perf/x86/intel: Add mem-loads/stores support for Haswell

mem-loads is basically the same as Sandy Bridge,
but we use a separate string for changes later.

Haswell doesn't support the full precise store mode,
so we emulate it using the "DataLA" facility.
This allows to do everything, but for data sources we
can only detect L1 hit or not.

There is no explicit enable bit anymore, so we have
to tie it to a perf internal only flag.

The address is supported for all memory related PEBS
events with DataLA. Instead of only logging for the
load and store events we allow logging it for all
(it will be simply 0 if the current event does not
support it)

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Andi Kleen <ak@linux.jf.intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Link: http://lkml.kernel.org/r/1371515812-9646-7-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>

arch/x86/kernel/cpu/perf_event.h

@@ -67,6 +67,7 @@ struct event_constraint {
  */
 #define PERF_X86_EVENT_PEBS_LDLAT	0x1 /* ld+ldlat data address sampling */
 #define PERF_X86_EVENT_PEBS_ST		0x2 /* st data address sampling */
+#define PERF_X86_EVENT_PEBS_ST_HSW	0x4 /* haswell style st data sampling */
 
 struct amd_nb {
 	int nb_id;  /* NorthBridge id */
@@ -250,6 +251,11 @@ struct cpu_hw_events {
 	__EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK, \
 			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST)
 
+/* DataLA version of store sampling without extra enable bit. */
+#define INTEL_PST_HSW_CONSTRAINT(c, n)	\
+	__EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST_HSW)
+
 #define EVENT_CONSTRAINT_END		\
 	EVENT_CONSTRAINT(0, 0, 0)
 

arch/x86/kernel/cpu/perf_event_intel.c

@@ -2036,6 +2036,15 @@ static __init void intel_nehalem_quirk(void)
 	}
 }
 
+EVENT_ATTR_STR(mem-loads,      mem_ld_hsw,     "event=0xcd,umask=0x1,ldlat=3");
+EVENT_ATTR_STR(mem-stores,     mem_st_hsw,     "event=0xd0,umask=0x82")
+
+static struct attribute *hsw_events_attrs[] = {
+	EVENT_PTR(mem_ld_hsw),
+	EVENT_PTR(mem_st_hsw),
+	NULL
+};
+
 __init int intel_pmu_init(void)
 {
 	union cpuid10_edx edx;
@@ -2279,6 +2288,7 @@ __init int intel_pmu_init(void)
 
 		x86_pmu.hw_config = hsw_hw_config;
 		x86_pmu.get_event_constraints = hsw_get_event_constraints;
+		x86_pmu.cpu_events = hsw_events_attrs;
 		pr_cont("Haswell events, ");
 		break;
 

arch/x86/kernel/cpu/perf_event_intel_ds.c

@@ -107,6 +107,19 @@ static u64 precise_store_data(u64 status)
 	return val;
 }
 
+static u64 precise_store_data_hsw(u64 status)
+{
+	union perf_mem_data_src dse;
+
+	dse.val = 0;
+	dse.mem_op = PERF_MEM_OP_STORE;
+	dse.mem_lvl = PERF_MEM_LVL_NA;
+	if (status & 1)
+		dse.mem_lvl = PERF_MEM_LVL_L1;
+	/* Nothing else supported. Sorry. */
+	return dse.val;
+}
+
 static u64 load_latency_data(u64 status)
 {
 	union intel_x86_pebs_dse dse;
@@ -566,13 +579,13 @@ struct event_constraint intel_ivb_pebs_event_constraints[] = {
 
 struct event_constraint intel_hsw_pebs_event_constraints[] = {
 	INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
-	INTEL_UEVENT_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
+	INTEL_PST_HSW_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
 	INTEL_UEVENT_CONSTRAINT(0x02c2, 0xf), /* UOPS_RETIRED.RETIRE_SLOTS */
 	INTEL_EVENT_CONSTRAINT(0xc4, 0xf),    /* BR_INST_RETIRED.* */
 	INTEL_UEVENT_CONSTRAINT(0x01c5, 0xf), /* BR_MISP_RETIRED.CONDITIONAL */
 	INTEL_UEVENT_CONSTRAINT(0x04c5, 0xf), /* BR_MISP_RETIRED.ALL_BRANCHES */
 	INTEL_UEVENT_CONSTRAINT(0x20c5, 0xf), /* BR_MISP_RETIRED.NEAR_TAKEN */
-	INTEL_EVENT_CONSTRAINT(0xcd, 0x8),    /* MEM_TRANS_RETIRED.* */
+	INTEL_PLD_CONSTRAINT(0x01cd, 0x8),    /* MEM_TRANS_RETIRED.* */
 	/* MEM_UOPS_RETIRED.STLB_MISS_LOADS */
 	INTEL_UEVENT_CONSTRAINT(0x11d0, 0xf),
 	/* MEM_UOPS_RETIRED.STLB_MISS_STORES */
@@ -582,7 +595,7 @@ struct event_constraint intel_hsw_pebs_event_constraints[] = {
 	/* MEM_UOPS_RETIRED.SPLIT_STORES */
 	INTEL_UEVENT_CONSTRAINT(0x42d0, 0xf),
 	INTEL_UEVENT_CONSTRAINT(0x81d0, 0xf), /* MEM_UOPS_RETIRED.ALL_LOADS */
-	INTEL_UEVENT_CONSTRAINT(0x82d0, 0xf), /* MEM_UOPS_RETIRED.ALL_STORES */
+	INTEL_PST_HSW_CONSTRAINT(0x82d0, 0xf), /* MEM_UOPS_RETIRED.ALL_STORES */
 	INTEL_UEVENT_CONSTRAINT(0x01d1, 0xf), /* MEM_LOAD_UOPS_RETIRED.L1_HIT */
 	INTEL_UEVENT_CONSTRAINT(0x02d1, 0xf), /* MEM_LOAD_UOPS_RETIRED.L2_HIT */
 	INTEL_UEVENT_CONSTRAINT(0x04d1, 0xf), /* MEM_LOAD_UOPS_RETIRED.L3_HIT */
@@ -759,7 +772,8 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 		return;
 
 	fll = event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT;
-	fst = event->hw.flags & PERF_X86_EVENT_PEBS_ST;
+	fst = event->hw.flags & (PERF_X86_EVENT_PEBS_ST |
+				 PERF_X86_EVENT_PEBS_ST_HSW);
 
 	perf_sample_data_init(&data, 0, event->hw.last_period);
 
@@ -770,9 +784,6 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 	 * if PEBS-LL or PreciseStore
 	 */
 	if (fll || fst) {
-		if (sample_type & PERF_SAMPLE_ADDR)
-			data.addr = pebs->dla;
-
 		/*
 		 * Use latency for weight (only avail with PEBS-LL)
 		 */
@@ -785,6 +796,9 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 		if (sample_type & PERF_SAMPLE_DATA_SRC) {
 			if (fll)
 				data.data_src.val = load_latency_data(pebs->dse);
+			else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST_HSW)
+				data.data_src.val =
+					precise_store_data_hsw(pebs->dse);
 			else
 				data.data_src.val = precise_store_data(pebs->dse);
 		}
@@ -814,6 +828,10 @@ static void __intel_pmu_pebs_event(struct perf_event *event,
 	else
 		regs.flags &= ~PERF_EFLAGS_EXACT;
 
+	if ((event->attr.sample_type & PERF_SAMPLE_ADDR) &&
+		x86_pmu.intel_cap.pebs_format >= 1)
+		data.addr = pebs->dla;
+
 	if (has_branch_stack(event))
 		data.br_stack = &cpuc->lbr_stack;