[patch] inotify: locking

From: Robert Love (rml_at_novell.com)
Date: 09/30/04

  • Next message: Alan Cox: "Re: PATCH: Kconfig for EDD" 
    To: John McCutchan <ttb@tentacle.dhs.org>
Date:	Thu, 30 Sep 2004 15:01:26 -0400

    I finally got around to reviewing the locking in inotify, again, in
    response to Andrew's points.

    There are still three TODO items:

            - Look into replacing i_lock with i_sem.
            - dev->lock nests inode->i_lock. Preferably i_lock should
              remain an outermost lock. Maybe i_sem can fix this.
            - inotify_inode_is_dead() should always have the inode's
              i_lock locked when called. I have not yet reviewed the
              VFS paths that call it to ensure this is the case.

    Anyhow, this patch does the following

            - More locking documentation/comments
            - Respect lock ranking when locking two different
              inodes' i_lock
            - Don't grab dentry->d_lock and just use dget_parent(). [1]
            - Respect lock ranking with dev->lock vs. inode->i_lock.
            - inotify_release_all_watches() needed dev->lock.
            - misc. cleanup

    Patch is on top of 0.11.0.

            Robert Love

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  • Next message: Alan Cox: "Re: PATCH: Kconfig for EDD"

    Relevant Pages