* FAILED: patch "[PATCH] ocfs2: validate move extents against a stable EOF" failed to apply to 5.10-stable tree
@ 2026-11-10 1:27 Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 1/3] ocfs2: miscellaneous spelling fixes Sasha Levin
0 siblings, 1 reply; 4+ messages in thread
From: Sasha Levin @ 2026-11-10 1:27 UTC (permalink / raw)
To: zzzccc427, akpm, joseph.qi, mark, jlbec, heming.zhao,
baijiaju1990, jjzuming
Cc: stable
The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x 84db6619cb702c37cf848b450c21b9c9706b0775
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '20261109202709-failed-5.10-84db6619cb70@kernel.org' --subject-prefix 'PATCH 5.10.y' 'HEAD^..'
Thanks,
Sasha
------------------ original commit in Linus's tree ------------------
From 84db6619cb702c37cf848b450c21b9c9706b0775 Mon Sep 17 00:00:00 2001
From: Cen Zhang <zzzccc427@gmail.com>
Date: Mon, 26 Oct 2026 19:04:00 +0800
Subject: [PATCH] ocfs2: validate move extents against a stable EOF
The move-extents range must be bounded by the inode size used for extent
movement. ocfs2_ioctl_move_extents() checks and clamps the request before
taking the local inode and cluster locks in ocfs2_move_extents(). It also
uses separate i_size_read() calls for the range check and subtraction.
A concurrent truncate can invalidate the first check and make the later
unsigned subtraction underflow:
Move-extents ioctl Truncate
------------------ --------
check me_start <= i_size
take inode and cluster locks
shrink EOF below me_start
release locks
compare me_start + me_len with EOF
set me_len = EOF - me_start
[unsigned underflow]
acquire inode and cluster locks
select extents using the bad range
A truncate after clamping can also leave the range extending beyond EOF.
Conversely, the cached size may be smaller than the size refreshed when
the cluster inode lock is acquired. Early validation can then reject a
valid start or permanently shorten the requested length and threshold.
The me_start + me_len comparison can also wrap for a large input length.
Validate and adjust the original request once, after inode_lock(), the
exclusive rw lock and ocfs2_inode_lock() have stabilized and refreshed the
size. Read EOF once, reject a start beyond it, and bound the length by
EOF minus start. Normalize the threshold using the same EOF snapshot.
This keeps zero as the default-threshold sentinel until validation uses
the size that will govern extent movement.
Move goal validation after range adjustment as well. Its allocation-group
capacity check uses me_len and must see the EOF-clamped length. Both
validation failures use the existing inode-lock unwind path before any
extent movement or transaction starts.
Range errors now reach the ioctl's normal copyout path, as goal errors
already did. Unlike the old early rejection, a failed copyout can
therefore override -EINVAL with -EFAULT.
Link: https://lore.kernel.org/pm-ocfs2-move-extents-isize-rrw-v6-395bdee84ea81149068e@gmail.com
Fixes: 53069d4e7695 ("Ocfs2/move_extents: move/defrag extents within a certain range.")
Signed-off-by: Cen Zhang <zzzccc427@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Suggested-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Assisted-by: LLM
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Heming Zhao <heming.zhao@suse.com>
Cc: Jia-Ju Bai <baijiaju1990@gmail.com>
Cc: <jjzuming@gmail.com>
Cc: <stable@vger.kernel.org>
---
fs/ocfs2/move_extents.c | 70 +++++++++++++++++++++++------------------
1 file changed, 39 insertions(+), 31 deletions(-)
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index ad1678ee7cc46..93b45531bf7d8 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -907,6 +907,30 @@ static int __ocfs2_move_extents_range(struct buffer_head *di_bh,
return ret;
}
+static int ocfs2_validate_and_adjust_move_range(struct inode *inode,
+ struct ocfs2_move_extents *range)
+{
+ u64 i_size = i_size_read(inode);
+
+ if (range->me_start > i_size)
+ return -EINVAL;
+
+ if (range->me_len > i_size - range->me_start)
+ range->me_len = i_size - range->me_start;
+
+ /*
+ * The default threshold for defragmentation is 1 MiB, which
+ * matches the maximum OCFS2 cluster size.
+ */
+ if (!range->me_threshold)
+ range->me_threshold = 1024 * 1024;
+
+ if (range->me_threshold > i_size)
+ range->me_threshold = i_size;
+
+ return 0;
+}
+
static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
{
int status;
@@ -936,6 +960,21 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
goto out_rw_unlock;
}
+ status = ocfs2_validate_and_adjust_move_range(inode, context->range);
+ if (status) {
+ mlog_errno(status);
+ goto out_inode_unlock;
+ }
+
+ /* The goal check depends on the adjusted range length. */
+ if (!context->auto_defrag) {
+ status = ocfs2_validate_and_adjust_move_goal(inode, context->range);
+ if (status) {
+ mlog_errno(status);
+ goto out_inode_unlock;
+ }
+ }
+
/*
* remember ip_xattr_sem also needs to be held if necessary
*/
@@ -1028,27 +1067,8 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
goto out_free;
}
- if (range.me_start > i_size_read(inode)) {
- status = -EINVAL;
- goto out_free;
- }
-
- if (range.me_start + range.me_len > i_size_read(inode))
- range.me_len = i_size_read(inode) - range.me_start;
-
context->range = ⦥
- /*
- * ok, the default threshold for the defragmentation
- * is 1M, since our maximum clustersize was 1M also.
