aboutsummaryrefslogtreecommitdiffstats
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/libxfs/xfs_bmap.c13
-rw-r--r--fs/xfs/scrub/common.h5
-rw-r--r--fs/xfs/scrub/inode_repair.c12
-rw-r--r--fs/xfs/scrub/repair.h11
-rw-r--r--fs/xfs/scrub/scrub.c12
-rw-r--r--fs/xfs/xfs_aops.c41
-rw-r--r--fs/xfs/xfs_buf.c36
-rw-r--r--fs/xfs/xfs_buf.h1
-rw-r--r--fs/xfs/xfs_exchrange.c71
-rw-r--r--fs/xfs/xfs_inode.c7
-rw-r--r--fs/xfs/xfs_iomap.c6
-rw-r--r--fs/xfs/xfs_qm_bhv.c55
-rw-r--r--fs/xfs/xfs_super.c8
13 files changed, 172 insertions, 106 deletions
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 40ad22fb808b..0ef19f1469ec 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3563,12 +3563,12 @@ xfs_bmap_btalloc_at_eof(
int error;
/*
- * If there are already extents in the file, try an exact EOF block
- * allocation to extend the file as a contiguous extent. If that fails,
- * or it's the first allocation in a file, just try for a stripe aligned
- * allocation.
+ * If there are already extents in the file, and xfs_bmap_adjacent() has
+ * given a better blkno, try an exact EOF block allocation to extend the
+ * file as a contiguous extent. If that fails, or it's the first
+ * allocation in a file, just try for a stripe aligned allocation.
*/
- if (ap->offset) {
+ if (ap->eof) {
xfs_extlen_t nextminlen = 0;
/*
@@ -3736,7 +3736,8 @@ xfs_bmap_btalloc_best_length(
int error;
ap->blkno = XFS_INO_TO_FSB(args->mp, ap->ip->i_ino);
- xfs_bmap_adjacent(ap);
+ if (!xfs_bmap_adjacent(ap))
+ ap->eof = false;
/*
* Search for an allocation group with a single extent large enough for
diff --git a/fs/xfs/scrub/common.h b/fs/xfs/scrub/common.h
index bdcd40f0ec74..19877d99f255 100644
--- a/fs/xfs/scrub/common.h
+++ b/fs/xfs/scrub/common.h
@@ -224,7 +224,6 @@ static inline bool xchk_skip_xref(struct xfs_scrub_metadata *sm)
bool xchk_dir_looks_zapped(struct xfs_inode *dp);
bool xchk_pptr_looks_zapped(struct xfs_inode *ip);
-#ifdef CONFIG_XFS_ONLINE_REPAIR
/* Decide if a repair is required. */
static inline bool xchk_needs_repair(const struct xfs_scrub_metadata *sm)
{
@@ -244,10 +243,6 @@ static inline bool xchk_could_repair(const struct xfs_scrub *sc)
return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR) &&
!(sc->flags & XREP_ALREADY_FIXED);
}
-#else
-# define xchk_needs_repair(sc) (false)
-# define xchk_could_repair(sc) (false)
-#endif /* CONFIG_XFS_ONLINE_REPAIR */
int xchk_metadata_inode_forks(struct xfs_scrub *sc);
diff --git a/fs/xfs/scrub/inode_repair.c b/fs/xfs/scrub/inode_repair.c
index 2f641b6d663e..13ff1c933cb8 100644
--- a/fs/xfs/scrub/inode_repair.c
+++ b/fs/xfs/scrub/inode_repair.c
@@ -1055,9 +1055,17 @@ xrep_dinode_check_dfork(
return true;
break;
case S_IFREG:
- if (fmt == XFS_DINODE_FMT_LOCAL)
+ switch (fmt) {
+ case XFS_DINODE_FMT_LOCAL:
return true;
- fallthrough;
+ case XFS_DINODE_FMT_EXTENTS:
+ case XFS_DINODE_FMT_BTREE:
+ case XFS_DINODE_FMT_META_BTREE:
+ break;
+ default:
+ return true;
+ }
+ break;
case S_IFLNK:
case S_IFDIR:
switch (fmt) {
diff --git a/fs/xfs/scrub/repair.h b/fs/xfs/scrub/repair.h
index 823c00d1a502..af0a3a9e5ed9 100644
--- a/fs/xfs/scrub/repair.h
+++ b/fs/xfs/scrub/repair.h
@@ -191,7 +191,16 @@ int xrep_reset_metafile_resv(struct xfs_scrub *sc);
#else
#define xrep_ino_dqattach(sc) (0)
-#define xrep_will_attempt(sc) (false)
+
+/*
+ * When online repair is not built into the kernel, we still want to attempt
+ * the repair so that the stub xrep_attempt below will return EOPNOTSUPP.
