Mulyadi Santosa schrieb:
Hi...
Sorry if my earlier post confused some people. What I wanted to say
was actually that spinlock is SMP safe, however, in multicore it makes
code running on a cpu might spin to wait for other code running on
different cpu.
so, to avoid this, perhaps the original poster could gain better
scalability by adopting per CPU data structure. Therefore, spin lock
usage is reduced only on cases e.g data must be global across CPU or
blocking wait is not an option
regards,
Mulyadi.
Thanks for the infos. I will have a look at these per CPU Data
structures. But i think what I need to avoid a possible error is
something like this:
while(spinlock_is_locked(...)){wait(...)}
Because I'm developing for a custom platform, I don't care about several
processors that much...
Regards,
Ole
