Hi Hubertus,
I could not get my kernel booted when you guys moved forward from the
version 1 CPU controller. It appeared to croak very early on, which turns
out to be in sched_init(). Finally tracked down why this was the case. It
appears that the do_div() operation in ckrm_cpu_monitor.c would cause a FPE
because the total field == 0. Unfortunately, it is too early in the boot
process for the kernel to print such a message to the screen. What I've
figured out is that you need to initialize the demand_stat structure of the
first process, as shown by the oneline patch below (against 2.6.8.1 + e16rc1
cpu controller v7). Am very curious why you guys didn't have this
problem---naturally I am concerned that I missed something in my merge.
*** sched.c 2004-09-28 01:47:08.252312059 -0400
--- sched.c.new 2004-09-28 01:47:13.191440261 -0400
***************
*** 4556,4561 ****
--- 4556,4562 ----
rq->idle = current;
set_task_cpu(current, smp_processor_id());
#ifdef CONFIG_CKRM_CPU_SCHEDULE
+ cpu_demand_event(&(current)->demand_stat,CPU_DEMAND_INIT,0);
current->cpu_class = get_default_cpu_class();
current->array = NULL;
#endif
Marc
-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech
