jou
 	journal_submit_data_buffers(journal, commit_transaction);
 
 	/*
-	 * Wait for all previously submitted IO to complete.
+	 * Wait for all previously submitted IO to complete if commit
+	 * record is to be written synchronously.
 	 */
 	spin_lock(&journal->j_list_lock);
-	while (commit_transaction->t_locked_list) {
-		struct buffer_head *bh;
+	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT))
+		err = journal_wait_on_locked_list(journal,
+						commit_transaction);
 
-		jh = commit_transaction->t_locked_list->b_tprev;
-		bh = jh2bh(jh);
-		get_bh(bh);
-		if (buffer_locked(bh)) {
-			spin_unlock(&journal->j_list_lock);
-			wait_on_buffer(bh);
-			if (unlikely(!buffer_uptodate(bh)))
-				err = -EIO;
-			spin_lock(&journal->j_list_lock);
-		}
-		if (!inverted_lock(journal, bh)) {
-			put_bh(bh);
-			spin_lock(&journal->j_list_lock);
-			continue;
-		}
-		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
-			__jbd2_journal_unfile_buffer(jh);
-			jbd_unlock_bh_state(bh);
-			jbd2_journal_remove_journal_head(bh);
-			put_bh(bh);
-		} else {
-			jbd_unlock_bh_state(bh);
-		}
-		put_bh(bh);
-		cond_resched_lock(&journal->j_list_lock);
-	}
 	spin_unlock(&journal->j_list_lock);
 
 	if (err)
@@ -640,6 +696,16 @@ void jbd2_journal_commit_transaction(jou
 start_journal_io:
 			for (i = 0; i < bufs; i++) {
 				struct buffer_head *bh = wbuf[i];
+				/*
+				 * Compute checksum.
+				 */
+				if (JBD2_HAS_COMPAT_FEATURE(journal,
+					JBD2_FEATURE_COMPAT_CHECKSUM)) {
+					crc32_sum = crc32_be(crc32_sum,
+							(void *)bh->b_data,
+							bh->b_size);
+				}
+
 				lock_buffer(bh);
 				clear_buffer_dirty(bh);
 				set_buffer_uptodate(bh);
@@ -655,6 +721,23 @@ start_journal_io:
 		}
 	}
 
+	/* Done it all: now write the commit record asynchronously. */
+
+	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
+		err = journal_submit_commit_record(journal, commit_transaction,
+						 &cbh, crc32_sum);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+
+		spin_lock(&journal->j_list_lock);
+		err = journal_wait_on_locked_list(journal,
+						commit_transaction);
+		spin_unlock(&journal->j_list_lock);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+	}
+
 	/* Lo and behold: we have just managed to send a transaction to
            the log.  Before we can commit it, wait for the IO so far to
            complete.  Control buffers being written are on the
@@ -753,9 +836,15 @@ wait_for_iobuf:
 	}
 
 	jbd_debug(3, "JBD: commit phase 6\n");
-
-	if (journal_write_commit_record(journal, commit_transaction))
-		err = -EIO;
+	 	
+	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
+		JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
+		err = journal_submit_commit_record(journal, commit_transaction,
+						&cbh, crc32_sum);
+		if (err)
+			__jbd2_journal_abort_hard(journal);
+	}
+	err = journal_wait_on_commit_record(cbh);
 
 	if (err)
 		__jbd2_journal_abort_hard(journal);
Index: linux-2.6.20/include/linux/jbd2.h
===================================================================
--- linux-2.6.20.orig/include/linux/jbd2.h
+++ linux-2.6.20/include/linux/jbd2.h
@@ -148,6 +148,29 @@ typedef struct journal_header_s
 	__be32		h_sequence;
 } journal_header_t;
 
+/*
+ * Checksum types.
+ */
+#define JBD2_CRC32_CHKSUM   1
+#define JBD2_MD5_CHKSUM     2
+#define JBD2_SHA1_CHKSUM    3
+
+#define JBD2_CRC32_CHKSUM_SIZE 4
+
+#define JBD2_CHECKSUM_BYTES (32 / sizeof(u32))
+/*
+ * Commit block header for storing transactional checksums:
+ */
+struct commit_header
+{
+	__be32		h_magic;
+	__be32          h_blocktype;
+	__be32          h_sequence;
+	unsigned char   h_chksum_type;
+	unsigned char   h_chksum_size;
+	unsigned char 	h_padding[2];
+	__u32 		h_chksum[JBD2_CHECKSUM_BYTES];
+};
 
 /*
  * The block tag: used to describe a single buffer in the journal.
@@ -241,14 +264,18 @@ typedef struct journal_superblock_s
 	((j)->j_format_version >= 2 &&					\
 	 ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask))))
 
-#define JBD2_FEATURE_INCOMPAT_REVOKE	0x00000001
-#define JBD2_FEATURE_INCOMPAT_64BIT	0x00000002
+#define JBD2_FEATURE_COMPAT_CHECKSUM	0x00000001
+
+#define JBD2_FEATURE_INCOMPAT_REVOKE		0x00000001
+#define JBD2_FEATURE_INCOMPAT_64BIT		0x00000002
+#define JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT	0x00000004
 
 /* Features known to this kernel version: */
-#define JBD2_KNOWN_COMPAT_FEATURES	0
+#define JBD2_KNOWN_COMPAT_FEATURES	JBD2_FEATURE_COMPAT_CHECKSUM
 #define JBD2_KNOWN_ROCOMPAT_FEATURES	0
 #define JBD2_KNOWN_INCOMPAT_FEATURES	(JBD2_FEATURE_INCOMPAT_REVOKE | \
-					 JBD2_FEATURE_INCOMPAT_64BIT)
+					JBD2_FEATURE_INCOMPAT_64BIT | \
+ 					JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)
 
 #ifdef __KERNEL__
 
@@ -931,6 +958,8 @@ extern int	   jbd2_journal_check_availab
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   jbd2_journal_set_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
+extern int	   jbd2_journal_clear_features
+		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   jbd2_journal_create     (journal_t *);
 extern int	   jbd2_journal_load       (journal_t *journal);
 extern void	   jbd2_journal_destroy    (journal_t *);
Index: linux-2.6.20/fs/ext4/super.c
===================================================================
--- linux-2.6.20.orig/fs/ext4/super.c
+++ linux-2.6.20/fs/ext4/super.c
@@ -724,6 +724,7 @@ enum {
 	Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl,
 	Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_bh,
 	Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
+	Opt_journal_checksum, Opt_journal_async_commit,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
@@ -763,6 +764,8 @@ static match_table_t tokens = {
 	{Opt_journal_update, "journal=update"},
 	{Opt_journal_inum, "journal=%u"},
 	{Opt_journal_dev, "journal_dev=%u"},
+	{Opt_journal_checksum, "journal_checksum"},
+	{Opt_journal_async_commit, "journal_async_commit"},
 	{Opt_abort, "abort"},
 	{Opt_data_journal, "data=journal"},
 	{Opt_data_ordered, "data=ordered"},
@@ -949,6 +952,13 @@ static int parse_options (char *options,
 				return 0;
 			*journal_devnum = option;
 			break;
+		case Opt_journal_checksum:
+			set_opt (sbi->s_mount_opt, JOURNAL_CHECKSUM);
+			break;
+		case Opt_journal_async_commit:
+			set_opt (sbi->s_mount_opt, JOURNAL_ASYNC_COMMIT);
+			set_opt (sbi->s_mount_opt, JOURNAL_CHECKSUM);
+			break;
 		case Opt_noload:
 			set_opt (sbi->s_mount_opt, NOLOAD);
 			break;
@@ -1793,6 +1803,21 @@ static int ext4_fill_super (struct super
 		goto failed_mount3;
 	}
 
+	if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
+		jbd2_journal_set_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	} else if (test_opt(sb, JOURNAL_CHECKSUM)) {
+		jbd2_journal_set_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0);
+		jbd2_journal_clear_features(sbi->s_journal, 0, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	} else {
+		jbd2_journal_clear_features(sbi->s_journal,
+				JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+				JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+	}
+
 	/* We have now updated the journal if required, so we can
 	 * validate the data journaling mode. */
 	switch (test_opt(sb, DATA_FLAGS)) {
Index: linux-2.6.20/include/linux/ext4_fs.h
===================================================================
--- linux-2.6.20.orig/include/linux/ext4_fs.h
+++ linux-2.6.20/include/linux/ext4_fs.h
@@ -400,6 +400,8 @@ struct ext4_inode {
 #define EXT4_MOUNT_USRQUOTA		0x100000 /* "old" user quota */
 #define EXT4_MOUNT_GRPQUOTA		0x200000 /* "old" group quota */
 #define EXT4_MOUNT_EXTENTS		0x400000 /* Extents support */
+#define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
+#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
 
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
Index: linux-2.6.20/fs/jbd2/recovery.c
===================================================================
--- linux-2.6.20.orig/fs/jbd2/recovery.c
+++ linux-2.6.20/fs/jbd2/recovery.c
@@ -21,6 +21,7 @@
 #include <linux/jbd2.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
+#include <linux/crc32.h>
 #endif
 
 /*
@@ -316,6 +317,37 @@ static inline unsigned long long read_ta
 	return block;
 }
 
+/* 
+ * cal_chksums calculates the checksums for the blocks described in the
+ * descriptor block.
+ */
+static int cal_chksums(journal_t *journal, struct buffer_head *bh,
+		       unsigned long *next_log_block, __u32 *crc32_sum)
+{
+	int i, num_blks, err;
+	unsigned long io_block;
+	struct buffer_head *obh;
+
+	num_blks = count_tags(journal, bh);
+	/* Calculate checksum of the descriptor block. */
+	*crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size);
+
+	for (i = 0; i < num_blks; i++) {
+		io_block = (*next_log_block)++;
+		wrap(journal, *next_log_block);
+		err = jread(&obh, journal, io_block);
+		if (err) {
+			printk(KERN_ERR "JBD: IO error %d recovering block "
+				"%ld in log\n", err, io_block);
+			return 1;
+		} else {
+			*crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data,
+				     obh->b_size);
+		}
+	}
+	return 0;
+}
+
 static int do_one_pass(journal_t *journal,
 			struct recovery_info *info, enum passtype pass)
 {
@@ -328,6 +360,7 @@ static int do_one_pass(journal_t *journa
 	unsigned int		sequence;
 	int			blocktype;
 	int			tag_bytes = journal_tag_bytes(journal);
+	__u32			crc32_sum = ~0; /* Transactional Checksums */
 
 	/* Precompute the maximum metadata descriptors in a descriptor block */
 	int			MAX_BLOCKS_PER_DESC;
@@ -419,9 +452,23 @@ static int do_one_pass(journal_t *journa
 		switch(blocktype) {
 		case JBD2_DESCRIPTOR_BLOCK:
 			/* If it is a valid descriptor block, replay it
-			 * in pass REPLAY; otherwise, just skip over the
-			 * blocks it describes. */
+			 * in pass REPLAY; if journal_checksums enabled, then
+			 * calculate checksums in PASS_SCAN, otherwise,
+			 * just skip over the blocks it describes. */
 			if (pass != PASS_REPLAY) {
+				if (pass == PASS_SCAN &&
+				    JBD2_HAS_COMPAT_FEATURE(journal,
+					    JBD2_FEATURE_COMPAT_CHECKSUM) &&
+				    !info->end_transaction) {
+					if (cal_chksums(journal, bh,
+							&next_log_block,
+							&crc32_sum)) {
+						brelse(bh);
+       						break;
+					}
+					brelse(bh);
+					continue;
+				}
 				next_log_block += count_tags(journal, bh);
 				wrap(journal, next_log_block);
 				brelse(bh);
@@ -516,9 +563,71 @@ static int do_one_pass(journal_t *journa
 			continue;
 
 		case JBD2_COMMIT_BLOCK:
-			/* Found an expected commit block: not much to
-			 * do other than move on to the next sequence
+			/*     How to differentiate between interrupted commit
+			 *               and journal corruption ?
+			 *
+			 * {nth transaction}
+			 *        Checksum Verification Failed
+			 *			 |
+			 *		 ____________________
+			 *		|		     |
+			 * 	async_commit             sync_commit
+			 *     		|                    |
+			 *		| GO TO NEXT    "Journal Corruption"
+			 *		| TRANSACTION
+			 *		|
+			 * {(n+1)th transanction}
+		         *		|
+			 * 	 _______|______________
+			 * 	|	 	      |
+			 * Commit block found	Commit block not found
+		         *      |		      |
+		         * "Journal Corruption"       |
+		      	 *		 _____________|__________
+       			 *     		|	           	|
+		     	 *	nth trans corrupt	OR   nth trans
+		 	 *	and (n+1)th interrupted     interrupted	
+			 *	before commit block
+			 *      could reach the disk.
+			 *	(Cannot find the difference in above
+			 *	 mentioned conditions. Hence assume
+			 *	 "Interrupted Commit".)
+			 */
+
+			/* Found an expected commit block: if checksums
+			 * are present verify them in PASS_SCAN; else not
+			 * much to do other than move on to the next sequence
 			 * number. */
+			if (pass == PASS_SCAN &&
+			    JBD2_HAS_COMPAT_FEATURE(journal,
+				    JBD2_FEATURE_COMPAT_CHECKSUM)) {
+				struct commit_header *cbh =
+	  				(struct commit_header *)bh->b_data;
+
+				if (info->end_transaction) {
+					printk(KERN_ERR "JBD: Transaction %u "
+						"found to be corrupt.\n",
+						next_commit_ID - 1);
+					brelse(bh);
+					break;
+				}
+
+				if (crc32_sum != cbh->h_chksum[0]) {
+					info->end_transaction = next_commit_ID;
+
+					if (!JBD2_HAS_COMPAT_FEATURE(journal,
+					    JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT))
+					{
+						printk(KERN_ERR
+						       "JBD: Transaction %u "
+       						       "found to be corrupt.\n",
+       						       next_commit_ID);
+						brelse(bh);
+						break;
+					}
+				}
+				crc32_sum = ~0;
+			}
 			brelse(bh);
 			next_commit_ID++;
 			continue;
@@ -554,9 +663,10 @@ static int do_one_pass(journal_t *journa
 	 * transaction marks the end of the valid log.
 	 */
 
-	if (pass == PASS_SCAN)
-		info->end_transaction = next_commit_ID;
-	else {
+	if (pass == PASS_SCAN) {
+ 		if (!info->end_transaction)
+       		       info->end_transaction = next_commit_ID;
+	} else {
 		/* It's really bad news if different passes end up at
 		 * different places (but possible due to IO errors). */
 		if (info->end_transaction != next_commit_ID) {
Index: linux-2.6.20/fs/jbd2/journal.c
===================================================================
--- linux-2.6.20.orig/fs/jbd2/journal.c
+++ linux-2.6.20/fs/jbd2/journal.c
@@ -1268,6 +1268,33 @@ int jbd2_journal_set_features (journal_t
 	return 1;
 }
 
+/**
+ * int jbd2_journal_clear_features () - Clear a given journal feature in the superblock
+ * @journal: Journal to act on.
+ * @compat: bitmask of compatible features
+ * @ro: bitmask of features that force read-only mount
+ * @incompat: bitmask of incompatible features
+ *
+ * Clear a given journal feature as present on the
+ * superblock.  Returns true if the requested features could be reset.
+ *
+ */
+int jbd2_journal_clear_features (journal_t *journal, unsigned long compat,
+			  unsigned long ro, unsigned long incompat)
+{
+	journal_superblock_t *sb;
+
+	jbd_debug(1, "Clear features 0x%lx/0x%lx/0x%lx\n",
+		  compat, ro, incompat);
+
+	sb = journal->j_superblock;
+
+	sb->s_feature_compat    &= ~cpu_to_be32(compat);
+	sb->s_feature_ro_compat &= ~cpu_to_be32(ro);
+	sb->s_feature_incompat  &= ~cpu_to_be32(incompat);
+
+	return 1;
+}
 
 /**
  * int jbd2_journal_update_format () - Update on-disk journal structure.

[-- Attachment #3: e2fsprogs-journal_chksum.patch --]
[-- Type: text/x-patch, Size: 42578 bytes --]

Index: e2fsprogs-1.39/lib/ext2fs/kernel-jbd.h
===================================================================
--- e2fsprogs-1.39.orig/lib/ext2fs/kernel-jbd.h
+++ e2fsprogs-1.39/lib/ext2fs/kernel-jbd.h
@@ -108,7 +108,29 @@ typedef struct journal_header_s
 	__u32		h_sequence;
 } journal_header_t;
 
+/*
+ * Checksum types.
+ */
+#define JFS_CRC32_CHKSUM   1
+#define JFS_MD5_CHKSUM     2
+#define JFS_SHA1_CHKSUM    3
+
+#define JFS_CRC32_CHKSUM_SIZE 4
 
+#define JFS_CHECKSUM_BYTES (32 / sizeof(__u32))
+/*
+ * Commit block header for storing transactional checksums:
+ */
+struct commit_header
+{
+	__u32		h_magic;
+	__u32		h_blocktype;
+	__u32		h_sequence;
+	unsigned char   h_chksum_type;
+	unsigned char   h_chksum_size;
+	unsigned char   h_padding[2];
+	__u32		h_chksum[JFS_CHECKSUM_BYTES];
+};
 /* 
  * The block tag: used to describe a single buffer in the journal 
  */
@@ -194,12 +216,17 @@ typedef struct journal_superblock_s
 	((j)->j_format_version >= 2 &&					\
 	 ((j)->j_superblock->s_feature_incompat & cpu_to_be32((mask))))
 
-#define JFS_FEATURE_INCOMPAT_REVOKE	0x00000001
+#define JFS_FEATURE_COMPAT_CHECKSUM    0x00000001
+
+#define JFS_FEATURE_INCOMPAT_REVOKE		0x00000001
+/*#define JFS_FEATURE_INCOMPAT_64BIT             0x00000002*/
+#define JFS_FEATURE_INCOMPAT_ASYNC_COMMIT	0x00000004
 
 /* Features known to this kernel version: */
-#define JFS_KNOWN_COMPAT_FEATURES	0
+#define JFS_KNOWN_COMPAT_FEATURES	JFS_FEATURE_COMPAT_CHECKSUM
 #define JFS_KNOWN_ROCOMPAT_FEATURES	0
-#define JFS_KNOWN_INCOMPAT_FEATURES	JFS_FEATURE_INCOMPAT_REVOKE
+#define JFS_KNOWN_INCOMPAT_FEATURES	(JFS_FEATURE_INCOMPAT_REVOKE| \
+					JFS_FEATURE_INCOMPAT_ASYNC_COMMIT)
 
