On Thu, Nov 2, 2023 at 2:27 AM Jan Kara wrote:
>
> We removed all codepaths where s_umount is taken beneath open_mutex and
> bd_holder_lock so don't make things more complicated than they need to
> be and hold s_umount over block device opening.
>
> CC: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxx>
> CC: linux-nilfs@xxxxxxxxxxxxxxx
> Signed-off-by: Jan Kara <jack@xxxxxxx>
> ---
> fs/nilfs2/super.c | 8 --------
> 1 file changed, 8 deletions(-)
Acked-by: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxx>
I had missed that cleanup patch set. Thank you!
>
> Hi Christian, I think you've missed this simplification in your cleanup
> patches. Can you merge it please?
Thank you in advance, Christian.
Ryusuke Konishi
>
> diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
> index a5d1fa4e7552..df8674173b22 100644
> --- a/fs/nilfs2/super.c
> +++ b/fs/nilfs2/super.c
> @@ -1314,15 +1314,7 @@ nilfs_mount(struct file_system_type *fs_type, int flags,
> return ERR_CAST(s);
>
> if (!s->s_root) {
> - /*
> - * We drop s_umount here because we need to open the bdev and
> - * bdev->open_mutex ranks above s_umount (blkdev_put() ->
> - * __invalidate_device()). It is safe because we have active sb
> - * reference and SB_BORN is not set yet.
> - */
> - up_write(&s->s_umount);
> err = setup_bdev_super(s, flags, NULL);
> - down_write(&s->s_umount);
> if (!err)
> err = nilfs_fill_super(s, data,
> flags & SB_SILENT ? 1 : 0);
> --
> 2.35.3
>
