Re: poor cyclictest results with 5.0 series rt for arm64
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2019-08-15 08:29:58
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: 2019-08-15 08:29:58
On 2019-08-14 13:04:21 [-0400], Paul Thomas wrote:
Hi Sebastian,
Hi Paul,
The attachment is here: https://bugzilla.kernel.org/attachment.cgi?id=284391
both traces have `printk' in it. Could you please explain what it is doing? It should print something to the console. The printk code in v4.19 vs later changed and now (by default) ERR levels are printed immediately and will cause latency spikes. This would explain the spike but I don't understand how lazy preempt is related here.
We don't actually need Lazy Preempt, so now that we know what the issues is we can just turn it off in the config? However I suspect everyone wants to get to the bottom of what's going on at this point?
There is no option for lazy-preemption. And yes, we should figure out what is causing this.
thanks, Paul
Sebastian