I've been running with this patch on ARM (dra7 SOC, which is a
dual-core A15) with a 4.1.18-rt17 kernel for a while without any
noticeable problems. Boosting the ktimersoftd processes on both cores
to a higher priority fixed some timing problems my userspace
application was seeing too.
The userspa
The softirqd runs in -RT with SCHED_FIFO (prio 1) and deals mostly with
timer wakeup which can not happen in hardirq context. The prio has been
risen from the normal SCHED_OTHER so the timer wakeup does not happen
too late.
With enough networking load it is possible that the system never goes
idle
2 matches
Mail list logo