Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf fixes from Ingo Molnar:
 "Most of the kernel diffstat relates to a group of Intel P6 and KNC
  (Xeon-Phi Knights Corner) PMU driver fixes, neither of which is in
  heavy use, so we took the fixes.

  The rest is diverse smallish fixes to the tooling and kernel side."

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86: Remove unused variable in nhmex_rbox_alter_er()
  perf/x86: Enable overflow on Intel KNC with a custom knc_pmu_handle_irq()
  perf/x86: Remove cpuc->enable check on Intl KNC event enable/disable
  perf/x86: Make Intel KNC use full 40-bit width of counters
  perf/x86/uncore: Handle pci_read_config_dword() errors
  perf/x86: Remove P6 cpuc->enabled check
  perf/x86: Update/fix generic events on P6 PMU
  perf/x86: Fix P6 FP_ASSIST event constraint
  perf, cpu hotplug: Use cached value of smp_processor_id()
  perf, cpu hotplug: Run CPU_STARTING notifiers with irqs disabled
  x86/perf: Fix virtualization sanity check
  perf test: Fix exclude_guest parse events tests
  perf tools: do not flush maps on COMM for perf report
  perf help: Fix --help for builtins
  perf trace: Check if sample raw_data field is set
  perf trace: Validate syscall id before growing syscall table

4 files changed, 228 insertions, 47 deletions

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 3373f84d1397..4a3374e61a93 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -208,12 +208,14 @@ static bool check_hw_exists(void)
 	}
 
 	/*
-	 * Now write a value and read it back to see if it matches,
-	 * this is needed to detect certain hardware emulators (qemu/kvm)
-	 * that don't trap on the MSR access and always return 0s.
+	 * Read the current value, change it and read it back to see if it
+	 * matches, this is needed to detect certain hardware emulators
+	 * (qemu/kvm) that don't trap on the MSR access and always return 0s.
 	 */
-	val = 0xabcdUL;
 	reg = x86_pmu_event_addr(0);
+	if (rdmsrl_safe(reg, &val))
+		goto msr_fail;
+	val ^= 0xffffUL;
 	ret = wrmsrl_safe(reg, val);
 	ret |= rdmsrl_safe(reg, &val_new);
 	if (ret || val != val_new)
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
index 5df8d32ba91e..3cf3d97cce3a 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
@@ -118,22 +118,24 @@ static void snbep_uncore_pci_disable_box(struct intel_uncore_box *box)
 {
 	struct pci_dev *pdev = box->pci_dev;
 	int box_ctl = uncore_pci_box_ctl(box);
-	u32 config;
+	u32 config = 0;
 
-	pci_read_config_dword(pdev, box_ctl, &config);
-	config |= SNBEP_PMON_BOX_CTL_FRZ;
-	pci_write_config_dword(pdev, box_ctl, config);
+	if (!pci_read_config_dword(pdev, box_ctl, &config)) {
+		config |= SNBEP_PMON_BOX_CTL_FRZ;
+		pci_write_config_dword(pdev, box_ctl, config);
+	}
 }
 
 static void snbep_uncore_pci_enable_box(struct intel_uncore_box *box)
 {
 	struct pci_dev *pdev = box->pci_dev;
 	int box_ctl = uncore_pci_box_ctl(box);
-	u32 config;
+	u32 config = 0;
 
-	pci_read_config_dword(pdev, box_ctl, &config);
-	config &= ~SNBEP_PMON_BOX_CTL_FRZ;
-	pci_write_config_dword(pdev, box_ctl, config);
+	if (!pci_read_config_dword(pdev, box_ctl, &config)) {
+		config &= ~SNBEP_PMON_BOX_CTL_FRZ;
+		pci_write_config_dword(pdev, box_ctl, config);
+	}
 }
 
