Re: [RFC][PATCH 12/13] stop_machine: Remove lglock

2015-06-22 Thread Oleg Nesterov
On 06/22, Peter Zijlstra wrote: > > By having stop_two_cpus() acquire two cpu_stopper::locks we gain full > order against the global stop_machine which takes each of these locks > in order. Yes, but stop_machine() locks/unlocs cpu_stopper->lock sequentially, it never holds more than 1 ->lock, so

Re: [RFC][PATCH 12/13] stop_machine: Remove lglock

2015-06-22 Thread Oleg Nesterov
On 06/22, Peter Zijlstra wrote: By having stop_two_cpus() acquire two cpu_stopper::locks we gain full order against the global stop_machine which takes each of these locks in order. Yes, but stop_machine() locks/unlocs cpu_stopper-lock sequentially, it never holds more than 1 -lock, so

<    1   2