Re: [PATCH 1/2] devpts: remove unneeded inode_lock in mknod_ptmx

On Mon, Jan 28, 2019 at 06:11:32PM +0100, Christian Brauner wrote: > Afaict, the mknod_ptmx() call is a no-op on subsequent calls and the first > call is done before we unlock the suberblock. If I'm not mistaken this is > exactly parallel to what Al pointed out for binderfs 29ef1c8e16a > ("binderfs

[PATCH 1/2] devpts: remove unneeded inode_lock in mknod_ptmx

Afaict, the mknod_ptmx() call is a no-op on subsequent calls and the first call is done before we unlock the suberblock. If I'm not mistaken this is exactly parallel to what Al pointed out for binderfs 29ef1c8e16a ("binderfs: drop lock in binderfs_binder_ctl_create"). In both filesystems it should