/*
- * linux/fs/recovery.c
+ * linux/fs/jbd2/recovery.c
*
* Written by Stephen C. Tweedie <sct@redhat.com>, 1999
*
#else
#include <linux/time.h>
#include <linux/fs.h>
-#include <linux/jbd.h>
+#include <linux/jbd2.h>
#include <linux/errno.h>
#include <linux/slab.h>
+#include <linux/crc32.h>
#endif
/*
{
int err;
unsigned int max, nbufs, next;
- unsigned long blocknr;
+ unsigned long long blocknr;
struct buffer_head *bh;
struct buffer_head * bufs[MAXBUF];
nbufs = 0;
for (next = start; next < max; next++) {
- err = journal_bmap(journal, next, &blocknr);
+ err = jbd2_journal_bmap(journal, next, &blocknr);
if (err) {
printk (KERN_ERR "JBD: bad block at offset %u\n",
unsigned int offset)
{
int err;
- unsigned long blocknr;
+ unsigned long long blocknr;
struct buffer_head *bh;
*bhp = NULL;
return -EIO;
}
- err = journal_bmap(journal, offset, &blocknr);
+ err = jbd2_journal_bmap(journal, offset, &blocknr);
if (err) {
printk (KERN_ERR "JBD: bad block at offset %u\n",
* Count the number of in-use tags in a journal descriptor block.
*/
-static int count_tags(struct buffer_head *bh, int size)
+static int count_tags(journal_t *journal, struct buffer_head *bh)
{
char * tagp;
journal_block_tag_t * tag;
- int nr = 0;
+ int nr = 0, size = journal->j_blocksize;
+ int tag_bytes = journal_tag_bytes(journal);
tagp = &bh->b_data[sizeof(journal_header_t)];
- while ((tagp - bh->b_data + sizeof(journal_block_tag_t)) <= size) {
+ while ((tagp - bh->b_data + tag_bytes) <= size) {
tag = (journal_block_tag_t *) tagp;
nr++;
- tagp += sizeof(journal_block_tag_t);
- if (!(tag->t_flags & cpu_to_be32(JFS_FLAG_SAME_UUID)))
+ tagp += tag_bytes;
+ if (!(tag->t_flags & cpu_to_be32(JBD2_FLAG_SAME_UUID)))
tagp += 16;
- if (tag->t_flags & cpu_to_be32(JFS_FLAG_LAST_TAG))
+ if (tag->t_flags & cpu_to_be32(JBD2_FLAG_LAST_TAG))
break;
}
} while (0)
/**
- * journal_recover - recovers a on-disk journal
+ * jbd2_journal_recover - recovers a on-disk journal
* @journal: the journal to recover
*
* The primary function for recovering the log contents when mounting a
* blocks. In the third and final pass, we replay any un-revoked blocks
* in the log.
*/
-int journal_recover(journal_t *journal)
+int jbd2_journal_recover(journal_t *journal)
{
- int err;
+ int err, err2;
journal_superblock_t * sb;
struct recovery_info info;
if (!err)
err = do_one_pass(journal, &info, PASS_REPLAY);
- jbd_debug(0, "JBD: recovery, exit status %d, "
+ jbd_debug(1, "JBD: recovery, exit status %d, "
"recovered transactions %u to %u\n",
err, info.start_transaction, info.end_transaction);
- jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n",
+ jbd_debug(1, "JBD: Replayed %d and revoked %d/%d blocks\n",
info.nr_replays, info.nr_revoke_hits, info.nr_revokes);
/* Restart the log at the next transaction ID, thus invalidating
* any existing commit records in the log. */
journal->j_transaction_sequence = ++info.end_transaction;
- journal_clear_revoke(journal);
- sync_blockdev(journal->j_fs_dev);
+ jbd2_journal_clear_revoke(journal);
+ err2 = sync_blockdev(journal->j_fs_dev);
+ if (!err)
+ err = err2;
+
return err;
}
/**
- * journal_skip_recovery - Start journal and wipe exiting records
+ * jbd2_journal_skip_recovery - Start journal and wipe exiting records
* @journal: journal to startup
*
* Locate any valid recovery information from the journal and set up the
* much recovery information is being erased, and to let us initialise
* the journal transaction sequence numbers to the next unused ID.
*/
-int journal_skip_recovery(journal_t *journal)
+int jbd2_journal_skip_recovery(journal_t *journal)
{
int err;
journal_superblock_t * sb;
printk(KERN_ERR "JBD: error %d scanning journal\n", err);
++journal->j_transaction_sequence;
} else {
-#ifdef CONFIG_JBD_DEBUG
+#ifdef CONFIG_JBD2_DEBUG
int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence);
#endif
- jbd_debug(0,
+ jbd_debug(1,
"JBD: ignoring %d transaction%s from the journal.\n",
dropped, (dropped == 1) ? "" : "s");
journal->j_transaction_sequence = ++info.end_transaction;
return err;
}
+static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag_t *tag)
+{
+ unsigned long long block = be32_to_cpu(tag->t_blocknr);
+ if (tag_bytes > JBD2_TAG_SIZE32)
+ block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32;
+ return block;
+}
+
+/*
+ * calc_chksums calculates the checksums for the blocks described in the
+ * descriptor block.
+ */
+static int calc_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 "
+ "%lu in log\n", err, io_block);
+ return 1;
+ } else {
+ *crc32_sum = crc32_be(*crc32_sum, (void *)obh->b_data,
+ obh->b_size);
+ }
+ put_bh(obh);
+ }
+ return 0;
+}
+
static int do_one_pass(journal_t *journal,
struct recovery_info *info, enum passtype pass)
{
struct buffer_head * bh;
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;
MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t))
- / sizeof(journal_block_tag_t));
+ / tag_bytes);
/*
* First thing is to establish what we expect to find in the log
struct buffer_head * obh;
struct buffer_head * nbh;
- cond_resched(); /* We're under lock_kernel() */
+ cond_resched();
/* If we already know where to stop the log traversal,
* check right now that we haven't gone past the end of
tmp = (journal_header_t *)bh->b_data;
- if (tmp->h_magic != cpu_to_be32(JFS_MAGIC_NUMBER)) {
+ if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) {
brelse(bh);
break;
}
* to do with it? That depends on the pass... */
switch(blocktype) {
- case JFS_DESCRIPTOR_BLOCK:
+ 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) {
- next_log_block +=
- count_tags(bh, journal->j_blocksize);
+ if (pass == PASS_SCAN &&
+ JBD2_HAS_COMPAT_FEATURE(journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM) &&
+ !info->end_transaction) {
+ if (calc_chksums(journal, bh,
+ &next_log_block,
+ &crc32_sum)) {
+ put_bh(bh);
+ break;
+ }
+ put_bh(bh);
+ continue;
+ }
+ next_log_block += count_tags(journal, bh);
wrap(journal, next_log_block);
- brelse(bh);
+ put_bh(bh);
continue;
}
* getting done here! */
tagp = &bh->b_data[sizeof(journal_header_t)];
- while ((tagp - bh->b_data +sizeof(journal_block_tag_t))
+ while ((tagp - bh->b_data + tag_bytes)
<= journal->j_blocksize) {
unsigned long io_block;
"block %ld in log\n",
err, io_block);
} else {
- unsigned long blocknr;
+ unsigned long long blocknr;
J_ASSERT(obh != NULL);
- blocknr = be32_to_cpu(tag->t_blocknr);
+ blocknr = read_tag_block(tag_bytes,
+ tag);
/* If the block has been
* revoked, then we're all done
* here. */
- if (journal_test_revoke
+ if (jbd2_journal_test_revoke
(journal, blocknr,
next_commit_ID)) {
brelse(obh);
lock_buffer(nbh);
memcpy(nbh->b_data, obh->b_data,
journal->j_blocksize);
- if (flags & JFS_FLAG_ESCAPE) {