 static void snbep_uncore_pci_enable_event(struct intel_uncore_box *box, struct perf_event *event)
@@ -156,7 +158,7 @@ static u64 snbep_uncore_pci_read_counter(struct intel_uncore_box *box, struct pe
 {
 	struct pci_dev *pdev = box->pci_dev;
 	struct hw_perf_event *hwc = &event->hw;
-	u64 count;
+	u64 count = 0;
 
 	pci_read_config_dword(pdev, hwc->event_base, (u32 *)&count);
 	pci_read_config_dword(pdev, hwc->event_base + 4, (u32 *)&count + 1);
@@ -603,11 +605,12 @@ static struct pci_driver snbep_uncore_pci_driver = {
 /*
  * build pci bus to socket mapping
  */
-static void snbep_pci2phy_map_init(void)
+static int snbep_pci2phy_map_init(void)
 {
 	struct pci_dev *ubox_dev = NULL;
 	int i, bus, nodeid;
-	u32 config;
+	int err = 0;
+	u32 config = 0;
 
 	while (1) {
 		/* find the UBOX device */
@@ -618,10 +621,14 @@ static void snbep_pci2phy_map_init(void)
 			break;
 		bus = ubox_dev->bus->number;
 		/* get the Node ID of the local register */
-		pci_read_config_dword(ubox_dev, 0x40, &config);
+		err = pci_read_config_dword(ubox_dev, 0x40, &config);
+		if (err)
+			break;
 		nodeid = config;
 		/* get the Node ID mapping */
-		pci_read_config_dword(ubox_dev, 0x54, &config);
+		err = pci_read_config_dword(ubox_dev, 0x54, &config);
+		if (err)
+			break;
 		/*
 		 * every three bits in the Node ID mapping register maps
 		 * to a particular node.
@@ -633,7 +640,11 @@ static void snbep_pci2phy_map_init(void)
 			}
 		}
 	};
-	return;
+
+	if (ubox_dev)
+		pci_dev_put(ubox_dev);
+
+	return err ? pcibios_err_to_errno(err) : 0;
 }
 /* end of Sandy Bridge-EP uncore support */
 
@@ -1547,7 +1558,6 @@ void nhmex_rbox_alter_er(struct intel_uncore_box *box, struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	struct hw_perf_event_extra *reg1 = &hwc->extra_reg;
-	int port;
 
 	/* adjust the main event selector and extra register index */
 	if (reg1->idx % 2) {
@@ -1559,7 +1569,6 @@ void nhmex_rbox_alter_er(struct intel_uncore_box *box, struct perf_event *event)
 	}
 
 	/* adjust extra register config */
-	port = reg1->idx / 6 + box->pmu->pmu_idx * 4;
 	switch (reg1->idx % 6) {
 	case 2:
 		/* shift the 8~15 bits to the 0~7 bits */
@@ -2578,9 +2587,11 @@ static int __init uncore_pci_init(void)
 
 	switch (boot_cpu_data.x86_model) {
 	case 45: /* Sandy Bridge-EP */
+		ret = snbep_pci2phy_map_init();
+		if (ret)
+			return ret;
 		pci_uncores = snbep_pci_uncores;
 		uncore_pci_driver = &snbep_uncore_pci_driver;
-		snbep_pci2phy_map_init();
 		break;
 	default:
 		return 0;
diff --git a/arch/x86/kernel/cpu/perf_event_knc.c b/arch/x86/kernel/cpu/perf_event_knc.c
index 7c46bfdbc373..4b7731bf23a8 100644
--- a/arch/x86/kernel/cpu/perf_event_knc.c
+++ b/arch/x86/kernel/cpu/perf_event_knc.c
@@ -3,6 +3,8 @@
 #include <linux/perf_event.h>
 #include <linux/types.h>
 
+#include <asm/hardirq.h>
+
 #include "perf_event.h"
 
 static const u64 knc_perfmon_event_map[] =
@@ -173,30 +175,100 @@ static void knc_pmu_enable_all(int added)
 static inline void
 knc_pmu_disable_event(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	u64 val;
 
 	val = hwc->config;
-	if (cpuc->enabled)
-		val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+	val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
 