 #ifdef __KERNEL__
 
Index: e2fsprogs-1.39/lib/ext2fs/crc32.h
===================================================================
--- /dev/null
+++ e2fsprogs-1.39/lib/ext2fs/crc32.h
@@ -0,0 +1,28 @@
+/*
+ * crc32.h
+ * See crc32.c for license and changes
+ */
+#ifndef _LINUX_CRC32_H
+#define _LINUX_CRC32_H
+
+typedef unsigned int u32;
+typedef unsigned char u8;
+
+extern u32 crc32_le(u32 crc, unsigned char const *p, size_t len);
+extern u32 crc32_be(u32 crc, unsigned char const *p, size_t len);
+extern u32 bitreverse(u32 in);
+
+#define crc32(seed, data, length)  crc32_le(seed, (unsigned char const *)data, length)
+
+/*
+ * Helpers for hash table generation of ethernet nics:
+ *
+ * Ethernet sends the least significant bit of a byte first, thus crc32_le
+ * is used. The output of crc32_le is bit reversed [most significant bit
+ * is in bit nr 0], thus it must be reversed before use. Except for
+ * nics that bit swap the result internally...
+ */
+#define ether_crc(length, data)    bitreverse(crc32_le(~0, data, length))
+#define ether_crc_le(length, data) crc32_le(~0, data, length)
+
+#endif /* _LINUX_CRC32_H */
Index: e2fsprogs-1.39/lib/ext2fs/crc32.c
===================================================================
--- /dev/null
+++ e2fsprogs-1.39/lib/ext2fs/crc32.c
@@ -0,0 +1,519 @@
+/*
+ * Oct 15, 2000 Matt Domsch <Matt_Domsch@dell.com>
+ * Nicer crc32 functions/docs submitted by linux@horizon.com.  Thanks!
+ * Code was from the public domain, copyright abandoned.  Code was
+ * subsequently included in the kernel, thus was re-licensed under the
+ * GNU GPL v2.
+ *
+ * Oct 12, 2000 Matt Domsch <Matt_Domsch@dell.com>
+ * Same crc32 function was used in 5 other places in the kernel.
+ * I made one version, and deleted the others.
+ * There are various incantations of crc32().  Some use a seed of 0 or ~0.
+ * Some xor at the end with ~0.  The generic crc32() function takes
+ * seed as an argument, and doesn't xor at the end.  Then individual
+ * users can do whatever they need.
+ *   drivers/net/smc9194.c uses seed ~0, doesn't xor with ~0.
+ *   fs/jffs2 uses seed 0, doesn't xor with ~0.
+ *   fs/partitions/efi.c uses seed ~0, xor's with ~0.
+ *
+ * This source code is licensed under the GNU General Public License,
+ * Version 2.  See the file COPYING for more details.
+ */
+
+#include <stdlib.h>
+#include "crc32_user.h"
+#include "crc32.h"
+#include "crc32defs.h"
+#if CRC_LE_BITS == 8
+#define tole(x) __constant_cpu_to_le32(x)
+#define tobe(x) __constant_cpu_to_be32(x)
+#else
+#define tole(x) (x)
+#define tobe(x) (x)
+#endif
+#include "crc32table.h"
+
+
+#if CRC_LE_BITS == 1
+/*
+ * In fact, the table-based code will work in this case, but it can be
+ * simplified by inlining the table in ?: form.
+ */
+
+/**
+ * crc32_le() - Calculate bitwise little-endian Ethernet AUTODIN II CRC32
+ * @crc - seed value for computation.  ~0 for Ethernet, sometimes 0 for
+ *        other uses, or the previous crc32 value if computing incrementally.
+ * @p   - pointer to buffer over which CRC is run
+ * @len - length of buffer @p
+ *
+ */
+u32  crc32_le(u32 crc, unsigned char const *p, size_t len)
+{
+	int i;
+	while (len--) {
+		crc ^= *p++;
+		for (i = 0; i < 8; i++)
+			crc = (crc >> 1) ^ ((crc & 1) ? CRCPOLY_LE : 0);
+	}
+	return crc;
+}
+#else				/* Table-based approach */
+
+/**
+ * crc32_le() - Calculate bitwise little-endian Ethernet AUTODIN II CRC32
+ * @crc - seed value for computation.  ~0 for Ethernet, sometimes 0 for
+ *        other uses, or the previous crc32 value if computing incrementally.
+ * @p   - pointer to buffer over which CRC is run
+ * @len - length of buffer @p
+ *
+ */
+u32  crc32_le(u32 crc, unsigned char const *p, size_t len)
+{
+# if CRC_LE_BITS == 8
+	const u32      *b =(u32 *)p;
+	const u32      *tab = crc32table_le;
+
+# if __BYTE_ORDER == __LITTLE_ENDIAN
+#  define DO_CRC(x) crc = tab[ (crc ^ (x)) & 255 ] ^ (crc>>8)
+# else
+#  define DO_CRC(x) crc = tab[ ((crc >> 24) ^ (x)) & 255] ^ (crc<<8)
+# endif
+
+	crc = __cpu_to_le32(crc);
+	/* Align it */
+	if(unlikely(((long)b)&3 && len)){
+		do {
+			u8 *p = (u8 *)b;
+			DO_CRC(*p++);
+			b = (void *)p;
+		} while ((--len) && ((long)b)&3 );
+	}
+	if(likely(len >= 4)){
+		/* load data 32 bits wide, xor data 32 bits wide. */
+		size_t save_len = len & 3;
+	        len = len >> 2;
+		--b; /* use pre increment below(*++b) for speed */
+		do {
+			crc ^= *++b;
+			DO_CRC(0);
+			DO_CRC(0);
+			DO_CRC(0);
+			DO_CRC(0);
+		} while (--len);
+		b++; /* point to next byte(s) */
+		len = save_len;
+	}
+	/* And the last few bytes */
+	if(len){
+		do {
+			u8 *p = (u8 *)b;
+			DO_CRC(*p++);
+			b = (void *)p;
+		} while (--len);
+	}
+
+	return __le32_to_cpu(crc);
+#undef ENDIAN_SHIFT
+#undef DO_CRC
+
+# elif CRC_LE_BITS == 4
+	while (len--) {
+		crc ^= *p++;
+		crc = (crc >> 4) ^ crc32table_le[crc & 15];
+		crc = (crc >> 4) ^ crc32table_le[crc & 15];
+	}
+	return crc;
+# elif CRC_LE_BITS == 2
+	while (len--) {
+		crc ^= *p++;
+		crc = (crc >> 2) ^ crc32table_le[crc & 3];
+		crc = (crc >> 2) ^ crc32table_le[crc & 3];
+		crc = (crc >> 2) ^ crc32table_le[crc & 3];
+		crc = (crc >> 2) ^ crc32table_le[crc & 3];
+	}
+	return crc;
+# endif
+}
+#endif
+
+#if CRC_BE_BITS == 1
+/*
+ * In fact, the table-based code will work in this case, but it can be
+ * simplified by inlining the table in ?: form.
+ */
+
+/**
+ * crc32_be() - Calculate bitwise big-endian Ethernet AUTODIN II CRC32
+ * @crc - seed value for computation.  ~0 for Ethernet, sometimes 0 for
+ *        other uses, or the previous crc32 value if computing incrementally.
+ * @p   - pointer to buffer over which CRC is run
+ * @len - length of buffer @p
+ *
+ */
+u32  crc32_be(u32 crc, unsigned char const *p, size_t len)
+{
+	int i;
+	printf("CRC %u\n",crc);
+	while (len--) {
+		crc ^= *p++ << 24;
+		for (i = 0; i < 8; i++)
+			crc =
+			    (crc << 1) ^ ((crc & 0x80000000) ? CRCPOLY_BE :
+					  0);
+	}
+	return crc;
+}
+
+#else				/* Table-based approach */
+/**
+ * crc32_be() - Calculate bitwise big-endian Ethernet AUTODIN II CRC32
+ * @crc - seed value for computation.  ~0 for Ethernet, sometimes 0 for
+ *        other uses, or the previous crc32 value if computing incrementally.
+ * @p   - pointer to buffer over which CRC is run
+ * @len - length of buffer @p
+ *
+ */
+u32  crc32_be(u32 crc, unsigned char const *p, size_t len)
+{
+# if CRC_BE_BITS == 8
+	const u32      *b =(u32 *)p;
+	const u32      *tab = crc32table_be;
+
+# if __BYTE_ORDER == __LITTLE_ENDIAN
+#  define DO_CRC(x) crc = tab[ (crc ^ (x)) & 255 ] ^ (crc>>8)
+# else
+#  define DO_CRC(x) crc = tab[ ((crc >> 24) ^ (x)) & 255] ^ (crc<<8)
+# endif
+	crc = __cpu_to_be32(crc);
+	/* Align it */
+	if(unlikely(((long)b)&3 && len)){
+		do {
+			u8 *p = (u8 *)b;
+			DO_CRC(*p++);
+			b = (u32 *)p;
+		} while ((--len) && ((long)b)&3 );
+	}
+
+	if(likely(len >= 4)){
+		/* load data 32 bits wide, xor data 32 bits wide. */
+		size_t save_len = len & 3;
+	        len = len >> 2;
+		--b; /* use pre increment below(*++b) for speed */
+		do {
+			crc ^= *++b;
+			DO_CRC(0);
+			DO_CRC(0);
+			DO_CRC(0);
+			DO_CRC(0);
+		} while (--len);
+		b++; /* point to next byte(s) */
+		len = save_len;
+	}
+	/* And the last few bytes */
+	if(len){
+		do {
+			u8 *p = (u8 *)b;
+			DO_CRC(*p++);
+			b = (void *)p;
+		} while (--len);
+	}
+	return __be32_to_cpu(crc);
+#undef ENDIAN_SHIFT
+#undef DO_CRC
+
+# elif CRC_BE_BITS == 4
+	while (len--) {
+		crc ^= *p++ << 24;
+		crc = (crc << 4) ^ crc32table_be[crc >> 28];
+		crc = (crc << 4) ^ crc32table_be[crc >> 28];
+	}
+	return crc;
+# elif CRC_BE_BITS == 2
+	while (len--) {
+		crc ^= *p++ << 24;
+		crc = (crc << 2) ^ crc32table_be[crc >> 30];
+		crc = (crc << 2) ^ crc32table_be[crc >> 30];
+		crc = (crc << 2) ^ crc32table_be[crc >> 30];
+		crc = (crc << 2) ^ crc32table_be[crc >> 30];
+	}
+	return crc;
+# endif
+}
+#endif
+
+u32 bitreverse(u32 x)
+{
+	x = (x >> 16) | (x << 16);
+	x = (x >> 8 & 0x00ff00ff) | (x << 8 & 0xff00ff00);
+	x = (x >> 4 & 0x0f0f0f0f) | (x << 4 & 0xf0f0f0f0);
+	x = (x >> 2 & 0x33333333) | (x << 2 & 0xcccccccc);
+	x = (x >> 1 & 0x55555555) | (x << 1 & 0xaaaaaaaa);
+	return x;
+}
+
+
+/*
+ * A brief CRC tutorial.
+ *
+ * A CRC is a long-division remainder.  You add the CRC to the message,
+ * and the whole thing (message+CRC) is a multiple of the given
+ * CRC polynomial.  To check the CRC, you can either check that the
+ * CRC matches the recomputed value, *or* you can check that the
+ * remainder computed on the message+CRC is 0.  This latter approach
+ * is used by a lot of hardware implementations, and is why so many
+ * protocols put the end-of-frame flag after the CRC.
+ *
+ * It's actually the same long division you learned in school, except that
+ * - We're working in binary, so the digits are only 0 and 1, and
+ * - When dividing polynomials, there are no carries.  Rather than add and
+ *   subtract, we just xor.  Thus, we tend to get a bit sloppy about
+ *   the difference between adding and subtracting.
+ *
+ * A 32-bit CRC polynomial is actually 33 bits long.  But since it's
+ * 33 bits long, bit 32 is always going to be set, so usually the CRC
+ * is written in hex with the most significant bit omitted.  (If you're
+ * familiar with the IEEE 754 floating-point format, it's the same idea.)
+ *
+ * Note that a CRC is computed over a string of *bits*, so you have
+ * to decide on the endianness of the bits within each byte.  To get
+ * the best error-detecting properties, this should correspond to the
+ * order they're actually sent.  For example, standard RS-232 serial is
+ * little-endian; the most significant bit (sometimes used for parity)
+ * is sent last.  And when appending a CRC word to a message, you should
+ * do it in the right order, matching the endianness.
+ *
+ * Just like with ordinary division, the remainder is always smaller than
+ * the divisor (the CRC polynomial) you're dividing by.  Each step of the
+ * division, you take one more digit (bit) of the dividend and append it
+ * to the current remainder.  Then you figure out the appropriate multiple
+ * of the divisor to subtract to being the remainder back into range.
+ * In binary, it's easy - it has to be either 0 or 1, and to make the
+ * XOR cancel, it's just a copy of bit 32 of the remainder.
+ *
+ * When computing a CRC, we don't care about the quotient, so we can
+ * throw the quotient bit away, but subtract the appropriate multiple of
