Example #1
0
static int count_tags(journal_t *journal, struct buffer_head *bh)
{
	char *			tagp;
	journal_block_tag_t *	tag;
	int			nr = 0, size = journal->j_blocksize;
	int			tag_bytes = journal_tag_bytes(journal);

	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2))
		size -= sizeof(struct jbd2_journal_block_tail);

	tagp = &bh->b_data[sizeof(journal_header_t)];

	while ((tagp - bh->b_data + tag_bytes) <= size) {
		tag = (journal_block_tag_t *) tagp;

		nr++;
		tagp += tag_bytes;
		if (!(tag->t_flags & cpu_to_be16(JBD2_FLAG_SAME_UUID)))
			tagp += 16;

		if (tag->t_flags & cpu_to_be16(JBD2_FLAG_LAST_TAG))
			break;
	}

	return nr;
}
Example #2
0
static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
			       tid_t sequence, struct recovery_info *info)
{
	jbd2_journal_revoke_header_t *header;
	int offset, max;
	int record_len = 4;

	header = (jbd2_journal_revoke_header_t *) bh->b_data;
	offset = sizeof(jbd2_journal_revoke_header_t);
	max = be32_to_cpu(header->r_count);

	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT))
		record_len = 8;

	while (offset + record_len <= max) {
		unsigned long long blocknr;
		int err;

		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;
	}
	return 0;
}
Example #3
0
static void write_one_revoke_record(journal_t *journal,
				    transaction_t *transaction,
				    struct journal_head **descriptorp,
				    int *offsetp,
				    struct jbd2_revoke_record_s *record)
{
	struct journal_head *descriptor;
	int offset;
	journal_header_t *header;

	/* If we are already aborting, this all becomes a noop.  We
           still need to go round the loop in
           jbd2_journal_write_revoke_records in order to free all of the
           revoke records: only the IO to the journal is omitted. */
	if (is_journal_aborted(journal))
		return;

	descriptor = *descriptorp;
	offset = *offsetp;

	/* Make sure we have a descriptor with space left for the record */
	if (descriptor) {
		if (offset == journal->j_blocksize) {
			flush_descriptor(journal, descriptor, offset);
			descriptor = NULL;
		}
	}

	if (!descriptor) {
		descriptor = jbd2_journal_get_descriptor_buffer(journal);
		if (!descriptor)
			return;
		header = (journal_header_t *) &jh2bh(descriptor)->b_data[0];
		header->h_magic     = cpu_to_be32(JBD2_MAGIC_NUMBER);
		header->h_blocktype = cpu_to_be32(JBD2_REVOKE_BLOCK);
		header->h_sequence  = cpu_to_be32(transaction->t_tid);

		/* Record it so that we can wait for IO completion later */
		JBUFFER_TRACE(descriptor, "file as BJ_LogCtl");
		jbd2_journal_file_buffer(descriptor, transaction, BJ_LogCtl);

		offset = sizeof(jbd2_journal_revoke_header_t);
		*descriptorp = descriptor;
	}

	if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) {
		* ((__be64 *)(&jh2bh(descriptor)->b_data[offset])) =
			cpu_to_be64(record->blocknr);
		offset += 8;

	} else {
		* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) =
			cpu_to_be32(record->blocknr);
		offset += 4;
	}

	*offsetp = offset;
}
Example #4
0
static int jbd2_block_tag_csum_verify(journal_t *j, journal_block_tag_t *tag,
				      void *buf, __u32 sequence)
{
	__u32 provided, calculated;

	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
		return 1;

	sequence = cpu_to_be32(sequence);
	calculated = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence,
				 sizeof(sequence));
	calculated = jbd2_chksum(j, calculated, buf, j->j_blocksize);
	provided = be32_to_cpu(tag->t_checksum);

	return provided == cpu_to_be32(calculated);
}
Example #5
0
static int jbd2_commit_block_csum_verify(journal_t *j, void *buf)
{
	struct commit_header *h;
	__u32 provided, calculated;

