On Wed, Apr 02, 2008 at 04:16:28PM -0700, Sunil Mushran wrote:
> This patch adds code to track the number of times the fs takes
> various cluster locks as well as the times associated with it.

Ok, this is definitely some usefull stuff. I'm concerned about how much it
grows struct ocfs2_lock_res though - I count 52 bytes. Even if the overhead
doesn't technically push it out to another slab, that's still a lot. Can we
wrap enough of this in OCFS2_DEBUG_FS so that the overhead can be easily
compiled out?
        --Mark

--
Mark Fasheh

_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
http://oss.oracle.com/mailman/listinfo/ocfs2-devel

Reply via email to