- * any thought?
- */
- if (!range.me_threshold)
- range.me_threshold = 1024 * 1024;
-
- if (range.me_threshold > i_size_read(inode))
- range.me_threshold = i_size_read(inode);
-
if (range.me_flags & ~(OCFS2_MOVE_EXT_FL_AUTO_DEFRAG |
OCFS2_MOVE_EXT_FL_PART_DEFRAG)) {
status = -EINVAL;
@@ -1060,23 +1080,11 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
if (range.me_flags & OCFS2_MOVE_EXT_FL_PART_DEFRAG)
context->partial = 1;
- } else {
- /*
- * first best-effort attempt to validate and adjust the goal
- * (physical address in block), while it can't guarantee later
- * operation can succeed all the time since global_bitmap may
- * change a bit over time.
- */
-
- status = ocfs2_validate_and_adjust_move_goal(inode, &range);
- if (status)
- goto out_copy;
}
status = ocfs2_move_extents(context);
if (status)
mlog_errno(status);
-out_copy:
/*
* movement/defragmentation may end up being partially completed,
* that's the reason why we need to return userspace the finished
--
2.53.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH 5.10.y 1/3] ocfs2: miscellaneous spelling fixes
2026-11-10 1:27 FAILED: patch "[PATCH] ocfs2: validate move extents against a stable EOF" failed to apply to 5.10-stable tree Sasha Levin
@ 2026-11-15 15:00 ` Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 2/3] ocfs2: add extra flags check in ocfs2_ioctl_move_extents() Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 3/3] ocfs2: validate move extents against a stable EOF Sasha Levin
0 siblings, 2 replies; 4+ messages in thread
From: Sasha Levin @ 2026-11-15 15:00 UTC (permalink / raw)
To: stable
Cc: Dmitry Antipov, Joseph Qi, Mark Fasheh, Joel Becker, Junxiao Bi,
Changwei Ge, Jun Piao, Andrew Morton, Sasha Levin
From: Dmitry Antipov <dmantipov@yandex.ru>
[ Upstream commit a0f8a9a9636e19a11ab80db4a45fb80a230c6c5a ]
Correct spelling here and there as suggested by codespell.
Link: https://lkml.kernel.org/r/20241115151013.1404929-1-dmantipov@yandex.ru
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Acked-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Jun Piao <piaojun@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Stable-dep-of: 84db6619cb70 ("ocfs2: validate move extents against a stable EOF")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/ocfs2/alloc.c | 10 +++++-----
fs/ocfs2/aops.c | 2 +-
fs/ocfs2/cluster/heartbeat.c | 2 +-
fs/ocfs2/cluster/masklog.h | 2 +-
fs/ocfs2/cluster/quorum.c | 6 +++---
fs/ocfs2/cluster/tcp.c | 8 ++++----
fs/ocfs2/dlm/dlmapi.h | 2 +-
fs/ocfs2/dlm/dlmrecovery.c | 6 +++---
fs/ocfs2/dlmglue.c | 12 ++++++------
fs/ocfs2/inode.c | 4 ++--
fs/ocfs2/ioctl.c | 2 +-
fs/ocfs2/journal.c | 2 +-
fs/ocfs2/move_extents.c | 8 ++++----
fs/ocfs2/ocfs2_fs.h | 8 ++++----
fs/ocfs2/ocfs2_ioctl.h | 2 +-
fs/ocfs2/ocfs2_lockid.h | 2 +-
fs/ocfs2/refcounttree.c | 6 +++---
fs/ocfs2/reservations.h | 4 ++--
fs/ocfs2/stack_o2cb.c | 2 +-
fs/ocfs2/stackglue.h | 2 +-
fs/ocfs2/super.c | 2 +-
fs/ocfs2/xattr.c | 10 +++++-----
22 files changed, 52 insertions(+), 52 deletions(-)
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index cb92a653429e1..84acaabc8e56c 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -568,7 +568,7 @@ static void ocfs2_adjust_rightmost_records(handle_t *handle,
struct ocfs2_path *path,
struct ocfs2_extent_rec *insert_rec);
/*
- * Reset the actual path elements so that we can re-use the structure
+ * Reset the actual path elements so that we can reuse the structure
* to build another path. Generally, this involves freeing the buffer
* heads.
*/
@@ -1177,7 +1177,7 @@ static int ocfs2_add_branch(handle_t *handle,
/*
* If there is a gap before the root end and the real end
- * of the righmost leaf block, we need to remove the gap
+ * of the rightmost leaf block, we need to remove the gap
* between new_cpos and root_end first so that the tree
* is consistent after we add a new branch(it will start
* from new_cpos).
@@ -1233,7 +1233,7 @@ static int ocfs2_add_branch(handle_t *handle,
/* Note: new_eb_bhs[new_blocks - 1] is the guy which will be
* linked with the rest of the tree.
- * conversly, new_eb_bhs[0] is the new bottommost leaf.
+ * conversely, new_eb_bhs[0] is the new bottommost leaf.
*
* when we leave the loop, new_last_eb_blk will point to the
* newest leaf, and next_blkno will point to the topmost extent
@@ -3718,7 +3718,7 @@ static int ocfs2_try_to_merge_extent(handle_t *handle,
* update split_index here.