	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
		return 1;

	h = buf;
	provided = h->h_chksum[0];
	h->h_chksum[0] = 0;
	calculated = jbd2_chksum(j, j->j_csum_seed, buf, j->j_blocksize);
	h->h_chksum[0] = provided;

	provided = be32_to_cpu(provided);
	return provided == calculated;
}
Example #6
0
static int jbd2_revoke_block_csum_verify(journal_t *j,
					 void *buf)
{
	struct jbd2_journal_revoke_tail *tail;
	__u32 provided, calculated;

	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
		return 1;

	tail = (struct jbd2_journal_revoke_tail *)(buf + j->j_blocksize -
			sizeof(struct jbd2_journal_revoke_tail));
	provided = tail->r_checksum;
	tail->r_checksum = 0;
	calculated = jbd2_chksum(j, j->j_csum_seed, buf, j->j_blocksize);
	tail->r_checksum = provided;

	provided = be32_to_cpu(provided);
	return provided == calculated;
}
Example #7
0
static int do_one_pass(journal_t *journal,
			struct recovery_info *info, enum passtype pass)
{
	unsigned int		first_commit_ID, next_commit_ID;
	unsigned long		next_log_block;
	int			err, success = 0;
	journal_superblock_t *	sb;
	journal_header_t *	tmp;
	struct buffer_head *	bh;
	unsigned int		sequence;
	int			blocktype;
	int			tag_bytes = journal_tag_bytes(journal);
	__u32			crc32_sum = ~0; /* Transactional Checksums */

	/*
	 * First thing is to establish what we expect to find in the log
	 * (in terms of transaction IDs), and where (in terms of log
	 * block offsets): query the superblock.
	 */

	sb = journal->j_superblock;
	next_commit_ID = be32_to_cpu(sb->s_sequence);
	next_log_block = be32_to_cpu(sb->s_start);

	first_commit_ID = next_commit_ID;
	if (pass == PASS_SCAN)
		info->start_transaction = first_commit_ID;

	jbd_debug(1, "Starting recovery pass %d\n", pass);

	/*
	 * Now we walk through the log, transaction by transaction,
	 * making sure that each transaction has a commit block in the
	 * expected place.  Each complete transaction gets replayed back
	 * into the main filesystem.
	 */

	while (1) {
		int			flags;
		char *			tagp;
		journal_block_tag_t *	tag;
		struct buffer_head *	obh;
		struct buffer_head *	nbh;

		cond_resched();

		/* If we already know where to stop the log traversal,
		 * check right now that we haven't gone past the end of
		 * the log. */

		if (pass != PASS_SCAN)
			if (tid_geq(next_commit_ID, info->end_transaction))
				break;

		jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n",
			  next_commit_ID, next_log_block, journal->j_last);

		/* Skip over each chunk of the transaction looking
		 * either the next descriptor block or the final commit
		 * record. */

		jbd_debug(3, "JBD2: checking block %ld\n", next_log_block);
		err = jread(&bh, journal, next_log_block);
		if (err)
			goto failed;

		next_log_block++;
		wrap(journal, next_log_block);

		/* What kind of buffer is it?
		 *
		 * If it is a descriptor block, check that it has the
		 * expected sequence number.  Otherwise, we're all done
		 * here. */

		tmp = (journal_header_t *)bh->b_data;

		if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) {
			brelse(bh);
			break;
		}

		blocktype = be32_to_cpu(tmp->h_blocktype);
		sequence = be32_to_cpu(tmp->h_sequence);
		jbd_debug(3, "Found magic %d, sequence %d\n",
			  blocktype, sequence);

		if (sequence != next_commit_ID) {
			brelse(bh);
			break;
		}

		/* OK, we have a valid descriptor block which matches
		 * all of the sequence number checks.  What are we going
		 * to do with it?  That depends on the pass... */

		switch(blocktype) {
		case JBD2_DESCRIPTOR_BLOCK:
			/* If it is a valid descriptor block, replay it
			 * 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 (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);
				put_bh(bh);
				continue;
			}