- *((__be32 *)bh->b_data) =
- cpu_to_be32(JFS_MAGIC_NUMBER);
+ if (flags & JBD2_FLAG_ESCAPE) {
+ *((__be32 *)nbh->b_data) =
+ cpu_to_be32(JBD2_MAGIC_NUMBER);
}
BUFFER_TRACE(nbh, "marking dirty");
}
skip_write:
- tagp += sizeof(journal_block_tag_t);
- if (!(flags & JFS_FLAG_SAME_UUID))
+ tagp += tag_bytes;
+ if (!(flags & JBD2_FLAG_SAME_UUID))
tagp += 16;
- if (flags & JFS_FLAG_LAST_TAG)
+ if (flags & JBD2_FLAG_LAST_TAG)
break;
}
brelse(bh);
continue;
- case JFS_COMMIT_BLOCK:
- /* Found an expected commit block: not much to
- * do other than move on to the next sequence
+ case JBD2_COMMIT_BLOCK:
+ /* 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)) {
+ int chksum_err, chksum_seen;
+ struct commit_header *cbh =
+ (struct commit_header *)bh->b_data;
+ unsigned found_chksum =
+ be32_to_cpu(cbh->h_chksum[0]);
+
+ chksum_err = chksum_seen = 0;
+
+ if (info->end_transaction) {
+ journal->j_failed_commit =
+ info->end_transaction;
+ brelse(bh);
+ break;
+ }
+
+ if (crc32_sum == found_chksum &&
+ cbh->h_chksum_type == JBD2_CRC32_CHKSUM &&
+ cbh->h_chksum_size ==
+ JBD2_CRC32_CHKSUM_SIZE)
+ chksum_seen = 1;
+ else if (!(cbh->h_chksum_type == 0 &&
+ cbh->h_chksum_size == 0 &&
+ found_chksum == 0 &&
+ !chksum_seen))
+ /*
+ * If fs is mounted using an old kernel and then
+ * kernel with journal_chksum is used then we
+ * get a situation where the journal flag has
+ * checksum flag set but checksums are not
+ * present i.e chksum = 0, in the individual
+ * commit blocks.
+ * Hence to avoid checksum failures, in this
+ * situation, this extra check is added.
+ */
+ chksum_err = 1;
+
+ if (chksum_err) {
+ info->end_transaction = next_commit_ID;
+
+ if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
+ JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)){
+ journal->j_failed_commit =
+ next_commit_ID;
+ brelse(bh);
+ break;
+ }
+ }
+ crc32_sum = ~0;
+ }
brelse(bh);
next_commit_ID++;
continue;
- case JFS_REVOKE_BLOCK:
+ case JBD2_REVOKE_BLOCK:
/* If we aren't in the REVOKE pass, then we can
* just skip over this block. */
if (pass != PASS_REVOKE) {
* 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) {
static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
tid_t sequence, struct recovery_info *info)
{
- journal_revoke_header_t *header;
+ jbd2_journal_revoke_header_t *header;
int offset, max;
+ int record_len = 4;
- header = (journal_revoke_header_t *) bh->b_data;
- offset = sizeof(journal_revoke_header_t);
+ header = (jbd2_journal_revoke_header_t *) bh->b_data;
+ offset = sizeof(jbd2_journal_revoke_header_t);
max = be32_to_cpu(header->r_count);
- while (offset < max) {
- unsigned long blocknr;
+ if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT))
+ record_len = 8;
+
+ while (offset + record_len <= max) {
+ unsigned long long blocknr;
int err;
- blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset)));
- offset += 4;
- err = journal_set_revoke(journal, blocknr, sequence);
+ if (record_len == 4)
+ blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset)));
+ else
+ blocknr = be64_to_cpu(* ((__be64 *) (bh->b_data+offset)));
+ offset += record_len;
+ err = jbd2_journal_set_revoke(journal, blocknr, sequence);
if (err)
return err;
++info->nr_revokes;