Re: [PATCH V5 04/25] perf/x86/intel: Hybrid PMU support for perf capabilities

[Liang, Kan]

On 4/8/2021 1:00 PM, Peter Zijlstra wrote:
> On Mon, Apr 05, 2021 at 08:10:46AM -0700, kan.liang@xxxxxxxxxxxxxxx wrote:
> > +#define is_hybrid()			(!!x86_pmu.num_hybrid_pmus)
>
> Given this is sprinkled all over the place, can you make this a
> static_key_false + static_branch_unlikely() such that the hybrid case is
> out-of-line?

Sure, I will add a new static_key_false "perf_is_hybrid" to indicate a 
hybrid system as below (not test yet).

diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index f8d1222..bd6412e 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -54,6 +54,7 @@ DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {

 DEFINE_STATIC_KEY_FALSE(rdpmc_never_available_key);
 DEFINE_STATIC_KEY_FALSE(rdpmc_always_available_key);
+DEFINE_STATIC_KEY_FALSE(perf_is_hybrid);

 /*
  * This here uses DEFINE_STATIC_CALL_NULL() to get a static_call defined
diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 2b553d9..7cef3cd 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -6119,6 +6119,7 @@ __init int intel_pmu_init(void)
 					     GFP_KERNEL);
 		if (!x86_pmu.hybrid_pmu)
 			return -ENOMEM;
+		static_branch_enable(&perf_is_hybrid);
 		x86_pmu.num_hybrid_pmus = X86_HYBRID_NUM_PMUS;

 		x86_pmu.late_ack = true;
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index bfbecde..d6383d1 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -663,8 +663,8 @@ static __always_inline struct x86_hybrid_pmu 
*hybrid_pmu(struct pmu *pmu)
 	return container_of(pmu, struct x86_hybrid_pmu, pmu);
 }

-/* The number of hybrid PMUs implies whether it's a hybrid system */
-#define is_hybrid()			(!!x86_pmu.num_hybrid_pmus)
+extern struct static_key_false perf_is_hybrid;
+#define is_hybrid()		static_branch_unlikely(&perf_is_hybrid)

 #define hybrid(_pmu, _field)				\
 ({							\

Thanks,
Kan