			/* A descriptor block: we can now write all of
			 * the data blocks.  Yay, useful work is finally
			 * getting done here! */

			tagp = &bh->b_data[sizeof(journal_header_t)];
			while ((tagp - bh->b_data + tag_bytes)
			       <= journal->j_blocksize) {
				unsigned long io_block;

				tag = (journal_block_tag_t *) tagp;
				flags = be32_to_cpu(tag->t_flags);

				io_block = next_log_block++;
				wrap(journal, next_log_block);
				err = jread(&obh, journal, io_block);
				if (err) {
					/* Recover what we can, but
					 * report failure at the end. */
					success = err;
					printk(KERN_ERR
						"JBD2: IO error %d recovering "
						"block %ld in log\n",
						err, io_block);
				} else {
					unsigned long long blocknr;

					J_ASSERT(obh != NULL);
					blocknr = read_tag_block(tag_bytes,
								 tag);

					/* If the block has been
					 * revoked, then we're all done
					 * here. */
					if (jbd2_journal_test_revoke
					    (journal, blocknr,
					     next_commit_ID)) {
						brelse(obh);
						++info->nr_revoke_hits;
						goto skip_write;
					}

					/* Find a buffer for the new
					 * data being restored */
					nbh = __getblk(journal->j_fs_dev,
							blocknr,
							journal->j_blocksize);
					if (nbh == NULL) {
						printk(KERN_ERR
						       "JBD2: Out of memory "
						       "during recovery.\n");
						err = -ENOMEM;
						brelse(bh);
						brelse(obh);
						goto failed;
					}

					lock_buffer(nbh);
					memcpy(nbh->b_data, obh->b_data,
							journal->j_blocksize);
					if (flags & JBD2_FLAG_ESCAPE) {
						*((__be32 *)nbh->b_data) =
						cpu_to_be32(JBD2_MAGIC_NUMBER);
					}

					BUFFER_TRACE(nbh, "marking dirty");
					set_buffer_uptodate(nbh);
					mark_buffer_dirty(nbh);
					BUFFER_TRACE(nbh, "marking uptodate");
					++info->nr_replays;
					/* ll_rw_block(WRITE, 1, &nbh); */
					unlock_buffer(nbh);
					brelse(obh);
					brelse(nbh);
				}

			skip_write:
				tagp += tag_bytes;
				if (!(flags & JBD2_FLAG_SAME_UUID))
					tagp += 16;

				if (flags & JBD2_FLAG_LAST_TAG)
					break;
			}

			brelse(bh);
			continue;

		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 JBD2_REVOKE_BLOCK:
			/* If we aren't in the REVOKE pass, then we can
			 * just skip over this block. */
			if (pass != PASS_REVOKE) {
				brelse(bh);
				continue;
			}

			err = scan_revoke_records(journal, bh,
						  next_commit_ID, info);
			brelse(bh);
			if (err)
				goto failed;
			continue;

		default:
			jbd_debug(3, "Unrecognised magic %d, end of scan.\n",
				  blocktype);
			brelse(bh);
			goto done;
		}
	}

 done:
	/*
	 * We broke out of the log scan loop: either we came to the
	 * known end of the log or we found an unexpected block in the
	 * log.  If the latter happened, then we know that the "current"
	 * transaction marks the end of the valid log.
	 */

	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) {
			printk(KERN_ERR "JBD2: recovery pass %d ended at "
				"transaction %u, expected %u\n",
				pass, next_commit_ID, info->end_transaction);
			if (!success)
				success = -EIO;
		}
	}

	return success;

 failed:
	return err;
}
Example #8
0
static int do_one_pass(journal_t *journal,
			struct recovery_info *info, enum passtype pass)
{
	unsigned int		first_commit_ID, next_commit_ID;
	unsigned long		next_log_block;
	int			err, success = 0;
	journal_superblock_t *	sb;
	journal_header_t *	tmp;
	struct buffer_head *	bh;
	unsigned int		sequence;
	int			blocktype;
	int			tag_bytes = journal_tag_bytes(journal);
	__u32			crc32_sum = ~0; 


	sb = journal->j_superblock;
	next_commit_ID = be32_to_cpu(sb->s_sequence);
	next_log_block = be32_to_cpu(sb->s_start);

	first_commit_ID = next_commit_ID;
	if (pass == PASS_SCAN)
		info->start_transaction = first_commit_ID;

	jbd_debug(1, "Starting recovery pass %d\n", pass);


	while (1) {
		int			flags;
		char *			tagp;
		journal_block_tag_t *	tag;
		struct buffer_head *	obh;
		struct buffer_head *	nbh;