*
* When the split_index is zero, we need to merge it to the
- * prevoius extent block. It is more efficient and easier
+ * previous extent block. It is more efficient and easier
* if we do merge_right first and merge_left later.
*/
ret = ocfs2_merge_rec_right(path, handle, et, split_rec,
@@ -4523,7 +4523,7 @@ static void ocfs2_figure_contig_type(struct ocfs2_extent_tree *et,
}
/*
- * This should only be called against the righmost leaf extent list.
+ * This should only be called against the rightmost leaf extent list.
*
* ocfs2_figure_appending_type() will figure out whether we'll have to
* insert at the tail of the rightmost leaf.
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index cd54477529a08..ac797b59612ee 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -310,7 +310,7 @@ static int ocfs2_readpage(struct file *file, struct page *page)
}
/*
- * i_size might have just been updated as we grabed the meta lock. We
+ * i_size might have just been updated as we grabbed the meta lock. We
* might now be discovering a truncate that hit on another node.
* block_read_full_page->get_block freaks out if it is asked to read
* beyond the end of a file, so we check here. Callers
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 938425dce6e9a..2b4b0d6019b64 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1032,7 +1032,7 @@ static int o2hb_check_slot(struct o2hb_region *reg,
if (list_empty(&slot->ds_live_item))
goto out;
- /* live nodes only go dead after enough consequtive missed
+ /* live nodes only go dead after enough consecutive missed
* samples.. reset the missed counter whenever we see
* activity */
if (slot->ds_equal_samples >= o2hb_dead_threshold || gen_changed) {
diff --git a/fs/ocfs2/cluster/masklog.h b/fs/ocfs2/cluster/masklog.h
index 446e452ac7a64..d145fde4e633d 100644
--- a/fs/ocfs2/cluster/masklog.h
+++ b/fs/ocfs2/cluster/masklog.h
@@ -31,7 +31,7 @@
* just calling printk() so that this can eventually make its way through
* relayfs along with the debugging messages. Everything else gets KERN_DEBUG.
* The inline tests and macro dance give GCC the opportunity to quite cleverly
- * only emit the appropriage printk() when the caller passes in a constant
+ * only emit the appropriate printk() when the caller passes in a constant
* mask, as is almost always the case.
*
* All this bitmask nonsense is managed from the files under
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c
index 341f62a2f5bc6..6c4c8ba4c3570 100644
--- a/fs/ocfs2/cluster/quorum.c
+++ b/fs/ocfs2/cluster/quorum.c
@@ -25,7 +25,7 @@
* race between when we see a node start heartbeating and when we connect
* to it.
*
- * So nodes that are in this transtion put a hold on the quorum decision
+ * So nodes that are in this transition put a hold on the quorum decision
* with a counter. As they fall out of this transition they drop the count
* and if they're the last, they fire off the decision.
*/
@@ -191,7 +191,7 @@ static void o2quo_clear_hold(struct o2quo_state *qs, u8 node)
}
/* as a node comes up we delay the quorum decision until we know the fate of
- * the connection. the hold will be droped in conn_up or hb_down. it might be
+ * the connection. the hold will be dropped in conn_up or hb_down. it might be
* perpetuated by con_err until hb_down. if we already have a conn, we might
* be dropping a hold that conn_up got. */
void o2quo_hb_up(u8 node)
@@ -258,7 +258,7 @@ void o2quo_hb_still_up(u8 node)
}
/* This is analogous to hb_up. as a node's connection comes up we delay the
- * quorum decision until we see it heartbeating. the hold will be droped in
+ * quorum decision until we see it heartbeating. the hold will be dropped in
* hb_up or hb_down. it might be perpetuated by con_err until hb_down. if
* it's already heartbeating we might be dropping a hold that conn_up got.
* */
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index e02b32e91b048..401a42caf6390 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -7,13 +7,13 @@
*
* ----
*
- * Callers for this were originally written against a very simple synchronus
+ * Callers for this were originally written against a very simple synchronous
* API. This implementation reflects those simple callers. Some day I'm sure
* we'll need to move to a more robust posting/callback mechanism.
*
* Transmit calls pass in kernel virtual addresses and block copying this into
* the socket's tx buffers via a usual blocking sendmsg. They'll block waiting
- * for a failed socket to timeout. TX callers can also pass in a poniter to an
+ * for a failed socket to timeout. TX callers can also pass in a pointer to an
* 'int' which gets filled with an errno off the wire in response to the
* message they send.
*
@@ -102,7 +102,7 @@ static struct socket *o2net_listen_sock;
* o2net_wq. teardown detaches the callbacks before destroying the workqueue.
* quorum work is queued as sock containers are shutdown.. stop_listening
* tears down all the node's sock containers, preventing future shutdowns
- * and queued quroum work, before canceling delayed quorum work and
+ * and queued quorum work, before canceling delayed quorum work and
* destroying the work queue.