 	(void)wrmsrl_safe(hwc->config_base + hwc->idx, val);
 }
 
 static void knc_pmu_enable_event(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	u64 val;
 
 	val = hwc->config;
-	if (cpuc->enabled)
-		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
+	val |= ARCH_PERFMON_EVENTSEL_ENABLE;
 
 	(void)wrmsrl_safe(hwc->config_base + hwc->idx, val);
 }
 
+static inline u64 knc_pmu_get_status(void)
+{
+	u64 status;
+
+	rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_STATUS, status);
+
+	return status;
+}
+
+static inline void knc_pmu_ack_status(u64 ack)
+{
+	wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_OVF_CONTROL, ack);
+}
+
+static int knc_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc;
+	int handled = 0;
+	int bit, loops;
+	u64 status;
+
+	cpuc = &__get_cpu_var(cpu_hw_events);
+
+	knc_pmu_disable_all();
+
+	status = knc_pmu_get_status();
+	if (!status) {
+		knc_pmu_enable_all(0);
+		return handled;
+	}
+
+	loops = 0;
+again:
+	knc_pmu_ack_status(status);
+	if (++loops > 100) {
+		WARN_ONCE(1, "perf: irq loop stuck!\n");
+		perf_event_print_debug();
+		goto done;
+	}
+
+	inc_irq_stat(apic_perf_irqs);
+
+	for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) {
+		struct perf_event *event = cpuc->events[bit];
+
+		handled++;
+
+		if (!test_bit(bit, cpuc->active_mask))
+			continue;
+
+		if (!intel_pmu_save_and_restart(event))
+			continue;
+
+		perf_sample_data_init(&data, 0, event->hw.last_period);
+
+		if (perf_event_overflow(event, &data, regs))
+			x86_pmu_stop(event, 0);
+	}
+
+	/*
+	 * Repeat if there is more work to be done:
+	 */
+	status = knc_pmu_get_status();
+	if (status)
+		goto again;
+
+done:
+	knc_pmu_enable_all(0);
+
+	return handled;
+}
+
+
 PMU_FORMAT_ATTR(event,	"config:0-7"	);
 PMU_FORMAT_ATTR(umask,	"config:8-15"	);
 PMU_FORMAT_ATTR(edge,	"config:18"	);
@@ -214,7 +286,7 @@ static struct attribute *intel_knc_formats_attr[] = {
 
