From: Christian Brauner brauner@kernel.org
[ Upstream commit 62a2175ddf7e72941868f164b7c1f92e00f213bd ]
The filesystem's freeze/thaw functions can be called from contexts where the holder isn't userspace but the kernel, e.g., during systemd suspend/hibernate. So pass through the freeze/thaw flags from the VFS instead of hard-coding them.
Signed-off-by: Christian Brauner brauner@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org ---
**YES** This commit should be backported to stable kernel trees. **Extensive explanation:** **1. Fixes a real user-affecting bug:** The commit addresses a concrete functional issue where GFS2 filesystems could malfunction during system suspend/hibernate operations. When systemd or other kernel components initiate freeze/thaw operations with `FREEZE_HOLDER_KERNEL`, but GFS2 internally hard-codes `FREEZE_HOLDER_USERSPACE`, this creates a mismatch that can cause freeze/thaw operations to fail or behave incorrectly. **2. Small and contained change:** The fix is minimal and surgical: - Only changes parameter passing through the call chain - No algorithmic or architectural changes - Changes are confined to the GFS2 subsystem - Simply replaces hard-coded `FREEZE_HOLDER_USERSPACE` with the correct `who` parameter **3. Low regression risk:** - The change maintains exact same behavior for userspace-initiated freezes (the most common case) - Only affects the less common kernel-initiated freeze scenarios - No changes to core logic, just parameter propagation - All existing userspace freeze/thaw workflows remain unchanged **4. Critical subsystem impact:** Freeze/thaw operations are fundamental to system suspend/hibernate functionality. A broken implementation can prevent proper system power management, which is a critical feature for laptops and mobile devices. **5. Follows stable tree criteria:** - **Fixes important functionality**: System suspend/hibernate with GFS2 filesystems - **Minimal scope**: Changes only parameter passing, no new features - **Well-contained**: Limited to fs/gfs2/super.c - **Clear purpose**: Explicitly described fix for kernel vs userspace freeze holder mismatch **6. Comparison with reference commits:** This matches the pattern of **Similar Commit #1** and **Similar Commit #5** which were marked YES: - Simple functional fix - Small, targeted change - No architectural modifications - Addresses specific user-visible issue Unlike the NO commits which involved more complex structural changes, cleanups, or major refactoring, this is a straightforward bug fix. **7. Evidence from kernel context:** The examination of `/home/sasha/linux/include/linux/fs.h` confirms that `FREEZE_HOLDER_KERNEL` is a legitimate and expected freeze holder type used by kernel components, making this fix necessary for correct operation.
fs/gfs2/super.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 44e5658b896c8..a3fe95e519cbc 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -674,7 +674,7 @@ static int gfs2_sync_fs(struct super_block *sb, int wait) return sdp->sd_log_error; }
-static int gfs2_do_thaw(struct gfs2_sbd *sdp) +static int gfs2_do_thaw(struct gfs2_sbd *sdp, enum freeze_holder who) { struct super_block *sb = sdp->sd_vfs; int error; @@ -682,7 +682,7 @@ static int gfs2_do_thaw(struct gfs2_sbd *sdp) error = gfs2_freeze_lock_shared(sdp); if (error) goto fail; - error = thaw_super(sb, FREEZE_HOLDER_USERSPACE); + error = thaw_super(sb, who); if (!error) return 0;
@@ -710,7 +710,7 @@ void gfs2_freeze_func(struct work_struct *work) gfs2_freeze_unlock(sdp); set_bit(SDF_FROZEN, &sdp->sd_flags);
- error = gfs2_do_thaw(sdp); + error = gfs2_do_thaw(sdp, FREEZE_HOLDER_USERSPACE); if (error) goto out;
@@ -728,6 +728,7 @@ void gfs2_freeze_func(struct work_struct *work) /** * gfs2_freeze_super - prevent further writes to the filesystem * @sb: the VFS structure for the filesystem + * @who: freeze flags * */
@@ -744,7 +745,7 @@ static int gfs2_freeze_super(struct super_block *sb, enum freeze_holder who) }
for (;;) { - error = freeze_super(sb, FREEZE_HOLDER_USERSPACE); + error = freeze_super(sb, who); if (error) { fs_info(sdp, "GFS2: couldn't freeze filesystem: %d\n", error); @@ -758,7 +759,7 @@ static int gfs2_freeze_super(struct super_block *sb, enum freeze_holder who) break; }
- error = gfs2_do_thaw(sdp); + error = gfs2_do_thaw(sdp, who); if (error) goto out;
@@ -796,6 +797,7 @@ static int gfs2_freeze_fs(struct super_block *sb) /** * gfs2_thaw_super - reallow writes to the filesystem * @sb: the VFS structure for the filesystem + * @who: freeze flags * */
@@ -814,7 +816,7 @@ static int gfs2_thaw_super(struct super_block *sb, enum freeze_holder who) atomic_inc(&sb->s_active); gfs2_freeze_unlock(sdp);
- error = gfs2_do_thaw(sdp); + error = gfs2_do_thaw(sdp, who);
if (!error) { clear_bit(SDF_FREEZE_INITIATOR, &sdp->sd_flags);