Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-25 Thread Kirill Tkhai
25.02.2014, 18:14, "Juri Lelli" : > On Sat, 22 Feb 2014 04:56:59 +0400 > Kirill Tkhai wrote: > >>  On 21.02.2014 20:36, Juri Lelli wrote: >>>  On Fri, 21 Feb 2014 11:37:15 +0100 >>>  Peter Zijlstra wrote:  On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: >  Since deadline

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-25 Thread Juri Lelli
On Sat, 22 Feb 2014 04:56:59 +0400 Kirill Tkhai wrote: > On 21.02.2014 20:36, Juri Lelli wrote: > > On Fri, 21 Feb 2014 11:37:15 +0100 > > Peter Zijlstra wrote: > > > >> On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > >>> Since deadline tasks share rt bandwidth, we must care abo

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Kirill Tkhai
On 21.02.2014 20:36, Juri Lelli wrote: > On Fri, 21 Feb 2014 11:37:15 +0100 > Peter Zijlstra wrote: > >> On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: >>> Since deadline tasks share rt bandwidth, we must care about >>> bandwidth timer set. Otherwise rt_time may grow up to infinity

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Kirill Tkhai
21.02.2014, 20:52, "Juri Lelli" : > On Fri, 21 Feb 2014 17:36:41 +0100 > Juri Lelli wrote: > >>  On Fri, 21 Feb 2014 11:37:15 +0100 >>  Peter Zijlstra wrote: >>>  On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote:  Since deadline tasks share rt bandwidth, we must care about

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Juri Lelli
On Fri, 21 Feb 2014 17:36:41 +0100 Juri Lelli wrote: > On Fri, 21 Feb 2014 11:37:15 +0100 > Peter Zijlstra wrote: > > > On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > > > Since deadline tasks share rt bandwidth, we must care about > > > bandwidth timer set. Otherwise rt_time ma

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Juri Lelli
On Fri, 21 Feb 2014 11:37:15 +0100 Peter Zijlstra wrote: > On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > > Since deadline tasks share rt bandwidth, we must care about > > bandwidth timer set. Otherwise rt_time may grow up to infinity > > in update_curr_dl(), if there are no othe

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Kirill Tkhai
21.02.2014, 16:44, "Juri Lelli" : > On Fri, 21 Feb 2014 16:09:25 +0400 > Kirill Tkhai wrote: > >>  21.02.2014, 15:39, "Kirill Tkhai" : >>>  21.02.2014, 14:37, "Peter Zijlstra" :   On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: >    Since deadline tasks share rt bandwidth,

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Juri Lelli
On Fri, 21 Feb 2014 16:09:25 +0400 Kirill Tkhai wrote: > > > 21.02.2014, 15:39, "Kirill Tkhai" : > > 21.02.2014, 14:37, "Peter Zijlstra" : > > > >>  On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > >>>   Since deadline tasks share rt bandwidth, we must care about > >>>   bandwidt

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Kirill Tkhai
21.02.2014, 15:39, "Kirill Tkhai" : > 21.02.2014, 14:37, "Peter Zijlstra" : > >>  On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: >>>   Since deadline tasks share rt bandwidth, we must care about >>>   bandwidth timer set. Otherwise rt_time may grow up to infinity >>>   in update_cu

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Kirill Tkhai
21.02.2014, 14:37, "Peter Zijlstra" : > On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > >>  Since deadline tasks share rt bandwidth, we must care about >>  bandwidth timer set. Otherwise rt_time may grow up to infinity >>  in update_curr_dl(), if there are no other available RT ta

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-21 Thread Peter Zijlstra
On Thu, Feb 20, 2014 at 02:16:00AM +0400, Kirill Tkhai wrote: > Since deadline tasks share rt bandwidth, we must care about > bandwidth timer set. Otherwise rt_time may grow up to infinity > in update_curr_dl(), if there are no other available RT tasks > on top level bandwidth. > > I'm going to de

Re: [RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-20 Thread Steven Rostedt
Peter, Juri, Have you seen this? -- Steve On Thu, 20 Feb 2014 02:16:00 +0400 Kirill Tkhai wrote: > Since deadline tasks share rt bandwidth, we must care about > bandwidth timer set. Otherwise rt_time may grow up to infinity > in update_curr_dl(), if there are no other available RT tasks > on

[RFC] sched/deadline: Prevent rt_time growth to infinity

2014-02-19 Thread Kirill Tkhai
Since deadline tasks share rt bandwidth, we must care about bandwidth timer set. Otherwise rt_time may grow up to infinity in update_curr_dl(), if there are no other available RT tasks on top level bandwidth. I'm going to decide the problem the way below. Almost untested because of I skipped almos