> If I understand correctly that would always look for a valid "cpu"
> even when none is needed. Not quite what I proposed but should
> work. Just some wasted cycles in a non critical path.
Reinette,
Sorry for misunderstanding. You are right. Assignment to msr_param.dom
can also be deferred to the same point. Like this:
list_for_each_entry(d, &r->domains, list) {
hw_dom = resctrl_to_arch_dom(d);
msr_param.res = NULL;
for (t = 0; t < CDP_NUM_TYPES; t++) {
cfg = &hw_dom->d_resctrl.staged_config[t];
if (!cfg->have_new_ctrl)
continue;
idx = get_config_index(closid, t);
if (cfg->new_ctrl == hw_dom->ctrl_val[idx])
continue;
hw_dom->ctrl_val[idx] = cfg->new_ctrl;
if (!msr_param.res) {
msr_param.low = idx;
msr_param.high = msr_param.low + 1;
msr_param.res = r;
msr_param.dom = d;
cpu = cpumask_any(&d->cpu_mask);
} else {
msr_param.low = min(msr_param.low, idx);
msr_param.high = max(msr_param.high, idx + 1);
}
}
if (msr_param.res)
smp_call_function_single(cpu, rdt_ctrl_update, &msr_param, 1);
}
-Tony
