perf_counter, x86: speed up the scheduling fast-path
We have to set up the LVT entry only at counter init time, not at
every switch-in time.
There's friction between NMI and non-NMI use here - we'll probably
remove the per counter configurability of it - but until then, dont
slow down things ...
[ Impact: micro-optimization ]
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 5bfd30a..c109819 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -285,6 +285,7 @@
return -EACCES;
hwc->nmi = 1;
}
+ perf_counters_lapic_init(hwc->nmi);
if (!hwc->irq_period)
hwc->irq_period = x86_pmu.max_period;
@@ -603,8 +604,6 @@
hwc->counter_base = x86_pmu.perfctr;
}
- perf_counters_lapic_init(hwc->nmi);
-
x86_pmu.disable(hwc, idx);
cpuc->counters[idx] = counter;
@@ -1054,7 +1053,7 @@
pr_info("... counter mask: %016Lx\n", perf_counter_mask);
- perf_counters_lapic_init(0);
+ perf_counters_lapic_init(1);
register_die_notifier(&perf_counter_nmi_notifier);
}