xfs: Ensure we force all busy extents in range to disk
When we search for and find a busy extent during allocation we force the log out to ensure the extent free transaction is on disk before the allocation transaction. The current implementation has a subtle bug in it--it does not handle multiple overlapping ranges. That is, if we free lots of little extents into a single contiguous extent, then allocate the contiguous extent, the busy search code stops searching at the first extent it finds that overlaps the allocated range. It then uses the commit LSN of the transaction to force the log out to. Unfortunately, the other busy ranges might have more recent commit LSNs than the first busy extent that is found, and this results in xfs_alloc_search_busy() returning before all the extent free transactions are on disk for the range being allocated. This can lead to potential metadata corruption or stale data exposure after a crash because log replay won't replay all the extent free transactions that cover the allocation range. Modified-by: Alex Elder <aelder@sgi.com> (Dropped the "found" argument from the xfs_alloc_busysearch trace event.) Signed-off-by: Dave Chinner <david@fromorbit.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
This commit is contained in:
parent
44e08c45cc
commit
fd45e47841
@ -1079,28 +1079,28 @@ TRACE_EVENT(xfs_alloc_unbusy,
|
|||||||
|
|
||||||
TRACE_EVENT(xfs_alloc_busysearch,
|
TRACE_EVENT(xfs_alloc_busysearch,
|
||||||
TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
|
TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
|
||||||
xfs_extlen_t len, int found),
|
xfs_extlen_t len, xfs_lsn_t lsn),
|
||||||
TP_ARGS(mp, agno, agbno, len, found),
|
TP_ARGS(mp, agno, agbno, len, lsn),
|
||||||
TP_STRUCT__entry(
|
TP_STRUCT__entry(
|
||||||
__field(dev_t, dev)
|
__field(dev_t, dev)
|
||||||
__field(xfs_agnumber_t, agno)
|
__field(xfs_agnumber_t, agno)
|
||||||
__field(xfs_agblock_t, agbno)
|
__field(xfs_agblock_t, agbno)
|
||||||
__field(xfs_extlen_t, len)
|
__field(xfs_extlen_t, len)
|
||||||
__field(int, found)
|
__field(xfs_lsn_t, lsn)
|
||||||
),
|
),
|
||||||
TP_fast_assign(
|
TP_fast_assign(
|
||||||
__entry->dev = mp->m_super->s_dev;
|
__entry->dev = mp->m_super->s_dev;
|
||||||
__entry->agno = agno;
|
__entry->agno = agno;
|
||||||
__entry->agbno = agbno;
|
__entry->agbno = agbno;
|
||||||
__entry->len = len;
|
__entry->len = len;
|
||||||
__entry->found = found;
|
__entry->lsn = lsn;
|
||||||
),
|
),
|
||||||
TP_printk("dev %d:%d agno %u agbno %u len %u %s",
|
TP_printk("dev %d:%d agno %u agbno %u len %u force lsn 0x%llx",
|
||||||
MAJOR(__entry->dev), MINOR(__entry->dev),
|
MAJOR(__entry->dev), MINOR(__entry->dev),
|
||||||
__entry->agno,
|
__entry->agno,
|
||||||
__entry->agbno,
|
__entry->agbno,
|
||||||
__entry->len,
|
__entry->len,
|
||||||
__print_symbolic(__entry->found, XFS_BUSY_STATES))
|
__entry->lsn)
|
||||||
);
|
);
|
||||||
|
|
||||||
TRACE_EVENT(xfs_agf,
|
TRACE_EVENT(xfs_agf,
|
||||||
|
@ -2563,43 +2563,41 @@ xfs_alloc_search_busy(xfs_trans_t *tp,
|
|||||||
xfs_mount_t *mp;
|
xfs_mount_t *mp;
|
||||||
xfs_perag_busy_t *bsy;
|
xfs_perag_busy_t *bsy;
|
||||||
xfs_agblock_t uend, bend;
|
xfs_agblock_t uend, bend;
|
||||||
xfs_lsn_t lsn;
|
xfs_lsn_t lsn = 0;
|
||||||
int cnt;
|
int cnt;
|
||||||
|
|
||||||
mp = tp->t_mountp;
|
mp = tp->t_mountp;
|
||||||
|
|
||||||
spin_lock(&mp->m_perag[agno].pagb_lock);
|
spin_lock(&mp->m_perag[agno].pagb_lock);
|
||||||
cnt = mp->m_perag[agno].pagb_count;
|
|
||||||
|
|
||||||
uend = bno + len - 1;
|
uend = bno + len - 1;
|
||||||
|
|
||||||
/* search pagb_list for this slot, skipping open slots */
|
/*
|
||||||
for (bsy = mp->m_perag[agno].pagb_list; cnt; bsy++) {
|
* search pagb_list for this slot, skipping open slots. We have to
|
||||||
|
* search the entire array as there may be multiple overlaps and
|
||||||
|
* we have to get the most recent LSN for the log force to push out
|
||||||
|
* all the transactions that span the range.
|
||||||
|
*/
|
||||||
|
for (cnt = 0; cnt < mp->m_perag[agno].pagb_count; cnt++) {
|
||||||
|
bsy = &mp->m_perag[agno].pagb_list[cnt];
|
||||||
|
if (!bsy->busy_tp)
|
||||||
|
continue;
|
||||||
|
|
||||||
/*
|
bend = bsy->busy_start + bsy->busy_length - 1;
|
||||||
* (start1,length1) within (start2, length2)
|
if (bno > bend || uend < bsy->busy_start)
|
||||||
*/
|
continue;
|
||||||
if (bsy->busy_tp != NULL) {
|
|
||||||
bend = bsy->busy_start + bsy->busy_length - 1;
|
/* (start1,length1) within (start2, length2) */
|
||||||
if ((bno > bend) || (uend < bsy->busy_start)) {
|
if (XFS_LSN_CMP(bsy->busy_tp->t_commit_lsn, lsn) > 0)
|
||||||
cnt--;
|
lsn = bsy->busy_tp->t_commit_lsn;
|
||||||
} else {
|
|
||||||
break;
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
spin_unlock(&mp->m_perag[agno].pagb_lock);
|
||||||
trace_xfs_alloc_busysearch(mp, agno, bno, len, !!cnt);
|
trace_xfs_alloc_busysearch(tp->t_mountp, agno, bno, len, lsn);
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* If a block was found, force the log through the LSN of the
|
* If a block was found, force the log through the LSN of the
|
||||||
* transaction that freed the block
|
* transaction that freed the block
|
||||||
*/
|
*/
|
||||||
if (cnt) {
|
if (lsn)
|
||||||
lsn = bsy->busy_tp->t_commit_lsn;
|
|
||||||
spin_unlock(&mp->m_perag[agno].pagb_lock);
|
|
||||||
xfs_log_force(mp, lsn, XFS_LOG_FORCE|XFS_LOG_SYNC);
|
xfs_log_force(mp, lsn, XFS_LOG_FORCE|XFS_LOG_SYNC);
|
||||||
} else {
|
|
||||||
spin_unlock(&mp->m_perag[agno].pagb_lock);
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
Loading…
Reference in New Issue
Block a user