Hi Ingo/Peter,
I am trying to understand the complex scheduler code and just found something incorrect (maybe i am not reading it well):
File: kernel/sched/fair.c
static void prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) { if (!p->se.on_rq) return;
/* * Reschedule if we are currently running on this runqueue and * our priority decreased, or if we are not currently running on * this runqueue and our priority is higher than the current's */ if (rq->curr == p) { if (p->prio > oldprio) resched_task(rq->curr); } else check_preempt_curr(rq, p, 0); }
Comment says that we must mark the task to be rescheduled, if we are currently running and our priority has decreased. But in code we are checking (p->prio > oldprio). i.e. reschedule if we were currently running and our priority increased.
Sorry if i am wrong :(
-- viresh
On 11/07/2012 03:49 PM, Viresh Kumar wrote: Hi, Viresh
Hi Ingo/Peter,
I am trying to understand the complex scheduler code and just found something incorrect (maybe i am not reading it well):
File: kernel/sched/fair.c
static void prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) { if (!p->se.on_rq) return;
/* * Reschedule if we are currently running on this runqueue and * our priority decreased, or if we are not currently running on * this runqueue and our priority is higher than the current's */ if (rq->curr == p) { if (p->prio > oldprio) resched_task(rq->curr); } else check_preempt_curr(rq, p, 0); }
Comment says that we must mark the task to be rescheduled, if we are currently running and our priority has decreased. But in code we are checking (p->prio > oldprio). i.e. reschedule if we were currently running and our priority increased.
It's the user nice value I suppose, so it should be reversed when we are talking about weight.
Regards, Michael Wang
Sorry if i am wrong :(
-- viresh -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
On 7 November 2012 13:26, Michael Wang wangyun@linux.vnet.ibm.com wrote:
It's the user nice value I suppose, so it should be reversed when we are talking about weight.
Ahh.. I knew it .. How can i miss it.
Sorry for the noise :(
-- viresh