Re: [patch] CFS scheduler, -v15

[Srivatsa Vaddagiri <vatsa@in.ibm.com>]

On Wed, Jun 06, 2007 at 09:01:43AM +0200, Ingo Molnar wrote:
> [...] and my tree already contains the fixes for rt task's 
> exec_start. 

Can I have this snapshot pls? I have to deal with the same issue when 
the current task switches groups and I was planning to fix it by
introducing a set_curr_task() method in fair_sched_class which
initializes exec_start and other fields for that task.

> > Also, real-time tasks are currently influencing fair clock 
> > calculations (because there is only one raw_weighted_load field per 
> > runqueue). Is that intentional?
> 
> yep - the 'fair clock' in essence freezes when RT tasks are running (but 
> not completely). Can you think of any bad effect from this perhaps?

hmm ..i was thinking of the case where currently running task belongs to 
fair_sched class but the raw_weighted_load field of runqueue includes
weight from real time tasks (which would have affected the delta_mine
calculation). Now I realize that it is not possible because real-time
tasks always preempt fair-sched tasks. So possibly this is not an issue.

> It's useful to have unified load-balancing (smpnice-driven) between the RT 
> and fair classes.

Sure ..the unified weight number is usefull to pick the busiest
group/queue during load balance. I am retaining this aspect in my
patches (which I am preparing to send against -v15 atm).

-- 
Regards,
vatsa