On 15 April 2014 01:22, Stephen Boyd sboyd@codeaurora.org wrote:
On 04/14/14 09:23, Viresh Kumar wrote:
We have already checked if 'cpu' is online or not and so don't need to recheck it.
Signed-off-by: Viresh Kumar viresh.kumar@linaro.org
Hm... doing some git archeology shows fa116ea35ec7 (nohz: no softirq pending warnings for offline cpus, 2008-12-11), where the cpu_online() check was added. Before that commit we already checked cpu_online() similar to how the code is today. Perhaps we need to add a comment here?
Okay, I didn't understood what you want completely. You want me to update commit logs or drop this patch and update code?
But yes there is one more thing worth mentioning:
commit f7ea0fd6 (tick: Don't invoke tick_nohz_stop_sched_tick() if the cpu is offline, May 13 2013), which did this:
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 0eed1db..0121421 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -469,6 +469,7 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts) if (unlikely(!cpu_online(cpu))) { if (cpu == tick_do_timer_cpu) tick_do_timer_cpu = TICK_DO_TIMER_NONE; + return false; }
if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
So, when the patch you mentioned was added, we didn't had this and a recheck of cpu_online() was perhaps required. But it is surely not required anymore.