How about "virtually nmi" so that the RT side would ensure that
whenever linux itself ran again, the flush tlb would happen
first?
Flushing tlb may be fast, but it introduces some long delays.
Some of the people who want mp rt-linux will most likely almost
monopolize one processor with real-time stuff and making that
processor do tlb invalidates is an interesting problem. On PPC and
Alpha this should be easy. Not sure about SPARC, but I'm ignorant
about the interrupt controller on SPARCs.
>> C) Am I correct in assuming that the critical material to flush,
>> fter bootup, will be in user space.
>
>Most of the time, yes. That is the case you should optimize for, but you
>cannot _assume_ it - there are kernel page tables that get updated too by
>vmalloc() and friends. Those need to do cross-CPU invalidates as well (it
>doesn't happen often, and you may or may not be able to special case
>that).
Will look into it.
victor