*/
static struct workqueue_struct *o2net_wq;
@@ -1419,7 +1419,7 @@ static int o2net_advance_rx(struct o2net_sock_container *sc)
return ret;
}
-/* this work func is triggerd by data ready. it reads until it can read no
+/* this work func is triggered by data ready. it reads until it can read no
* more. it interprets 0, eof, as fatal. if data_ready hits while we're doing
* our work the work struct will be marked and we'll be called again. */
static void o2net_rx_until_empty(struct work_struct *work)
diff --git a/fs/ocfs2/dlm/dlmapi.h b/fs/ocfs2/dlm/dlmapi.h
index 6456c0fbcbb2b..fba3b843c50ec 100644
--- a/fs/ocfs2/dlm/dlmapi.h
+++ b/fs/ocfs2/dlm/dlmapi.h
@@ -122,7 +122,7 @@ struct dlm_lockstatus {
#define LKM_VALBLK 0x00000100 /* lock value block request */
#define LKM_NOQUEUE 0x00000200 /* non blocking request */
#define LKM_CONVERT 0x00000400 /* conversion request */
-#define LKM_NODLCKWT 0x00000800 /* this lock wont deadlock (U) */
+#define LKM_NODLCKWT 0x00000800 /* this lock won't deadlock (U) */
#define LKM_UNLOCK 0x00001000 /* deallocate this lock */
#define LKM_CANCEL 0x00002000 /* cancel conversion request */
#define LKM_DEQALL 0x00004000 /* remove all locks held by proc (U) */
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 3d945a5f731af..db9fd939cf01f 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -216,7 +216,7 @@ void dlm_complete_recovery_thread(struct dlm_ctxt *dlm)
* 1) all recovery threads cluster wide will work on recovering
* ONE node at a time
* 2) negotiate who will take over all the locks for the dead node.
- * thats right... ALL the locks.
+ * that's right... ALL the locks.
* 3) once a new master is chosen, everyone scans all locks
* and moves aside those mastered by the dead guy
* 4) each of these locks should be locked until recovery is done
@@ -1487,7 +1487,7 @@ int dlm_mig_lockres_handler(struct o2net_msg *msg, u32 len, void *data,
* The first one is handled at the end of this function. The
* other two are handled in the worker thread after locks have
* been attached. Yes, we don't wait for purge time to match
- * kref_init. The lockres will still have atleast one ref
+ * kref_init. The lockres will still have at least one ref
* added because it is in the hash __dlm_insert_lockres() */
extra_refs++;
@@ -1753,7 +1753,7 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data,
spin_unlock(&res->spinlock);
}
} else {
- /* put.. incase we are not the master */
+ /* put.. in case we are not the master */
spin_unlock(&res->spinlock);
dlm_lockres_put(res);
}
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 8d51575c348d2..8ae0cd74d6176 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -788,7 +788,7 @@ void ocfs2_lock_res_free(struct ocfs2_lock_res *res)
/*
* Keep a list of processes who have interest in a lockres.
- * Note: this is now only uesed for check recursive cluster locking.
+ * Note: this is now only used for check recursive cluster locking.
*/
static inline void ocfs2_add_holder(struct ocfs2_lock_res *lockres,
struct ocfs2_lock_holder *oh)
@@ -2516,7 +2516,7 @@ int ocfs2_inode_lock_full_nested(struct inode *inode,
* locks while holding a page lock and the downconvert thread which
* blocks dlm lock acquiry while acquiring page locks.
*
- * ** These _with_page variantes are only intended to be called from aop
+ * ** These _with_page variants are only intended to be called from aop
* methods that hold page locks and return a very specific *positive* error
* code that aop methods pass up to the VFS -- test for errors with != 0. **
*
@@ -2614,7 +2614,7 @@ void ocfs2_inode_unlock(struct inode *inode,
}
/*
- * This _tracker variantes are introduced to deal with the recursive cluster
+ * This _tracker variants are introduced to deal with the recursive cluster
* locking issue. The idea is to keep track of a lock holder on the stack of
* the current process. If there's a lock holder on the stack, we know the
* task context is already protected by cluster locking. Currently, they're
@@ -2719,7 +2719,7 @@ void ocfs2_inode_unlock_tracker(struct inode *inode,
struct ocfs2_lock_res *lockres;
lockres = &OCFS2_I(inode)->ip_inode_lockres;
- /* had_lock means that the currect process already takes the cluster
+ /* had_lock means that the current process already takes the cluster
* lock previously.
* If had_lock is 1, we have nothing to do here.
* If had_lock is 0, we will release the lock.
@@ -3819,9 +3819,9 @@ static int ocfs2_unblock_lock(struct ocfs2_super *osb,
* set when the ast is received for an upconvert just before the
* OCFS2_LOCK_BUSY flag is cleared. Now if the fs received a bast
* on the heels of the ast, we want to delay the downconvert just
- * enough to allow the up requestor to do its task. Because this
+ * enough to allow the up requester to do its task. Because this
* lock is in the blocked queue, the lock will be downconverted
- * as soon as the requestor is done with the lock.
+ * as soon as the requester is done with the lock.
*/
if (lockres->l_flags & OCFS2_LOCK_UPCONVERT_FINISHING)
goto leave_requeue;
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index ddbedc9aacf45..a001c4ec89974 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -1157,7 +1157,7 @@ static void ocfs2_clear_inode(struct inode *inode)
dquot_drop(inode);
- /* To preven remote deletes we hold open lock before, now it
+ /* To prevent remote deletes we hold open lock before, now it
* is time to unlock PR and EX open locks. */
ocfs2_open_unlock(inode);
@@ -1583,7 +1583,7 @@ static int ocfs2_filecheck_validate_inode_block(struct super_block *sb,
* Call ocfs2_validate_meta_ecc() first since it has ecc repair
* function, but we should not return error immediately when ecc
* validation fails, because the reason is quite likely the invalid
- * inode number inputed.
+ * inode number inputted.