+ */
+static inline bool xrep_will_attempt(const struct xfs_scrub *sc)
+{
+ return (sc->sm->sm_flags & XFS_SCRUB_IFLAG_FORCE_REBUILD) ||
+ xchk_needs_repair(sc->sm);
+}
static inline int
xrep_attempt(
diff --git a/fs/xfs/scrub/scrub.c b/fs/xfs/scrub/scrub.c
index 7567dd5cad14..6fa9e3e5bab7 100644
--- a/fs/xfs/scrub/scrub.c
+++ b/fs/xfs/scrub/scrub.c
@@ -149,6 +149,18 @@ xchk_probe(
if (xchk_should_terminate(sc, &error))
return error;
+ /*
+ * If the caller is probing to see if repair works but repair isn't
+ * built into the kernel, return EOPNOTSUPP because that's the signal
+ * that userspace expects. If online repair is built in, set the
+ * CORRUPT flag (without any of the usual tracing/logging) to force us
+ * into xrep_probe.
+ */
+ if (xchk_could_repair(sc)) {
+ if (!IS_ENABLED(CONFIG_XFS_ONLINE_REPAIR))
+ return -EOPNOTSUPP;
+ sc->sm->sm_flags |= XFS_SCRUB_OFLAG_CORRUPT;
+ }
return 0;
}
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 67877c36ed11..6d9965b546cb 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -19,6 +19,7 @@
#include "xfs_reflink.h"
#include "xfs_errortag.h"
#include "xfs_error.h"
+#include "xfs_icache.h"
struct xfs_writepage_ctx {
struct iomap_writepage_ctx ctx;
@@ -528,12 +529,44 @@ xfs_vm_readahead(
}
static int
-xfs_iomap_swapfile_activate(
+xfs_vm_swap_activate(
struct swap_info_struct *sis,
struct file *swap_file,
sector_t *span)
{
- sis->bdev = xfs_inode_buftarg(XFS_I(file_inode(swap_file)))->bt_bdev;
+ struct xfs_inode *ip = XFS_I(file_inode(swap_file));
+
+ /*
+ * Swap file activation can race against concurrent shared extent
+ * removal in files that have been cloned. If this happens,
+ * iomap_swapfile_iter() can fail because it encountered a shared
+ * extent even though an operation is in progress to remove those
+ * shared extents.
+ *
+ * This race becomes problematic when we defer extent removal
+ * operations beyond the end of a syscall (i.e. use async background
+ * processing algorithms). Users think the extents are no longer
+ * shared, but iomap_swapfile_iter() still sees them as shared
+ * because the refcountbt entries for the extents being removed have
+ * not yet been updated. Hence the swapon call fails unexpectedly.
+ *
+ * The race condition is currently most obvious from the unlink()
+ * operation as extent removal is deferred until after the last
+ * reference to the inode goes away. We then process the extent
+ * removal asynchronously, hence triggers the "syscall completed but
+ * work not done" condition mentioned above. To close this race
+ * window, we need to flush any pending inodegc operations to ensure
+ * they have updated the refcountbt records before we try to map the
+ * swapfile.
+ */
+ xfs_inodegc_flush(ip->i_mount);
+
+ /*
+ * Direct the swap code to the correct block device when this file
+ * sits on the RT device.
+ */
+ sis->bdev = xfs_inode_buftarg(ip)->bt_bdev;
+
return iomap_swapfile_activate(sis, swap_file, span,
&xfs_read_iomap_ops);
}
@@ -549,11 +582,11 @@ const struct address_space_operations xfs_address_space_operations = {
.migrate_folio = filemap_migrate_folio,
.is_partially_uptodate = iomap_is_partially_uptodate,
.error_remove_folio = generic_error_remove_folio,
- .swap_activate = xfs_iomap_swapfile_activate,
+ .swap_activate = xfs_vm_swap_activate,
};
const struct address_space_operations xfs_dax_aops = {
.writepages = xfs_dax_writepages,
.dirty_folio = noop_dirty_folio,
- .swap_activate = xfs_iomap_swapfile_activate,
+ .swap_activate = xfs_vm_swap_activate,
};
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index d1d4a0a22e13..15bb790359f8 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -41,8 +41,7 @@ struct kmem_cache *xfs_buf_cache;
*
* xfs_buf_rele:
* b_lock
- * pag_buf_lock
- * lru_lock
+ * lru_lock
*
* xfs_buftarg_drain_rele
* lru_lock
@@ -220,23 +219,25 @@ _xfs_buf_alloc(
*/
flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD);
- spin_lock_init(&bp->b_lock);
+ /*
+ * A new buffer is held and locked by the owner. This ensures that the
+ * buffer is owned by the caller and racing RCU lookups right after
+ * inserting into the hash table are safe (and will have to wait for
+ * the unlock to do anything non-trivial).
+ */
bp->b_hold = 1;
+ sema_init(&bp->b_sema, 0); /* held, no waiters */
+
+ spin_lock_init(&bp->b_lock);
atomic_set(&bp->b_lru_ref, 1);
init_completion(&bp->b_iowait);
INIT_LIST_HEAD(&bp->b_lru);
INIT_LIST_HEAD(&bp->b_list);
INIT_LIST_HEAD(&bp->b_li_list);
- sema_init(&bp->b_sema, 0); /* held, no waiters */
bp->b_target = target;
bp->b_mount = target->bt_mount;
bp->b_flags = flags;
- /*
- * Set length and io_length to the same value initially.
- * I/O routines should use io_length, which will be the same in
- * most cases but may be reset (e.g. XFS recovery).
- */
error = xfs_buf_get_maps(bp, nmaps);
if (error) {
kmem_cache_free(xfs_buf_cache, bp);
@@ -502,7 +503,6 @@ int
xfs_buf_cache_init(
struct xfs_buf_cache *bch)
{
- spin_lock_init(&bch->bc_lock);
return rhashtable_init(&bch->bc_hash, &xfs_buf_hash_params);
}
@@ -652,17 +652,20 @@ xfs_buf_find_insert(
if (error)
goto out_free_buf;
- spin_lock(&bch->bc_lock);
+ /* The new buffer keeps the perag reference until it is freed. */
+ new_bp->b_pag = pag;
+
+ rcu_read_lock();
bp = rhashtable_lookup_get_insert_fast(&bch->bc_hash,
&new_bp->b_rhash_head, xfs_buf_hash_params);
if (IS_ERR(bp)) {
+ rcu_read_unlock();
error = PTR_ERR(bp);
- spin_unlock(&bch->bc_lock);
goto out_free_buf;
}
if (bp && xfs_buf_try_hold(bp)) {
/* found an existing buffer */
- spin_unlock(&bch->bc_lock);
+ rcu_read_unlock();
error = xfs_buf_find_lock(bp, flags);
if (error)
xfs_buf_rele(bp);
@@ -670,10 +673,8 @@ xfs_buf_find_insert(
*bpp = bp;
goto out_free_buf;
}
+ rcu_read_unlock();
- /* The new buffer keeps the perag reference until it is freed. */
- new_bp->b_pag = pag;
- spin_unlock(&bch->bc_lock);
*bpp = new_bp;
return 0;
@@ -1090,7 +1091,6 @@ xfs_buf_rele_cached(
}
/* we are asked to drop the last reference */
- spin_lock(&bch->bc_lock);
__xfs_buf_ioacct_dec(bp);
if (!(bp->b_flags & XBF_STALE) && atomic_read(&bp->b_lru_ref)) {
/*
@@ -1102,7 +1102,6 @@ xfs_buf_rele_cached(
bp->b_state &= ~XFS_BSTATE_DISPOSE;
else
bp->b_hold--;
- spin_unlock(&bch->bc_lock);
} else {
bp->b_hold--;
/*
@@ -1120,7 +1119,6 @@ xfs_buf_rele_cached(
ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
rhashtable_remove_fast(&bch->bc_hash, &bp->b_rhash_head,
xfs_buf_hash_params);
- spin_unlock(&bch->bc_lock);
if (pag)
xfs_perag_put(pag);
freebuf = true;
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 7e73663c5d4a..3b4ed42e11c0 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -80,7 +80,6 @@ typedef unsigned int xfs_buf_flags_t;
#define XFS_BSTATE_IN_FLIGHT (1 << 1) /* I/O in flight */
struct xfs_buf_cache {
- spinlock_t bc_lock;
struct rhashtable bc_hash;
};
diff --git a/fs/xfs/xfs_exchrange.c b/fs/xfs/xfs_exchrange.c
index f340a2015c4c..0b41bdfecdfb 100644
--- a/fs/xfs/xfs_exchrange.c
+++ b/fs/xfs/xfs_exchrange.c
@@ -329,22 +329,6 @@ out_trans_cancel:
* successfully but before locks are dropped.
*/
-/* Verify that we have security clearance to perform this operation. */
-static int
-xfs_exchange_range_verify_area(
- struct xfs_exchrange *fxr)
-{
- int ret;
-
- ret = remap_verify_area(fxr->file1, fxr->file1_offset, fxr->length,
- true);
- if (ret)
- return ret;
-
- return remap_verify_area(fxr->file2, fxr->file2_offset, fxr->length,
- true);
-}
-
/*
* Performs necessary checks before doing a range exchange, having stabilized
* mutable inode attributes via i_rwsem.
@@ -355,11 +339,13 @@ xfs_exchange_range_checks(
unsigned int alloc_unit)
{
struct inode *inode1 = file_inode(fxr->file1);
+ loff_t size1 = i_size_read(inode1);
struct inode *inode2 = file_inode(fxr->file2);
+ loff_t size2 = i_size_read(inode2);
uint64_t allocmask = alloc_unit - 1;
int64_t test_len;
uint64_t blen;
- loff_t size1, size2, tmp;
+ loff_t tmp;
int error;
/* Don't touch certain kinds of inodes */
@@ -368,24 +354,25 @@ xfs_exchange_range_checks(
if (IS_SWAPFILE(inode1) || IS_SWAPFILE(inode2))
return -ETXTBSY;
- size1 = i_size_read(inode1);
- size2 = i_size_read(inode2);
-
/* Ranges cannot start after EOF. */
if (fxr->file1_offset > size1 || fxr->file2_offset > size2)
return -EINVAL;
- /*
- * If the caller said to exchange to EOF, we set the length of the
- * request large enough to cover everything to the end of both files.
- */
if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) {
+ /*
+ * If the caller said to exchange to EOF, we set the length of
+ * the request large enough to cover everything to the end of
+ * both files.
+ */
fxr->length = max_t(int64_t, size1 - fxr->file1_offset,
size2 - fxr->file2_offset);
-
- error = xfs_exchange_range_verify_area(fxr);
- if (error)
- return error;
+ } else {
+ /*
+ * Otherwise we require both ranges to end within EOF.
+ */
+ if (fxr->file1_offset + fxr->length > size1 ||
+ fxr->file2_offset + fxr->length > size2)
+ return -EINVAL;
}
/*
@@ -402,15 +389,6 @@ xfs_exchange_range_checks(
return -EINVAL;
/*
- * We require both ranges to end within EOF, unless we're exchanging
- * to EOF.
- */
- if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF) &&
- (fxr->file1_offset + fxr->length > size1 ||
- fxr->file2_offset + fxr->length > size2))
- return -EINVAL;
-
- /*
* Make sure we don't hit any file size limits. If we hit any size
* limits such that test_length was adjusted, we abort the whole
* operation.
@@ -747,6 +725,7 @@ xfs_exchange_range(
{
struct inode *inode1 = file_inode(fxr->file1);
struct inode *inode2 = file_inode(fxr->file2);
+ loff_t check_len = fxr->length;
int ret;
BUILD_BUG_ON(XFS_EXCHANGE_RANGE_ALL_FLAGS &
@@ -779,14 +758,18 @@ xfs_exchange_range(
return -EBADF;
/*
- * If we're not exchanging to EOF, we can check the areas before
- * stabilizing both files' i_size.
+ * If we're exchanging to EOF we can't calculate the length until taking
+ * the iolock. Pass a 0 length to remap_verify_area similar to the
+ * FICLONE and FICLONERANGE ioctls that support cloning to EOF as well.
