gfs2: mark the VFS superblock read-only before quota cleanup
PATCH 1/1 gfs2: mark the VFS superblock read-only before quota cleanup
During unmount, gfs2_put_super() calls gfs2_make_fs_ro(), which shuts down quota/statfs activity and then frees the quota bitmap via gfs2_quota_cleanup().
However, gfs2_make_fs_ro() did not mark the VFS superblock read-only before that cleanup. As a result, later inode eviction could still enter the deallocation path through gfs2_evict_inode() -> gfs2_dinode_dealloc() -> gfs2_quota_hold() -> slot_get(), even though sd_quota_bitmap had already been freed and cleared. That leads to a NULL pointer dereference in find_first_zero_bit().
Set SB_RDONLY after the final quota/statfs sync and log shutdown, but before gfs2_quota_cleanup(). This keeps subsequent inode eviction from re-entering write-side quota/deallocation paths once quota state has been torn down.
This fixes a crash reported by syzbot:
Oops: general protection fault in _find_first_zero_bit KASAN: null-ptr-deref in slot_get()
--- fs/gfs2/super.c | 8 ++++++++ 1 file changed, 8 insertions(+)
fs/gfs2/super.c +8 -0
@@ -565,6 +565,14 @@ void gfs2_make_fs_ro(struct gfs2_sbd *sdp) HZ * 5);gfs2_assert_warn (sdp,gfs2_log_is_empty (sdp)); } + +/* +* Once the final quota/statfs sync and log shutdown are complete, the VFS +* must see the filesystem as read-only so later inode eviction cannot +* re-enter deallocation paths that still expect live quota state. +*/ + sdp->sd_vfs->s_flags |=SB_RDONLY ; +gfs2_quota_cleanup (sdp); }
Signature · 1 line
2.34.1
Andreas Gruenbacher
maintainerQuoted 48 lines · Tao Yu
During unmount,
gfs2_put_super()callsgfs2_make_fs_ro(), which shuts down quota/statfs activity and then frees the quota bitmap viagfs2_quota_cleanup().However,
gfs2_make_fs_ro()did not mark the VFS superblock read-only before that cleanup. As a result, later inode eviction could still enter the deallocation path throughgfs2_evict_inode()->gfs2_dinode_dealloc()->gfs2_quota_hold()->slot_get(), even thoughsd_quota_bitmaphad already been freed and cleared. That leads to a NULL pointer dereference infind_first_zero_bit().Set
SB_RDONLYafter the final quota/statfs sync and log shutdown, but beforegfs2_quota_cleanup(). This keeps subsequent inode eviction from re-entering write-side quota/deallocation paths once quota state has been torn down.This fixes a crash reported by syzbot:
Oops: general protection fault in _find_first_zero_bit KASAN: null-ptr-deref in slot_get()Reported-by [email protected]Signed-off-by Tao Yu <[email protected]>--- fs/gfs2/super.c | 8 ++++++++ 1 file changed, 8 insertions(+)fs/gfs2/super.c +8
-1@@ -565,6 +565,14 @@ void gfs2_make_fs_ro(struct gfs2_sbd *sdp) HZ * 5);gfs2_assert_warn (sdp,gfs2_log_is_empty (sdp)); } + +/* +* Once the final quota/statfs sync and log shutdown are complete, the VFS +* must see the filesystem as read-only so later inode eviction cannot +* re-enter deallocation paths that still expect live quota state. +*/ + sdp->sd_vfs->s_flags |=SB_RDONLY ; +gfs2_quota_cleanup (sdp); } --
Quoted 2 lines
2.34.1
Thanks for the patch but no, that's not it. Instead, I've added this:
https://lore.kernel.org/gfs2/[email protected]
Andreas