Re: [RFC PATCH RT] push waiting rt tasks to cpus with lower prios.

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Steven Rostedt
Date: Tuesday, October 9, 2007 - 1:50 pm

--
On Tue, 9 Oct 2007, mike kravetz wrote:


You mean to keep the copy for the next two tries?


I was a bit scared of grabing the lock anyway, because that's another
cache hit (write side). So only grabbing the lock when needed would save
us from dirting the runqueue lock for each CPU.


Yes indeed. The aim for now is to solve the latencies that you've been
seeing. But really, there is still holes (small ones) that can cause a
latency if a schedule happened "just right". Hopefully the final result of
this work will close them too.

-- Steve

-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 1/5] RT - fix for scheduling issue, Gregory Haskins, (Tue Oct 9, 7:25 am)
[PATCH 2/5] RT - fix reschedule IPI, Gregory Haskins, (Tue Oct 9, 7:25 am)
[PATCH 3/5] RT - fix mistargeted RESCHED_IPI, Gregory Haskins, (Tue Oct 9, 7:25 am)
[PATCH 4/5] RT: Add a per-cpu rt_overload indication, Gregory Haskins, (Tue Oct 9, 7:26 am)
Re: [RFC PATCH RT] push waiting rt tasks to cpus with lowe ..., Steven Rostedt, (Tue Oct 9, 1:50 pm)
Re: [RFC PATCH RT] push waiting rt tasks to cpus with lowe ..., Girish kathalagiri, (Tue Oct 9, 7:12 pm)