> Subject: sched, autogroup: Fix failure to set cpu.rt_runtime_us
> From: Peter Zijlstra
> Date: Mon Feb 9 11:53:18 CET 2015
>
> Because task_group() uses a cache of autogroup_task_group(), whoes
> output depends on sched_class, switching classes can generate
> problems.
>
> In particular, when
On Mon, Feb 09, 2015 at 12:22:37PM +0100, Peter Zijlstra wrote:
> Indeed, setting runtime=0 for the root group is a very bad thing
> regardless of this patch. It would disallow the kernel from creating RT
> threads, which it needs for 'correct' operation in a number of cases.
>
> But lets make tha
On Sat, Feb 07, 2015 at 03:02:11PM +0800, Zefan Li wrote:
> Before 8323f26ce342 "sched: Fix race in task_group()", task_group() of
> those RT tasks always return root_task_group, but the escape can't happen.
Ah, yes, I'm an idiot. I'm not sure what I was thinking, but I seemed
to have confused mys
>> See this commit:
>>
>> commit f44937718ce3b8360f72f6c68c9481712517a867
>> Author: Mike Galbraith
>> Date: Thu Jan 13 04:54:50 2011 +0100
>>
>> sched, autogroup: Fix CONFIG_RT_GROUP_SCHED sched_setscheduler() failure
>> ...
>> + /*
>> +* Autogroup RT tasks are redirected to t
On Fri, Feb 06, 2015 at 09:30:51AM +0800, Zefan Li wrote:
> After running the test program, we have:
>
> root
>|
> autogroup <-- the RT test program is now here
>
> root.rt_runtime = 95
> autogroup.rt_runtime = 0
>
> Now if you try to change the rt_runtime of the root cgroup, it re
On 2015/2/5 22:25, Peter Zijlstra wrote:
> On Thu, Feb 05, 2015 at 04:33:24PM +0800, Zefan Li wrote:
>> This is how to reproduce the bug:
>>
>> int main() {
>> struct sched_param param = {.sched_priority=1};
>>
>> if (fork() > 0)
>> exit(0);
>>
>> setsid();
>>
>> if (sched_s
On Thu, Feb 05, 2015 at 04:33:24PM +0800, Zefan Li wrote:
> This is how to reproduce the bug:
>
> int main() {
> struct sched_param param = {.sched_priority=1};
>
> if (fork() > 0)
> exit(0);
>
> setsid();
>
> if (sched_setscheduler(0, SCHED_RR, ¶m) < 0){
> perro
7 matches
Mail list logo