+ * the polynomial from the remainder and we're back to where we started,
+ * ready to process the next bit.
+ *
+ * A big-endian CRC written this way would be coded like:
+ * for (i = 0; i < input_bits; i++) {
+ * 	multiple = remainder & 0x80000000 ? CRCPOLY : 0;
+ * 	remainder = (remainder << 1 | next_input_bit()) ^ multiple;
+ * }
+ * Notice how, to get at bit 32 of the shifted remainder, we look
+ * at bit 31 of the remainder *before* shifting it.
+ *
+ * But also notice how the next_input_bit() bits we're shifting into
+ * the remainder don't actually affect any decision-making until
+ * 32 bits later.  Thus, the first 32 cycles of this are pretty boring.
+ * Also, to add the CRC to a message, we need a 32-bit-long hole for it at
+ * the end, so we have to add 32 extra cycles shifting in zeros at the
+ * end of every message,
+ *
+ * So the standard trick is to rearrage merging in the next_input_bit()
+ * until the moment it's needed.  Then the first 32 cycles can be precomputed,
+ * and merging in the final 32 zero bits to make room for the CRC can be
+ * skipped entirely.
+ * This changes the code to:
+ * for (i = 0; i < input_bits; i++) {
+ *      remainder ^= next_input_bit() << 31;
+ * 	multiple = (remainder & 0x80000000) ? CRCPOLY : 0;
+ * 	remainder = (remainder << 1) ^ multiple;
+ * }
+ * With this optimization, the little-endian code is simpler:
+ * for (i = 0; i < input_bits; i++) {
+ *      remainder ^= next_input_bit();
+ * 	multiple = (remainder & 1) ? CRCPOLY : 0;
+ * 	remainder = (remainder >> 1) ^ multiple;
+ * }
+ *
+ * Note that the other details of endianness have been hidden in CRCPOLY
+ * (which must be bit-reversed) and next_input_bit().
+ *
+ * However, as long as next_input_bit is returning the bits in a sensible
+ * order, we can actually do the merging 8 or more bits at a time rather
+ * than one bit at a time:
+ * for (i = 0; i < input_bytes; i++) {
+ * 	remainder ^= next_input_byte() << 24;
+ * 	for (j = 0; j < 8; j++) {
+ * 		multiple = (remainder & 0x80000000) ? CRCPOLY : 0;
+ * 		remainder = (remainder << 1) ^ multiple;
+ * 	}
+ * }
+ * Or in little-endian:
+ * for (i = 0; i < input_bytes; i++) {
+ * 	remainder ^= next_input_byte();
+ * 	for (j = 0; j < 8; j++) {
+ * 		multiple = (remainder & 1) ? CRCPOLY : 0;
+ * 		remainder = (remainder << 1) ^ multiple;
+ * 	}
+ * }
+ * If the input is a multiple of 32 bits, you can even XOR in a 32-bit
+ * word at a time and increase the inner loop count to 32.
+ *
+ * You can also mix and match the two loop styles, for example doing the
+ * bulk of a message byte-at-a-time and adding bit-at-a-time processing
+ * for any fractional bytes at the end.
+ *
+ * The only remaining optimization is to the byte-at-a-time table method.
+ * Here, rather than just shifting one bit of the remainder to decide
+ * in the correct multiple to subtract, we can shift a byte at a time.
+ * This produces a 40-bit (rather than a 33-bit) intermediate remainder,
+ * but again the multiple of the polynomial to subtract depends only on
+ * the high bits, the high 8 bits in this case.
+ *
+ * The multile we need in that case is the low 32 bits of a 40-bit
+ * value whose high 8 bits are given, and which is a multiple of the
+ * generator polynomial.  This is simply the CRC-32 of the given
+ * one-byte message.
+ *
+ * Two more details: normally, appending zero bits to a message which
+ * is already a multiple of a polynomial produces a larger multiple of that
+ * polynomial.  To enable a CRC to detect this condition, it's common to
+ * invert the CRC before appending it.  This makes the remainder of the
+ * message+crc come out not as zero, but some fixed non-zero value.
+ *
+ * The same problem applies to zero bits prepended to the message, and
+ * a similar solution is used.  Instead of starting with a remainder of
+ * 0, an initial remainder of all ones is used.  As long as you start
+ * the same way on decoding, it doesn't make a difference.
+ */
+
+#ifdef UNITTEST
+
+#include <stdlib.h>
+#include <stdio.h>
+
+#if 0				/*Not used at present */
+static void
+buf_dump(char const *prefix, unsigned char const *buf, size_t len)
+{
+	fputs(prefix, stdout);
+	while (len--)
+		printf(" %02x", *buf++);
+	putchar('\n');
+
+}
+#endif
+
+static void bytereverse(unsigned char *buf, size_t len)
+{
+	while (len--) {
+		unsigned char x = *buf;
+		x = (x >> 4) | (x << 4);
+		x = (x >> 2 & 0x33) | (x << 2 & 0xcc);
+		x = (x >> 1 & 0x55) | (x << 1 & 0xaa);
+		*buf++ = x;
+	}
+}
+
+static void random_garbage(unsigned char *buf, size_t len)
+{
+	while (len--)
+		*buf++ = (unsigned char) random();
+}
+
+#if 0				/* Not used at present */
+static void store_le(u32 x, unsigned char *buf)
+{
+	buf[0] = (unsigned char) x;
+	buf[1] = (unsigned char) (x >> 8);
+	buf[2] = (unsigned char) (x >> 16);
+	buf[3] = (unsigned char) (x >> 24);
+}
+#endif
+
+static void store_be(u32 x, unsigned char *buf)
+{
+	buf[0] = (unsigned char) (x >> 24);
+	buf[1] = (unsigned char) (x >> 16);
+	buf[2] = (unsigned char) (x >> 8);
+	buf[3] = (unsigned char) x;
+}
+
+/*
+ * This checks that CRC(buf + CRC(buf)) = 0, and that
+ * CRC commutes with bit-reversal.  This has the side effect
+ * of bytewise bit-reversing the input buffer, and returns
+ * the CRC of the reversed buffer.
+ */
+static u32 test_step(u32 init, unsigned char *buf, size_t len)
+{
+	u32 crc1, crc2;
+	size_t i;
+
+	crc1 = crc32_be(init, buf, len);
+	store_be(crc1, buf + len);
+	crc2 = crc32_be(init, buf, len + 4);
+	if (crc2)
+		printf("\nCRC cancellation fail: 0x%08x should be 0\n",
+		       crc2);
+
+	for (i = 0; i <= len + 4; i++) {
+		crc2 = crc32_be(init, buf, i);
+		crc2 = crc32_be(crc2, buf + i, len + 4 - i);
+		if (crc2)
+			printf("\nCRC split fail: 0x%08x\n", crc2);
+	}
+
+	/* Now swap it around for the other test */
+
+	bytereverse(buf, len + 4);
+	init = bitreverse(init);
+	crc2 = bitreverse(crc1);
+	if (crc1 != bitreverse(crc2))
+		printf("\nBit reversal fail: 0x%08x -> %0x08x -> 0x%08x\n",
+		       crc1, crc2, bitreverse(crc2));
+	crc1 = crc32_le(init, buf, len);
+	if (crc1 != crc2)
+		printf("\nCRC endianness fail: 0x%08x != 0x%08x\n", crc1,
+		       crc2);
+	crc2 = crc32_le(init, buf, len + 4);
+	if (crc2)
+		printf("\nCRC cancellation fail: 0x%08x should be 0\n",
+		       crc2);
+
+	for (i = 0; i <= len + 4; i++) {
+		crc2 = crc32_le(init, buf, i);
+		crc2 = crc32_le(crc2, buf + i, len + 4 - i);
+		if (crc2)
+			printf("\nCRC split fail: 0x%08x\n", crc2);
+	}
+
+	return crc1;
+}
+
+#define SIZE 64
+#define INIT1 0
+#define INIT2 0
+
+int main(void)
+{
+	unsigned char buf1[SIZE + 4];
+	unsigned char buf2[SIZE + 4];
+	unsigned char buf3[SIZE + 4];
+	int i, j;
+	u32 crc1, crc2, crc3;
+
+	for (i = 0; i <= SIZE; i++) {
+		printf("\rTesting length %d...", i);
+		fflush(stdout);
+		random_garbage(buf1, i);
+		random_garbage(buf2, i);
+		for (j = 0; j < i; j++)
+			buf3[j] = buf1[j] ^ buf2[j];
+
+		crc1 = test_step(INIT1, buf1, i);
+		crc2 = test_step(INIT2, buf2, i);
+		/* Now check that CRC(buf1 ^ buf2) = CRC(buf1) ^ CRC(buf2) */
+		crc3 = test_step(INIT1 ^ INIT2, buf3, i);
+		if (crc3 != (crc1 ^ crc2))
+			printf("CRC XOR fail: 0x%08x != 0x%08x ^ 0x%08x\n",
+			       crc3, crc1, crc2);
+	}
+	printf("\nAll test complete.  No failures expected.\n");
+	return 0;
+}
+
+#endif				/* UNITTEST */
Index: e2fsprogs-1.39/lib/ext2fs/crc32defs.h
===================================================================
--- /dev/null
+++ e2fsprogs-1.39/lib/ext2fs/crc32defs.h
@@ -0,0 +1,32 @@
+/*
+ * There are multiple 16-bit CRC polynomials in common use, but this is
+ * *the* standard CRC-32 polynomial, first popularized by Ethernet.
+ * x^32+x^26+x^23+x^22+x^16+x^12+x^11+x^10+x^8+x^7+x^5+x^4+x^2+x^1+x^0
+ */
+#define CRCPOLY_LE 0xedb88320
+#define CRCPOLY_BE 0x04c11db7
+
+/* How many bits at a time to use.  Requires a table of 4<<CRC_xx_BITS bytes. */
+/* For less performance-sensitive, use 4 */
+#ifndef CRC_LE_BITS
+# define CRC_LE_BITS 8
+#endif
+#ifndef CRC_BE_BITS
+# define CRC_BE_BITS 8
+#endif
+
+/*
+ * Little-endian CRC computation.  Used with serial bit streams sent
+ * lsbit-first.  Be sure to use cpu_to_le32() to append the computed CRC.
+ */
+#if CRC_LE_BITS > 8 || CRC_LE_BITS < 1 || CRC_LE_BITS & CRC_LE_BITS-1
+# error CRC_LE_BITS must be a power of 2 between 1 and 8
+#endif
+
+/*
+ * Big-endian CRC computation.  Used with serial bit streams sent
+ * msbit-first.  Be sure to use cpu_to_be32() to append the computed CRC.
+ */
+#if CRC_BE_BITS > 8 || CRC_BE_BITS < 1 || CRC_BE_BITS & CRC_BE_BITS-1
+# error CRC_BE_BITS must be a power of 2 between 1 and 8
+#endif
Index: e2fsprogs-1.39/lib/ext2fs/crc32table.h
===================================================================
--- /dev/null
+++ e2fsprogs-1.39/lib/ext2fs/crc32table.h
@@ -0,0 +1,135 @@
+/* this file is generated - do not edit */
+
+static const u32 crc32table_le[] = {
+tole(0x00000000L), tole(0x77073096L), tole(0xee0e612cL), tole(0x990951baL),
+tole(0x076dc419L), tole(0x706af48fL), tole(0xe963a535L), tole(0x9e6495a3L),
+tole(0x0edb8832L), tole(0x79dcb8a4L), tole(0xe0d5e91eL), tole(0x97d2d988L),
+tole(0x09b64c2bL), tole(0x7eb17cbdL), tole(0xe7b82d07L), tole(0x90bf1d91L),
+tole(0x1db71064L), tole(0x6ab020f2L), tole(0xf3b97148L), tole(0x84be41deL),
+tole(0x1adad47dL), tole(0x6ddde4ebL), tole(0xf4d4b551L), tole(0x83d385c7L),
+tole(0x136c9856L), tole(0x646ba8c0L), tole(0xfd62f97aL), tole(0x8a65c9ecL),
+tole(0x14015c4fL), tole(0x63066cd9L), tole(0xfa0f3d63L), tole(0x8d080df5L),
+tole(0x3b6e20c8L), tole(0x4c69105eL), tole(0xd56041e4L), tole(0xa2677172L),
+tole(0x3c03e4d1L), tole(0x4b04d447L), tole(0xd20d85fdL), tole(0xa50ab56bL),
+tole(0x35b5a8faL), tole(0x42b2986cL), tole(0xdbbbc9d6L), tole(0xacbcf940L),
+tole(0x32d86ce3L), tole(0x45df5c75L), tole(0xdcd60dcfL), tole(0xabd13d59L),
+tole(0x26d930acL), tole(0x51de003aL), tole(0xc8d75180L), tole(0xbfd06116L),
+tole(0x21b4f4b5L), tole(0x56b3c423L), tole(0xcfba9599L), tole(0xb8bda50fL),
+tole(0x2802b89eL), tole(0x5f058808L), tole(0xc60cd9b2L), tole(0xb10be924L),
+tole(0x2f6f7c87L), tole(0x58684c11L), tole(0xc1611dabL), tole(0xb6662d3dL),
+tole(0x76dc4190L), tole(0x01db7106L), tole(0x98d220bcL), tole(0xefd5102aL),
+tole(0x71b18589L), tole(0x06b6b51fL), tole(0x9fbfe4a5L), tole(0xe8b8d433L),
+tole(0x7807c9a2L), tole(0x0f00f934L), tole(0x9609a88eL), tole(0xe10e9818L),
+tole(0x7f6a0dbbL), tole(0x086d3d2dL), tole(0x91646c97L), tole(0xe6635c01L),
+tole(0x6b6b51f4L), tole(0x1c6c6162L), tole(0x856530d8L), tole(0xf262004eL),
+tole(0x6c0695edL), tole(0x1b01a57bL), tole(0x8208f4c1L), tole(0xf50fc457L),
+tole(0x65b0d9c6L), tole(0x12b7e950L), tole(0x8bbeb8eaL), tole(0xfcb9887cL),
+tole(0x62dd1ddfL), tole(0x15da2d49L), tole(0x8cd37cf3L), tole(0xfbd44c65L),
+tole(0x4db26158L), tole(0x3ab551ceL), tole(0xa3bc0074L), tole(0xd4bb30e2L),
+tole(0x4adfa541L), tole(0x3dd895d7L), tole(0xa4d1c46dL), tole(0xd3d6f4fbL),
+tole(0x4369e96aL), tole(0x346ed9fcL), tole(0xad678846L), tole(0xda60b8d0L),
+tole(0x44042d73L), tole(0x33031de5L), tole(0xaa0a4c5fL), tole(0xdd0d7cc9L),
+tole(0x5005713cL), tole(0x270241aaL), tole(0xbe0b1010L), tole(0xc90c2086L),
+tole(0x5768b525L), tole(0x206f85b3L), tole(0xb966d409L), tole(0xce61e49fL),
+tole(0x5edef90eL), tole(0x29d9c998L), tole(0xb0d09822L), tole(0xc7d7a8b4L),
+tole(0x59b33d17L), tole(0x2eb40d81L), tole(0xb7bd5c3bL), tole(0xc0ba6cadL),
+tole(0xedb88320L), tole(0x9abfb3b6L), tole(0x03b6e20cL), tole(0x74b1d29aL),
+tole(0xead54739L), tole(0x9dd277afL), tole(0x04db2615L), tole(0x73dc1683L),
+tole(0xe3630b12L), tole(0x94643b84L), tole(0x0d6d6a3eL), tole(0x7a6a5aa8L),
+tole(0xe40ecf0bL), tole(0x9309ff9dL), tole(0x0a00ae27L), tole(0x7d079eb1L),
+tole(0xf00f9344L), tole(0x8708a3d2L), tole(0x1e01f268L), tole(0x6906c2feL),
+tole(0xf762575dL), tole(0x806567cbL), tole(0x196c3671L), tole(0x6e6b06e7L),
+tole(0xfed41b76L), tole(0x89d32be0L), tole(0x10da7a5aL), tole(0x67dd4accL),
+tole(0xf9b9df6fL), tole(0x8ebeeff9L), tole(0x17b7be43L), tole(0x60b08ed5L),
+tole(0xd6d6a3e8L), tole(0xa1d1937eL), tole(0x38d8c2c4L), tole(0x4fdff252L),
+tole(0xd1bb67f1L), tole(0xa6bc5767L), tole(0x3fb506ddL), tole(0x48b2364bL),
+tole(0xd80d2bdaL), tole(0xaf0a1b4cL), tole(0x36034af6L), tole(0x41047a60L),
+tole(0xdf60efc3L), tole(0xa867df55L), tole(0x316e8eefL), tole(0x4669be79L),
+tole(0xcb61b38cL), tole(0xbc66831aL), tole(0x256fd2a0L), tole(0x5268e236L),
+tole(0xcc0c7795L), tole(0xbb0b4703L), tole(0x220216b9L), tole(0x5505262fL),
+tole(0xc5ba3bbeL), tole(0xb2bd0b28L), tole(0x2bb45a92L), tole(0x5cb36a04L),
+tole(0xc2d7ffa7L), tole(0xb5d0cf31L), tole(0x2cd99e8bL), tole(0x5bdeae1dL),
+tole(0x9b64c2b0L), tole(0xec63f226L), tole(0x756aa39cL), tole(0x026d930aL),
+tole(0x9c0906a9L), tole(0xeb0e363fL), tole(0x72076785L), tole(0x05005713L),
+tole(0x95bf4a82L), tole(0xe2b87a14L), tole(0x7bb12baeL), tole(0x0cb61b38L),
+tole(0x92d28e9bL), tole(0xe5d5be0dL), tole(0x7cdcefb7L), tole(0x0bdbdf21L),
+tole(0x86d3d2d4L), tole(0xf1d4e242L), tole(0x68ddb3f8L), tole(0x1fda836eL),
+tole(0x81be16cdL), tole(0xf6b9265bL), tole(0x6fb077e1L), tole(0x18b74777L),
+tole(0x88085ae6L), tole(0xff0f6a70L), tole(0x66063bcaL), tole(0x11010b5cL),
+tole(0x8f659effL), tole(0xf862ae69L), tole(0x616bffd3L), tole(0x166ccf45L),
+tole(0xa00ae278L), tole(0xd70dd2eeL), tole(0x4e048354L), tole(0x3903b3c2L),
+tole(0xa7672661L), tole(0xd06016f7L), tole(0x4969474dL), tole(0x3e6e77dbL),
+tole(0xaed16a4aL), tole(0xd9d65adcL), tole(0x40df0b66L), tole(0x37d83bf0L),
+tole(0xa9bcae53L), tole(0xdebb9ec5L), tole(0x47b2cf7fL), tole(0x30b5ffe9L),
+tole(0xbdbdf21cL), tole(0xcabac28aL), tole(0x53b39330L), tole(0x24b4a3a6L),
+tole(0xbad03605L), tole(0xcdd70693L), tole(0x54de5729L), tole(0x23d967bfL),
+tole(0xb3667a2eL), tole(0xc4614ab8L), tole(0x5d681b02L), tole(0x2a6f2b94L),
+tole(0xb40bbe37L), tole(0xc30c8ea1L), tole(0x5a05df1bL), tole(0x2d02ef8dL)
+};
+
+static const u32 crc32table_be[] = {
+tobe(0x00000000L), tobe(0x04c11db7L), tobe(0x09823b6eL), tobe(0x0d4326d9L),
+tobe(0x130476dcL), tobe(0x17c56b6bL), tobe(0x1a864db2L), tobe(0x1e475005L),
+tobe(0x2608edb8L), tobe(0x22c9f00fL), tobe(0x2f8ad6d6L), tobe(0x2b4bcb61L),
+tobe(0x350c9b64L), tobe(0x31cd86d3L), tobe(0x3c8ea00aL), tobe(0x384fbdbdL),
+tobe(0x4c11db70L), tobe(0x48d0c6c7L), tobe(0x4593e01eL), tobe(0x4152fda9L),
+tobe(0x5f15adacL), tobe(0x5bd4b01bL), tobe(0x569796c2L), tobe(0x52568b75L),
+tobe(0x6a1936c8L), tobe(0x6ed82b7fL), tobe(0x639b0da6L), tobe(0x675a1011L),
+tobe(0x791d4014L), tobe(0x7ddc5da3L), tobe(0x709f7b7aL), tobe(0x745e66cdL),
+tobe(0x9823b6e0L), tobe(0x9ce2ab57L), tobe(0x91a18d8eL), tobe(0x95609039L),
+tobe(0x8b27c03cL), tobe(0x8fe6dd8bL), tobe(0x82a5fb52L), tobe(0x8664e6e5L),
+tobe(0xbe2b5b58L), tobe(0xbaea46efL), tobe(0xb7a96036L), tobe(0xb3687d81L),
+tobe(0xad2f2d84L), tobe(0xa9ee3033L), tobe(0xa4ad16eaL), tobe(0xa06c0b5dL),
+tobe(0xd4326d90L), tobe(0xd0f37027L), tobe(0xddb056feL), tobe(0xd9714b49L),
+tobe(0xc7361b4cL), tobe(0xc3f706fbL), tobe(0xceb42022L), tobe(0xca753d95L),
+tobe(0xf23a8028L), tobe(0xf6fb9d9fL), tobe(0xfbb8bb46L), tobe(0xff79a6f1L),
+tobe(0xe13ef6f4L), tobe(0xe5ffeb43L), tobe(0xe8bccd9aL), tobe(0xec7dd02dL),
+tobe(0x34867077L), tobe(0x30476dc0L), tobe(0x3d044b19L), tobe(0x39c556aeL),
+tobe(0x278206abL), tobe(0x23431b1cL), tobe(0x2e003dc5L), tobe(0x2ac12072L),
+tobe(0x128e9dcfL), tobe(0x164f8078L), tobe(0x1b0ca6a1L), tobe(0x1fcdbb16L),
+tobe(0x018aeb13L), tobe(0x054bf6a4L), tobe(0x0808d07dL), tobe(0x0cc9cdcaL),
+tobe(0x7897ab07L), tobe(0x7c56b6b0L), tobe(0x71159069L), tobe(0x75d48ddeL),
+tobe(0x6b93dddbL), tobe(0x6f52c06cL), tobe(0x6211e6b5L), tobe(0x66d0fb02L),
+tobe(0x5e9f46bfL), tobe(0x5a5e5b08L), tobe(0x571d7dd1L), tobe(0x53dc6066L),
+tobe(0x4d9b3063L), tobe(0x495a2dd4L), tobe(0x44190b0dL), tobe(0x40d816baL),
+tobe(0xaca5c697L), tobe(0xa864db20L), tobe(0xa527fdf9L), tobe(0xa1e6e04eL),
+tobe(0xbfa1b04bL), tobe(0xbb60adfcL), tobe(0xb6238b25L), tobe(0xb2e29692L),
+tobe(0x8aad2b2fL), tobe(0x8e6c3698L), tobe(0x832f1041L), tobe(0x87ee0df6L),
+tobe(0x99a95df3L), tobe(0x9d684044L), tobe(0x902b669dL), tobe(0x94ea7b2aL),
+tobe(0xe0b41de7L), tobe(0xe4750050L), tobe(0xe9362689L), tobe(0xedf73b3eL),
+tobe(0xf3b06b3bL), tobe(0xf771768cL), tobe(0xfa325055L), tobe(0xfef34de2L),
+tobe(0xc6bcf05fL), tobe(0xc27dede8L), tobe(0xcf3ecb31L), tobe(0xcbffd686L),
+tobe(0xd5b88683L), tobe(0xd1799b34L), tobe(0xdc3abdedL), tobe(0xd8fba05aL),
+tobe(0x690ce0eeL), tobe(0x6dcdfd59L), tobe(0x608edb80L), tobe(0x644fc637L),
+tobe(0x7a089632L), tobe(0x7ec98b85L), tobe(0x738aad5cL), tobe(0x774bb0ebL),
+tobe(0x4f040d56L), tobe(0x4bc510e1L), tobe(0x46863638L), tobe(0x42472b8fL),
+tobe(0x5c007b8aL), tobe(0x58c1663dL), tobe(0x558240e4L), tobe(0x51435d53L),
+tobe(0x251d3b9eL), tobe(0x21dc2629L), tobe(0x2c9f00f0L), tobe(0x285e1d47L),
+tobe(0x36194d42L), tobe(0x32d850f5L), tobe(0x3f9b762cL), tobe(0x3b5a6b9bL),
+tobe(0x0315d626L), tobe(0x07d4cb91L), tobe(0x0a97ed48L), tobe(0x0e56f0ffL),
+tobe(0x1011a0faL), tobe(0x14d0bd4dL), tobe(0x19939b94L), tobe(0x1d528623L),
+tobe(0xf12f560eL), tobe(0xf5ee4bb9L), tobe(0xf8ad6d60L), tobe(0xfc6c70d7L),
+tobe(0xe22b20d2L), tobe(0xe6ea3d65L), tobe(0xeba91bbcL), tobe(0xef68060bL),
+tobe(0xd727bbb6L), tobe(0xd3e6a601L), tobe(0xdea580d8L), tobe(0xda649d6fL),
+tobe(0xc423cd6aL), tobe(0xc0e2d0ddL), tobe(0xcda1f604L), tobe(0xc960ebb3L),
+tobe(0xbd3e8d7eL), tobe(0xb9ff90c9L), tobe(0xb4bcb610L), tobe(0xb07daba7L),
+tobe(0xae3afba2L), tobe(0xaafbe615L), tobe(0xa7b8c0ccL), tobe(0xa379dd7bL),
+tobe(0x9b3660c6L), tobe(0x9ff77d71L), tobe(0x92b45ba8L), tobe(0x9675461fL),
+tobe(0x8832161aL), tobe(0x8cf30badL), tobe(0x81b02d74L), tobe(0x857130c3L),
+tobe(0x5d8a9099L), tobe(0x594b8d2eL), tobe(0x5408abf7L), tobe(0x50c9b640L),
+tobe(0x4e8ee645L), tobe(0x4a4ffbf2L), tobe(0x470cdd2bL), tobe(0x43cdc09cL),
+tobe(0x7b827d21L), tobe(0x7f436096L), tobe(0x7200464fL), tobe(0x76c15bf8L),
+tobe(0x68860bfdL), tobe(0x6c47164aL), tobe(0x61043093L), tobe(0x65c52d24L),
+tobe(0x119b4be9L), tobe(0x155a565eL), tobe(0x18197087L), tobe(0x1cd86d30L),
+tobe(0x029f3d35L), tobe(0x065e2082L), tobe(0x0b1d065bL), tobe(0x0fdc1becL),
+tobe(0x3793a651L), tobe(0x3352bbe6L), tobe(0x3e119d3fL), tobe(0x3ad08088L),
+tobe(0x2497d08dL), tobe(0x2056cd3aL), tobe(0x2d15ebe3L), tobe(0x29d4f654L),
+tobe(0xc5a92679L), tobe(0xc1683bceL), tobe(0xcc2b1d17L), tobe(0xc8ea00a0L),
+tobe(0xd6ad50a5L), tobe(0xd26c4d12L), tobe(0xdf2f6bcbL), tobe(0xdbee767cL),
+tobe(0xe3a1cbc1L), tobe(0xe760d676L), tobe(0xea23f0afL), tobe(0xeee2ed18L),
+tobe(0xf0a5bd1dL), tobe(0xf464a0aaL), tobe(0xf9278673L), tobe(0xfde69bc4L),
+tobe(0x89b8fd09L), tobe(0x8d79e0beL), tobe(0x803ac667L), tobe(0x84fbdbd0L),
+tobe(0x9abc8bd5L), tobe(0x9e7d9662L), tobe(0x933eb0bbL), tobe(0x97ffad0cL),
+tobe(0xafb010b1L), tobe(0xab710d06L), tobe(0xa6322bdfL), tobe(0xa2f33668L),
+tobe(0xbcb4666dL), tobe(0xb8757bdaL), tobe(0xb5365d03L), tobe(0xb1f740b4L)
+};
Index: e2fsprogs-1.39/lib/ext2fs/crc32_user.h
===================================================================
--- /dev/null
+++ e2fsprogs-1.39/lib/ext2fs/crc32_user.h
@@ -0,0 +1,45 @@
+/*
+ * Defines macros and types required by crc32 code undefined in user space.
+ */
+#ifndef _LINUX_CRC32_USER_H
+#define _LINUX_CRC32_USER_H
+#include <linux/types.h>
+
+#define likely(x)       __builtin_expect(!!(x), 1)
+#define unlikely(x)     __builtin_expect(!!(x), 0)
+
+#define __swab32(x) \
+({ \
+        __u32 __x = (x); \
+        ((__u32)( \
+                (((__u32)(__x) & (__u32)0x000000ffUL) << 24) | \
+                (((__u32)(__x) & (__u32)0x0000ff00UL) <<  8) | \
+                (((__u32)(__x) & (__u32)0x00ff0000UL) >>  8) | \
+                (((__u32)(__x) & (__u32)0xff000000UL) >> 24) )); \
+})
+
+#define ___constant_swab32(x) \
+        ((__u32)( \
+                (((__u32)(x) & (__u32)0x000000ffUL) << 24) | \
+                (((__u32)(x) & (__u32)0x0000ff00UL) <<  8) | \
+                (((__u32)(x) & (__u32)0x0000ff00UL) <<  8) | \
+                (((__u32)(x) & (__u32)0x00ff0000UL) >>  8) | \
+                (((__u32)(x) & (__u32)0xff000000UL) >> 24) ))
+
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+#define __le32_to_cpu(x) ((__u32)(x))
+#define __cpu_to_le32(x) ((__u32)(x))
+#define __be32_to_cpu(x) __swab32((x))
+#define __cpu_to_be32(x) __swab32((x))
+#define __constant_cpu_to_le32(x) ((__u32)(x))
+#define __constant_cpu_to_be32(x) (( __u32)___constant_swab32((x)))
+#else
+#define __le32_to_cpu(x) __swab32((x))
+#define __cpu_to_le32(x) __swab32((x))
+#define __be32_to_cpu(x) ((__u32)(x))
+#define __cpu_to_be32(x) ((__u32)(x))
+#define __constant_cpu_to_le32(x) ___constant_swab32((x))
+#define __constant_cpu_to_be32(x) ((__u32)(x))
+#endif
+
+#endif /* _LINUX_CRC32_USER_H */
Index: e2fsprogs-1.39/lib/ext2fs/Makefile.in
===================================================================
--- e2fsprogs-1.39.orig/lib/ext2fs/Makefile.in
+++ e2fsprogs-1.39/lib/ext2fs/Makefile.in
@@ -68,6 +68,7 @@ OBJS= $(DEBUGFS_LIB_OBJS) $(RESIZE_LIB_O
 	unlink.o \
 	valid_blk.o \
 	version.o \
+	crc32.o \
 	crc16.o \
 	csum.o
 
