bp->b_io_length = nbblks;
bp->b_error = 0;
- if (XFS_FORCED_SHUTDOWN(log->l_mp))
- return -EIO;
-
- xfs_buf_iorequest(bp);
- error = xfs_buf_iowait(bp);
- if (error)
+ error = xfs_buf_submit_wait(bp);
+ if (error && !XFS_FORCED_SHUTDOWN(log->l_mp))
xfs_buf_ioerror_alert(bp, __func__);
return error;
}
* We're not going to bother about retrying
* this during recovery. One strike!
*/
- xfs_buf_ioerror_alert(bp, __func__);
- xfs_force_shutdown(bp->b_target->bt_mount,
- SHUTDOWN_META_IO_ERROR);
+ if (!XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) {
+ xfs_buf_ioerror_alert(bp, __func__);
+ xfs_force_shutdown(bp->b_target->bt_mount,
+ SHUTDOWN_META_IO_ERROR);
+ }
}
bp->b_iodone = NULL;
- xfs_buf_ioend(bp, 0);
+ xfs_buf_ioend(bp);
}
/*
******************************************************************************
*/
-STATIC xlog_recover_t *
-xlog_recover_find_tid(
- struct hlist_head *head,
- xlog_tid_t tid)
-{
- xlog_recover_t *trans;
-
- hlist_for_each_entry(trans, head, r_list) {
- if (trans->r_log_tid == tid)
- return trans;
- }
- return NULL;
-}
-
-STATIC void
-xlog_recover_new_tid(
- struct hlist_head *head,
- xlog_tid_t tid,
- xfs_lsn_t lsn)
-{
- xlog_recover_t *trans;
-
- trans = kmem_zalloc(sizeof(xlog_recover_t), KM_SLEEP);
- trans->r_log_tid = tid;
- trans->r_lsn = lsn;
- INIT_LIST_HEAD(&trans->r_itemq);
-
- INIT_HLIST_NODE(&trans->r_list);
- hlist_add_head(&trans->r_list, head);
-}
-
-STATIC void
-xlog_recover_add_item(
- struct list_head *head)
-{
- xlog_recover_item_t *item;
-
- item = kmem_zalloc(sizeof(xlog_recover_item_t), KM_SLEEP);
- INIT_LIST_HEAD(&item->ri_list);
- list_add_tail(&item->ri_list, head);
-}
-
-STATIC int
-xlog_recover_add_to_cont_trans(
- struct xlog *log,
- struct xlog_recover *trans,
- xfs_caddr_t dp,
- int len)
-{
- xlog_recover_item_t *item;
- xfs_caddr_t ptr, old_ptr;
- int old_len;
-
- if (list_empty(&trans->r_itemq)) {
- /* finish copying rest of trans header */
- xlog_recover_add_item(&trans->r_itemq);
- ptr = (xfs_caddr_t) &trans->r_theader +
- sizeof(xfs_trans_header_t) - len;
- memcpy(ptr, dp, len); /* d, s, l */
- return 0;
- }
- /* take the tail entry */
- item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
-
- old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
- old_len = item->ri_buf[item->ri_cnt-1].i_len;
-
- ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP);
- memcpy(&ptr[old_len], dp, len); /* d, s, l */
- item->ri_buf[item->ri_cnt-1].i_len += len;
- item->ri_buf[item->ri_cnt-1].i_addr = ptr;
- trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
- return 0;
-}
-
-/*
- * The next region to add is the start of a new region. It could be
- * a whole region or it could be the first part of a new region. Because
- * of this, the assumption here is that the type and size fields of all
- * format structures fit into the first 32 bits of the structure.
- *
- * This works because all regions must be 32 bit aligned. Therefore, we
- * either have both fields or we have neither field. In the case we have
- * neither field, the data part of the region is zero length. We only have
- * a log_op_header and can throw away the header since a new one will appear
- * later. If we have at least 4 bytes, then we can determine how many regions
- * will appear in the current log item.
