lkml.org 
[lkml]   [2013]   [Apr]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH v6 1/2] cpufreq: split the cpufreq_driver_lock and use the rcu
From
On 3 April 2013 04:27, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> On Tuesday, April 02, 2013 08:29:12 PM Viresh Kumar wrote:
>> On 2 April 2013 20:25, Nathan Zimmer <nzimmer@sgi.com> wrote:
>> > The lock is unneeded if we expect register and unregister driver to not be
>> > called from muliple threads at once. I didn't make that assumption.
>>
>> Hmm.. But doesn't rcu part take care of that too?? Two writers
>> updating stuff simultaneously?
>
> RCU doesn't cover that in general. Additional locking is needed to provide
> synchronization between writers.

Hmm.. I read the same from rcu documentation now...

Nathan, What about using a single spinlock (instead of two) that will take care
of all locking requirements of cpufreq.c ... i.e. both cpufreq_cpu_data and
cpufreq_driver_{register|unregister}... We don't need two locks actually.


\
 
 \ /
  Last update: 2013-04-03 08:01    [W:0.083 / U:1.816 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site