@@ -141,6 +142,7 @@ SRCS= ext2_err.c \
 	$(srcdir)/tst_types.c \
 	$(srcdir)/tst_iscan.c \
 	$(srcdir)/tst_csum.c \
+	$(srcdir)/crc32.c \
 	$(srcdir)/crc16.c \
 	$(srcdir)/csum.c
 
@@ -382,6 +384,8 @@ cmp_bitmaps.o: $(srcdir)/cmp_bitmaps.c $
  $(srcdir)/ext2_io.h $(top_builddir)/lib/ext2fs/ext2_err.h $(srcdir)/bitops.h
 crc16.o: $(srcdir)/crc16.c $(srcdir)/ext2_fs.h $(srcdir)/crc16.h \
  $(top_builddir)/lib/ext2fs/ext2_types.h $(srcdir)/ext2fs.h $(srcdir)/ext2_fs.h
+crc32.o: $(srcdir)/crc32.c $(srcdir)/ext2_fs.h $(srcdir)/crc16.h \
+ $(top_builddir)/lib/ext2fs/ext2_types.h $(srcdir)/ext2fs.h $(srcdir)/ext2_fs.h
 csum.o: $(srcdir)/csum.c $(srcdir)/ext2_fs.h \
  $(top_builddir)/lib/ext2fs/ext2_types.h $(srcdir)/ext2fs.h $(srcdir)/ext2_fs.h
 dblist.o: $(srcdir)/dblist.c $(srcdir)/ext2_fs.h \
Index: e2fsprogs-1.39/e2fsck/recovery.c
===================================================================
--- e2fsprogs-1.39.orig/e2fsck/recovery.c
+++ e2fsprogs-1.39/e2fsck/recovery.c
@@ -21,6 +21,7 @@
 #include <linux/jbd.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
+#include <linux/crc32.h>
 #endif
 