- */
-STATIC int
-xlog_recover_add_to_trans(
- struct xlog *log,
- struct xlog_recover *trans,
- xfs_caddr_t dp,
- int len)
-{
- xfs_inode_log_format_t *in_f; /* any will do */
- xlog_recover_item_t *item;
- xfs_caddr_t ptr;
-
- if (!len)
- return 0;
- if (list_empty(&trans->r_itemq)) {
- /* we need to catch log corruptions here */
- if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
- xfs_warn(log->l_mp, "%s: bad header magic number",
- __func__);
- ASSERT(0);
- return -EIO;
- }
- if (len == sizeof(xfs_trans_header_t))
- xlog_recover_add_item(&trans->r_itemq);
- memcpy(&trans->r_theader, dp, len); /* d, s, l */
- return 0;
- }
-
- ptr = kmem_alloc(len, KM_SLEEP);
- memcpy(ptr, dp, len);
- in_f = (xfs_inode_log_format_t *)ptr;
-
- /* take the tail entry */
- item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
- if (item->ri_total != 0 &&
- item->ri_total == item->ri_cnt) {
- /* tail item is in use, get a new one */
- xlog_recover_add_item(&trans->r_itemq);
- item = list_entry(trans->r_itemq.prev,
- xlog_recover_item_t, ri_list);
- }
-
- if (item->ri_total == 0) { /* first region to be added */
- if (in_f->ilf_size == 0 ||
- in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
- xfs_warn(log->l_mp,
- "bad number of regions (%d) in inode log format",
- in_f->ilf_size);
- ASSERT(0);
- kmem_free(ptr);
- return -EIO;
- }
-
- item->ri_total = in_f->ilf_size;
- item->ri_buf =
- kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
- KM_SLEEP);
- }
- ASSERT(item->ri_total > item->ri_cnt);
- /* Description region is ri_buf[0] */
- item->ri_buf[item->ri_cnt].i_addr = ptr;
- item->ri_buf[item->ri_cnt].i_len = len;
- item->ri_cnt++;
- trace_xfs_log_recover_item_add(log, trans, item, 0);
- return 0;
-}
-
/*
* Sort the log items in the transaction.
*
return 0;
}
-/*
- * Free up any resources allocated by the transaction
- *
- * Remember that EFIs, EFDs, and IUNLINKs are handled later.
- */
-STATIC void
-xlog_recover_free_trans(
- struct xlog_recover *trans)
-{
- xlog_recover_item_t *item, *n;
- int i;
-
- list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) {
- /* Free the regions in the item. */
- list_del(&item->ri_list);
- for (i = 0; i < item->ri_cnt; i++)
- kmem_free(item->ri_buf[i].i_addr);
- /* Free the item itself */
- kmem_free(item->ri_buf);
- kmem_free(item);
- }
- /* Free the transaction recover structure */
- kmem_free(trans);
-}
-
STATIC void
xlog_recover_buffer_ra_pass2(
struct xlog *log,
if (!list_empty(&done_list))
list_splice_init(&done_list, &trans->r_itemq);
- xlog_recover_free_trans(trans);
-
error2 = xfs_buf_delwri_submit(&buffer_list);
return error ? error : error2;
}
+STATIC void
+xlog_recover_add_item(
+ struct list_head *head)
+{
+ xlog_recover_item_t *item;
+
+ item = kmem_zalloc(sizeof(xlog_recover_item_t), KM_SLEEP);
+ INIT_LIST_HEAD(&item->ri_list);
+ list_add_tail(&item->ri_list, head);
+}
+
+STATIC int
+xlog_recover_add_to_cont_trans(
+ struct xlog *log,
+ struct xlog_recover *trans,
+ xfs_caddr_t dp,
+ int len)
+{
+ xlog_recover_item_t *item;
+ xfs_caddr_t ptr, old_ptr;
+ int old_len;
+
+ if (list_empty(&trans->r_itemq)) {
+ /* finish copying rest of trans header */
+ xlog_recover_add_item(&trans->r_itemq);
+ ptr = (xfs_caddr_t) &trans->r_theader +
+ sizeof(xfs_trans_header_t) - len;
+ memcpy(ptr, dp, len);
+ return 0;
+ }
+ /* take the tail entry */
+ item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
+
+ old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
+ old_len = item->ri_buf[item->ri_cnt-1].i_len;
+
+ ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP);
+ memcpy(&ptr[old_len], dp, len);
+ item->ri_buf[item->ri_cnt-1].i_len += len;
+ item->ri_buf[item->ri_cnt-1].i_addr = ptr;
+ trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
+ return 0;
+}
+
+/*
+ * The next region to add is the start of a new region. It could be
+ * a whole region or it could be the first part of a new region. Because
+ * of this, the assumption here is that the type and size fields of all
+ * format structures fit into the first 32 bits of the structure.
+ *
+ * This works because all regions must be 32 bit aligned. Therefore, we
+ * either have both fields or we have neither field. In the case we have
+ * neither field, the data part of the region is zero length. We only have
+ * a log_op_header and can throw away the header since a new one will appear
+ * later. If we have at least 4 bytes, then we can determine how many regions
+ * will appear in the current log item.
+ */
+STATIC int
+xlog_recover_add_to_trans(
+ struct xlog *log,
+ struct xlog_recover *trans,
+ xfs_caddr_t dp,
+ int len)
+{
+ xfs_inode_log_format_t *in_f; /* any will do */
+ xlog_recover_item_t *item;
+ xfs_caddr_t ptr;
+
+ if (!len)
+ return 0;
+ if (list_empty(&trans->r_itemq)) {
+ /* we need to catch log corruptions here */
+ if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
+ xfs_warn(log->l_mp, "%s: bad header magic number",
+ __func__);
+ ASSERT(0);
+ return -EIO;
+ }
+ if (len == sizeof(xfs_trans_header_t))
+ xlog_recover_add_item(&trans->r_itemq);
+ memcpy(&trans->r_theader, dp, len);
+ return 0;
+ }
+
+ ptr = kmem_alloc(len, KM_SLEEP);
+ memcpy(ptr, dp, len);
+ in_f = (xfs_inode_log_format_t *)ptr;
+
+ /* take the tail entry */
+ item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
+ if (item->ri_total != 0 &&
+ item->ri_total == item->ri_cnt) {
+ /* tail item is in use, get a new one */
+ xlog_recover_add_item(&trans->r_itemq);
+ item = list_entry(trans->r_itemq.prev,
+ xlog_recover_item_t, ri_list);
+ }
+
+ if (item->ri_total == 0) { /* first region to be added */
+ if (in_f->ilf_size == 0 ||
+ in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
+ xfs_warn(log->l_mp,
+ "bad number of regions (%d) in inode log format",
+ in_f->ilf_size);
+ ASSERT(0);
+ kmem_free(ptr);
+ return -EIO;
+ }
+
+ item->ri_total = in_f->ilf_size;
+ item->ri_buf =
+ kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
+ KM_SLEEP);
+ }
+ ASSERT(item->ri_total > item->ri_cnt);
+ /* Description region is ri_buf[0] */
+ item->ri_buf[item->ri_cnt].i_addr = ptr;
+ item->ri_buf[item->ri_cnt].i_len = len;
+ item->ri_cnt++;
+ trace_xfs_log_recover_item_add(log, trans, item, 0);
+ return 0;
+}
+
+/*
+ * Free up any resources allocated by the transaction
+ *
+ * Remember that EFIs, EFDs, and IUNLINKs are handled later.
+ */
+STATIC void
+xlog_recover_free_trans(
+ struct xlog_recover *trans)
+{
+ xlog_recover_item_t *item, *n;
+ int i;
+
+ list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) {
+ /* Free the regions in the item. */
+ list_del(&item->ri_list);
+ for (i = 0; i < item->ri_cnt; i++)
+ kmem_free(item->ri_buf[i].i_addr);
+ /* Free the item itself */
+ kmem_free(item->ri_buf);
+ kmem_free(item);
+ }
+ /* Free the transaction recover structure */
+ kmem_free(trans);
+}
+
+/*
+ * On error or completion, trans is freed.
