Description of problem: Upon unmounting an xfs-formatted file-backed partition, I was greeted with the following spew: ===================================== [ BUG: bad unlock balance detected! ] ------------------------------------- umount/25787 is trying to release lock (&(&ip->i_iolock)->mr_lock) at: [<f9035f23>] xfs_iunlock+0x2d/0x6f [xfs] but there are no more locks to release! other info that might help us debug this: 2 locks held by umount/25787: #0: (&type->s_umount_key#22){----}, at: [<c048007e>] deactivate_super+0x58/0x6f #1: (&type->s_lock_key#13){--..}, at: [<c0614eba>] mutex_lock+0x21/0x24 stack backtrace: [<c04061e9>] show_trace_log_lvl+0x1a/0x2f [<c04067ad>] show_trace+0x12/0x14 [<c0406831>] dump_stack+0x16/0x18 [<c0440d67>] print_unlock_inbalance_bug+0xec/0xf9 [<c0442bc8>] lock_release_non_nested+0x9e/0x162 [<c0442dc9>] lock_release+0x13d/0x159 [<c043b879>] up_read+0x16/0x29 [<f9035f23>] xfs_iunlock+0x2d/0x6f [xfs] [<f903619b>] xfs_ireclaim+0x78/0x82 [xfs] [<f9052537>] xfs_finish_reclaim+0x124/0x12e [xfs] [<f9052661>] xfs_reclaim+0x6b/0xe0 [xfs] [<f905fc11>] xfs_fs_clear_inode+0x97/0xba [xfs] [<c048f5d5>] clear_inode+0xd3/0x122 [<c048f852>] generic_drop_inode+0x11e/0x130 [<c048ed32>] iput+0x63/0x66 [<f90505ed>] xfs_unmount+0xdd/0x158 [xfs] [<f906033c>] vfs_unmount+0x1a/0x1e [xfs] [<f905f666>] xfs_fs_put_super+0x2e/0x69 [xfs] [<c047ff3a>] generic_shutdown_super+0x55/0xbe [<c047ffc3>] kill_block_super+0x20/0x32 [<c0480083>] deactivate_super+0x5d/0x6f [<c0491401>] mntput_no_expire+0x42/0x72 [<c0484516>] path_release_on_umount+0x15/0x18 [<c0491ac1>] sys_umount+0x1e3/0x217 [<c0491b0e>] sys_oldumount+0x19/0x1b [<c0405078>] syscall_call+0x7/0xb ======================= Version-Release number of selected component (if applicable): kernel-2.6.21-1.3116.fc7.i686 (will try to reproduce with a more recent one as well) How reproducible: # dd if=/dev/zero of=/scratch/testfile bs=4k count=100000 # mkfs.xfs /scratch/testfile # mkdir /mnt/testfile # mount /scratch/testfile /mnt/testfile # umount /mnt/testfile
So far, unable to reproduce with kernel 2.6.21-1.3163.fc7.i686... I've got mount/unmount running in a loop, will let it go for a while...
Couldn't ever reproduce, appears it was just temporary rawhide breakage.