 /*
@@ -304,6 +305,36 @@ int journal_skip_recovery(journal_t *jou
 	return err;
 }
 
+/* cal_chksums calculates the checksums for the blocks described in the
+ * descriptor block.
+ */
+static int cal_chksums(journal_t *journal, struct buffer_head *bh,
+			unsigned long *next_log_block, __u32 *crc32_sum)
+{
+	int i, num_blks, err;
+	unsigned long io_block;
+	struct buffer_head *obh;
+
+	num_blks = count_tags(bh, journal->j_blocksize);
+	/* Calculate checksum of the descriptor block. */
+	*crc32_sum = crc32_be(*crc32_sum, (void *)bh->b_data, bh->b_size);
+	for (i = 0; i < num_blks; i++) {
+		io_block = (*next_log_block)++;
+		wrap(journal, *next_log_block);
+
+		err = jread(&obh, journal, io_block);
+		if (err) {
+			printk (KERN_ERR "JBD: IO error %d recovering block "
+				"%ld in log\n", err, io_block);
+			return 1;
+		} else {
+			*crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data,
+				     obh->b_size);
+		}
+	}
+	return 0;
+}
+
 static int do_one_pass(journal_t *journal,
 			struct recovery_info *info, enum passtype pass)
 {
@@ -315,6 +346,7 @@ static int do_one_pass(journal_t *journa
 	struct buffer_head *	bh;
 	unsigned int		sequence;
 	int			blocktype;
+	__u32			crc32_sum = ~0; /* Transactional Checksums */
 	
 	/* Precompute the maximum metadata descriptors in a descriptor block */
 	int			MAX_BLOCKS_PER_DESC;
@@ -404,9 +436,24 @@ static int do_one_pass(journal_t *journa
 		switch(blocktype) {
 		case JFS_DESCRIPTOR_BLOCK:
 			/* If it is a valid descriptor block, replay it
-			 * in pass REPLAY; otherwise, just skip over the
-			 * blocks it describes. */
+			 * in pass REPLAY; if journal_checksums enabled, then
+			 * calculate checksums in PASS_SCAN, otherwise,
+			 * just skip over the blocks it describes. */
 			if (pass != PASS_REPLAY) {
+				if (pass == PASS_SCAN &&
+				    JFS_HAS_COMPAT_FEATURE(journal,
+					    JFS_FEATURE_COMPAT_CHECKSUM) &&
+				    !info->end_transaction) {
+					if (cal_chksums(journal, bh,
+							&next_log_block,
+							&crc32_sum)) {
+						brelse(bh);
+						break;
+					}
+					brelse(bh);
+					continue;
+				}
+
 				next_log_block +=
 					count_tags(bh, journal->j_blocksize);
 				wrap(journal, next_log_block);
@@ -501,9 +548,79 @@ static int do_one_pass(journal_t *journa
 			continue;
 
 		case JFS_COMMIT_BLOCK:
-			/* Found an expected commit block: not much to
-			 * do other than move on to the next sequence
+			/*     How to differentiate between interrupted commit
+			 *               and journal corruption ?
+			 *
+			 * {nth transaction}
+			 *        Checksum Verification Failed
+			 *			 |
+			 *		 ____________________
+			 *		|		     |
+			 * 	async_commit             sync_commit
+			 *     		|                    |
+			 *		| GO TO NEXT    "Journal Corruption"
+			 *		| TRANSACTION
+			 *		|
+			 * {(n+1)th transanction}
+		         *		|
+			 * 	 _______|______________
+			 * 	|	 	      |
+			 * Commit block found	Commit block not found
+		         *      |		      |
+		         * "Journal Corruption"       |
+		      	 *		 _____________|__________
+       			 *     		|	           	|
+		     	 *	nth trans corrupt	OR   nth trans
+		 	 *	and (n+1)th interrupted     interrupted
+			 *	before commit block
+			 *      could reach the disk.
+			 *	(Cannot find the difference in above
+			 *	 mentioned conditions. Hence assume
+			 *	 "Interrupted Commit".)
+			 */
+
+			/* Found an expected commit block: if checksums
+			 * are present verify them in PASS_SCAN; else not
+			 * much to do other than move on to the next sequence
 			 * number. */
+			if (pass == PASS_SCAN &&
+			    JFS_HAS_COMPAT_FEATURE(journal,
+				    JFS_FEATURE_COMPAT_CHECKSUM)) {
+				struct commit_header *cbh =
+	  				(struct commit_header *)bh->b_data;
+				__u32 found_chksum = ntohl(cbh->h_chksum[0]);
+#if 1
+				printk("Sequence %u, Found Chksum %u, "
+				       "Calculated Chksum %u\n",
+				       sequence,
+			     	       found_chksum,
+		      		       crc32_sum);
+#endif
+
+				if (info->end_transaction) {
+					printk(KERN_ERR "JBD: Transaction %u "
+						"found to be corrupt.\n",
+						next_commit_ID - 1);
+					brelse(bh);
+					break;
+				}
+
+				if (crc32_sum != found_chksum) {
+					info->end_transaction = next_commit_ID;
+
+					if (!JFS_HAS_COMPAT_FEATURE(journal,
+					    JFS_FEATURE_INCOMPAT_ASYNC_COMMIT))
+					{
+						printk(KERN_ERR
+						       "JBD: Transaction %u "
+       						       "found to be corrupt.\n",
+       						       next_commit_ID);
+						brelse(bh);
+						break;
+					}
+				}
+				crc32_sum = ~0;
+			}
 			brelse(bh);
 			next_commit_ID++;
 			continue;
@@ -539,9 +656,10 @@ static int do_one_pass(journal_t *journa
 	 * transaction marks the end of the valid log.
 	 */
 	
-	if (pass == PASS_SCAN)
-		info->end_transaction = next_commit_ID;
-	else {
+	if (pass == PASS_SCAN) {
+ 		if (!info->end_transaction)
+       		       info->end_transaction = next_commit_ID;
+	} else {
 		/* It's really bad news if different passes end up at
 		 * different places (but possible due to IO errors). */
 		if (info->end_transaction != next_commit_ID) {
Index: e2fsprogs-1.39/lib/ext2fs/tst_csum.c
===================================================================
--- e2fsprogs-1.39.orig/lib/ext2fs/tst_csum.c
+++ e2fsprogs-1.39/lib/ext2fs/tst_csum.c
@@ -1,16 +1,17 @@
 /*
- * This testing program verifies checksumming operations
- *
- * Copyright (C) 2006, 2007 by Andreas Dilger <adilger@clusterfs.com>
- *
- * %Begin-Header%
- * This file may be redistributed under the terms of the GNU Public
- * License.
- * %End-Header%
- */
+* This testing program verifies checksumming operations
+*
+* Copyright (C) 2006, 2007 by Andreas Dilger <adilger@clusterfs.com>
+*
+* %Begin-Header%
+* This file may be redistributed under the terms of the GNU Public
+* License.
+* %End-Header%
+*/
 
