Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-28 Thread Srivatsa S. Bhat
On 01/24/2014 04:31 AM, Rusty Russell wrote: > "Srivatsa S. Bhat" writes: >> On 01/23/2014 07:59 AM, Rusty Russell wrote: >>> "Srivatsa S. Bhat" writes: On 01/22/2014 02:00 PM, Srivatsa S. Bhat wrote: > Hi Paul, >>> >>> I find an old patch for register_allcpu_notifier(), but the "bool >>

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-28 Thread Srivatsa S. Bhat
On 01/23/2014 10:32 PM, Oleg Nesterov wrote: > On 01/23, Srivatsa S. Bhat wrote: >> >> On 01/23/2014 12:48 AM, Oleg Nesterov wrote: >>> On 01/22, Srivatsa S. Bhat wrote: Wait a min, that _will_ actually work for all cases because I have provided an option to invoke _any_ arbitrary fu

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-23 Thread Rusty Russell
"Srivatsa S. Bhat" writes: > On 01/23/2014 07:59 AM, Rusty Russell wrote: >> "Srivatsa S. Bhat" writes: >>> On 01/22/2014 02:00 PM, Srivatsa S. Bhat wrote: Hi Paul, >> >> I find an old patch for register_allcpu_notifier(), but the "bool >> replay_history" should be eliminated (always true):

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-23 Thread Oleg Nesterov
On 01/23, Srivatsa S. Bhat wrote: > > On 01/23/2014 12:48 AM, Oleg Nesterov wrote: > > On 01/22, Srivatsa S. Bhat wrote: > >> > >> Wait a min, that _will_ actually work for all cases because I have provided > >> an option to invoke _any_ arbitrary function as the "setup" routine. > > > > And probab

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Srivatsa S. Bhat
On 01/23/2014 07:59 AM, Rusty Russell wrote: > "Srivatsa S. Bhat" writes: >> On 01/22/2014 02:00 PM, Srivatsa S. Bhat wrote: >>> Hi Paul, > > I find an old patch for register_allcpu_notifier(), but the "bool > replay_history" should be eliminated (always true): it's too weird. > Sorry, I didn't

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Rusty Russell
"Srivatsa S. Bhat" writes: > On 01/22/2014 02:00 PM, Srivatsa S. Bhat wrote: >> Hi Paul, I find an old patch for register_allcpu_notifier(), but the "bool replay_history" should be eliminated (always true): it's too weird. Then we should get rid of register_cpu_notifier, or at least hide it. Th

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Srivatsa S. Bhat
On 01/23/2014 12:48 AM, Oleg Nesterov wrote: > On 01/22, Srivatsa S. Bhat wrote: >> >> Wait a min, that _will_ actually work for all cases because I have provided >> an option to invoke _any_ arbitrary function as the "setup" routine. > > And probably the generic solution makes sense. I am not sur

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Oleg Nesterov
On 01/22, Srivatsa S. Bhat wrote: > > Wait a min, that _will_ actually work for all cases because I have provided > an option to invoke _any_ arbitrary function as the "setup" routine. And probably the generic solution makes sense. I am not sure I actually understand the semantics of register_allc

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Srivatsa S. Bhat
On 01/22/2014 02:00 PM, Srivatsa S. Bhat wrote: > Hi Paul, > > On 01/22/2014 11:22 AM, Paul Mackerras wrote: >> This arises out of a report from a tester that offlining a CPU never >> finished on a system they were testing. This was on a POWER8 running >> a 3.10.x kernel, but the issue is still p

Re: Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-22 Thread Srivatsa S. Bhat
Hi Paul, On 01/22/2014 11:22 AM, Paul Mackerras wrote: > This arises out of a report from a tester that offlining a CPU never > finished on a system they were testing. This was on a POWER8 running > a 3.10.x kernel, but the issue is still present in mainline AFAICS. > > What I found when I looke

Deadlock between cpu_hotplug_begin and cpu_add_remove_lock

2014-01-21 Thread Paul Mackerras
This arises out of a report from a tester that offlining a CPU never finished on a system they were testing. This was on a POWER8 running a 3.10.x kernel, but the issue is still present in mainline AFAICS. What I found when I looked at the system was this: * There was a ppc64_cpu process stuck i