*/
- if (!(fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)) {
- ret = xfs_exchange_range_verify_area(fxr);
- if (ret)
- return ret;
- }
+ if (fxr->flags & XFS_EXCHANGE_RANGE_TO_EOF)
+ check_len = 0;
+ ret = remap_verify_area(fxr->file1, fxr->file1_offset, check_len, true);
+ if (ret)
+ return ret;
+ ret = remap_verify_area(fxr->file2, fxr->file2_offset, check_len, true);
+ if (ret)
+ return ret;
/* Update cmtime if the fd/inode don't forbid it. */
if (!(fxr->file1->f_mode & FMODE_NOCMTIME) && !IS_NOCMTIME(inode1))
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index c95fe1b1de4e..b1f9f156ec88 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1404,8 +1404,11 @@ xfs_inactive(
goto out;
/* Try to clean out the cow blocks if there are any. */
- if (xfs_inode_has_cow_data(ip))
- xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+ if (xfs_inode_has_cow_data(ip)) {
+ error = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, true);
+ if (error)
+ goto out;
+ }
if (VFS_I(ip)->i_nlink != 0) {
/*
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 50fa3ef89f6c..d61460309a78 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -976,10 +976,8 @@ xfs_dax_write_iomap_end(
if (!xfs_is_cow_inode(ip))
return 0;
- if (!written) {
- xfs_reflink_cancel_cow_range(ip, pos, length, true);
- return 0;
- }
+ if (!written)
+ return xfs_reflink_cancel_cow_range(ip, pos, length, true);
return xfs_reflink_end_cow(ip, pos, written);
}
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index 37f1230e7584..245d754f382a 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -78,6 +78,28 @@ xfs_qm_statvfs(
}
}
+STATIC int
+xfs_qm_validate_state_change(
+ struct xfs_mount *mp,
+ uint uqd,
+ uint gqd,
+ uint pqd)
+{
+ int state;
+
+ /* Is quota state changing? */
+ state = ((uqd && !XFS_IS_UQUOTA_ON(mp)) ||
+ (!uqd && XFS_IS_UQUOTA_ON(mp)) ||
+ (gqd && !XFS_IS_GQUOTA_ON(mp)) ||
+ (!gqd && XFS_IS_GQUOTA_ON(mp)) ||
+ (pqd && !XFS_IS_PQUOTA_ON(mp)) ||
+ (!pqd && XFS_IS_PQUOTA_ON(mp)));
+
+ return state &&
+ (xfs_dev_is_read_only(mp, "changing quota state") ||
+ xfs_has_norecovery(mp));
+}
+
int
xfs_qm_newmount(
xfs_mount_t *mp,
@@ -97,24 +119,25 @@ xfs_qm_newmount(
}
/*
- * If the device itself is read-only, we can't allow
- * the user to change the state of quota on the mount -
- * this would generate a transaction on the ro device,
- * which would lead to an I/O error and shutdown
+ * If the device itself is read-only and/or in norecovery
+ * mode, we can't allow the user to change the state of
+ * quota on the mount - this would generate a transaction
+ * on the ro device, which would lead to an I/O error and
+ * shutdown.
*/
- if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) ||
- (!uquotaondisk && XFS_IS_UQUOTA_ON(mp)) ||
- (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) ||
- (!gquotaondisk && XFS_IS_GQUOTA_ON(mp)) ||
- (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) ||
- (!pquotaondisk && XFS_IS_PQUOTA_ON(mp))) &&
- xfs_dev_is_read_only(mp, "changing quota state")) {
- xfs_warn(mp, "please mount with%s%s%s%s.",
- (!quotaondisk ? "out quota" : ""),
- (uquotaondisk ? " usrquota" : ""),
- (gquotaondisk ? " grpquota" : ""),
- (pquotaondisk ? " prjquota" : ""));
+ if (xfs_qm_validate_state_change(mp, uquotaondisk,
+ gquotaondisk, pquotaondisk)) {
+
+ if (xfs_has_metadir(mp))
+ xfs_warn(mp,
+ "metadir enabled, please mount without any quota mount options");
+ else
+ xfs_warn(mp, "please mount with%s%s%s%s.",
+ (!quotaondisk ? "out quota" : ""),
+ (uquotaondisk ? " usrquota" : ""),
+ (gquotaondisk ? " grpquota" : ""),
+ (pquotaondisk ? " prjquota" : ""));
return -EPERM;
}
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index d92d7a07ea89..0055066fb1d9 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -1661,8 +1661,12 @@ xfs_fs_fill_super(
#endif
}
- /* Filesystem claims it needs repair, so refuse the mount. */
- if (xfs_has_needsrepair(mp)) {
+ /*
+ * Filesystem claims it needs repair, so refuse the mount unless
+ * norecovery is also specified, in which case the filesystem can
+ * be mounted with no risk of further damage.
+ */
+ if (xfs_has_needsrepair(mp) && !xfs_has_norecovery(mp)) {
xfs_warn(mp, "Filesystem needs repair. Please run xfs_repair.");
error = -EFSCORRUPTED;
goto out_free_sb;
generated by cgit v1.2.3 (git 2.39.1) at 2026-01-03 05:40:21 +0000