lkml.org 
[lkml]   [2019]   [Aug]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v1] cpuidle-haltpoll: vcpu hotplug support
On Thu, Aug 29, 2019 at 03:24:31PM +0100, Joao Martins wrote:
> On 8/29/19 2:50 PM, Joao Martins wrote:
> > On 8/29/19 12:56 PM, Marcelo Tosatti wrote:
> >> Hi Joao,
> >>
> >> On Wed, Aug 28, 2019 at 07:56:50PM +0100, Joao Martins wrote:
> >>> +static void haltpoll_uninit(void)
> >>> +{
> >>> + unsigned int cpu;
> >>> +
> >>> + cpus_read_lock();
> >>> +
> >>> + for_each_online_cpu(cpu) {
> >>> + struct cpuidle_device *dev =
> >>> + per_cpu_ptr(haltpoll_cpuidle_devices, cpu);
> >>> +
> >>> + if (!dev->registered)
> >>> + continue;
> >>> +
> >>> + arch_haltpoll_disable(cpu);
> >>> + cpuidle_unregister_device(dev);
> >>> + }
> >>
> >> 1)
> >>
> >>> +
> >>> + cpuidle_unregister(&haltpoll_driver);
> >>
> >> cpuidle_unregister_driver.
> >
> > Will fix -- this was an oversight.
> >
> >>
> >>> + free_percpu(haltpoll_cpuidle_devices);
> >>> + haltpoll_cpuidle_devices = NULL;
> >>> +
> >>> + cpus_read_unlock();
> >>
> >> Any reason you can't cpus_read_unlock() at 1) ?
> >>
> > No, let me adjust that too.
> >
> >> Looks good otherwise.
> >>
>
> BTW, should I take this as a Acked-by, Reviewed-by, or neither? :)
>
> Joao

I'll ACK -v2 once you send it.

\
 
 \ /
  Last update: 2019-08-29 16:43    [W:0.081 / U:0.948 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site