*/
rc = ocfs2_validate_meta_ecc(sb, bh->b_data, &di->i_check);
if (rc) {
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index b43d6c3586e0a..838c7baf8063e 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -807,7 +807,7 @@ static int ocfs2_get_request_ptr(struct ocfs2_info *info, int idx,
/*
* OCFS2_IOC_INFO handles an array of requests passed from userspace.
*
- * ocfs2_info_handle() recevies a large info aggregation, grab and
+ * ocfs2_info_handle() receives a large info aggregation, grab and
* validate the request count from header, then break it into small
* pieces, later specific handlers can handle them one by one.
*
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index da355eb757bac..cf25621a87898 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1948,7 +1948,7 @@ int ocfs2_mark_dead_nodes(struct ocfs2_super *osb)
/*
* Scan timer should get fired every ORPHAN_SCAN_SCHEDULE_TIMEOUT. Add some
- * randomness to the timeout to minimize multple nodes firing the timer at the
+ * randomness to the timeout to minimize multiple nodes firing the timer at the
* same time.
*/
static inline unsigned long ocfs2_orphan_scan_timeout(void)
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index a16c132188eae..ef5ad7a04f643 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -500,7 +500,7 @@ static int ocfs2_validate_and_adjust_move_goal(struct inode *inode,
bg = (struct ocfs2_group_desc *)gd_bh->b_data;
/*
- * moving goal is not allowd to start with a group desc blok(#0 blk)
+ * moving goal is not allowed to start with a group desc blok(#0 blk)
* let's compromise to the latter cluster.
*/
if (range->me_goal == le64_to_cpu(bg->bg_blkno))
@@ -668,7 +668,7 @@ static int ocfs2_move_extent(struct ocfs2_move_extents_context *context,
/*
* probe the victim cluster group to find a proper
- * region to fit wanted movement, it even will perfrom
+ * region to fit wanted movement, it even will perform
* a best-effort attempt by compromising to a threshold
* around the goal.
*/
@@ -935,7 +935,7 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
}
/*
- * rememer ip_xattr_sem also needs to be held if necessary
+ * remember ip_xattr_sem also needs to be held if necessary
*/
down_write(&OCFS2_I(inode)->ip_alloc_sem);
@@ -1037,7 +1037,7 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
context->range = ⦥
/*
- * ok, the default theshold for the defragmentation
+ * ok, the default threshold for the defragmentation
* is 1M, since our maximum clustersize was 1M also.
* any thought?
*/
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index 19137c6d087bc..6f448065630c2 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -134,7 +134,7 @@
* well as the name of the cluster being joined.
* mount.ocfs2 must pass in a matching stack name.
*
- * If not set, the classic stack will be used. This is compatbile with
+ * If not set, the classic stack will be used. This is compatible with
* all older versions.
*/
#define OCFS2_FEATURE_INCOMPAT_USERSPACE_STACK 0x0080
@@ -145,7 +145,7 @@
/* Support for extended attributes */
#define OCFS2_FEATURE_INCOMPAT_XATTR 0x0200
-/* Support for indexed directores */
+/* Support for indexed directories */
#define OCFS2_FEATURE_INCOMPAT_INDEXED_DIRS 0x0400
/* Metadata checksum and error correction */
@@ -158,7 +158,7 @@
#define OCFS2_FEATURE_INCOMPAT_DISCONTIG_BG 0x2000
/*
- * Incompat bit to indicate useable clusterinfo with stackflags for all
+ * Incompat bit to indicate usable clusterinfo with stackflags for all
* cluster stacks (userspace adnd o2cb). If this bit is set,
* INCOMPAT_USERSPACE_STACK becomes superfluous and thus should not be set.
*/
@@ -1084,7 +1084,7 @@ struct ocfs2_xattr_block {
struct ocfs2_xattr_header xb_header; /* xattr header if this
block contains xattr */
struct ocfs2_xattr_tree_root xb_root;/* xattr tree root if this
- block cotains xattr
+ block contains xattr
tree. */
} xb_attrs;
};
diff --git a/fs/ocfs2/ocfs2_ioctl.h b/fs/ocfs2/ocfs2_ioctl.h
index d7b31734f6be4..b9a7287c3dba7 100644
--- a/fs/ocfs2/ocfs2_ioctl.h
+++ b/fs/ocfs2/ocfs2_ioctl.h
@@ -225,7 +225,7 @@ struct ocfs2_move_extents {
movement less likely
to fail, may make fs
even more fragmented */
-#define OCFS2_MOVE_EXT_FL_COMPLETE (0x00000004) /* Move or defragmenation
+#define OCFS2_MOVE_EXT_FL_COMPLETE (0x00000004) /* Move or defragmentation
completely gets done.
*/
diff --git a/fs/ocfs2/ocfs2_lockid.h b/fs/ocfs2/ocfs2_lockid.h
index b4be84956bc10..c8e123d5d7972 100644
--- a/fs/ocfs2/ocfs2_lockid.h
+++ b/fs/ocfs2/ocfs2_lockid.h
@@ -95,7 +95,7 @@ static char *ocfs2_lock_type_strings[] = {
[OCFS2_LOCK_TYPE_DATA] = "Data",
[OCFS2_LOCK_TYPE_SUPER] = "Super",
[OCFS2_LOCK_TYPE_RENAME] = "Rename",
- /* Need to differntiate from [R]ename.. serializing writes is the
+ /* Need to differentiate from [R]ename.. serializing writes is the
* important job it does, anyway. */
[OCFS2_LOCK_TYPE_RW] = "Write/Read",
[OCFS2_LOCK_TYPE_DENTRY] = "Dentry",
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index c00012b7b250c..540594ae6969f 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -2459,7 +2459,7 @@ static int ocfs2_calc_refcount_meta_credits(struct super_block *sb,
*
* If we will insert a new one, this is easy and only happens
* during adding refcounted flag to the extent, so we don't
- * have a chance of spliting. We just need one record.