 #include "ext2fs/ext2_fs.h"
 #include "ext2fs/ext2fs.h"
+#include "ext2fs/crc32.h"
 #include "ext2fs/crc16.h"
 #include "uuid/uuid.h"
 
@@ -64,28 +65,61 @@ int main(int argc, char **argv)
 						    0x4b, 0xae, 0xec, 0xdb } };
 	__u16 csum1, csum2, csum_known = 0xd3a4;
 	char data[8] = { 0x10, 0x20, 0x30, 0x40, 0xf1, 0xb2, 0xc3, 0xd4 };
-	__u16 data_crc[8] =   { 0xcc01, 0x180c, 0x1118, 0xfa10,
-				0x483a, 0x6648, 0x6726, 0x85e6 };
-	__u16 data_crc0[8] =  { 0x8cbe, 0xa80d, 0xd169, 0xde10,
-				0x481e, 0x7d48, 0x673d, 0x8ea6 };
+	__u16 data_crc16[8] =   { 0xcc01, 0x180c, 0x1118, 0xfa10,
+				  0x483a, 0x6648, 0x6726, 0x85e6 };
+	__u16 data_crc16_0[8] =  { 0x8cbe, 0xa80d, 0xd169, 0xde10,
+				   0x481e, 0x7d48, 0x673d, 0x8ea6 };
+	__u32 data_crc32[8] =  {
+				 0x4c11db70, 0x88722df3, 0xe91b93c6, 0xc8756001,
+				 0x839b9c9f, 0x4b6fef27, 0x20eb2a56, 0x7196ddd5
+			       };
+	__u32 data_crc32_0[8] = {
+				  0x21964c4, 0x88c5498e, 0x5e7feec6, 0xf71bd7a,
+				  0xc48b2703, 0x71155355, 0xa1efe310, 0x1892668c
+       				};
 	int i;
 
 	for (i = 0; i < sizeof(data); i++) {
 		csum1 = crc16(0, data, i + 1);
-		printf("crc16(0): data[%d]: %04x=%04x\n", i, csum1,data_crc[i]);
-		if (csum1 != data_crc[i]) {
+		printf("crc16(0): data[%d]: %04x=%04x\n", i, csum1,
+		       data_crc16[i]);
+		if (csum1 != data_crc16[i]) {
 			printf("error: crc16(0) for data[%d] should be %04x\n",
-			       i, data_crc[i]);
+			       i, data_crc16[i]);
 			exit(1);
 		}
 	}
 
 	for (i = 0; i < sizeof(data); i++) {
 		csum1 = crc16(~0, data, i + 1);
-		printf("crc16(~0): data[%d]: %04x=%04x\n",i,csum1,data_crc0[i]);
-		if (csum1 != data_crc0[i]) {
+		printf("crc16(~0): data[%d]: %04x=%04x\n",i,csum1,
+		       data_crc16_0[i]);
+		if (csum1 != data_crc16_0[i]) {
 			printf("error: crc16(~0) for data[%d] should be %04x\n",
-			       i, data_crc0[i]);
+			       i, data_crc16_0[i]);
+			exit(1);
+		}
+	}
+        for (i = 0; i < sizeof(data); i++) {
+		__u32 csum32;
+		csum32 = crc32_be(0, data, i + 1);
+		printf("crc32(0): data[%d]: %04x=%04x\n", i, csum32,
+		       data_crc32[i]);
+		if (csum32 != data_crc32[i]) {
+			printf("error: crc32(0) for data[%d] should be %04x\n",
+			       i, data_crc32[i]);
+			exit(1);
+		}
+	}
+
+	for (i = 0; i < sizeof(data); i++) {
+		__u32 csum32;
+		csum32 = crc32_be(~0U, data, i + 1);
+		printf("crc32(~0): data[%d]: %04x=%04x\n", i, csum32,
+		       data_crc32_0[i]);
+		if (csum32 != data_crc32_0[i]) {
+			printf("error: crc32(~0) for data[%d] should be %04x\n",
+			       i, data_crc32_0[i]);
 			exit(1);
 		}
 	}

Re: Updated patches for journal checksums.

[Alex Tomas]

say, at mount time we fund transaction logged. this means part of it can be
on a disk. what do we do if transaction in the journal is found with wrong
checksum? leave partial transaction in-place?

thanks, Alex

Girish Shilamkar wrote:
> Hi,
> 	The journal checksums patches for ext4 and e2fsprogs sent on 30th May
> had run into problems on big-endian machines. I have made the required
> changes. 
> The major updates are:
> 1. Checksum was written in little endian format in the commit header,
> changed it to big-endian.
> 2. The crc32 code ported to userspace (which is used by e2fsprogs) had
> problem with #ifdef, fixed that.
> 3. The crc32 user space code used i386 assembly code for swabbing
> changed it to generic one.
> 4. Removed various #includes, which caused compilation errors on ppc
> machine.
> 
> Any comments/suggestions are always welcome.
> 
> Thanks,
> Girish.
> 

Re: Updated patches for journal checksums.

[Girish Shilamkar]

On Tue, 2007-06-19 at 12:03 +0400, Alex Tomas wrote:
> say, at mount time we fund transaction logged. this means part of it can be
> on a disk. 
I am not sure I understand this completely. Still I hope the following
answers your question.

> what do we do if transaction in the journal is found with wrong
> checksum? leave partial transaction in-place?
> 
The sanity of the transaction is checked in PASS_SCAN. And if checksum
is found to be incorrect for nth transaction then last transaction which
is written to disk is (n - 1).



Regards,
Girish

Re: Updated patches for journal checksums.

[Andreas Dilger]

On Jun 19, 2007  13:45 +0530, Girish Shilamkar wrote:
> On Tue, 2007-06-19 at 12:03 +0400, Alex Tomas wrote:
> > say, at mount time we fund transaction logged. this means part of it can be
> > on a disk. 
>
> I am not sure I understand this completely. Still I hope the following
> answers your question.

I _think_ Alex is asking "what happens if during a transaction undergoing
checkpoint of blocks to filesystem (not the last one in the journal) is
interrupted by a crash and upon restart the partially-checkpointed
transaction is found to have a checksum error?"

> > what do we do if transaction in the journal is found with wrong
> > checksum? leave partial transaction in-place?
>
> The sanity of the transaction is checked in PASS_SCAN. And if checksum
> is found to be incorrect for nth transaction then last transaction which
> is written to disk is (n - 1).

The recovery.c code (AFAIK) does not do replay for any transaction that
does not have a valid checksum, or transactions beyond that.  If the
bad transaction had already started chekpoint (i.e. isn't the last
committed transaction) then the journal _should_ return an error up to
the filesystem, so it can call ext4_error() at startup.  For e2fsck
(which normally does journal replay & recovery) it can do a full
filesystem check at this point.

Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.

Re: Updated patches for journal checksums.

[Alex Tomas]

Andreas Dilger wrote:
> I _think_ Alex is asking "what happens if during a transaction undergoing
> checkpoint of blocks to filesystem (not the last one in the journal) is
> interrupted by a crash and upon restart the partially-checkpointed
> transaction is found to have a checksum error?"

yup, thanks for clarification.

>>> what do we do if transaction in the journal is found with wrong
>>> checksum? leave partial transaction in-place?
>> The sanity of the transaction is checked in PASS_SCAN. And if checksum
>> is found to be incorrect for nth transaction then last transaction which
>> is written to disk is (n - 1).
> 
> The recovery.c code (AFAIK) does not do replay for any transaction that
> does not have a valid checksum, or transactions beyond that.  If the
> bad transaction had already started chekpoint (i.e. isn't the last
> committed transaction) then the journal _should_ return an error up to
> the filesystem, so it can call ext4_error() at startup.  For e2fsck
> (which normally does journal replay & recovery) it can do a full
> filesystem check at this point.

hmm. it actually can be last transaction (following no activity?)

thanks, Alex

Re: Updated patches for journal checksums.

[Andreas Dilger]

On Jun 19, 2007  13:10 +0400, Alex Tomas wrote:
> Andreas Dilger wrote:
> >I _think_ Alex is asking "what happens if during a transaction undergoing
> >checkpoint of blocks to filesystem (not the last one in the journal) is
> >interrupted by a crash and upon restart the partially-checkpointed
> >transaction is found to have a checksum error?"
> 
> yup, thanks for clarification.
> 
> >>>what do we do if transaction in the journal is found with wrong
> >>>checksum? leave partial transaction in-place?
> >>The sanity of the transaction is checked in PASS_SCAN. And if checksum
> >>is found to be incorrect for nth transaction then last transaction which
> >>is written to disk is (n - 1).
> >
> >The recovery.c code (AFAIK) does not do replay for any transaction that
> >does not have a valid checksum, or transactions beyond that.  If the
> >bad transaction had already started chekpoint (i.e. isn't the last
> >committed transaction) then the journal _should_ return an error up to
> >the filesystem, so it can call ext4_error() at startup.  For e2fsck
> >(which normally does journal replay & recovery) it can do a full
> >filesystem check at this point.
> 
> hmm. it actually can be last transaction (following no activity?)

The current implementation is that if you are running with sync commits
(i.e. the current 2-phase write {data blocks} {wait} {commit block}
{wait}) and you detect a checksum error in the last transaction this
SHOULD result in the filesystem being marked in error (I haven't
verified this is in the most recent version of the code).  This is
what JBD_FEATURE_COMPAT_CHECKSUM does - it only adds checksums to the
commit blocks.

However, in the case of async commits (JBD_FEATURE_INCOMPAT_ASYNC_COMMIT
- write {data blocks} {commit block} {wait}, which is what gives the
performance gain) there isn't any way to tell the difference between a
checksum error in the last transaction and the case of an interrupted
commit.  In the case of an interrupted commit there would not have been
any checkpointing into the filesystem yet, so this case shouldn't matter.

This leaves only the "last transaction, finished async commit, started
checkpoint, crash, corruption, checksum error" case.  I agree this could
cause data corruption.  One possibility is to limit the corruption to
at most a single block by adding checksums to the transaction descriptor
blocks (which already hold the transaction ID to verify they are not
stale) and cover the block numbers.  This would seem to be the only place
that we could get error "fan out" and then we would limit the corruption
to the block(s) that were actually corrupted.  This wouldn't be any
different than having random corruption of disk blocks inside the fs.
This would probably be a separate INCOMPAT flag.


Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.