		cond_resched();


		if (pass != PASS_SCAN)
			if (tid_geq(next_commit_ID, info->end_transaction))
				break;

		jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n",
			  next_commit_ID, next_log_block, journal->j_last);


		jbd_debug(3, "JBD2: checking block %ld\n", next_log_block);
		err = jread(&bh, journal, next_log_block);
		if (err)
			goto failed;

		next_log_block++;
		wrap(journal, next_log_block);


		tmp = (journal_header_t *)bh->b_data;

		if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) {
			brelse(bh);
			break;
		}

		blocktype = be32_to_cpu(tmp->h_blocktype);
		sequence = be32_to_cpu(tmp->h_sequence);
		jbd_debug(3, "Found magic %d, sequence %d\n",
			  blocktype, sequence);

		if (sequence != next_commit_ID) {
			brelse(bh);
			break;
		}


		switch(blocktype) {
		case JBD2_DESCRIPTOR_BLOCK:
			if (pass != PASS_REPLAY) {
				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);
				put_bh(bh);
				continue;
			}


			tagp = &bh->b_data[sizeof(journal_header_t)];
			while ((tagp - bh->b_data + tag_bytes)
			       <= journal->j_blocksize) {
				unsigned long io_block;

				tag = (journal_block_tag_t *) tagp;
				flags = be32_to_cpu(tag->t_flags);

				io_block = next_log_block++;
				wrap(journal, next_log_block);
				err = jread(&obh, journal, io_block);
				if (err) {
					success = err;
					printk(KERN_ERR
						"JBD2: IO error %d recovering "
						"block %ld in log\n",
						err, io_block);
				} else {
					unsigned long long blocknr;

					J_ASSERT(obh != NULL);
					blocknr = read_tag_block(tag_bytes,
								 tag);

					if (jbd2_journal_test_revoke
					    (journal, blocknr,
					     next_commit_ID)) {
						brelse(obh);
						++info->nr_revoke_hits;
						goto skip_write;
					}

					nbh = __getblk(journal->j_fs_dev,
							blocknr,
							journal->j_blocksize);
					if (nbh == NULL) {
						printk(KERN_ERR
						       "JBD2: Out of memory "
						       "during recovery.\n");
						err = -ENOMEM;
						brelse(bh);
						brelse(obh);
						goto failed;
					}

					lock_buffer(nbh);
					memcpy(nbh->b_data, obh->b_data,
							journal->j_blocksize);
					if (flags & JBD2_FLAG_ESCAPE) {
						*((__be32 *)nbh->b_data) =
						cpu_to_be32(JBD2_MAGIC_NUMBER);
					}

					BUFFER_TRACE(nbh, "marking dirty");
					set_buffer_uptodate(nbh);
					mark_buffer_dirty(nbh);
					BUFFER_TRACE(nbh, "marking uptodate");
					++info->nr_replays;
					
					unlock_buffer(nbh);
					brelse(obh);
					brelse(nbh);
				}

			skip_write:
				tagp += tag_bytes;
				if (!(flags & JBD2_FLAG_SAME_UUID))
					tagp += 16;

				if (flags & JBD2_FLAG_LAST_TAG)
					break;
			}

			brelse(bh);
			continue;

		case JBD2_COMMIT_BLOCK:

			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))
						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 JBD2_REVOKE_BLOCK:
			if (pass != PASS_REVOKE) {
				brelse(bh);
				continue;
			}

			err = scan_revoke_records(journal, bh,
						  next_commit_ID, info);
			brelse(bh);
			if (err)
				goto failed;
			continue;

		default:
			jbd_debug(3, "Unrecognised magic %d, end of scan.\n",
				  blocktype);
			brelse(bh);
			goto done;
		}
	}

 done:

	if (pass == PASS_SCAN) {
		if (!info->end_transaction)
			info->end_transaction = next_commit_ID;
	} else {
		if (info->end_transaction != next_commit_ID) {
			printk(KERN_ERR "JBD2: recovery pass %d ended at "
				"transaction %u, expected %u\n",
				pass, next_commit_ID, info->end_transaction);
			if (!success)
				success = -EIO;
		}
	}

	return success;

 failed:
	return err;
}