+ * have a chance of splitting. We just need one record.
*
* If the refcount rec already exists, that would be a little
* complicated. we may have to:
@@ -2649,11 +2649,11 @@ static inline unsigned int ocfs2_cow_align_length(struct super_block *sb,
/*
* Calculate out the start and number of virtual clusters we need to to CoW.
*
- * cpos is vitual start cluster position we want to do CoW in a
+ * cpos is virtual start cluster position we want to do CoW in a
* file and write_len is the cluster length.
* max_cpos is the place where we want to stop CoW intentionally.
*
- * Normal we will start CoW from the beginning of extent record cotaining cpos.
+ * Normal we will start CoW from the beginning of extent record containing cpos.
* We try to break up extents on boundaries of MAX_CONTIG_BYTES so that we
* get good I/O from the resulting extent tree.
*/
diff --git a/fs/ocfs2/reservations.h b/fs/ocfs2/reservations.h
index 6ac88122896d8..4dfc4fe53924b 100644
--- a/fs/ocfs2/reservations.h
+++ b/fs/ocfs2/reservations.h
@@ -33,7 +33,7 @@ struct ocfs2_alloc_reservation {
#define OCFS2_RESV_FLAG_INUSE 0x01 /* Set when r_node is part of a btree */
#define OCFS2_RESV_FLAG_TMP 0x02 /* Temporary reservation, will be
- * destroyed immedately after use */
+ * destroyed immediately after use */
#define OCFS2_RESV_FLAG_DIR 0x04 /* Reservation is for an unindexed
* directory btree */
@@ -132,7 +132,7 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
/**
* ocfs2_resmap_claimed_bits() - Tell the reservation code that bits were used.
* @resmap: reservations bitmap
- * @resv: optional reservation to recalulate based on new bitmap
+ * @resv: optional reservation to recalculate based on new bitmap
* @cstart: start of allocation in clusters
* @clen: end of allocation in clusters.
*
diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index 19d73998757b5..ce63df14bde24 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -229,7 +229,7 @@ static int o2cb_dlm_lock_status(struct ocfs2_dlm_lksb *lksb)
}
/*
- * o2dlm aways has a "valid" LVB. If the dlm loses track of the LVB
+ * o2dlm always has a "valid" LVB. If the dlm loses track of the LVB
* contents, it will zero out the LVB. Thus the caller can always trust
* the contents.
*/
diff --git a/fs/ocfs2/stackglue.h b/fs/ocfs2/stackglue.h
index e9d26cbeb3b82..a81c0c1b36ec5 100644
--- a/fs/ocfs2/stackglue.h
+++ b/fs/ocfs2/stackglue.h
@@ -212,7 +212,7 @@ struct ocfs2_stack_operations {
struct file_lock *fl);
/*
- * This is an optoinal debugging hook. If provided, the
+ * This is an optional debugging hook. If provided, the
* stack can dump debugging information about this lock.
*/
void (*dump_lksb)(struct ocfs2_dlm_lksb *lksb);
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 91a9b86389160..3c952a682439c 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1874,7 +1874,7 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
osb = OCFS2_SB(sb);
BUG_ON(!osb);
- /* Remove file check sysfs related directores/files,
+ /* Remove file check sysfs related directories/files,
* and wait for the pending file check operations */
ocfs2_filecheck_remove_sysfs(osb);
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index ccec77decefe2..b4221cc8af1e2 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -647,7 +647,7 @@ void ocfs2_calc_xattr_init(struct inode *dir, umode_t mode,
* 256(name) + 80(value) + 16(entry) = 352 bytes,
* The max space of acl xattr taken inline is
* 80(value) + 16(entry) * 2(if directory) = 192 bytes,
- * when blocksize = 512, may reserve one more cluser for
+ * when blocksize = 512, may reserve one more cluster for
* xattr bucket, otherwise reserve one metadata block
* for them is ok.
* If this is a new directory with inline data,
@@ -4462,7 +4462,7 @@ static int cmp_xe_offset(const void *a, const void *b)
/*
* defrag a xattr bucket if we find that the bucket has some
- * holes beteen name/value pairs.
+ * holes between name/value pairs.
* We will move all the name/value pairs to the end of the bucket
* so that we can spare some space for insertion.
*/
@@ -5110,7 +5110,7 @@ static int ocfs2_divide_xattr_cluster(struct inode *inode,
* 2. If cluster_size == bucket_size:
* a) If the previous extent rec has more than one cluster and the insert
* place isn't in the last cluster, copy the entire last cluster to the
- * new one. This time, we don't need to upate the first_bh and header_bh
+ * new one. This time, we don't need to update the first_bh and header_bh
* since they will not be moved into the new cluster.
* b) Otherwise, move the bottom half of the xattrs in the last cluster into
* the new one. And we set the extend flag to zero if the insert place is
@@ -6316,7 +6316,7 @@ struct ocfs2_xattr_reflink {
/*
* Given a xattr header and xe offset,
* return the proper xv and the corresponding bh.
