cpufreq: schedutil: Use idle_calls counter of the remote CPU
authorJoel Fernandes <joelaf@google.com>
Thu, 21 Dec 2017 01:22:45 +0000 (02:22 +0100)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Thu, 28 Dec 2017 11:26:54 +0000 (12:26 +0100)
commit466a2b42d67644447a1765276259a3ea5531ddff
treeca896ab907badd58efe681d139fdcecab1491c70
parent464e1d5f23cca236b930ef068c328a64cab78fb1
cpufreq: schedutil: Use idle_calls counter of the remote CPU

Since the recent remote cpufreq callback work, its possible that a cpufreq
update is triggered from a remote CPU. For single policies however, the current
code uses the local CPU when trying to determine if the remote sg_cpu entered
idle or is busy. This is incorrect. To remedy this, compare with the nohz tick
idle_calls counter of the remote CPU.

Fixes: 674e75411fc2 (sched: cpufreq: Allow remote cpufreq callbacks)
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Joel Fernandes <joelaf@google.com>
Cc: 4.14+ <stable@vger.kernel.org> # 4.14+
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
include/linux/tick.h
kernel/sched/cpufreq_schedutil.c
kernel/time/tick-sched.c