Re: [PATCH v2] stop_machine: Make migration_cpu_stop() does useful works for CONFIG_PREEMPT_NONE

2016-09-09 Thread chengchao
--in-reply-to doesn't work? v1 is : https://lkml.org/lkml/2016/9/7/819 on 09/09/2016 04:13 PM, Cheng Chao wrote: > For CONFIG_PREEMPT_NONE=y, when sched_exec() needs migration, sched_exec() > calls stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg). > > If the migration_cpu_stop() can not mi

Re: [lkp] [sched/core] 3d26b7622f: BUG: unable to handle kernel NULL pointer dereference at 00000001

2016-09-08 Thread chengchao
/2016 10:26 AM, Ye Xiaolong wrote: > Hi, cheng hao, > > On 09/09, chengchao wrote: >> Hi, xiaolong >> >> where can I find the commit 3d26b7622f3bab689696900ffd33c6dd7849d7c2? > > This is the commit by 0Day bot for your original email patch sent on Sep 05 >

Re: [lkp] [sched/core] 3d26b7622f: BUG: unable to handle kernel NULL pointer dereference at 00000001

2016-09-08 Thread chengchao
Hi, xiaolong where can I find the commit 3d26b7622f3bab689696900ffd33c6dd7849d7c2? config-4.8.0-rc5-1-g3d26b76: # CONFIG_PREEMPT_NONE is not set CONFIG_PREEMPT_VOLUNTARY=y # CONFIG_PREEMPT is not set the patch ("sched/core: simpler function for sched_exec migration") only is on if CON

Re: [PATCH] sched/core: simpler function for sched_exec migration

2016-09-07 Thread chengchao
Oled, thank you for moving this patch on. on 09/07/2016 08:35 PM, Oleg Nesterov wrote: > On 09/07, chengchao wrote: >> >> Oleg, thank you very much. >> >> on 09/06/2016 11:22 PM, Oleg Nesterov wrote: >>> On 09/06, chengchao wrote: >>>>

Re: [PATCH] sched/core: simpler function for sched_exec migration

2016-09-06 Thread chengchao
Oleg, thank you very much. on 09/06/2016 11:22 PM, Oleg Nesterov wrote: > On 09/06, chengchao wrote: >> >> the key point is for CONFIG_PREEMPT_NONE=y, >> ... >> it is too much overhead for one task(fork()+exec()), isn't it? > > Yes, yes, I see, this is su

Re: [PATCH] sched/core: simpler function for sched_exec migration

2016-09-05 Thread chengchao
Oleg, thank you. the key point is for CONFIG_PREEMPT_NONE=y, if sched_exec needs migrate the current, migration_cpu_stop doesn't migrate the task(current) at all, it means that the stopper thread does some unuseful works in this scenario. finally,the stopper thread calls cpu_stop_signal_done()