Messages in this thread | | | Subject | Re: [PATCH] sched: remove resetting exec_start in put_prev_task_rt() | From | Peter Zijlstra <> | Date | Fri, 10 Jun 2011 12:19:06 +0200 |
| |
On Thu, 2011-06-09 at 22:41 -0400, Steven Rostedt wrote: > On Fri, 2011-06-10 at 10:38 +0800, Yong Zhang wrote: > > On Thu, Jun 9, 2011 at 8:34 PM, Steven Rostedt <rostedt@goodmis.org> wrote: > > > I disagree. Yes the exec_start is reset there, but I like the fact that > > > it's 0 when not running. > > > > Actually this depends on how we look at the code: > > if we set exec_start to 0 explicitly, as you said the code is more direct and > > readable. > > if we don't set exec_start to 0, we can save one instruction though > > it's minor. > > > > I have no strong opinion on either of them :) > > I don't have any real strong opinion on this either, so I'll just let > Peter decide :)
Yay! So IFF its correct (I didn't check) then sure it saves a whole store :-), I don't think sched_fair clears exec_start on de-schedule either.
| |