* Jens Koehler | 2016-06-01 09:46:56 [+0200]: >Hi, Hi, >I am running a linux kernel with rt preempt patch and made also a rt >application. Application works good but sometimes I a get a message >like "hrtimer: interrupt took 10252 ns" or "NOHZ: >local_softirq_pending 102". What does it mean and is it a problem? The hrtimer message tries to inform you that for some reasons you had one or more hrtimers which expired and processing them to around 10us. At a single point. This means either you had a lot of them or one in particular took a very long time to finish. This might be problem because you had a window of 10us in which you were not able to react to external interrupts. The NOHZ message means that you had a TIMER_SOFTIRQ and HRTIMER_SOFTIRQ event pending but you rather went idle than process them. I've seen that with network on an (almost) idle machine that the softirq was blocked due to CPU pinning and so there was nothing else to do then to go idle. There was one report around v3.8-RT I think and back the the debug code didn't cover the scenario. >Thanks and regards, >Jens Sebastian -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
