summaryrefslogtreecommitdiff
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/libxfs/xfs_shared.h1
-rw-r--r--fs/xfs/libxfs/xfs_trans_space.h2
-rw-r--r--fs/xfs/scrub/bmap.c22
-rw-r--r--fs/xfs/xfs_bmap_util.c18
-rw-r--r--fs/xfs/xfs_qm.c1
-rw-r--r--fs/xfs/xfs_reflink.c21
-rw-r--r--fs/xfs/xfs_sysfs.h6
-rw-r--r--fs/xfs/xfs_trans.c19
-rw-r--r--fs/xfs/xfs_trans_dquot.c2
9 files changed, 69 insertions, 23 deletions
diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index c45acbd3add9..708feb8eac76 100644
--- a/fs/xfs/libxfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
@@ -65,6 +65,7 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp,
#define XFS_TRANS_DQ_DIRTY 0x10 /* at least one dquot in trx dirty */
#define XFS_TRANS_RESERVE 0x20 /* OK to use reserved data blocks */
#define XFS_TRANS_NO_WRITECOUNT 0x40 /* do not elevate SB writecount */
+#define XFS_TRANS_RES_FDBLKS 0x80 /* reserve newly freed blocks */
/*
* LOWMODE is used by the allocator to activate the lowspace algorithm - when
* free space is running low the extent allocator may choose to allocate an
diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h
index 88221c7a04cc..c6df01a2a158 100644
--- a/fs/xfs/libxfs/xfs_trans_space.h
+++ b/fs/xfs/libxfs/xfs_trans_space.h
@@ -57,7 +57,7 @@
XFS_DAREMOVE_SPACE_RES(mp, XFS_DATA_FORK)
#define XFS_IALLOC_SPACE_RES(mp) \
(M_IGEO(mp)->ialloc_blks + \
- (xfs_sb_version_hasfinobt(&mp->m_sb) ? 2 : 1 * \
+ ((xfs_sb_version_hasfinobt(&mp->m_sb) ? 2 : 1) * \
(M_IGEO(mp)->inobt_maxlevels - 1)))
/*
diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c
index 7badd6dfe544..955302e7cdde 100644
--- a/fs/xfs/scrub/bmap.c
+++ b/fs/xfs/scrub/bmap.c
@@ -45,9 +45,27 @@ xchk_setup_inode_bmap(
*/
if (S_ISREG(VFS_I(sc->ip)->i_mode) &&
sc->sm->sm_type == XFS_SCRUB_TYPE_BMBTD) {
+ struct address_space *mapping = VFS_I(sc->ip)->i_mapping;
+
inode_dio_wait(VFS_I(sc->ip));
- error = filemap_write_and_wait(VFS_I(sc->ip)->i_mapping);
- if (error)
+
+ /*
+ * Try to flush all incore state to disk before we examine the
+ * space mappings for the data fork. Leave accumulated errors
+ * in the mapping for the writer threads to consume.
+ *
+ * On ENOSPC or EIO writeback errors, we continue into the
+ * extent mapping checks because write failures do not
+ * necessarily imply anything about the correctness of the file
+ * metadata. The metadata and the file data could be on
+ * completely separate devices; a media failure might only
+ * affect a subset of the disk, etc. We can handle delalloc
+ * extents in the scrubber, so leaving them in memory is fine.
+ */
+ error = filemap_fdatawrite(mapping);
+ if (!error)
+ error = filemap_fdatawait_keep_errors(mapping);
+ if (error && (error != -ENOSPC && error != -EIO))
goto out;
}
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index f37f5cc4b19f..afdc7f8e0e70 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -1567,6 +1567,7 @@ xfs_swap_extents(
int lock_flags;
uint64_t f;
int resblks = 0;
+ unsigned int flags = 0;
/*
* Lock the inodes against other IO, page faults and truncate to
@@ -1630,17 +1631,16 @@ xfs_swap_extents(
resblks += XFS_SWAP_RMAP_SPACE_RES(mp, tipnext, w);
/*
- * Handle the corner case where either inode might straddle the
- * btree format boundary. If so, the inode could bounce between
- * btree <-> extent format on unmap -> remap cycles, freeing and
- * allocating a bmapbt block each time.
+ * If either inode straddles a bmapbt block allocation boundary,
+ * the rmapbt algorithm triggers repeated allocs and frees as
+ * extents are remapped. This can exhaust the block reservation
+ * prematurely and cause shutdown. Return freed blocks to the
+ * transaction reservation to counter this behavior.
*/
- if (ipnext == (XFS_IFORK_MAXEXT(ip, w) + 1))
- resblks += XFS_IFORK_MAXEXT(ip, w);
- if (tipnext == (XFS_IFORK_MAXEXT(tip, w) + 1))
- resblks += XFS_IFORK_MAXEXT(tip, w);
+ flags |= XFS_TRANS_RES_FDBLKS;
}
- error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp);
+ error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, flags,
+ &tp);
if (error)
goto out_unlock;
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index d6cd83317344..938023dd8ce5 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -148,6 +148,7 @@ xfs_qm_dqpurge(
error = xfs_bwrite(bp);
xfs_buf_relse(bp);
} else if (error == -EAGAIN) {
+ dqp->dq_flags &= ~XFS_DQ_FREEING;
goto out_unlock;
}
xfs_dqflock(dqp);
diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
index 107bf2a2f344..d89201d40891 100644
--- a/fs/xfs/xfs_reflink.c
+++ b/fs/xfs/xfs_reflink.c
@@ -1003,6 +1003,7 @@ xfs_reflink_remap_extent(
xfs_filblks_t rlen;
xfs_filblks_t unmap_len;
xfs_off_t newlen;
+ int64_t qres;
int error;
unmap_len = irec->br_startoff + irec->br_blockcount - destoff;
@@ -1025,13 +1026,19 @@ xfs_reflink_remap_extent(
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
- /* If we're not just clearing space, then do we have enough quota? */
- if (real_extent) {
- error = xfs_trans_reserve_quota_nblks(tp, ip,
- irec->br_blockcount, 0, XFS_QMOPT_RES_REGBLKS);
- if (error)
- goto out_cancel;
- }
+ /*
+ * Reserve quota for this operation. We don't know if the first unmap
+ * in the dest file will cause a bmap btree split, so we always reserve
+ * at least enough blocks for that split. If the extent being mapped
+ * in is written, we need to reserve quota for that too.
+ */
+ qres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
+ if (real_extent)
+ qres += irec->br_blockcount;
+ error = xfs_trans_reserve_quota_nblks(tp, ip, qres, 0,
+ XFS_QMOPT_RES_REGBLKS);
+ if (error)
+ goto out_cancel;
trace_xfs_reflink_remap(ip, irec->br_startoff,
irec->br_blockcount, irec->br_startblock);
diff --git a/fs/xfs/xfs_sysfs.h b/fs/xfs/xfs_sysfs.h
index e9f810fc6731..43585850f154 100644
--- a/fs/xfs/xfs_sysfs.h
+++ b/fs/xfs/xfs_sysfs.h
@@ -32,9 +32,11 @@ xfs_sysfs_init(
struct xfs_kobj *parent_kobj,
const char *name)
{
+ struct kobject *parent;
+
+ parent = parent_kobj ? &parent_kobj->kobject : NULL;
init_completion(&kobj->complete);
- return kobject_init_and_add(&kobj->kobject, ktype,
- &parent_kobj->kobject, "%s", name);
+ return kobject_init_and_add(&kobj->kobject, ktype, parent, "%s", name);
}
static inline void
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 3c94e5ff4316..0ad72a83edac 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -107,7 +107,8 @@ xfs_trans_dup(
ntp->t_flags = XFS_TRANS_PERM_LOG_RES |
(tp->t_flags & XFS_TRANS_RESERVE) |
- (tp->t_flags & XFS_TRANS_NO_WRITECOUNT);
+ (tp->t_flags & XFS_TRANS_NO_WRITECOUNT) |
+ (tp->t_flags & XFS_TRANS_RES_FDBLKS);
/* We gave our writer reference to the new transaction */
tp->t_flags |= XFS_TRANS_NO_WRITECOUNT;
ntp->t_ticket = xfs_log_ticket_get(tp->t_ticket);
@@ -272,6 +273,8 @@ xfs_trans_alloc(
*/
WARN_ON(resp->tr_logres > 0 &&
mp->m_super->s_writers.frozen == SB_FREEZE_COMPLETE);
+ ASSERT(!(flags & XFS_TRANS_RES_FDBLKS) ||
+ xfs_sb_version_haslazysbcount(&mp->m_sb));
tp->t_magic = XFS_TRANS_HEADER_MAGIC;
tp->t_flags = flags;
@@ -365,6 +368,20 @@ xfs_trans_mod_sb(
tp->t_blk_res_used += (uint)-delta;
if (tp->t_blk_res_used > tp->t_blk_res)
xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+ } else if (delta > 0 && (tp->t_flags & XFS_TRANS_RES_FDBLKS)) {
+ int64_t blkres_delta;
+
+ /*
+ * Return freed blocks directly to the reservation
+ * instead of the global pool, being careful not to
+ * overflow the trans counter. This is used to preserve
+ * reservation across chains of transaction rolls that
+ * repeatedly free and allocate blocks.
+ */
+ blkres_delta = min_t(int64_t, delta,
+ UINT_MAX - tp->t_blk_res);
+ tp->t_blk_res += blkres_delta;
+ delta -= blkres_delta;
}
tp->t_fdblocks_delta += delta;
if (xfs_sb_version_haslazysbcount(&mp->m_sb))
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index c0f73b82c055..ed0ce8b301b4 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -647,7 +647,7 @@ xfs_trans_dqresv(
}
}
if (ninos > 0) {
- total_count = be64_to_cpu(dqp->q_core.d_icount) + ninos;
+ total_count = dqp->q_res_icount + ninos;
timer = be32_to_cpu(dqp->q_core.d_itimer);
warns = be16_to_cpu(dqp->q_core.d_iwarns);
warnlimit = defq->iwarnlimit;
generated by cgit v1.2.3 (git 2.39.0) at 2025-04-30 11:22:07 +0300