+ */
STATIC int
xlog_recovery_process_trans(
struct xlog *log,
unsigned int flags,
int pass)
{
- int error = -EIO;
+ int error = 0;
+ bool freeit = false;
/* mask off ophdr transaction container flags */
flags &= ~XLOG_END_TRANS;
if (flags & XLOG_WAS_CONT_TRANS)
flags &= ~XLOG_CONTINUE_TRANS;
+ /*
+ * Callees must not free the trans structure. We'll decide if we need to
+ * free it or not based on the operation being done and it's result.
+ */
switch (flags) {
/* expected flag values */
case 0:
break;
case XLOG_COMMIT_TRANS:
error = xlog_recover_commit_trans(log, trans, pass);
+ /* success or fail, we are now done with this transaction. */
+ freeit = true;
break;
/* unexpected flag values */
case XLOG_UNMOUNT_TRANS:
+ /* just skip trans */
xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
- error = 0; /* just skip trans */
+ freeit = true;
break;
case XLOG_START_TRANS:
- xfs_warn(log->l_mp, "%s: bad transaction", __func__);
- ASSERT(0);
- break;
default:
xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags);
ASSERT(0);
+ error = -EIO;
break;
}
+ if (error || freeit)
+ xlog_recover_free_trans(trans);
return error;
}
+/*
+ * Lookup the transaction recovery structure associated with the ID in the
+ * current ophdr. If the transaction doesn't exist and the start flag is set in
+ * the ophdr, then allocate a new transaction for future ID matches to find.
+ * Either way, return what we found during the lookup - an existing transaction
+ * or nothing.
+ */
STATIC struct xlog_recover *
xlog_recover_ophdr_to_trans(
struct hlist_head rhash[],
tid = be32_to_cpu(ohead->oh_tid);
rhp = &rhash[XLOG_RHASH(tid)];
- trans = xlog_recover_find_tid(rhp, tid);
- if (trans)
- return trans;
+ hlist_for_each_entry(trans, rhp, r_list) {
+ if (trans->r_log_tid == tid)
+ return trans;
+ }
/*
- * If this is a new transaction, the ophdr only contains the
- * start record. In that case, the only processing we need to do
- * on this opheader is allocate a new recovery container to hold
- * the recovery ops that will follow.
+ * skip over non-start transaction headers - we could be
+ * processing slack space before the next transaction starts
+ */
+ if (!(ohead->oh_flags & XLOG_START_TRANS))
+ return NULL;
+
+ ASSERT(be32_to_cpu(ohead->oh_len) == 0);
+
+ /*
+ * This is a new transaction so allocate a new recovery container to
+ * hold the recovery ops that will follow.
+ */
+ trans = kmem_zalloc(sizeof(struct xlog_recover), KM_SLEEP);
+ trans->r_log_tid = tid;
+ trans->r_lsn = be64_to_cpu(rhead->h_lsn);
+ INIT_LIST_HEAD(&trans->r_itemq);
+ INIT_HLIST_NODE(&trans->r_list);
+ hlist_add_head(&trans->r_list, rhp);
+
+ /*
+ * Nothing more to do for this ophdr. Items to be added to this new
+ * transaction will be in subsequent ophdr containers.
*/
- if (ohead->oh_flags & XLOG_START_TRANS) {
- ASSERT(be32_to_cpu(ohead->oh_len) == 0);
- xlog_recover_new_tid(rhp, tid, be64_to_cpu(rhead->h_lsn));
- }
return NULL;
}
int pass)
{
struct xlog_recover *trans;
- int error;
unsigned int len;
/* Do we understand who wrote this op? */
return 0;
}
- error = xlog_recovery_process_trans(log, trans, dp, len,
- ohead->oh_flags, pass);
- if (error)
- xlog_recover_free_trans(trans);
- return error;
+ return xlog_recovery_process_trans(log, trans, dp, len,
+ ohead->oh_flags, pass);
}
/*
}
memset(rhash, 0, sizeof(rhash));
- if (tail_blk <= head_blk) {
- for (blk_no = tail_blk; blk_no < head_blk; ) {
- error = xlog_bread(log, blk_no, hblks, hbp, &offset);
- if (error)
- goto bread_err2;
-
- rhead = (xlog_rec_header_t *)offset;
- error = xlog_valid_rec_header(log, rhead, blk_no);
- if (error)
- goto bread_err2;
-
- /* blocks in data section */
- bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
- error = xlog_bread(log, blk_no + hblks, bblks, dbp,
- &offset);
- if (error)
- goto bread_err2;
-
- error = xlog_unpack_data(rhead, offset, log);
- if (error)
- goto bread_err2;
-
- error = xlog_recover_process_data(log,
- rhash, rhead, offset, pass);
- if (error)
- goto bread_err2;
- blk_no += bblks + hblks;
- }
- } else {
+ blk_no = tail_blk;
+ if (tail_blk > head_blk) {
/*
* Perform recovery around the end of the physical log.
