[PATCH 1/4] nohz: Only update sleeptime stats locally

2014-05-07 Thread Denys Vlasenko
From: Frederic Weisbecker The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency, for example when a governor calls a

[PATCH 1/4] nohz: Only update sleeptime stats locally

2014-05-07 Thread Denys Vlasenko
From: Frederic Weisbecker fweis...@gmail.com The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency, for example when a governor

[PATCH 1/4] nohz: Only update sleeptime stats locally

2014-04-24 Thread Denys Vlasenko
From: Frederic Weisbecker The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency, for example when a governor calls a

[PATCH 1/4] nohz: Only update sleeptime stats locally

2014-04-24 Thread Denys Vlasenko
From: Frederic Weisbecker fweis...@gmail.com The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency, for example when a governor

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-19 Thread Stratos Karafotis
On 08/18/2013 08:04 PM, Oleg Nesterov wrote: > Sorry for double post. forgot to cc cpufreq maintainers. > > On 08/16, Frederic Weisbecker wrote: >> >> To fix this, lets only update the sleeptime stats locally when the CPU >> exits from idle. > > I am in no position to ack the changes in this

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-19 Thread Stratos Karafotis
On 08/18/2013 08:04 PM, Oleg Nesterov wrote: Sorry for double post. forgot to cc cpufreq maintainers. On 08/16, Frederic Weisbecker wrote: To fix this, lets only update the sleeptime stats locally when the CPU exits from idle. I am in no position to ack the changes in this area, but I

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Frederic Weisbecker
On Sun, Aug 18, 2013 at 06:49:37PM +0200, Oleg Nesterov wrote: > On 08/16, Frederic Weisbecker wrote: > > > > To fix this, lets only update the sleeptime stats locally when the CPU > > exits from idle. > > I am in no position to ack the changes in this area, but I like this > change very much.

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Oleg Nesterov
Sorry for double post. forgot to cc cpufreq maintainers. On 08/16, Frederic Weisbecker wrote: > > To fix this, lets only update the sleeptime stats locally when the CPU > exits from idle. I am in no position to ack the changes in this area, but I like this change very much. Because, as a code

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Oleg Nesterov
On 08/16, Frederic Weisbecker wrote: > > To fix this, lets only update the sleeptime stats locally when the CPU > exits from idle. I am in no position to ack the changes in this area, but I like this change very much. Because, as a code reader, I was totally confused by if

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Oleg Nesterov
On 08/16, Frederic Weisbecker wrote: To fix this, lets only update the sleeptime stats locally when the CPU exits from idle. I am in no position to ack the changes in this area, but I like this change very much. Because, as a code reader, I was totally confused by if

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Oleg Nesterov
Sorry for double post. forgot to cc cpufreq maintainers. On 08/16, Frederic Weisbecker wrote: To fix this, lets only update the sleeptime stats locally when the CPU exits from idle. I am in no position to ack the changes in this area, but I like this change very much. Because, as a code

Re: [PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-18 Thread Frederic Weisbecker
On Sun, Aug 18, 2013 at 06:49:37PM +0200, Oleg Nesterov wrote: On 08/16, Frederic Weisbecker wrote: To fix this, lets only update the sleeptime stats locally when the CPU exits from idle. I am in no position to ack the changes in this area, but I like this change very much. Because, as

[PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-16 Thread Frederic Weisbecker
The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency , for example when a governor calls a get_cpu_*_time_us() API and the target

[PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-16 Thread Frederic Weisbecker
The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency , for example when a governor calls a get_cpu_*_time_us() API and the target

[PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-08 Thread Frederic Weisbecker
The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency , for example when a governor calls a get_cpu_*_time_us() API and the target

[PATCH 1/4] nohz: Only update sleeptime stats locally

2013-08-08 Thread Frederic Weisbecker
The idle and io sleeptime stats can be updated concurrently from callers of get_cpu_idle_time_us(), get_cpu_iowait_time_us() and tick_nohz_stop_idle(). Updaters can easily race and mess up with internal datas coherency , for example when a governor calls a get_cpu_*_time_us() API and the target