On Mon, Mar 28, 2011 at 06:48:13PM +0900, KOSAKI Motohiro wrote:

Hi

quoted

@@ -434,9 +452,17 @@ static int oom_kill_task(struct task_struct *p)
                K(get_mm_counter(p->mm, MM_FILEPAGES)));
        task_unlock(p);
 
-       p->rt.time_slice = HZ; <<---- THIS
+
        set_tsk_thread_flag(p, TIF_MEMDIE);
        force_sig(SIGKILL, p);
+
+       /*
+        * We give our sacrificial lamb high priority and access to
+        * all the memory it needs. That way it should be able to
+        * exit() and clear out its resources quickly...
+        */
+       boost_dying_task_prio(p, mem);
+
        return 0;
 }

At that time, I thought that routine is meaningless in non-RT scheduler.
So I Cced Peter but don't get the answer.
I just want to confirm it.

Do you still think it's meaningless? 

In short, yes.

quoted

so you remove it when you revert 93b43fa5508?
Then, this isn't just revert patch but revert + killing meaningless code patch.

If you want, I'd like to rename a patch title. That said, we can't revert
93b43fa5508 simple cleanly, several patches depend on it. therefore I
reverted it manualy. and at that time, I don't want to resurrect
meaningless logic. anyway it's no matter. Luis is preparing new patches.
therefore we will get the same end result. :)

I don't mind it, either. :)
I just want to make sure the meaningless logic.
Thanks.

-- 
Kind regards,
Minchan Kim

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>