* When the head is not on the same cycle number as the tail,
- * we can't do a sequential recovery as above.
+ * we can't do a sequential recovery.
*/
- blk_no = tail_blk;
while (blk_no < log->l_logBBsize) {
/*
* Check for header wrapping around physical end-of-log
ASSERT(blk_no >= log->l_logBBsize);
blk_no -= log->l_logBBsize;
+ }
- /* read first part of physical log */
- while (blk_no < head_blk) {
- error = xlog_bread(log, blk_no, hblks, hbp, &offset);
- if (error)
- goto bread_err2;
+ /* read first part of physical log */
+ while (blk_no < head_blk) {
+ error = xlog_bread(log, blk_no, hblks, hbp, &offset);
+ if (error)
+ goto bread_err2;
- rhead = (xlog_rec_header_t *)offset;
- error = xlog_valid_rec_header(log, rhead, blk_no);
- if (error)
- goto bread_err2;
+ rhead = (xlog_rec_header_t *)offset;
+ error = xlog_valid_rec_header(log, rhead, blk_no);
+ if (error)
+ goto bread_err2;
- bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
- error = xlog_bread(log, blk_no+hblks, bblks, dbp,
- &offset);
- if (error)
- goto bread_err2;
+ /* blocks in data section */
+ bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
+ error = xlog_bread(log, blk_no+hblks, bblks, dbp,
+ &offset);
+ if (error)
+ goto bread_err2;
- error = xlog_unpack_data(rhead, offset, log);
- if (error)
- goto bread_err2;
+ error = xlog_unpack_data(rhead, offset, log);
+ if (error)
+ goto bread_err2;
- error = xlog_recover_process_data(log, rhash,
- rhead, offset, pass);
- if (error)
- goto bread_err2;
- blk_no += bblks + hblks;
- }
+ error = xlog_recover_process_data(log, rhash,
+ rhead, offset, pass);
+ if (error)
+ goto bread_err2;
+ blk_no += bblks + hblks;
}
bread_err2:
XFS_BUF_UNASYNC(bp);
bp->b_ops = &xfs_sb_buf_ops;
- if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
- xfs_buf_relse(bp);
- return -EIO;
- }
-
- xfs_buf_iorequest(bp);
- error = xfs_buf_iowait(bp);
+ error = xfs_buf_submit_wait(bp);
if (error) {
- xfs_buf_ioerror_alert(bp, __func__);
- ASSERT(0);
+ if (!XFS_FORCED_SHUTDOWN(log->l_mp)) {
+ xfs_buf_ioerror_alert(bp, __func__);
+ ASSERT(0);
+ }
xfs_buf_relse(bp);
return error;
}
return -EINVAL;
}
+ /*
+ * Delay log recovery if the debug hook is set. This is debug
+ * instrumention to coordinate simulation of I/O failures with
+ * log recovery.
+ */
+ if (xfs_globals.log_recovery_delay) {
+ xfs_notice(log->l_mp,
+ "Delaying log recovery for %d seconds.",
+ xfs_globals.log_recovery_delay);
+ msleep(xfs_globals.log_recovery_delay * 1000);
+ }
+
xfs_notice(log->l_mp, "Starting recovery (logdev: %s)",
log->l_mp->m_logname ? log->l_mp->m_logname
: "internal");