Messages in this thread | | | Date | Sun, 24 Feb 2008 22:21:07 -0800 | From | "Bill Huey (hui)" <> | Subject | Re: [PATCH [RT] 11/14] optimize the !printk fastpath through the lock acquisition |
| |
[repost with all folks CCed]
On Sun, Feb 24, 2008 at 9:20 PM, Gregory Haskins <gregory.haskins.ml@gmail.com> wrote: > Are you saying that the modified logic that I introduced is broken? Or > that the original use of the might_sleep() annotation inside this > function is broken?
It's probably safe to use, but it's not what its original purpose was and you should use another function/macro. This is an annotation issue and your use of it is inconsistent with how it's used in voluntary preempt. I mentioned it before in a previous post. Folks will correct me if I'm wrong but you should use another macro or function.
bill
| |