Messages in this thread | | | Date | Wed, 12 Feb 2020 11:09:40 +0100 | From | Peter Zijlstra <> | Subject | Re: [PATCH 0/8] tracing vs rcu vs nmi |
| |
On Wed, Feb 12, 2020 at 10:32:10AM +0100, Peter Zijlstra wrote: > Hi all, > > These here patches are the result of Mathieu and Steve trying to get commit > 865e63b04e9b2 ("tracing: Add back in rcu_irq_enter/exit_irqson() for rcuidle > tracepoints") reverted again. > > One of the things discovered is that tracing MUST NOT happen before nmi_enter() > or after nmi_exit(). I've only fixed x86, but quickly gone through other > architectures and there is definitely more stuff to be fixed (simply grep for > nmi_enter in your arch).
Someone should probably look at the whole ARM FiQ stuff, I got a little lost, but probably handle_fiq_as_nmi() wants notrace and everything using set_fiq_handler() wants looking at.
| |