All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev, Filipe Manana <fdmanana@suse.com>,
	David Sterba <dsterba@suse.com>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.17 33/35] btrfs: use smp_mb__after_atomic() when forcing COW in create_pending_snapshot()
Date: Fri, 31 Oct 2025 15:01:41 +0100	[thread overview]
Message-ID: <20251031140044.429085886@linuxfoundation.org> (raw)
In-Reply-To: <20251031140043.564670400@linuxfoundation.org>

6.17-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Filipe Manana <fdmanana@suse.com>

[ Upstream commit 45c222468d33202c07c41c113301a4b9c8451b8f ]

After setting the BTRFS_ROOT_FORCE_COW flag on the root we are doing a
full write barrier, smp_wmb(), but we don't need to, all we need is a
smp_mb__after_atomic().  The use of the smp_wmb() is from the old days
when we didn't use a bit and used instead an int field in the root to
signal if cow is forced. After the int field was changed to a bit in
the root's state (flags field), we forgot to update the memory barrier
in create_pending_snapshot() to smp_mb__after_atomic(), but we did the
change in commit_fs_roots() after clearing BTRFS_ROOT_FORCE_COW. That
happened in commit 27cdeb7096b8 ("Btrfs: use bitfield instead of integer
data type for the some variants in btrfs_root"). On the reader side, in
should_cow_block(), we also use the counterpart smp_mb__before_atomic()
which generates further confusion.

So change the smp_wmb() to smp_mb__after_atomic(). In fact we don't
even need any barrier at all since create_pending_snapshot() is called
in the critical section of a transaction commit and therefore no one
can concurrently join/attach the transaction, or start a new one, until
the transaction is unblocked. By the time someone starts a new transaction
and enters should_cow_block(), a lot of implicit memory barriers already
took place by having acquired several locks such as fs_info->trans_lock
and extent buffer locks on the root node at least. Nevertlheless, for
consistency use smp_mb__after_atomic() after setting the force cow bit
in create_pending_snapshot().

Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/btrfs/transaction.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index c5c0d9cf1a808..a4e486a600bed 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -1806,7 +1806,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	}
 	/* see comments in should_cow_block() */
 	set_bit(BTRFS_ROOT_FORCE_COW, &root->state);
-	smp_wmb();
+	smp_mb__after_atomic();
 
 	btrfs_set_root_node(new_root_item, tmp);
 	/* record when the snapshot was created in key.offset */
-- 
2.51.0




  parent reply	other threads:[~2025-10-31 14:07 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-31 14:01 [PATCH 6.17 00/35] 6.17.7-rc1 review Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 01/35] sched_ext: Move internal type and accessor definitions to ext_internal.h Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 02/35] sched_ext: Put event_stats_cpu in struct scx_sched_pcpu Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 03/35] sched_ext: Sync error_irq_work before freeing scx_sched Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 04/35] timekeeping: Fix aux clocks sysfs initialization loop bound Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 05/35] x86/bugs: Report correct retbleed mitigation status Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 06/35] x86/bugs: Qualify RETBLEED_INTEL_MSG Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 07/35] genirq/chip: Add buslock back in to irq_set_handler() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 08/35] genirq/manage: Add buslock back in to __disable_irq_nosync() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 09/35] genirq/manage: Add buslock back in to enable_irq() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 10/35] audit: record fanotify event regardless of presence of rules Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 11/35] EDAC/ie31200: Add two more Intel Alder Lake-S SoCs for EDAC support Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 12/35] perf/x86/intel: Add ICL_FIXED_0_ADAPTIVE bit into INTEL_FIXED_BITS_MASK Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 13/35] perf: Use current->flags & PF_KTHREAD|PF_USER_WORKER instead of current->mm == NULL Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 14/35] perf: Have get_perf_callchain() return NULL if crosstask and user are set Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 15/35] perf: Skip user unwind if the task is a kernel thread Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 16/35] EDAC: Fix wrong executable file modes for C source files Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 17/35] seccomp: passthrough uprobe systemcall without filtering Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 18/35] sched_ext: Keep bypass on between enable failure and scx_disable_workfn() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 19/35] x86/bugs: Add attack vector controls for VMSCAPE Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 20/35] sched/fair: update_cfs_group() for throttled cfs_rqs Greg Kroah-Hartman
2025-11-02 11:07   ` Aaron Lu
2025-11-02 12:21     ` Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 21/35] x86/bugs: Fix reporting of LFENCE retpoline Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 22/35] EDAC/mc_sysfs: Increase legacy channel support to 16 Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 23/35] cpuset: Use new excpus for nocpu error check when enabling root partition Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 24/35] btrfs: abort transaction on specific error places when walking log tree Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 25/35] btrfs: abort transaction in the process_one_buffer() log tree walk callback Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 26/35] btrfs: zoned: return error from btrfs_zone_finish_endio() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 27/35] btrfs: zoned: refine extent allocator hint selection Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 28/35] btrfs: scrub: replace max_t()/min_t() with clamp() in scrub_throttle_dev_io() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 29/35] btrfs: always drop log root tree reference in btrfs_replay_log() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 30/35] btrfs: use level argument in log tree walk callback replay_one_buffer() Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 31/35] btrfs: abort transaction if we fail to update inode in log replay dir fixup Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 32/35] btrfs: tree-checker: add inode extref checks Greg Kroah-Hartman
2025-10-31 14:01 ` Greg Kroah-Hartman [this message]
2025-10-31 14:01 ` [PATCH 6.17 34/35] sched_ext: Make qmap dump operation non-destructive Greg Kroah-Hartman
2025-10-31 14:01 ` [PATCH 6.17 35/35] arch: Add the macro COMPILE_OFFSETS to all the asm-offsets.c Greg Kroah-Hartman
2025-10-31 14:58 ` [PATCH 6.17 00/35] 6.17.7-rc1 review Ronald Warsow
2025-10-31 16:59 ` Peter Schneider
2025-10-31 17:06 ` Dileep malepu
2025-10-31 19:35 ` Jon Hunter
2025-10-31 20:39 ` Pavel Machek
2025-10-31 22:35 ` Shuah Khan
2025-10-31 22:45 ` Achill Gilgenast
2025-10-31 22:58 ` Justin Forbes
2025-11-01  9:10 ` Naresh Kamboju
2025-11-01  9:56 ` Jeffrin Thalakkottoor
2025-11-01 11:37 ` Ron Economos
2025-11-01 19:31 ` Brett A C Sheffield
2025-11-01 21:16 ` Miguel Ojeda
2025-11-02  2:58 ` Takeshi Ogasawara
2025-11-03 16:50 ` Florian Fainelli

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251031140044.429085886@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=dsterba@suse.com \
    --cc=fdmanana@suse.com \
    --cc=patches@lists.linux.dev \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.