- * xattr in inode, block and xattr tree have different implementaions.
+ * xattr in inode, block and xattr tree have different implementations.
*/
typedef int (get_xattr_value_root)(struct super_block *sb,
struct buffer_head *bh,
@@ -6396,7 +6396,7 @@ static int ocfs2_get_xattr_value_root(struct super_block *sb,
}
/*
- * Lock the meta_ac and caculate how much credits we need for reflink xattrs.
+ * Lock the meta_ac and calculate how much credits we need for reflink xattrs.
* It is only used for inline xattr and xattr block.
*/
static int ocfs2_reflink_lock_xattr_allocators(struct ocfs2_super *osb,
--
2.53.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH 5.10.y 2/3] ocfs2: add extra flags check in ocfs2_ioctl_move_extents()
2026-11-15 15:00 ` [PATCH 5.10.y 1/3] ocfs2: miscellaneous spelling fixes Sasha Levin
@ 2026-11-15 15:00 ` Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 3/3] ocfs2: validate move extents against a stable EOF Sasha Levin
1 sibling, 0 replies; 4+ messages in thread
From: Sasha Levin @ 2026-11-15 15:00 UTC (permalink / raw)
To: stable
Cc: Dmitry Antipov, Joseph Qi, Changwei Ge, Joel Becker, Jun Piao,
Junxiao Bi, Mark Fasheh, Andrew Morton, Sasha Levin
From: Dmitry Antipov <dmantipov@yandex.ru>
[ Upstream commit 41a5e8777093f7946fade3737363ef045303b8dc ]
In 'ocfs2_ioctl_move_extents()', add extra check whether only actually
supported flags are passed via 'ioctl(..., OCFS2_IOC_MOVE_EXT, ...)',
and reject anything beyond OCFS2_MOVE_EXT_FL_AUTO_DEFRAG and
OCFS2_MOVE_EXT_FL_PART_DEFRAG with -EINVAL. In particular,
OCFS2_MOVE_EXT_FL_COMPLETE may be set by the kernel only and
should never be passed from userspace.
Link: https://lkml.kernel.org/r/20251009102349.181126-1-dmantipov@yandex.ru
Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Jun Piao <piaojun@huawei.com>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Mark Fasheh <mark@fasheh.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Stable-dep-of: 84db6619cb70 ("ocfs2: validate move extents against a stable EOF")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/ocfs2/move_extents.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index ef5ad7a04f643..f75fec5e1f6ca 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -1047,6 +1047,12 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
if (range.me_threshold > i_size_read(inode))
range.me_threshold = i_size_read(inode);
+ if (range.me_flags & ~(OCFS2_MOVE_EXT_FL_AUTO_DEFRAG |
+ OCFS2_MOVE_EXT_FL_PART_DEFRAG)) {
+ status = -EINVAL;
+ goto out_free;
+ }
+
if (range.me_flags & OCFS2_MOVE_EXT_FL_AUTO_DEFRAG) {
context->auto_defrag = 1;
--
2.53.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH 5.10.y 3/3] ocfs2: validate move extents against a stable EOF
2026-11-15 15:00 ` [PATCH 5.10.y 1/3] ocfs2: miscellaneous spelling fixes Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 2/3] ocfs2: add extra flags check in ocfs2_ioctl_move_extents() Sasha Levin
@ 2026-11-15 15:00 ` Sasha Levin
1 sibling, 0 replies; 4+ messages in thread
From: Sasha Levin @ 2026-11-15 15:00 UTC (permalink / raw)
To: stable
Cc: Cen Zhang, Andrew Morton, Joseph Qi, Mark Fasheh, Joel Becker,
Heming Zhao, Jia-Ju Bai, jjzuming, Sasha Levin
From: Cen Zhang <zzzccc427@gmail.com>
[ Upstream commit 84db6619cb702c37cf848b450c21b9c9706b0775 ]
The move-extents range must be bounded by the inode size used for extent
movement. ocfs2_ioctl_move_extents() checks and clamps the request before
taking the local inode and cluster locks in ocfs2_move_extents(). It also
uses separate i_size_read() calls for the range check and subtraction.
A concurrent truncate can invalidate the first check and make the later
unsigned subtraction underflow:
Move-extents ioctl Truncate
------------------ --------
check me_start <= i_size
take inode and cluster locks
shrink EOF below me_start
release locks
compare me_start + me_len with EOF
set me_len = EOF - me_start
[unsigned underflow]
acquire inode and cluster locks
select extents using the bad range
A truncate after clamping can also leave the range extending beyond EOF.
Conversely, the cached size may be smaller than the size refreshed when
the cluster inode lock is acquired. Early validation can then reject a
valid start or permanently shorten the requested length and threshold.
The me_start + me_len comparison can also wrap for a large input length.
Validate and adjust the original request once, after inode_lock(), the
exclusive rw lock and ocfs2_inode_lock() have stabilized and refreshed the
size. Read EOF once, reject a start beyond it, and bound the length by
EOF minus start. Normalize the threshold using the same EOF snapshot.
This keeps zero as the default-threshold sentinel until validation uses
the size that will govern extent movement.
Move goal validation after range adjustment as well. Its allocation-group
capacity check uses me_len and must see the EOF-clamped length. Both
validation failures use the existing inode-lock unwind path before any
extent movement or transaction starts.