 static __initconst struct x86_pmu knc_pmu = {
 	.name			= "knc",
-	.handle_irq		= x86_pmu_handle_irq,
+	.handle_irq		= knc_pmu_handle_irq,
 	.disable_all		= knc_pmu_disable_all,
 	.enable_all		= knc_pmu_enable_all,
 	.enable			= knc_pmu_enable_event,
@@ -226,12 +298,11 @@ static __initconst struct x86_pmu knc_pmu = {
 	.event_map		= knc_pmu_event_map,
 	.max_events             = ARRAY_SIZE(knc_perfmon_event_map),
 	.apic			= 1,
-	.max_period		= (1ULL << 31) - 1,
+	.max_period		= (1ULL << 39) - 1,
 	.version		= 0,
 	.num_counters		= 2,
-	/* in theory 40 bits, early silicon is buggy though */
-	.cntval_bits		= 32,
-	.cntval_mask		= (1ULL << 32) - 1,
+	.cntval_bits		= 40,
+	.cntval_mask		= (1ULL << 40) - 1,
 	.get_event_constraints	= x86_get_event_constraints,
 	.event_constraints	= knc_event_constraints,
 	.format_attrs		= intel_knc_formats_attr,
diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c
index e4dd0f7a0453..7d0270bd793e 100644
--- a/arch/x86/kernel/cpu/perf_event_p6.c
+++ b/arch/x86/kernel/cpu/perf_event_p6.c
@@ -8,13 +8,106 @@
  */
 static const u64 p6_perfmon_event_map[] =
 {
-  [PERF_COUNT_HW_CPU_CYCLES]		= 0x0079,
-  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
-  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x0f2e,
-  [PERF_COUNT_HW_CACHE_MISSES]		= 0x012e,
-  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
-  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
-  [PERF_COUNT_HW_BUS_CYCLES]		= 0x0062,
+  [PERF_COUNT_HW_CPU_CYCLES]		= 0x0079,	/* CPU_CLK_UNHALTED */
+  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,	/* INST_RETIRED     */
+  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x0f2e,	/* L2_RQSTS:M:E:S:I */
+  [PERF_COUNT_HW_CACHE_MISSES]		= 0x012e,	/* L2_RQSTS:I       */
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,	/* BR_INST_RETIRED  */
+  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,	/* BR_MISS_PRED_RETIRED */
+  [PERF_COUNT_HW_BUS_CYCLES]		= 0x0062,	/* BUS_DRDY_CLOCKS  */
+  [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x00a2,	/* RESOURCE_STALLS  */
+
+};
+
+static __initconst u64 p6_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0043,	/* DATA_MEM_REFS       */
+                [ C(RESULT_MISS)   ] = 0x0045,	/* DCU_LINES_IN        */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x0f29,	/* L2_LD:M:E:S:I       */
+	},
+        [ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+        },
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080,	/* IFU_IFETCH         */
+		[ C(RESULT_MISS)   ] = 0x0f28,	/* L2_IFETCH:M:E:S:I  */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x0025,	/* L2_M_LINES_INM     */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0043,	/* DATA_MEM_REFS      */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080,	/* IFU_IFETCH         */
+		[ C(RESULT_MISS)   ] = 0x0085,	/* ITLB_MISS          */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c4,	/* BR_INST_RETIRED      */
+		[ C(RESULT_MISS)   ] = 0x00c5,	/* BR_MISS_PRED_RETIRED */
+        },
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
 };
 
 static u64 p6_pmu_event_map(int hw_event)
@@ -34,7 +127,7 @@ static struct event_constraint p6_event_constraints[] =
 {
 	INTEL_EVENT_CONSTRAINT(0xc1, 0x1),	/* FLOPS */
 	INTEL_EVENT_CONSTRAINT(0x10, 0x1),	/* FP_COMP_OPS_EXE */
-	INTEL_EVENT_CONSTRAINT(0x11, 0x1),	/* FP_ASSIST */
+	INTEL_EVENT_CONSTRAINT(0x11, 0x2),	/* FP_ASSIST */
 	INTEL_EVENT_CONSTRAINT(0x12, 0x2),	/* MUL */
 	INTEL_EVENT_CONSTRAINT(0x13, 0x2),	/* DIV */
 	INTEL_EVENT_CONSTRAINT(0x14, 0x1),	/* CYCLES_DIV_BUSY */
@@ -64,25 +157,25 @@ static void p6_pmu_enable_all(int added)
 static inline void
 p6_pmu_disable_event(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	u64 val = P6_NOP_EVENT;
 
-	if (cpuc->enabled)
-		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
-
 	(void)wrmsrl_safe(hwc->config_base, val);
 }
 
 static void p6_pmu_enable_event(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	u64 val;
 
 	val = hwc->config;
-	if (cpuc->enabled)
-		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
+
+	/*
+	 * p6 only has a global event enable, set on PerfEvtSel0
+	 * We "disable" events by programming P6_NOP_EVENT
+	 * and we rely on p6_pmu_enable_all() being called
+	 * to actually enable the events.
+	 */
 
 	(void)wrmsrl_safe(hwc->config_base, val);
 }
@@ -158,5 +251,9 @@ __init int p6_pmu_init(void)
 
 	x86_pmu = p6_pmu;
 
+	memcpy(hw_cache_event_ids, p6_hw_cache_event_ids,
+		sizeof(hw_cache_event_ids));
+
+
 	return 0;
 }