Range errors now reach the ioctl's normal copyout path, as goal errors
already did. Unlike the old early rejection, a failed copyout can
therefore override -EINVAL with -EFAULT.
Link: https://lore.kernel.org/pm-ocfs2-move-extents-isize-rrw-v6-395bdee84ea81149068e@gmail.com
Fixes: 53069d4e7695 ("Ocfs2/move_extents: move/defrag extents within a certain range.")
Signed-off-by: Cen Zhang <zzzccc427@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Suggested-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Assisted-by: LLM
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Heming Zhao <heming.zhao@suse.com>
Cc: Jia-Ju Bai <baijiaju1990@gmail.com>
Cc: <jjzuming@gmail.com>
Cc: <stable@vger.kernel.org>
A truncate after clamping can also leave the range extending beyond EOF.
Conversely, the cached size may be smaller than the size refreshed when
the cluster inode lock is acquired. Early validation can then reject a
valid start or permanently shorten the requested length and threshold.
The me_start + me_len comparison can also wrap for a large input length.
Validate and adjust the original request once, after inode_lock(), the
exclusive rw lock and ocfs2_inode_lock() have stabilized and refreshed the
size. Read EOF once, reject a start beyond it, and bound the length by
EOF minus start. Normalize the threshold using the same EOF snapshot.
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/ocfs2/move_extents.c | 70 +++++++++++++++++++++++------------------
1 file changed, 39 insertions(+), 31 deletions(-)
diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
index f75fec5e1f6ca..00054d7bd24b7 100644
--- a/fs/ocfs2/move_extents.c
+++ b/fs/ocfs2/move_extents.c
@@ -905,6 +905,30 @@ static int __ocfs2_move_extents_range(struct buffer_head *di_bh,
return ret;
}
+static int ocfs2_validate_and_adjust_move_range(struct inode *inode,
+ struct ocfs2_move_extents *range)
+{
+ u64 i_size = i_size_read(inode);
+
+ if (range->me_start > i_size)
+ return -EINVAL;
+
+ if (range->me_len > i_size - range->me_start)
+ range->me_len = i_size - range->me_start;
+
+ /*
+ * The default threshold for defragmentation is 1 MiB, which
+ * matches the maximum OCFS2 cluster size.
+ */
+ if (!range->me_threshold)
+ range->me_threshold = 1024 * 1024;
+
+ if (range->me_threshold > i_size)
+ range->me_threshold = i_size;
+
+ return 0;
+}
+
static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
{
int status;
@@ -934,6 +958,21 @@ static int ocfs2_move_extents(struct ocfs2_move_extents_context *context)
goto out_rw_unlock;
}
+ status = ocfs2_validate_and_adjust_move_range(inode, context->range);
+ if (status) {
+ mlog_errno(status);
+ goto out_inode_unlock;
+ }
+
+ /* The goal check depends on the adjusted range length. */
+ if (!context->auto_defrag) {
+ status = ocfs2_validate_and_adjust_move_goal(inode, context->range);
+ if (status) {
+ mlog_errno(status);
+ goto out_inode_unlock;
+ }
+ }
+
/*
* remember ip_xattr_sem also needs to be held if necessary
*/
@@ -1026,27 +1065,8 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
goto out_free;
}
- if (range.me_start > i_size_read(inode)) {
- status = -EINVAL;
- goto out_free;
- }
-
- if (range.me_start + range.me_len > i_size_read(inode))
- range.me_len = i_size_read(inode) - range.me_start;
-
context->range = ⦥
- /*
- * ok, the default threshold for the defragmentation
- * is 1M, since our maximum clustersize was 1M also.
- * any thought?
- */
- if (!range.me_threshold)
- range.me_threshold = 1024 * 1024;
-
- if (range.me_threshold > i_size_read(inode))
- range.me_threshold = i_size_read(inode);
-
if (range.me_flags & ~(OCFS2_MOVE_EXT_FL_AUTO_DEFRAG |
OCFS2_MOVE_EXT_FL_PART_DEFRAG)) {
status = -EINVAL;
@@ -1058,23 +1078,11 @@ int ocfs2_ioctl_move_extents(struct file *filp, void __user *argp)
if (range.me_flags & OCFS2_MOVE_EXT_FL_PART_DEFRAG)
context->partial = 1;
- } else {
- /*
- * first best-effort attempt to validate and adjust the goal
- * (physical address in block), while it can't guarantee later
- * operation can succeed all the time since global_bitmap may
- * change a bit over time.
- */
-
- status = ocfs2_validate_and_adjust_move_goal(inode, &range);
- if (status)
- goto out_copy;
}
status = ocfs2_move_extents(context);
if (status)
mlog_errno(status);
-out_copy:
/*
* movement/defragmentation may end up being partially completed,
* that's the reason why we need to return userspace the finished
--
2.53.0
^ permalink raw reply related [flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-11-15 15:00 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-10 1:27 FAILED: patch "[PATCH] ocfs2: validate move extents against a stable EOF" failed to apply to 5.10-stable tree Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 1/3] ocfs2: miscellaneous spelling fixes Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 2/3] ocfs2: add extra flags check in ocfs2_ioctl_move_extents() Sasha Levin
2026-11-15 15:00 ` [PATCH 5.10.y 3/3] ocfs2: validate move extents against a stable EOF Sasha Levin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox