Adding upstream version 5.10.209.upstream/5.10.209upstream

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

33 files changed, 34916 insertions, 0 deletions

diff --git a/arch/x86/events/Kconfig b/arch/x86/events/Kconfig
new file mode 100644
index 000000000..39d9ded9e
--- /dev/null
+++ b/arch/x86/events/Kconfig
@@ -0,0 +1,37 @@
+# SPDX-License-Identifier: GPL-2.0
+menu "Performance monitoring"
+
+config PERF_EVENTS_INTEL_UNCORE
+	tristate "Intel uncore performance events"
+	depends on PERF_EVENTS && CPU_SUP_INTEL && PCI
+	default y
+	help
+	Include support for Intel uncore performance events. These are
+	available on NehalemEX and more modern processors.
+
+config PERF_EVENTS_INTEL_RAPL
+	tristate "Intel/AMD rapl performance events"
+	depends on PERF_EVENTS && (CPU_SUP_INTEL || CPU_SUP_AMD) && PCI
+	default y
+	help
+	Include support for Intel and AMD rapl performance events for power
+	monitoring on modern processors.
+
+config PERF_EVENTS_INTEL_CSTATE
+	tristate "Intel cstate performance events"
+	depends on PERF_EVENTS && CPU_SUP_INTEL && PCI
+	default y
+	help
+	Include support for Intel cstate performance events for power
+	monitoring on modern processors.
+
+config PERF_EVENTS_AMD_POWER
+	depends on PERF_EVENTS && CPU_SUP_AMD
+	tristate "AMD Processor Power Reporting Mechanism"
+	help
+	  Provide power reporting mechanism support for AMD processors.
+	  Currently, it leverages X86_FEATURE_ACC_POWER
+	  (CPUID Fn8000_0007_EDX[12]) interface to calculate the
+	  average power consumption on Family 15h processors.
+
+endmenu
diff --git a/arch/x86/events/Makefile b/arch/x86/events/Makefile
new file mode 100644
index 000000000..9933c0e8e
--- /dev/null
+++ b/arch/x86/events/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
+obj-y					+= core.o probe.o
+obj-$(CONFIG_PERF_EVENTS_INTEL_RAPL)	+= rapl.o
+obj-y					+= amd/
+obj-$(CONFIG_X86_LOCAL_APIC)            += msr.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= intel/
+obj-$(CONFIG_CPU_SUP_CENTAUR)		+= zhaoxin/
+obj-$(CONFIG_CPU_SUP_ZHAOXIN)		+= zhaoxin/
diff --git a/arch/x86/events/amd/Makefile b/arch/x86/events/amd/Makefile
new file mode 100644
index 000000000..fe8795a67
--- /dev/null
+++ b/arch/x86/events/amd/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_CPU_SUP_AMD)		+= core.o uncore.o
+obj-$(CONFIG_PERF_EVENTS_AMD_POWER)	+= power.o
+obj-$(CONFIG_X86_LOCAL_APIC)		+= ibs.o
+ifdef CONFIG_AMD_IOMMU
+obj-$(CONFIG_CPU_SUP_AMD)		+= iommu.o
+endif
+
diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c
new file mode 100644
index 000000000..afc955340
--- /dev/null
+++ b/arch/x86/events/amd/core.c
@@ -0,0 +1,1054 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/perf_event.h>
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/jiffies.h>
+#include <asm/apicdef.h>
+#include <asm/nmi.h>
+
+#include "../perf_event.h"
+
+static DEFINE_PER_CPU(unsigned long, perf_nmi_tstamp);
+static unsigned long perf_nmi_window;
+
+/* AMD Event 0xFFF: Merge.  Used with Large Increment per Cycle events */
+#define AMD_MERGE_EVENT ((0xFULL << 32) | 0xFFULL)
+#define AMD_MERGE_EVENT_ENABLE (AMD_MERGE_EVENT | ARCH_PERFMON_EVENTSEL_ENABLE)
+
+static __initconst const u64 amd_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) ] = 0x0040, /* Data Cache Accesses        */
+		[ C(RESULT_MISS)   ] = 0x0141, /* Data Cache Misses          */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts  */
+		[ C(RESULT_MISS)   ] = 0x0167, /* Data Prefetcher :cancelled */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction cache fetches  */
+		[ C(RESULT_MISS)   ] = 0x0081, /* Instruction cache misses   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */
+		[ C(RESULT_MISS)   ] = 0x037E, /* L2 Cache Misses : IC+DC     */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback           */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses        */
+		[ C(RESULT_MISS)   ] = 0x0746, /* L1_DTLB_AND_L2_DLTB_MISS.ALL */
+	},
+	[ 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, /* Instruction fecthes        */
+		[ C(RESULT_MISS)   ] = 0x0385, /* L1_ITLB_AND_L2_ITLB_MISS.ALL */
+	},
+	[ 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) ] = 0x00c2, /* Retired Branch Instr.      */
+		[ C(RESULT_MISS)   ] = 0x00c3, /* Retired Mispredicted BI    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0xb8e9, /* CPU Request to Memory, l+r */
+		[ C(RESULT_MISS)   ] = 0x98e9, /* CPU Request to Memory, r   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static __initconst const u64 amd_hw_cache_event_ids_f17h
+				[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)] = 0x0040, /* Data Cache Accesses */
+		[C(RESULT_MISS)]   = 0xc860, /* L2$ access from DC Miss */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0,
+		[C(RESULT_MISS)]   = 0,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0xff5a, /* h/w prefetch DC Fills */
+		[C(RESULT_MISS)]   = 0,
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0080, /* Instruction cache fetches  */
+		[C(RESULT_MISS)]   = 0x0081, /* Instruction cache misses   */
+	},
+	[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)]   = 0,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0,
+		[C(RESULT_MISS)]   = 0,
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0xff45, /* All L2 DTLB accesses */
+		[C(RESULT_MISS)]   = 0xf045, /* L2 DTLB misses (PT walks) */
+	},
+	[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)] = 0x0084, /* L1 ITLB misses, L2 ITLB hits */
+		[C(RESULT_MISS)]   = 0xff85, /* L1 ITLB misses, L2 misses */
+	},
+	[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)] = 0x00c2, /* Retired Branch Instr.      */
+		[C(RESULT_MISS)]   = 0x00c3, /* Retired Mispredicted BI    */
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)]   = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)]   = -1,
+	},
+},
+[C(NODE)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0,
+		[C(RESULT_MISS)]   = 0,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)]   = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)]   = -1,
+	},
+},
+};
+
+/*
+ * AMD Performance Monitor K7 and later, up to and including Family 16h:
+ */
+static const u64 amd_perfmon_event_map[PERF_COUNT_HW_MAX] =
+{
+	[PERF_COUNT_HW_CPU_CYCLES]		= 0x0076,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= 0x077d,
+	[PERF_COUNT_HW_CACHE_MISSES]		= 0x077e,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c2,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c3,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= 0x00d0, /* "Decoder empty" event */
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= 0x00d1, /* "Dispatch stalls" event */
+};
+
+/*
+ * AMD Performance Monitor Family 17h and later:
+ */
+static const u64 amd_f17h_perfmon_event_map[PERF_COUNT_HW_MAX] =
+{
+	[PERF_COUNT_HW_CPU_CYCLES]		= 0x0076,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= 0xff60,
+	[PERF_COUNT_HW_CACHE_MISSES]		= 0x0964,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c2,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c3,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= 0x0287,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= 0x0187,
+};
+
+static u64 amd_pmu_event_map(int hw_event)
+{
+	if (boot_cpu_data.x86 >= 0x17)
+		return amd_f17h_perfmon_event_map[hw_event];
+
+	return amd_perfmon_event_map[hw_event];
+}
+
+/*
+ * Previously calculated offsets
+ */
+static unsigned int event_offsets[X86_PMC_IDX_MAX] __read_mostly;
+static unsigned int count_offsets[X86_PMC_IDX_MAX] __read_mostly;
+
+/*
+ * Legacy CPUs:
+ *   4 counters starting at 0xc0010000 each offset by 1
+ *
+ * CPUs with core performance counter extensions:
+ *   6 counters starting at 0xc0010200 each offset by 2
+ */
+static inline int amd_pmu_addr_offset(int index, bool eventsel)
+{
+	int offset;
+
+	if (!index)
+		return index;
+
+	if (eventsel)
+		offset = event_offsets[index];
+	else
+		offset = count_offsets[index];
+
+	if (offset)
+		return offset;
+
+	if (!boot_cpu_has(X86_FEATURE_PERFCTR_CORE))
+		offset = index;
+	else
+		offset = index << 1;
+
+	if (eventsel)
+		event_offsets[index] = offset;
+	else
+		count_offsets[index] = offset;
+
+	return offset;
+}
+
+/*
+ * AMD64 events are detected based on their event codes.
+ */
+static inline unsigned int amd_get_event_code(struct hw_perf_event *hwc)
+{
+	return ((hwc->config >> 24) & 0x0f00) | (hwc->config & 0x00ff);
+}
+
+static inline bool amd_is_pair_event_code(struct hw_perf_event *hwc)
+{
+	if (!(x86_pmu.flags & PMU_FL_PAIR))
+		return false;
+
+	switch (amd_get_event_code(hwc)) {
+	case 0x003:	return true;	/* Retired SSE/AVX FLOPs */
+	default:	return false;
+	}
+}
+
+static int amd_core_hw_config(struct perf_event *event)
+{
+	if (event->attr.exclude_host && event->attr.exclude_guest)
+		/*
+		 * When HO == GO == 1 the hardware treats that as GO == HO == 0
+		 * and will count in both modes. We don't want to count in that
+		 * case so we emulate no-counting by setting US = OS = 0.
+		 */
+		event->hw.config &= ~(ARCH_PERFMON_EVENTSEL_USR |
+				      ARCH_PERFMON_EVENTSEL_OS);
+	else if (event->attr.exclude_host)
+		event->hw.config |= AMD64_EVENTSEL_GUESTONLY;
+	else if (event->attr.exclude_guest)
+		event->hw.config |= AMD64_EVENTSEL_HOSTONLY;
+
+	if ((x86_pmu.flags & PMU_FL_PAIR) && amd_is_pair_event_code(&event->hw))
+		event->hw.flags |= PERF_X86_EVENT_PAIR;
+
+	return 0;
+}
+
+static inline int amd_is_nb_event(struct hw_perf_event *hwc)
+{
+	return (hwc->config & 0xe0) == 0xe0;
+}
+
+static inline int amd_has_nb(struct cpu_hw_events *cpuc)
+{
+	struct amd_nb *nb = cpuc->amd_nb;
+
+	return nb && nb->nb_id != -1;
+}
+
+static int amd_pmu_hw_config(struct perf_event *event)
+{
+	int ret;
+
+	/* pass precise event sampling to ibs: */
+	if (event->attr.precise_ip && get_ibs_caps())
+		return forward_event_to_ibs(event);
+
+	if (has_branch_stack(event))
+		return -EOPNOTSUPP;
+
+	ret = x86_pmu_hw_config(event);
+	if (ret)
+		return ret;
+
+	if (event->attr.type == PERF_TYPE_RAW)
+		event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK;
+
+	return amd_core_hw_config(event);
+}
+
+static void __amd_put_nb_event_constraints(struct cpu_hw_events *cpuc,
+					   struct perf_event *event)
+{
+	struct amd_nb *nb = cpuc->amd_nb;
+	int i;
+
+	/*
+	 * need to scan whole list because event may not have
+	 * been assigned during scheduling
+	 *
+	 * no race condition possible because event can only
+	 * be removed on one CPU at a time AND PMU is disabled
+	 * when we come here
+	 */
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		if (cmpxchg(nb->owners + i, event, NULL) == event)
+			break;
+	}
+}
+
+ /*
+  * AMD64 NorthBridge events need special treatment because
+  * counter access needs to be synchronized across all cores
+  * of a package. Refer to BKDG section 3.12
+  *
+  * NB events are events measuring L3 cache, Hypertransport
+  * traffic. They are identified by an event code >= 0xe00.
+  * They measure events on the NorthBride which is shared
+  * by all cores on a package. NB events are counted on a
+  * shared set of counters. When a NB event is programmed
+  * in a counter, the data actually comes from a shared
+  * counter. Thus, access to those counters needs to be
+  * synchronized.
+  *
+  * We implement the synchronization such that no two cores
+  * can be measuring NB events using the same counters. Thus,
+  * we maintain a per-NB allocation table. The available slot
+  * is propagated using the event_constraint structure.
+  *
+  * We provide only one choice for each NB event based on
+  * the fact that only NB events have restrictions. Consequently,
+  * if a counter is available, there is a guarantee the NB event
+  * will be assigned to it. If no slot is available, an empty
+  * constraint is returned and scheduling will eventually fail
+  * for this event.
+  *
+  * Note that all cores attached the same NB compete for the same
+  * counters to host NB events, this is why we use atomic ops. Some
+  * multi-chip CPUs may have more than one NB.
+  *
+  * Given that resources are allocated (cmpxchg), they must be
+  * eventually freed for others to use. This is accomplished by
+  * calling __amd_put_nb_event_constraints()
+  *
+  * Non NB events are not impacted by this restriction.
+  */
+static struct event_constraint *
+__amd_get_nb_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
+			       struct event_constraint *c)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct amd_nb *nb = cpuc->amd_nb;
+	struct perf_event *old;
+	int idx, new = -1;
+
+	if (!c)
+		c = &unconstrained;
+
+	if (cpuc->is_fake)
+		return c;
+
+	/*
+	 * detect if already present, if so reuse
+	 *
+	 * cannot merge with actual allocation
+	 * because of possible holes
+	 *
+	 * event can already be present yet not assigned (in hwc->idx)
+	 * because of successive calls to x86_schedule_events() from
+	 * hw_perf_group_sched_in() without hw_perf_enable()
+	 */
+	for_each_set_bit(idx, c->idxmsk, x86_pmu.num_counters) {
+		if (new == -1 || hwc->idx == idx)
+			/* assign free slot, prefer hwc->idx */
+			old = cmpxchg(nb->owners + idx, NULL, event);
+		else if (nb->owners[idx] == event)
+			/* event already present */
+			old = event;
+		else
+			continue;
+
+		if (old && old != event)
+			continue;
+
+		/* reassign to this slot */
+		if (new != -1)
+			cmpxchg(nb->owners + new, event, NULL);
+		new = idx;
+
+		/* already present, reuse */
+		if (old == event)
+			break;
+	}
+
+	if (new == -1)
+		return &emptyconstraint;
+
+	return &nb->event_constraints[new];
+}
+
+static struct amd_nb *amd_alloc_nb(int cpu)
+{
+	struct amd_nb *nb;
+	int i;
+
+	nb = kzalloc_node(sizeof(struct amd_nb), GFP_KERNEL, cpu_to_node(cpu));
+	if (!nb)
+		return NULL;
+
+	nb->nb_id = -1;
+
+	/*
+	 * initialize all possible NB constraints
+	 */
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		__set_bit(i, nb->event_constraints[i].idxmsk);
+		nb->event_constraints[i].weight = 1;
+	}
+	return nb;
+}
+
+static int amd_pmu_cpu_prepare(int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+
+	WARN_ON_ONCE(cpuc->amd_nb);
+
+	if (!x86_pmu.amd_nb_constraints)
+		return 0;
+
+	cpuc->amd_nb = amd_alloc_nb(cpu);
+	if (!cpuc->amd_nb)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void amd_pmu_cpu_starting(int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+	void **onln = &cpuc->kfree_on_online[X86_PERF_KFREE_SHARED];
+	struct amd_nb *nb;
+	int i, nb_id;
+
+	cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+
+	if (!x86_pmu.amd_nb_constraints)
+		return;
+
+	nb_id = amd_get_nb_id(cpu);
+	WARN_ON_ONCE(nb_id == BAD_APICID);
+
+	for_each_online_cpu(i) {
+		nb = per_cpu(cpu_hw_events, i).amd_nb;
+		if (WARN_ON_ONCE(!nb))
+			continue;
+
+		if (nb->nb_id == nb_id) {
+			*onln = cpuc->amd_nb;
+			cpuc->amd_nb = nb;
+			break;
+		}
+	}
+
+	cpuc->amd_nb->nb_id = nb_id;
+	cpuc->amd_nb->refcnt++;
+}
+
+static void amd_pmu_cpu_dead(int cpu)
+{
+	struct cpu_hw_events *cpuhw;
+
+	if (!x86_pmu.amd_nb_constraints)
+		return;
+
+	cpuhw = &per_cpu(cpu_hw_events, cpu);
+
+	if (cpuhw->amd_nb) {
+		struct amd_nb *nb = cpuhw->amd_nb;
+
+		if (nb->nb_id == -1 || --nb->refcnt == 0)
+			kfree(nb);
+
+		cpuhw->amd_nb = NULL;
+	}
+}
+
+/*
+ * When a PMC counter overflows, an NMI is used to process the event and
+ * reset the counter. NMI latency can result in the counter being updated
+ * before the NMI can run, which can result in what appear to be spurious
+ * NMIs. This function is intended to wait for the NMI to run and reset
+ * the counter to avoid possible unhandled NMI messages.
+ */
+#define OVERFLOW_WAIT_COUNT	50
+
+static void amd_pmu_wait_on_overflow(int idx)
+{
+	unsigned int i;
+	u64 counter;
+
+	/*
+	 * Wait for the counter to be reset if it has overflowed. This loop
+	 * should exit very, very quickly, but just in case, don't wait
+	 * forever...
+	 */
+	for (i = 0; i < OVERFLOW_WAIT_COUNT; i++) {
+		rdmsrl(x86_pmu_event_addr(idx), counter);
+		if (counter & (1ULL << (x86_pmu.cntval_bits - 1)))
+			break;
+
+		/* Might be in IRQ context, so can't sleep */
+		udelay(1);
+	}
+}
+
+static void amd_pmu_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	x86_pmu_disable_all();
+
+	/*
+	 * This shouldn't be called from NMI context, but add a safeguard here
+	 * to return, since if we're in NMI context we can't wait for an NMI
+	 * to reset an overflowed counter value.
+	 */
+	if (in_nmi())
+		return;
+
+	/*
+	 * Check each counter for overflow and wait for it to be reset by the
+	 * NMI if it has overflowed. This relies on the fact that all active
+	 * counters are always enabled when this function is caled and
+	 * ARCH_PERFMON_EVENTSEL_INT is always set.
+	 */
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		amd_pmu_wait_on_overflow(idx);
+	}
+}
+
+static void amd_pmu_disable_event(struct perf_event *event)
+{
+	x86_pmu_disable_event(event);
+
+	/*
+	 * This can be called from NMI context (via x86_pmu_stop). The counter
+	 * may have overflowed, but either way, we'll never see it get reset
+	 * by the NMI if we're already in the NMI. And the NMI latency support
+	 * below will take care of any pending NMI that might have been
+	 * generated by the overflow.
+	 */
+	if (in_nmi())
+		return;
+
+	amd_pmu_wait_on_overflow(event->hw.idx);
+}
+
+/*
+ * Because of NMI latency, if multiple PMC counters are active or other sources
+ * of NMIs are received, the perf NMI handler can handle one or more overflowed
+ * PMC counters outside of the NMI associated with the PMC overflow. If the NMI
+ * doesn't arrive at the LAPIC in time to become a pending NMI, then the kernel
+ * back-to-back NMI support won't be active. This PMC handler needs to take into
+ * account that this can occur, otherwise this could result in unknown NMI
+ * messages being issued. Examples of this is PMC overflow while in the NMI
+ * handler when multiple PMCs are active or PMC overflow while handling some
+ * other source of an NMI.
+ *
+ * Attempt to mitigate this by creating an NMI window in which un-handled NMIs
+ * received during this window will be claimed. This prevents extending the
+ * window past when it is possible that latent NMIs should be received. The
+ * per-CPU perf_nmi_tstamp will be set to the window end time whenever perf has
+ * handled a counter. When an un-handled NMI is received, it will be claimed
+ * only if arriving within that window.
+ */
+static int amd_pmu_handle_irq(struct pt_regs *regs)
+{
+	int handled;
+
+	/* Process any counter overflows */
+	handled = x86_pmu_handle_irq(regs);
+
+	/*
+	 * If a counter was handled, record a timestamp such that un-handled
+	 * NMIs will be claimed if arriving within that window.
+	 */
+	if (handled) {
+		this_cpu_write(perf_nmi_tstamp, jiffies + perf_nmi_window);
+
+		return handled;
+	}
+
+	if (time_after(jiffies, this_cpu_read(perf_nmi_tstamp)))
+		return NMI_DONE;
+
+	return NMI_HANDLED;
+}
+
+static struct event_constraint *
+amd_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	/*
+	 * if not NB event or no NB, then no constraints
+	 */
+	if (!(amd_has_nb(cpuc) && amd_is_nb_event(&event->hw)))
+		return &unconstrained;
+
+	return __amd_get_nb_event_constraints(cpuc, event, NULL);
+}
+
+static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
+				      struct perf_event *event)
+{
+	if (amd_has_nb(cpuc) && amd_is_nb_event(&event->hw))
+		__amd_put_nb_event_constraints(cpuc, event);
+}
+
+PMU_FORMAT_ATTR(event,	"config:0-7,32-35");
+PMU_FORMAT_ATTR(umask,	"config:8-15"	);
+PMU_FORMAT_ATTR(edge,	"config:18"	);
+PMU_FORMAT_ATTR(inv,	"config:23"	);
+PMU_FORMAT_ATTR(cmask,	"config:24-31"	);
+
+static struct attribute *amd_format_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+/* AMD Family 15h */
+
+#define AMD_EVENT_TYPE_MASK	0x000000F0ULL
+
+#define AMD_EVENT_FP		0x00000000ULL ... 0x00000010ULL
+#define AMD_EVENT_LS		0x00000020ULL ... 0x00000030ULL
+#define AMD_EVENT_DC		0x00000040ULL ... 0x00000050ULL
+#define AMD_EVENT_CU		0x00000060ULL ... 0x00000070ULL
+#define AMD_EVENT_IC_DE		0x00000080ULL ... 0x00000090ULL
+#define AMD_EVENT_EX_LS		0x000000C0ULL
+#define AMD_EVENT_DE		0x000000D0ULL
+#define AMD_EVENT_NB		0x000000E0ULL ... 0x000000F0ULL
+
+/*
+ * AMD family 15h event code/PMC mappings:
+ *
+ * type = event_code & 0x0F0:
+ *
+ * 0x000	FP	PERF_CTL[5:3]
+ * 0x010	FP	PERF_CTL[5:3]
+ * 0x020	LS	PERF_CTL[5:0]
+ * 0x030	LS	PERF_CTL[5:0]
+ * 0x040	DC	PERF_CTL[5:0]
+ * 0x050	DC	PERF_CTL[5:0]
+ * 0x060	CU	PERF_CTL[2:0]
+ * 0x070	CU	PERF_CTL[2:0]
+ * 0x080	IC/DE	PERF_CTL[2:0]
+ * 0x090	IC/DE	PERF_CTL[2:0]
+ * 0x0A0	---
+ * 0x0B0	---
+ * 0x0C0	EX/LS	PERF_CTL[5:0]
+ * 0x0D0	DE	PERF_CTL[2:0]
+ * 0x0E0	NB	NB_PERF_CTL[3:0]
+ * 0x0F0	NB	NB_PERF_CTL[3:0]
+ *
+ * Exceptions:
+ *
+ * 0x000	FP	PERF_CTL[3], PERF_CTL[5:3] (*)
+ * 0x003	FP	PERF_CTL[3]
+ * 0x004	FP	PERF_CTL[3], PERF_CTL[5:3] (*)
+ * 0x00B	FP	PERF_CTL[3]
+ * 0x00D	FP	PERF_CTL[3]
+ * 0x023	DE	PERF_CTL[2:0]
+ * 0x02D	LS	PERF_CTL[3]
+ * 0x02E	LS	PERF_CTL[3,0]
+ * 0x031	LS	PERF_CTL[2:0] (**)
+ * 0x043	CU	PERF_CTL[2:0]
+ * 0x045	CU	PERF_CTL[2:0]
+ * 0x046	CU	PERF_CTL[2:0]
+ * 0x054	CU	PERF_CTL[2:0]
+ * 0x055	CU	PERF_CTL[2:0]
+ * 0x08F	IC	PERF_CTL[0]
+ * 0x187	DE	PERF_CTL[0]
+ * 0x188	DE	PERF_CTL[0]
+ * 0x0DB	EX	PERF_CTL[5:0]
+ * 0x0DC	LS	PERF_CTL[5:0]
+ * 0x0DD	LS	PERF_CTL[5:0]
+ * 0x0DE	LS	PERF_CTL[5:0]
+ * 0x0DF	LS	PERF_CTL[5:0]
+ * 0x1C0	EX	PERF_CTL[5:3]
+ * 0x1D6	EX	PERF_CTL[5:0]
+ * 0x1D8	EX	PERF_CTL[5:0]
+ *
+ * (*)  depending on the umask all FPU counters may be used
+ * (**) only one unitmask enabled at a time
+ */
+
+static struct event_constraint amd_f15_PMC0  = EVENT_CONSTRAINT(0, 0x01, 0);
+static struct event_constraint amd_f15_PMC20 = EVENT_CONSTRAINT(0, 0x07, 0);
+static struct event_constraint amd_f15_PMC3  = EVENT_CONSTRAINT(0, 0x08, 0);
+static struct event_constraint amd_f15_PMC30 = EVENT_CONSTRAINT_OVERLAP(0, 0x09, 0);
+static struct event_constraint amd_f15_PMC50 = EVENT_CONSTRAINT(0, 0x3F, 0);
+static struct event_constraint amd_f15_PMC53 = EVENT_CONSTRAINT(0, 0x38, 0);
+
+static struct event_constraint *
+amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, int idx,
+			       struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned int event_code = amd_get_event_code(hwc);
+
+	switch (event_code & AMD_EVENT_TYPE_MASK) {
+	case AMD_EVENT_FP:
+		switch (event_code) {
+		case 0x000:
+			if (!(hwc->config & 0x0000F000ULL))
+				break;
+			if (!(hwc->config & 0x00000F00ULL))
+				break;
+			return &amd_f15_PMC3;
+		case 0x004:
+			if (hweight_long(hwc->config & ARCH_PERFMON_EVENTSEL_UMASK) <= 1)
+				break;
+			return &amd_f15_PMC3;
+		case 0x003:
+		case 0x00B:
+		case 0x00D:
+			return &amd_f15_PMC3;
+		}
+		return &amd_f15_PMC53;
+	case AMD_EVENT_LS:
+	case AMD_EVENT_DC:
+	case AMD_EVENT_EX_LS:
+		switch (event_code) {
+		case 0x023:
+		case 0x043:
+		case 0x045:
+		case 0x046:
+		case 0x054:
+		case 0x055:
+			return &amd_f15_PMC20;
+		case 0x02D:
+			return &amd_f15_PMC3;
+		case 0x02E:
+			return &amd_f15_PMC30;
+		case 0x031:
+			if (hweight_long(hwc->config & ARCH_PERFMON_EVENTSEL_UMASK) <= 1)
+				return &amd_f15_PMC20;
+			return &emptyconstraint;
+		case 0x1C0:
+			return &amd_f15_PMC53;
+		default:
+			return &amd_f15_PMC50;
+		}
+	case AMD_EVENT_CU:
+	case AMD_EVENT_IC_DE:
+	case AMD_EVENT_DE:
+		switch (event_code) {
+		case 0x08F:
+		case 0x187:
+		case 0x188:
+			return &amd_f15_PMC0;
+		case 0x0DB ... 0x0DF:
+		case 0x1D6:
+		case 0x1D8:
+			return &amd_f15_PMC50;
+		default:
+			return &amd_f15_PMC20;
+		}
+	case AMD_EVENT_NB:
+		/* moved to uncore.c */
+		return &emptyconstraint;
+	default:
+		return &emptyconstraint;
+	}
+}
+
+static struct event_constraint pair_constraint;
+
+static struct event_constraint *
+amd_get_event_constraints_f17h(struct cpu_hw_events *cpuc, int idx,
+			       struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (amd_is_pair_event_code(hwc))
+		return &pair_constraint;
+
+	return &unconstrained;
+}
+
+static void amd_put_event_constraints_f17h(struct cpu_hw_events *cpuc,
+					   struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (is_counter_pair(hwc))
+		--cpuc->n_pair;
+}
+
+static ssize_t amd_event_sysfs_show(char *page, u64 config)
+{
+	u64 event = (config & ARCH_PERFMON_EVENTSEL_EVENT) |
+		    (config & AMD64_EVENTSEL_EVENT) >> 24;
+
+	return x86_event_sysfs_show(page, config, event);
+}
+
+static __initconst const struct x86_pmu amd_pmu = {
+	.name			= "AMD",
+	.handle_irq		= amd_pmu_handle_irq,
+	.disable_all		= amd_pmu_disable_all,
+	.enable_all		= x86_pmu_enable_all,
+	.enable			= x86_pmu_enable_event,
+	.disable		= amd_pmu_disable_event,
+	.hw_config		= amd_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_K7_EVNTSEL0,
+	.perfctr		= MSR_K7_PERFCTR0,
+	.addr_offset            = amd_pmu_addr_offset,
+	.event_map		= amd_pmu_event_map,
+	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
+	.num_counters		= AMD64_NUM_COUNTERS,
+	.cntval_bits		= 48,
+	.cntval_mask		= (1ULL << 48) - 1,
+	.apic			= 1,
+	/* use highest bit to detect overflow */
+	.max_period		= (1ULL << 47) - 1,
+	.get_event_constraints	= amd_get_event_constraints,
+	.put_event_constraints	= amd_put_event_constraints,
+
+	.format_attrs		= amd_format_attr,
+	.events_sysfs_show	= amd_event_sysfs_show,
+
+	.cpu_prepare		= amd_pmu_cpu_prepare,
+	.cpu_starting		= amd_pmu_cpu_starting,
+	.cpu_dead		= amd_pmu_cpu_dead,
+
+	.amd_nb_constraints	= 1,
+};
+
+static int __init amd_core_pmu_init(void)
+{
+	u64 even_ctr_mask = 0ULL;
+	int i;
+
+	if (!boot_cpu_has(X86_FEATURE_PERFCTR_CORE))
+		return 0;
+
+	/* Avoid calculating the value each time in the NMI handler */
+	perf_nmi_window = msecs_to_jiffies(100);
+
+	/*
+	 * If core performance counter extensions exists, we must use
+	 * MSR_F15H_PERF_CTL/MSR_F15H_PERF_CTR msrs. See also
+	 * amd_pmu_addr_offset().
+	 */
+	x86_pmu.eventsel	= MSR_F15H_PERF_CTL;
+	x86_pmu.perfctr		= MSR_F15H_PERF_CTR;
+	x86_pmu.num_counters	= AMD64_NUM_COUNTERS_CORE;
+	/*
+	 * AMD Core perfctr has separate MSRs for the NB events, see
+	 * the amd/uncore.c driver.
+	 */
+	x86_pmu.amd_nb_constraints = 0;
+
+	if (boot_cpu_data.x86 == 0x15) {
+		pr_cont("Fam15h ");
+		x86_pmu.get_event_constraints = amd_get_event_constraints_f15h;
+	}
+	if (boot_cpu_data.x86 >= 0x17) {
+		pr_cont("Fam17h+ ");
+		/*
+		 * Family 17h and compatibles have constraints for Large
+		 * Increment per Cycle events: they may only be assigned an
+		 * even numbered counter that has a consecutive adjacent odd
+		 * numbered counter following it.
+		 */
+		for (i = 0; i < x86_pmu.num_counters - 1; i += 2)
+			even_ctr_mask |= BIT_ULL(i);
+
+		pair_constraint = (struct event_constraint)
+				    __EVENT_CONSTRAINT(0, even_ctr_mask, 0,
+				    x86_pmu.num_counters / 2, 0,
+				    PERF_X86_EVENT_PAIR);
+
+		x86_pmu.get_event_constraints = amd_get_event_constraints_f17h;
+		x86_pmu.put_event_constraints = amd_put_event_constraints_f17h;
+		x86_pmu.perf_ctr_pair_en = AMD_MERGE_EVENT_ENABLE;
+		x86_pmu.flags |= PMU_FL_PAIR;
+	}
+
+	pr_cont("core perfctr, ");
+	return 0;
+}
+
+__init int amd_pmu_init(void)
+{
+	int ret;
+
+	/* Performance-monitoring supported from K7 and later: */
+	if (boot_cpu_data.x86 < 6)
+		return -ENODEV;
+
+	x86_pmu = amd_pmu;
+
+	ret = amd_core_pmu_init();
+	if (ret)
+		return ret;
+
+	if (num_possible_cpus() == 1) {
+		/*
+		 * No point in allocating data structures to serialize
+		 * against other CPUs, when there is only the one CPU.
+		 */
+		x86_pmu.amd_nb_constraints = 0;
+	}
+
+	if (boot_cpu_data.x86 >= 0x17)
+		memcpy(hw_cache_event_ids, amd_hw_cache_event_ids_f17h, sizeof(hw_cache_event_ids));
+	else
+		memcpy(hw_cache_event_ids, amd_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+
+	return 0;
+}
+
+void amd_pmu_enable_virt(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	cpuc->perf_ctr_virt_mask = 0;
+
+	/* Reload all events */
+	amd_pmu_disable_all();
+	x86_pmu_enable_all(0);
+}
+EXPORT_SYMBOL_GPL(amd_pmu_enable_virt);
+
+void amd_pmu_disable_virt(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * We only mask out the Host-only bit so that host-only counting works
+	 * when SVM is disabled. If someone sets up a guest-only counter when
+	 * SVM is disabled the Guest-only bits still gets set and the counter
+	 * will not count anything.
+	 */
+	cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+
+	/* Reload all events */
+	amd_pmu_disable_all();
+	x86_pmu_enable_all(0);
+}
+EXPORT_SYMBOL_GPL(amd_pmu_disable_virt);
diff --git a/arch/x86/events/amd/ibs.c b/arch/x86/events/amd/ibs.c
new file mode 100644
index 000000000..354d52e17
--- /dev/null
+++ b/arch/x86/events/amd/ibs.c
@@ -0,0 +1,1116 @@
+/*
+ * Performance events - AMD IBS
+ *
+ *  Copyright (C) 2011 Advanced Micro Devices, Inc., Robert Richter
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+#include <linux/init.h>
+#include <linux/export.h>
+#include <linux/pci.h>
+#include <linux/ptrace.h>
+#include <linux/syscore_ops.h>
+#include <linux/sched/clock.h>
+
+#include <asm/apic.h>
+
+#include "../perf_event.h"
+
+static u32 ibs_caps;
+
+#if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD)
+
+#include <linux/kprobes.h>
+#include <linux/hardirq.h>
+
+#include <asm/nmi.h>
+
+#define IBS_FETCH_CONFIG_MASK	(IBS_FETCH_RAND_EN | IBS_FETCH_MAX_CNT)
+#define IBS_OP_CONFIG_MASK	IBS_OP_MAX_CNT
+
+
+/*
+ * IBS states:
+ *
+ * ENABLED; tracks the pmu::add(), pmu::del() state, when set the counter is taken
+ * and any further add()s must fail.
+ *
+ * STARTED/STOPPING/STOPPED; deal with pmu::start(), pmu::stop() state but are
+ * complicated by the fact that the IBS hardware can send late NMIs (ie. after
+ * we've cleared the EN bit).
+ *
+ * In order to consume these late NMIs we have the STOPPED state, any NMI that
+ * happens after we've cleared the EN state will clear this bit and report the
+ * NMI handled (this is fundamentally racy in the face or multiple NMI sources,
+ * someone else can consume our BIT and our NMI will go unhandled).
+ *
+ * And since we cannot set/clear this separate bit together with the EN bit,
+ * there are races; if we cleared STARTED early, an NMI could land in
+ * between clearing STARTED and clearing the EN bit (in fact multiple NMIs
+ * could happen if the period is small enough), and consume our STOPPED bit
+ * and trigger streams of unhandled NMIs.
+ *
+ * If, however, we clear STARTED late, an NMI can hit between clearing the
+ * EN bit and clearing STARTED, still see STARTED set and process the event.
+ * If this event will have the VALID bit clear, we bail properly, but this
+ * is not a given. With VALID set we can end up calling pmu::stop() again
+ * (the throttle logic) and trigger the WARNs in there.
+ *
+ * So what we do is set STOPPING before clearing EN to avoid the pmu::stop()
+ * nesting, and clear STARTED late, so that we have a well defined state over
+ * the clearing of the EN bit.
+ *
+ * XXX: we could probably be using !atomic bitops for all this.
+ */
+
+enum ibs_states {
+	IBS_ENABLED	= 0,
+	IBS_STARTED	= 1,
+	IBS_STOPPING	= 2,
+	IBS_STOPPED	= 3,
+
+	IBS_MAX_STATES,
+};
+
+struct cpu_perf_ibs {
+	struct perf_event	*event;
+	unsigned long		state[BITS_TO_LONGS(IBS_MAX_STATES)];
+};
+
+struct perf_ibs {
+	struct pmu			pmu;
+	unsigned int			msr;
+	u64				config_mask;
+	u64				cnt_mask;
+	u64				enable_mask;
+	u64				valid_mask;
+	u64				max_period;
+	unsigned long			offset_mask[1];
+	int				offset_max;
+	unsigned int			fetch_count_reset_broken : 1;
+	unsigned int			fetch_ignore_if_zero_rip : 1;
+	struct cpu_perf_ibs __percpu	*pcpu;
+
+	struct attribute		**format_attrs;
+	struct attribute_group		format_group;
+	const struct attribute_group	*attr_groups[2];
+
+	u64				(*get_count)(u64 config);
+};
+
+struct perf_ibs_data {
+	u32		size;
+	union {
+		u32	data[0];	/* data buffer starts here */
+		u32	caps;
+	};
+	u64		regs[MSR_AMD64_IBS_REG_COUNT_MAX];
+};
+
+static int
+perf_event_set_period(struct hw_perf_event *hwc, u64 min, u64 max, u64 *hw_period)
+{
+	s64 left = local64_read(&hwc->period_left);
+	s64 period = hwc->sample_period;
+	int overflow = 0;
+
+	/*
+	 * If we are way outside a reasonable range then just skip forward:
+	 */
+	if (unlikely(left <= -period)) {
+		left = period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		overflow = 1;
+	}
+
+	if (unlikely(left < (s64)min)) {
+		left += period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		overflow = 1;
+	}
+
+	/*
+	 * If the hw period that triggers the sw overflow is too short
+	 * we might hit the irq handler. This biases the results.
+	 * Thus we shorten the next-to-last period and set the last
+	 * period to the max period.
+	 */
+	if (left > max) {
+		left -= max;
+		if (left > max)
+			left = max;
+		else if (left < min)
+			left = min;
+	}
+
+	*hw_period = (u64)left;
+
+	return overflow;
+}
+
+static  int
+perf_event_try_update(struct perf_event *event, u64 new_raw_count, int width)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int shift = 64 - width;
+	u64 prev_raw_count;
+	u64 delta;
+
+	/*
+	 * Careful: an NMI might modify the previous event value.
+	 *
+	 * Our tactic to handle this is to first atomically read and
+	 * exchange a new raw count - then add that new-prev delta
+	 * count to the generic event atomically:
+	 */
+	prev_raw_count = local64_read(&hwc->prev_count);
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+					new_raw_count) != prev_raw_count)
+		return 0;
+
+	/*
+	 * Now we have the new raw value and have updated the prev
+	 * timestamp already. We can now calculate the elapsed delta
+	 * (event-)time and add that to the generic event.
+	 *
+	 * Careful, not all hw sign-extends above the physical width
+	 * of the count.
+	 */
+	delta = (new_raw_count << shift) - (prev_raw_count << shift);
+	delta >>= shift;
+
+	local64_add(delta, &event->count);
+	local64_sub(delta, &hwc->period_left);
+
+	return 1;
+}
+
+static struct perf_ibs perf_ibs_fetch;
+static struct perf_ibs perf_ibs_op;
+
+static struct perf_ibs *get_ibs_pmu(int type)
+{
+	if (perf_ibs_fetch.pmu.type == type)
+		return &perf_ibs_fetch;
+	if (perf_ibs_op.pmu.type == type)
+		return &perf_ibs_op;
+	return NULL;
+}
+
+/*
+ * core pmu config -> IBS config
+ *
+ *  perf record -a -e cpu-cycles:p ...    # use ibs op counting cycle count
+ *  perf record -a -e r076:p ...          # same as -e cpu-cycles:p
+ *  perf record -a -e r0C1:p ...          # use ibs op counting micro-ops
+ *
+ * IbsOpCntCtl (bit 19) of IBS Execution Control Register (IbsOpCtl,
+ * MSRC001_1033) is used to select either cycle or micro-ops counting
+ * mode.
+ */
+static int core_pmu_ibs_config(struct perf_event *event, u64 *config)
+{
+	switch (event->attr.type) {
+	case PERF_TYPE_HARDWARE:
+		switch (event->attr.config) {
+		case PERF_COUNT_HW_CPU_CYCLES:
+			*config = 0;
+			return 0;
+		}
+		break;
+	case PERF_TYPE_RAW:
+		switch (event->attr.config) {
+		case 0x0076:
+			*config = 0;
+			return 0;
+		case 0x00C1:
+			*config = IBS_OP_CNT_CTL;
+			return 0;
+		}
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+/*
+ * The rip of IBS samples has skid 0. Thus, IBS supports precise
+ * levels 1 and 2 and the PERF_EFLAGS_EXACT is set. In rare cases the
+ * rip is invalid when IBS was not able to record the rip correctly.
+ * We clear PERF_EFLAGS_EXACT and take the rip from pt_regs then.
+ */
+int forward_event_to_ibs(struct perf_event *event)
+{
+	u64 config = 0;
+
+	if (!event->attr.precise_ip || event->attr.precise_ip > 2)
+		return -EOPNOTSUPP;
+
+	if (!core_pmu_ibs_config(event, &config)) {
+		event->attr.type = perf_ibs_op.pmu.type;
+		event->attr.config = config;
+	}
+	return -ENOENT;
+}
+
+static int perf_ibs_init(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct perf_ibs *perf_ibs;
+	u64 max_cnt, config;
+
+	perf_ibs = get_ibs_pmu(event->attr.type);
+	if (!perf_ibs)
+		return -ENOENT;
+
+	config = event->attr.config;
+
+	if (event->pmu != &perf_ibs->pmu)
+		return -ENOENT;
+
+	if (config & ~perf_ibs->config_mask)
+		return -EINVAL;
+
+	if (hwc->sample_period) {
+		if (config & perf_ibs->cnt_mask)
+			/* raw max_cnt may not be set */
+			return -EINVAL;
+		if (!event->attr.sample_freq && hwc->sample_period & 0x0f)
+			/*
+			 * lower 4 bits can not be set in ibs max cnt,
+			 * but allowing it in case we adjust the
+			 * sample period to set a frequency.
+			 */
+			return -EINVAL;
+		hwc->sample_period &= ~0x0FULL;
+		if (!hwc->sample_period)
+			hwc->sample_period = 0x10;
+	} else {
+		max_cnt = config & perf_ibs->cnt_mask;
+		config &= ~perf_ibs->cnt_mask;
+		event->attr.sample_period = max_cnt << 4;
+		hwc->sample_period = event->attr.sample_period;
+	}
+
+	if (!hwc->sample_period)
+		return -EINVAL;
+
+	/*
+	 * If we modify hwc->sample_period, we also need to update
+	 * hwc->last_period and hwc->period_left.
+	 */
+	hwc->last_period = hwc->sample_period;
+	local64_set(&hwc->period_left, hwc->sample_period);
+
+	hwc->config_base = perf_ibs->msr;
+	hwc->config = config;
+
+	/*
+	 * rip recorded by IbsOpRip will not be consistent with rsp and rbp
+	 * recorded as part of interrupt regs. Thus we need to use rip from
+	 * interrupt regs while unwinding call stack. Setting _EARLY flag
+	 * makes sure we unwind call-stack before perf sample rip is set to
+	 * IbsOpRip.
+	 */
+	if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+		event->attr.sample_type |= __PERF_SAMPLE_CALLCHAIN_EARLY;
+
+	return 0;
+}
+
+static int perf_ibs_set_period(struct perf_ibs *perf_ibs,
+			       struct hw_perf_event *hwc, u64 *period)
+{
+	int overflow;
+
+	/* ignore lower 4 bits in min count: */
+	overflow = perf_event_set_period(hwc, 1<<4, perf_ibs->max_period, period);
+	local64_set(&hwc->prev_count, 0);
+
+	return overflow;
+}
+
+static u64 get_ibs_fetch_count(u64 config)
+{
+	return (config & IBS_FETCH_CNT) >> 12;
+}
+
+static u64 get_ibs_op_count(u64 config)
+{
+	u64 count = 0;
+
+	/*
+	 * If the internal 27-bit counter rolled over, the count is MaxCnt
+	 * and the lower 7 bits of CurCnt are randomized.
+	 * Otherwise CurCnt has the full 27-bit current counter value.
+	 */
+	if (config & IBS_OP_VAL) {
+		count = (config & IBS_OP_MAX_CNT) << 4;
+		if (ibs_caps & IBS_CAPS_OPCNTEXT)
+			count += config & IBS_OP_MAX_CNT_EXT_MASK;
+	} else if (ibs_caps & IBS_CAPS_RDWROPCNT) {
+		count = (config & IBS_OP_CUR_CNT) >> 32;
+	}
+
+	return count;
+}
+
+static void
+perf_ibs_event_update(struct perf_ibs *perf_ibs, struct perf_event *event,
+		      u64 *config)
+{
+	u64 count = perf_ibs->get_count(*config);
+
+	/*
+	 * Set width to 64 since we do not overflow on max width but
+	 * instead on max count. In perf_ibs_set_period() we clear
+	 * prev count manually on overflow.
+	 */
+	while (!perf_event_try_update(event, count, 64)) {
+		rdmsrl(event->hw.config_base, *config);
+		count = perf_ibs->get_count(*config);
+	}
+}
+
+static inline void perf_ibs_enable_event(struct perf_ibs *perf_ibs,
+					 struct hw_perf_event *hwc, u64 config)
+{
+	u64 tmp = hwc->config | config;
+
+	if (perf_ibs->fetch_count_reset_broken)
+		wrmsrl(hwc->config_base, tmp & ~perf_ibs->enable_mask);
+
+	wrmsrl(hwc->config_base, tmp | perf_ibs->enable_mask);
+}
+
+/*
+ * Erratum #420 Instruction-Based Sampling Engine May Generate
+ * Interrupt that Cannot Be Cleared:
+ *
+ * Must clear counter mask first, then clear the enable bit. See
+ * Revision Guide for AMD Family 10h Processors, Publication #41322.
+ */
+static inline void perf_ibs_disable_event(struct perf_ibs *perf_ibs,
+					  struct hw_perf_event *hwc, u64 config)
+{
+	config &= ~perf_ibs->cnt_mask;
+	if (boot_cpu_data.x86 == 0x10)
+		wrmsrl(hwc->config_base, config);
+	config &= ~perf_ibs->enable_mask;
+	wrmsrl(hwc->config_base, config);
+}
+
+/*
+ * We cannot restore the ibs pmu state, so we always needs to update
+ * the event while stopping it and then reset the state when starting
+ * again. Thus, ignoring PERF_EF_RELOAD and PERF_EF_UPDATE flags in
+ * perf_ibs_start()/perf_ibs_stop() and instead always do it.
+ */
+static void perf_ibs_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
+	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+	u64 period, config = 0;
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+	hwc->state = 0;
+
+	perf_ibs_set_period(perf_ibs, hwc, &period);
+	if (perf_ibs == &perf_ibs_op && (ibs_caps & IBS_CAPS_OPCNTEXT)) {
+		config |= period & IBS_OP_MAX_CNT_EXT_MASK;
+		period &= ~IBS_OP_MAX_CNT_EXT_MASK;
+	}
+	config |= period >> 4;
+
+	/*
+	 * Set STARTED before enabling the hardware, such that a subsequent NMI
+	 * must observe it.
+	 */
+	set_bit(IBS_STARTED,    pcpu->state);
+	clear_bit(IBS_STOPPING, pcpu->state);
+	perf_ibs_enable_event(perf_ibs, hwc, config);
+
+	perf_event_update_userpage(event);
+}
+
+static void perf_ibs_stop(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
+	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+	u64 config;
+	int stopping;
+
+	if (test_and_set_bit(IBS_STOPPING, pcpu->state))
+		return;
+
+	stopping = test_bit(IBS_STARTED, pcpu->state);
+
+	if (!stopping && (hwc->state & PERF_HES_UPTODATE))
+		return;
+
+	rdmsrl(hwc->config_base, config);
+
+	if (stopping) {
+		/*
+		 * Set STOPPED before disabling the hardware, such that it
+		 * must be visible to NMIs the moment we clear the EN bit,
+		 * at which point we can generate an !VALID sample which
+		 * we need to consume.
+		 */
+		set_bit(IBS_STOPPED, pcpu->state);
+		perf_ibs_disable_event(perf_ibs, hwc, config);
+		/*
+		 * Clear STARTED after disabling the hardware; if it were
+		 * cleared before an NMI hitting after the clear but before
+		 * clearing the EN bit might think it a spurious NMI and not
+		 * handle it.
+		 *
+		 * Clearing it after, however, creates the problem of the NMI
+		 * handler seeing STARTED but not having a valid sample.
+		 */
+		clear_bit(IBS_STARTED, pcpu->state);
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	if (hwc->state & PERF_HES_UPTODATE)
+		return;
+
+	/*
+	 * Clear valid bit to not count rollovers on update, rollovers
+	 * are only updated in the irq handler.
+	 */
+	config &= ~perf_ibs->valid_mask;
+
+	perf_ibs_event_update(perf_ibs, event, &config);
+	hwc->state |= PERF_HES_UPTODATE;
+}
+
+static int perf_ibs_add(struct perf_event *event, int flags)
+{
+	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
+	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+
+	if (test_and_set_bit(IBS_ENABLED, pcpu->state))
+		return -ENOSPC;
+
+	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	pcpu->event = event;
+
+	if (flags & PERF_EF_START)
+		perf_ibs_start(event, PERF_EF_RELOAD);
+
+	return 0;
+}
+
+static void perf_ibs_del(struct perf_event *event, int flags)
+{
+	struct perf_ibs *perf_ibs = container_of(event->pmu, struct perf_ibs, pmu);
+	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+
+	if (!test_and_clear_bit(IBS_ENABLED, pcpu->state))
+		return;
+
+	perf_ibs_stop(event, PERF_EF_UPDATE);
+
+	pcpu->event = NULL;
+
+	perf_event_update_userpage(event);
+}
+
+static void perf_ibs_read(struct perf_event *event) { }
+
+PMU_FORMAT_ATTR(rand_en,	"config:57");
+PMU_FORMAT_ATTR(cnt_ctl,	"config:19");
+
+static struct attribute *ibs_fetch_format_attrs[] = {
+	&format_attr_rand_en.attr,
+	NULL,
+};
+
+static struct attribute *ibs_op_format_attrs[] = {
+	NULL,	/* &format_attr_cnt_ctl.attr if IBS_CAPS_OPCNT */
+	NULL,
+};
+
+static struct perf_ibs perf_ibs_fetch = {
+	.pmu = {
+		.task_ctx_nr	= perf_invalid_context,
+
+		.event_init	= perf_ibs_init,
+		.add		= perf_ibs_add,
+		.del		= perf_ibs_del,
+		.start		= perf_ibs_start,
+		.stop		= perf_ibs_stop,
+		.read		= perf_ibs_read,
+		.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+	},
+	.msr			= MSR_AMD64_IBSFETCHCTL,
+	.config_mask		= IBS_FETCH_CONFIG_MASK,
+	.cnt_mask		= IBS_FETCH_MAX_CNT,
+	.enable_mask		= IBS_FETCH_ENABLE,
+	.valid_mask		= IBS_FETCH_VAL,
+	.max_period		= IBS_FETCH_MAX_CNT << 4,
+	.offset_mask		= { MSR_AMD64_IBSFETCH_REG_MASK },
+	.offset_max		= MSR_AMD64_IBSFETCH_REG_COUNT,
+	.format_attrs		= ibs_fetch_format_attrs,
+
+	.get_count		= get_ibs_fetch_count,
+};
+
+static struct perf_ibs perf_ibs_op = {
+	.pmu = {
+		.task_ctx_nr	= perf_invalid_context,
+
+		.event_init	= perf_ibs_init,
+		.add		= perf_ibs_add,
+		.del		= perf_ibs_del,
+		.start		= perf_ibs_start,
+		.stop		= perf_ibs_stop,
+		.read		= perf_ibs_read,
+		.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+	},
+	.msr			= MSR_AMD64_IBSOPCTL,
+	.config_mask		= IBS_OP_CONFIG_MASK,
+	.cnt_mask		= IBS_OP_MAX_CNT | IBS_OP_CUR_CNT |
+				  IBS_OP_CUR_CNT_RAND,
+	.enable_mask		= IBS_OP_ENABLE,
+	.valid_mask		= IBS_OP_VAL,
+	.max_period		= IBS_OP_MAX_CNT << 4,
+	.offset_mask		= { MSR_AMD64_IBSOP_REG_MASK },
+	.offset_max		= MSR_AMD64_IBSOP_REG_COUNT,
+	.format_attrs		= ibs_op_format_attrs,
+
+	.get_count		= get_ibs_op_count,
+};
+
+static int perf_ibs_handle_irq(struct perf_ibs *perf_ibs, struct pt_regs *iregs)
+{
+	struct cpu_perf_ibs *pcpu = this_cpu_ptr(perf_ibs->pcpu);
+	struct perf_event *event = pcpu->event;
+	struct hw_perf_event *hwc;
+	struct perf_sample_data data;
+	struct perf_raw_record raw;
+	struct pt_regs regs;
+	struct perf_ibs_data ibs_data;
+	int offset, size, check_rip, offset_max, throttle = 0;
+	unsigned int msr;
+	u64 *buf, *config, period, new_config = 0;
+
+	if (!test_bit(IBS_STARTED, pcpu->state)) {
+fail:
+		/*
+		 * Catch spurious interrupts after stopping IBS: After
+		 * disabling IBS there could be still incoming NMIs
+		 * with samples that even have the valid bit cleared.
+		 * Mark all this NMIs as handled.
+		 */
+		if (test_and_clear_bit(IBS_STOPPED, pcpu->state))
+			return 1;
+
+		return 0;
+	}
+
+	if (WARN_ON_ONCE(!event))
+		goto fail;
+
+	hwc = &event->hw;
+	msr = hwc->config_base;
+	buf = ibs_data.regs;
+	rdmsrl(msr, *buf);
+	if (!(*buf++ & perf_ibs->valid_mask))
+		goto fail;
+
+	config = &ibs_data.regs[0];
+	perf_ibs_event_update(perf_ibs, event, config);
+	perf_sample_data_init(&data, 0, hwc->last_period);
+	if (!perf_ibs_set_period(perf_ibs, hwc, &period))
+		goto out;	/* no sw counter overflow */
+
+	ibs_data.caps = ibs_caps;
+	size = 1;
+	offset = 1;
+	check_rip = (perf_ibs == &perf_ibs_op && (ibs_caps & IBS_CAPS_RIPINVALIDCHK));
+	if (event->attr.sample_type & PERF_SAMPLE_RAW)
+		offset_max = perf_ibs->offset_max;
+	else if (check_rip)
+		offset_max = 3;
+	else
+		offset_max = 1;
+	do {
+		rdmsrl(msr + offset, *buf++);
+		size++;
+		offset = find_next_bit(perf_ibs->offset_mask,
+				       perf_ibs->offset_max,
+				       offset + 1);
+	} while (offset < offset_max);
+	/*
+	 * Read IbsBrTarget, IbsOpData4, and IbsExtdCtl separately
+	 * depending on their availability.
+	 * Can't add to offset_max as they are staggered
+	 */
+	if (event->attr.sample_type & PERF_SAMPLE_RAW) {
+		if (perf_ibs == &perf_ibs_op) {
+			if (ibs_caps & IBS_CAPS_BRNTRGT) {
+				rdmsrl(MSR_AMD64_IBSBRTARGET, *buf++);
+				size++;
+			}
+			if (ibs_caps & IBS_CAPS_OPDATA4) {
+				rdmsrl(MSR_AMD64_IBSOPDATA4, *buf++);
+				size++;
+			}
+		}
+		if (perf_ibs == &perf_ibs_fetch && (ibs_caps & IBS_CAPS_FETCHCTLEXTD)) {
+			rdmsrl(MSR_AMD64_ICIBSEXTDCTL, *buf++);
+			size++;
+		}
+	}
+	ibs_data.size = sizeof(u64) * size;
+
+	regs = *iregs;
+	if (check_rip && (ibs_data.regs[2] & IBS_RIP_INVALID)) {
+		regs.flags &= ~PERF_EFLAGS_EXACT;
+	} else {
+		/* Workaround for erratum #1197 */
+		if (perf_ibs->fetch_ignore_if_zero_rip && !(ibs_data.regs[1]))
+			goto out;
+
+		set_linear_ip(&regs, ibs_data.regs[1]);
+		regs.flags |= PERF_EFLAGS_EXACT;
+	}
+
+	if (event->attr.sample_type & PERF_SAMPLE_RAW) {
+		raw = (struct perf_raw_record){
+			.frag = {
+				.size = sizeof(u32) + ibs_data.size,
+				.data = ibs_data.data,
+			},
+		};
+		data.raw = &raw;
+	}
+
+	/*
+	 * rip recorded by IbsOpRip will not be consistent with rsp and rbp
+	 * recorded as part of interrupt regs. Thus we need to use rip from
+	 * interrupt regs while unwinding call stack.
+	 */
+	if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+		data.callchain = perf_callchain(event, iregs);
+
+	throttle = perf_event_overflow(event, &data, &regs);
+out:
+	if (throttle) {
+		perf_ibs_stop(event, 0);
+	} else {
+		if (perf_ibs == &perf_ibs_op) {
+			if (ibs_caps & IBS_CAPS_OPCNTEXT) {
+				new_config = period & IBS_OP_MAX_CNT_EXT_MASK;
+				period &= ~IBS_OP_MAX_CNT_EXT_MASK;
+			}
+			if ((ibs_caps & IBS_CAPS_RDWROPCNT) && (*config & IBS_OP_CNT_CTL))
+				new_config |= *config & IBS_OP_CUR_CNT_RAND;
+		}
+		new_config |= period >> 4;
+
+		perf_ibs_enable_event(perf_ibs, hwc, new_config);
+	}
+
+	perf_event_update_userpage(event);
+
+	return 1;
+}
+
+static int
+perf_ibs_nmi_handler(unsigned int cmd, struct pt_regs *regs)
+{
+	u64 stamp = sched_clock();
+	int handled = 0;
+
+	handled += perf_ibs_handle_irq(&perf_ibs_fetch, regs);
+	handled += perf_ibs_handle_irq(&perf_ibs_op, regs);
+
+	if (handled)
+		inc_irq_stat(apic_perf_irqs);
+
+	perf_sample_event_took(sched_clock() - stamp);
+
+	return handled;
+}
+NOKPROBE_SYMBOL(perf_ibs_nmi_handler);
+
+static __init int perf_ibs_pmu_init(struct perf_ibs *perf_ibs, char *name)
+{
+	struct cpu_perf_ibs __percpu *pcpu;
+	int ret;
+
+	pcpu = alloc_percpu(struct cpu_perf_ibs);
+	if (!pcpu)
+		return -ENOMEM;
+
+	perf_ibs->pcpu = pcpu;
+
+	/* register attributes */
+	if (perf_ibs->format_attrs[0]) {
+		memset(&perf_ibs->format_group, 0, sizeof(perf_ibs->format_group));
+		perf_ibs->format_group.name	= "format";
+		perf_ibs->format_group.attrs	= perf_ibs->format_attrs;
+
+		memset(&perf_ibs->attr_groups, 0, sizeof(perf_ibs->attr_groups));
+		perf_ibs->attr_groups[0]	= &perf_ibs->format_group;
+		perf_ibs->pmu.attr_groups	= perf_ibs->attr_groups;
+	}
+
+	ret = perf_pmu_register(&perf_ibs->pmu, name, -1);
+	if (ret) {
+		perf_ibs->pcpu = NULL;
+		free_percpu(pcpu);
+	}
+
+	return ret;
+}
+
+static __init int perf_event_ibs_init(void)
+{
+	struct attribute **attr = ibs_op_format_attrs;
+	int ret;
+
+	/*
+	 * Some chips fail to reset the fetch count when it is written; instead
+	 * they need a 0-1 transition of IbsFetchEn.
+	 */
+	if (boot_cpu_data.x86 >= 0x16 && boot_cpu_data.x86 <= 0x18)
+		perf_ibs_fetch.fetch_count_reset_broken = 1;
+
+	if (boot_cpu_data.x86 == 0x19 && boot_cpu_data.x86_model < 0x10)
+		perf_ibs_fetch.fetch_ignore_if_zero_rip = 1;
+
+	ret = perf_ibs_pmu_init(&perf_ibs_fetch, "ibs_fetch");
+	if (ret)
+		return ret;
+
+	if (ibs_caps & IBS_CAPS_OPCNT) {
+		perf_ibs_op.config_mask |= IBS_OP_CNT_CTL;
+		*attr++ = &format_attr_cnt_ctl.attr;
+	}
+
+	if (ibs_caps & IBS_CAPS_OPCNTEXT) {
+		perf_ibs_op.max_period  |= IBS_OP_MAX_CNT_EXT_MASK;
+		perf_ibs_op.config_mask	|= IBS_OP_MAX_CNT_EXT_MASK;
+		perf_ibs_op.cnt_mask    |= IBS_OP_MAX_CNT_EXT_MASK;
+	}
+
+	ret = perf_ibs_pmu_init(&perf_ibs_op, "ibs_op");
+	if (ret)
+		goto err_op;
+
+	ret = register_nmi_handler(NMI_LOCAL, perf_ibs_nmi_handler, 0, "perf_ibs");
+	if (ret)
+		goto err_nmi;
+
+	pr_info("perf: AMD IBS detected (0x%08x)\n", ibs_caps);
+	return 0;
+
+err_nmi:
+	perf_pmu_unregister(&perf_ibs_op.pmu);
+	free_percpu(perf_ibs_op.pcpu);
+	perf_ibs_op.pcpu = NULL;
+err_op:
+	perf_pmu_unregister(&perf_ibs_fetch.pmu);
+	free_percpu(perf_ibs_fetch.pcpu);
+	perf_ibs_fetch.pcpu = NULL;
+
+	return ret;
+}
+
+#else /* defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) */
+
+static __init int perf_event_ibs_init(void)
+{
+	return 0;
+}
+
+#endif
+
+/* IBS - apic initialization, for perf and oprofile */
+
+static __init u32 __get_ibs_caps(void)
+{
+	u32 caps;
+	unsigned int max_level;
+
+	if (!boot_cpu_has(X86_FEATURE_IBS))
+		return 0;
+
+	/* check IBS cpuid feature flags */
+	max_level = cpuid_eax(0x80000000);
+	if (max_level < IBS_CPUID_FEATURES)
+		return IBS_CAPS_DEFAULT;
+
+	caps = cpuid_eax(IBS_CPUID_FEATURES);
+	if (!(caps & IBS_CAPS_AVAIL))
+		/* cpuid flags not valid */
+		return IBS_CAPS_DEFAULT;
+
+	return caps;
+}
+
+u32 get_ibs_caps(void)
+{
+	return ibs_caps;
+}
+
+EXPORT_SYMBOL(get_ibs_caps);
+
+static inline int get_eilvt(int offset)
+{
+	return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1);
+}
+
+static inline int put_eilvt(int offset)
+{
+	return !setup_APIC_eilvt(offset, 0, 0, 1);
+}
+
+/*
+ * Check and reserve APIC extended interrupt LVT offset for IBS if available.
+ */
+static inline int ibs_eilvt_valid(void)
+{
+	int offset;
+	u64 val;
+	int valid = 0;
+
+	preempt_disable();
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	offset = val & IBSCTL_LVT_OFFSET_MASK;
+
+	if (!(val & IBSCTL_LVT_OFFSET_VALID)) {
+		pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n",
+		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
+		goto out;
+	}
+
+	if (!get_eilvt(offset)) {
+		pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n",
+		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
+		goto out;
+	}
+
+	valid = 1;
+out:
+	preempt_enable();
+
+	return valid;
+}
+
+static int setup_ibs_ctl(int ibs_eilvt_off)
+{
+	struct pci_dev *cpu_cfg;
+	int nodes;
+	u32 value = 0;
+
+	nodes = 0;
+	cpu_cfg = NULL;
+	do {
+		cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD,
+					 PCI_DEVICE_ID_AMD_10H_NB_MISC,
+					 cpu_cfg);
+		if (!cpu_cfg)
+			break;
+		++nodes;
+		pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
+				       | IBSCTL_LVT_OFFSET_VALID);
+		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
+		if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) {
+			pci_dev_put(cpu_cfg);
+			pr_debug("Failed to setup IBS LVT offset, IBSCTL = 0x%08x\n",
+				 value);
+			return -EINVAL;
+		}
+	} while (1);
+
+	if (!nodes) {
+		pr_debug("No CPU node configured for IBS\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/*
+ * This runs only on the current cpu. We try to find an LVT offset and
+ * setup the local APIC. For this we must disable preemption. On
+ * success we initialize all nodes with this offset. This updates then
+ * the offset in the IBS_CTL per-node msr. The per-core APIC setup of
+ * the IBS interrupt vector is handled by perf_ibs_cpu_notifier that
+ * is using the new offset.
+ */
+static void force_ibs_eilvt_setup(void)
+{
+	int offset;
+	int ret;
+
+	preempt_disable();
+	/* find the next free available EILVT entry, skip offset 0 */
+	for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) {
+		if (get_eilvt(offset))
+			break;
+	}
+	preempt_enable();
+
+	if (offset == APIC_EILVT_NR_MAX) {
+		pr_debug("No EILVT entry available\n");
+		return;
+	}
+
+	ret = setup_ibs_ctl(offset);
+	if (ret)
+		goto out;
+
+	if (!ibs_eilvt_valid())
+		goto out;
+
+	pr_info("LVT offset %d assigned\n", offset);
+
+	return;
+out:
+	preempt_disable();
+	put_eilvt(offset);
+	preempt_enable();
+	return;
+}
+
+static void ibs_eilvt_setup(void)
+{
+	/*
+	 * Force LVT offset assignment for family 10h: The offsets are
+	 * not assigned by the BIOS for this family, so the OS is
+	 * responsible for doing it. If the OS assignment fails, fall
+	 * back to BIOS settings and try to setup this.
+	 */
+	if (boot_cpu_data.x86 == 0x10)
+		force_ibs_eilvt_setup();
+}
+
+static inline int get_ibs_lvt_offset(void)
+{
+	u64 val;
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	if (!(val & IBSCTL_LVT_OFFSET_VALID))
+		return -EINVAL;
+
+	return val & IBSCTL_LVT_OFFSET_MASK;
+}
+
+static void setup_APIC_ibs(void)
+{
+	int offset;
+
+	offset = get_ibs_lvt_offset();
+	if (offset < 0)
+		goto failed;
+
+	if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0))
+		return;
+failed:
+	pr_warn("perf: IBS APIC setup failed on cpu #%d\n",
+		smp_processor_id());
+}
+
+static void clear_APIC_ibs(void)
+{
+	int offset;
+
+	offset = get_ibs_lvt_offset();
+	if (offset >= 0)
+		setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
+}
+
+static int x86_pmu_amd_ibs_starting_cpu(unsigned int cpu)
+{
+	setup_APIC_ibs();
+	return 0;
+}
+
+#ifdef CONFIG_PM
+
+static int perf_ibs_suspend(void)
+{
+	clear_APIC_ibs();
+	return 0;
+}
+
+static void perf_ibs_resume(void)
+{
+	ibs_eilvt_setup();
+	setup_APIC_ibs();
+}
+
+static struct syscore_ops perf_ibs_syscore_ops = {
+	.resume		= perf_ibs_resume,
+	.suspend	= perf_ibs_suspend,
+};
+
+static void perf_ibs_pm_init(void)
+{
+	register_syscore_ops(&perf_ibs_syscore_ops);
+}
+
+#else
+
+static inline void perf_ibs_pm_init(void) { }
+
+#endif
+
+static int x86_pmu_amd_ibs_dying_cpu(unsigned int cpu)
+{
+	clear_APIC_ibs();
+	return 0;
+}
+
+static __init int amd_ibs_init(void)
+{
+	u32 caps;
+
+	caps = __get_ibs_caps();
+	if (!caps)
+		return -ENODEV;	/* ibs not supported by the cpu */
+
+	ibs_eilvt_setup();
+
+	if (!ibs_eilvt_valid())
+		return -EINVAL;
+
+	perf_ibs_pm_init();
+
+	ibs_caps = caps;
+	/* make ibs_caps visible to other cpus: */
+	smp_mb();
+	/*
+	 * x86_pmu_amd_ibs_starting_cpu will be called from core on
+	 * all online cpus.
+	 */
+	cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_IBS_STARTING,
+			  "perf/x86/amd/ibs:starting",
+			  x86_pmu_amd_ibs_starting_cpu,
+			  x86_pmu_amd_ibs_dying_cpu);
+
+	return perf_event_ibs_init();
+}
+
+/* Since we need the pci subsystem to init ibs we can't do this earlier: */
+device_initcall(amd_ibs_init);
diff --git a/arch/x86/events/amd/iommu.c b/arch/x86/events/amd/iommu.c
new file mode 100644
index 000000000..2da6139b0
--- /dev/null
+++ b/arch/x86/events/amd/iommu.c
@@ -0,0 +1,488 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2013 Advanced Micro Devices, Inc.
+ *
+ * Author: Steven Kinney <Steven.Kinney@amd.com>
+ * Author: Suravee Suthikulpanit <Suraveee.Suthikulpanit@amd.com>
+ *
+ * Perf: amd_iommu - AMD IOMMU Performance Counter PMU implementation
+ */
+
+#define pr_fmt(fmt)	"perf/amd_iommu: " fmt
+
+#include <linux/perf_event.h>
+#include <linux/init.h>
+#include <linux/cpumask.h>
+#include <linux/slab.h>
+
+#include "../perf_event.h"
+#include "iommu.h"
+
+/* iommu pmu conf masks */
+#define GET_CSOURCE(x)     ((x)->conf & 0xFFULL)
+#define GET_DEVID(x)       (((x)->conf >> 8)  & 0xFFFFULL)
+#define GET_DOMID(x)       (((x)->conf >> 24) & 0xFFFFULL)
+#define GET_PASID(x)       (((x)->conf >> 40) & 0xFFFFFULL)
+
+/* iommu pmu conf1 masks */
+#define GET_DEVID_MASK(x)  ((x)->conf1  & 0xFFFFULL)
+#define GET_DOMID_MASK(x)  (((x)->conf1 >> 16) & 0xFFFFULL)
+#define GET_PASID_MASK(x)  (((x)->conf1 >> 32) & 0xFFFFFULL)
+
+#define IOMMU_NAME_SIZE 16
+
+struct perf_amd_iommu {
+	struct list_head list;
+	struct pmu pmu;
+	struct amd_iommu *iommu;
+	char name[IOMMU_NAME_SIZE];
+	u8 max_banks;
+	u8 max_counters;
+	u64 cntr_assign_mask;
+	raw_spinlock_t lock;
+};
+
+static LIST_HEAD(perf_amd_iommu_list);
+
+/*---------------------------------------------
+ * sysfs format attributes
+ *---------------------------------------------*/
+PMU_FORMAT_ATTR(csource,    "config:0-7");
+PMU_FORMAT_ATTR(devid,      "config:8-23");
+PMU_FORMAT_ATTR(domid,      "config:24-39");
+PMU_FORMAT_ATTR(pasid,      "config:40-59");
+PMU_FORMAT_ATTR(devid_mask, "config1:0-15");
+PMU_FORMAT_ATTR(domid_mask, "config1:16-31");
+PMU_FORMAT_ATTR(pasid_mask, "config1:32-51");
+
+static struct attribute *iommu_format_attrs[] = {
+	&format_attr_csource.attr,
+	&format_attr_devid.attr,
+	&format_attr_pasid.attr,
+	&format_attr_domid.attr,
+	&format_attr_devid_mask.attr,
+	&format_attr_pasid_mask.attr,
+	&format_attr_domid_mask.attr,
+	NULL,
+};
+
+static struct attribute_group amd_iommu_format_group = {
+	.name = "format",
+	.attrs = iommu_format_attrs,
+};
+
+/*---------------------------------------------
+ * sysfs events attributes
+ *---------------------------------------------*/
+static struct attribute_group amd_iommu_events_group = {
+	.name = "events",
+};
+
+struct amd_iommu_event_desc {
+	struct device_attribute attr;
+	const char *event;
+};
+
+static ssize_t _iommu_event_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct amd_iommu_event_desc *event =
+		container_of(attr, struct amd_iommu_event_desc, attr);
+	return sprintf(buf, "%s\n", event->event);
+}
+
+#define AMD_IOMMU_EVENT_DESC(_name, _event)			\
+{								\
+	.attr  = __ATTR(_name, 0444, _iommu_event_show, NULL),	\
+	.event = _event,					\
+}
+
+static struct amd_iommu_event_desc amd_iommu_v2_event_descs[] = {
+	AMD_IOMMU_EVENT_DESC(mem_pass_untrans,        "csource=0x01"),
+	AMD_IOMMU_EVENT_DESC(mem_pass_pretrans,       "csource=0x02"),
+	AMD_IOMMU_EVENT_DESC(mem_pass_excl,           "csource=0x03"),
+	AMD_IOMMU_EVENT_DESC(mem_target_abort,        "csource=0x04"),
+	AMD_IOMMU_EVENT_DESC(mem_trans_total,         "csource=0x05"),
+	AMD_IOMMU_EVENT_DESC(mem_iommu_tlb_pte_hit,   "csource=0x06"),
+	AMD_IOMMU_EVENT_DESC(mem_iommu_tlb_pte_mis,   "csource=0x07"),
+	AMD_IOMMU_EVENT_DESC(mem_iommu_tlb_pde_hit,   "csource=0x08"),
+	AMD_IOMMU_EVENT_DESC(mem_iommu_tlb_pde_mis,   "csource=0x09"),
+	AMD_IOMMU_EVENT_DESC(mem_dte_hit,             "csource=0x0a"),
+	AMD_IOMMU_EVENT_DESC(mem_dte_mis,             "csource=0x0b"),
+	AMD_IOMMU_EVENT_DESC(page_tbl_read_tot,       "csource=0x0c"),
+	AMD_IOMMU_EVENT_DESC(page_tbl_read_nst,       "csource=0x0d"),
+	AMD_IOMMU_EVENT_DESC(page_tbl_read_gst,       "csource=0x0e"),
+	AMD_IOMMU_EVENT_DESC(int_dte_hit,             "csource=0x0f"),
+	AMD_IOMMU_EVENT_DESC(int_dte_mis,             "csource=0x10"),
+	AMD_IOMMU_EVENT_DESC(cmd_processed,           "csource=0x11"),
+	AMD_IOMMU_EVENT_DESC(cmd_processed_inv,       "csource=0x12"),
+	AMD_IOMMU_EVENT_DESC(tlb_inv,                 "csource=0x13"),
+	AMD_IOMMU_EVENT_DESC(ign_rd_wr_mmio_1ff8h,    "csource=0x14"),
+	AMD_IOMMU_EVENT_DESC(vapic_int_non_guest,     "csource=0x15"),
+	AMD_IOMMU_EVENT_DESC(vapic_int_guest,         "csource=0x16"),
+	AMD_IOMMU_EVENT_DESC(smi_recv,                "csource=0x17"),
+	AMD_IOMMU_EVENT_DESC(smi_blk,                 "csource=0x18"),
+	{ /* end: all zeroes */ },
+};
+
+/*---------------------------------------------
+ * sysfs cpumask attributes
+ *---------------------------------------------*/
+static cpumask_t iommu_cpumask;
+
+static ssize_t _iommu_cpumask_show(struct device *dev,
+				   struct device_attribute *attr,
+				   char *buf)
+{
+	return cpumap_print_to_pagebuf(true, buf, &iommu_cpumask);
+}
+static DEVICE_ATTR(cpumask, S_IRUGO, _iommu_cpumask_show, NULL);
+
+static struct attribute *iommu_cpumask_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group amd_iommu_cpumask_group = {
+	.attrs = iommu_cpumask_attrs,
+};
+
+/*---------------------------------------------*/
+
+static int get_next_avail_iommu_bnk_cntr(struct perf_event *event)
+{
+	struct perf_amd_iommu *piommu = container_of(event->pmu, struct perf_amd_iommu, pmu);
+	int max_cntrs = piommu->max_counters;
+	int max_banks = piommu->max_banks;
+	u32 shift, bank, cntr;
+	unsigned long flags;
+	int retval;
+
+	raw_spin_lock_irqsave(&piommu->lock, flags);
+
+	for (bank = 0, shift = 0; bank < max_banks; bank++) {
+		for (cntr = 0; cntr < max_cntrs; cntr++) {
+			shift = bank + (bank*3) + cntr;
+			if (piommu->cntr_assign_mask & BIT_ULL(shift)) {
+				continue;
+			} else {
+				piommu->cntr_assign_mask |= BIT_ULL(shift);
+				event->hw.iommu_bank = bank;
+				event->hw.iommu_cntr = cntr;
+				retval = 0;
+				goto out;
+			}
+		}
+	}
+	retval = -ENOSPC;
+out:
+	raw_spin_unlock_irqrestore(&piommu->lock, flags);
+	return retval;
+}
+
+static int clear_avail_iommu_bnk_cntr(struct perf_amd_iommu *perf_iommu,
+					u8 bank, u8 cntr)
+{
+	unsigned long flags;
+	int max_banks, max_cntrs;
+	int shift = 0;
+
+	max_banks = perf_iommu->max_banks;
+	max_cntrs = perf_iommu->max_counters;
+
+	if ((bank > max_banks) || (cntr > max_cntrs))
+		return -EINVAL;
+
+	shift = bank + cntr + (bank*3);
+
+	raw_spin_lock_irqsave(&perf_iommu->lock, flags);
+	perf_iommu->cntr_assign_mask &= ~(1ULL<<shift);
+	raw_spin_unlock_irqrestore(&perf_iommu->lock, flags);
+
+	return 0;
+}
+
+static int perf_iommu_event_init(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/* test the event attr type check for PMU enumeration */
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/*
+	 * IOMMU counters are shared across all cores.
+	 * Therefore, it does not support per-process mode.
+	 * Also, it does not support event sampling mode.
+	 */
+	if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
+		return -EINVAL;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	/* update the hw_perf_event struct with the iommu config data */
+	hwc->conf  = event->attr.config;
+	hwc->conf1 = event->attr.config1;
+
+	return 0;
+}
+
+static inline struct amd_iommu *perf_event_2_iommu(struct perf_event *ev)
+{
+	return (container_of(ev->pmu, struct perf_amd_iommu, pmu))->iommu;
+}
+
+static void perf_iommu_enable_event(struct perf_event *ev)
+{
+	struct amd_iommu *iommu = perf_event_2_iommu(ev);
+	struct hw_perf_event *hwc = &ev->hw;
+	u8 bank = hwc->iommu_bank;
+	u8 cntr = hwc->iommu_cntr;
+	u64 reg = 0ULL;
+
+	reg = GET_CSOURCE(hwc);
+	amd_iommu_pc_set_reg(iommu, bank, cntr, IOMMU_PC_COUNTER_SRC_REG, &reg);
+
+	reg = GET_DEVID_MASK(hwc);
+	reg = GET_DEVID(hwc) | (reg << 32);
+	if (reg)
+		reg |= BIT(31);
+	amd_iommu_pc_set_reg(iommu, bank, cntr, IOMMU_PC_DEVID_MATCH_REG, &reg);
+
+	reg = GET_PASID_MASK(hwc);
+	reg = GET_PASID(hwc) | (reg << 32);
+	if (reg)
+		reg |= BIT(31);
+	amd_iommu_pc_set_reg(iommu, bank, cntr, IOMMU_PC_PASID_MATCH_REG, &reg);
+
+	reg = GET_DOMID_MASK(hwc);
+	reg = GET_DOMID(hwc) | (reg << 32);
+	if (reg)
+		reg |= BIT(31);
+	amd_iommu_pc_set_reg(iommu, bank, cntr, IOMMU_PC_DOMID_MATCH_REG, &reg);
+}
+
+static void perf_iommu_disable_event(struct perf_event *event)
+{
+	struct amd_iommu *iommu = perf_event_2_iommu(event);
+	struct hw_perf_event *hwc = &event->hw;
+	u64 reg = 0ULL;
+
+	amd_iommu_pc_set_reg(iommu, hwc->iommu_bank, hwc->iommu_cntr,
+			     IOMMU_PC_COUNTER_SRC_REG, &reg);
+}
+
+static void perf_iommu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+	hwc->state = 0;
+
+	/*
+	 * To account for power-gating, which prevents write to
+	 * the counter, we need to enable the counter
+	 * before setting up counter register.
+	 */
+	perf_iommu_enable_event(event);
+
+	if (flags & PERF_EF_RELOAD) {
+		u64 count = 0;
+		struct amd_iommu *iommu = perf_event_2_iommu(event);
+
+		/*
+		 * Since the IOMMU PMU only support counting mode,
+		 * the counter always start with value zero.
+		 */
+		amd_iommu_pc_set_reg(iommu, hwc->iommu_bank, hwc->iommu_cntr,
+				     IOMMU_PC_COUNTER_REG, &count);
+	}
+
+	perf_event_update_userpage(event);
+}
+
+static void perf_iommu_read(struct perf_event *event)
+{
+	u64 count;
+	struct hw_perf_event *hwc = &event->hw;
+	struct amd_iommu *iommu = perf_event_2_iommu(event);
+
+	if (amd_iommu_pc_get_reg(iommu, hwc->iommu_bank, hwc->iommu_cntr,
+				 IOMMU_PC_COUNTER_REG, &count))
+		return;
+
+	/* IOMMU pc counter register is only 48 bits */
+	count &= GENMASK_ULL(47, 0);
+
+	/*
+	 * Since the counter always start with value zero,
+	 * simply just accumulate the count for the event.
+	 */
+	local64_add(count, &event->count);
+}
+
+static void perf_iommu_stop(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->state & PERF_HES_UPTODATE)
+		return;
+
+	/*
+	 * To account for power-gating, in which reading the counter would
+	 * return zero, we need to read the register before disabling.
+	 */
+	perf_iommu_read(event);
+	hwc->state |= PERF_HES_UPTODATE;
+
+	perf_iommu_disable_event(event);
+	WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+	hwc->state |= PERF_HES_STOPPED;
+}
+
+static int perf_iommu_add(struct perf_event *event, int flags)
+{
+	int retval;
+
+	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	/* request an iommu bank/counter */
+	retval = get_next_avail_iommu_bnk_cntr(event);
+	if (retval)
+		return retval;
+
+	if (flags & PERF_EF_START)
+		perf_iommu_start(event, PERF_EF_RELOAD);
+
+	return 0;
+}
+
+static void perf_iommu_del(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct perf_amd_iommu *perf_iommu =
+			container_of(event->pmu, struct perf_amd_iommu, pmu);
+
+	perf_iommu_stop(event, PERF_EF_UPDATE);
+
+	/* clear the assigned iommu bank/counter */
+	clear_avail_iommu_bnk_cntr(perf_iommu,
+				   hwc->iommu_bank, hwc->iommu_cntr);
+
+	perf_event_update_userpage(event);
+}
+
+static __init int _init_events_attrs(void)
+{
+	int i = 0, j;
+	struct attribute **attrs;
+
+	while (amd_iommu_v2_event_descs[i].attr.attr.name)
+		i++;
+
+	attrs = kcalloc(i + 1, sizeof(*attrs), GFP_KERNEL);
+	if (!attrs)
+		return -ENOMEM;
+
+	for (j = 0; j < i; j++)
+		attrs[j] = &amd_iommu_v2_event_descs[j].attr.attr;
+
+	amd_iommu_events_group.attrs = attrs;
+	return 0;
+}
+
+static const struct attribute_group *amd_iommu_attr_groups[] = {
+	&amd_iommu_format_group,
+	&amd_iommu_cpumask_group,
+	&amd_iommu_events_group,
+	NULL,
+};
+
+static const struct pmu iommu_pmu __initconst = {
+	.event_init	= perf_iommu_event_init,
+	.add		= perf_iommu_add,
+	.del		= perf_iommu_del,
+	.start		= perf_iommu_start,
+	.stop		= perf_iommu_stop,
+	.read		= perf_iommu_read,
+	.task_ctx_nr	= perf_invalid_context,
+	.attr_groups	= amd_iommu_attr_groups,
+	.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+};
+
+static __init int init_one_iommu(unsigned int idx)
+{
+	struct perf_amd_iommu *perf_iommu;
+	int ret;
+
+	perf_iommu = kzalloc(sizeof(struct perf_amd_iommu), GFP_KERNEL);
+	if (!perf_iommu)
+		return -ENOMEM;
+
+	raw_spin_lock_init(&perf_iommu->lock);
+
+	perf_iommu->pmu          = iommu_pmu;
+	perf_iommu->iommu        = get_amd_iommu(idx);
+	perf_iommu->max_banks    = amd_iommu_pc_get_max_banks(idx);
+	perf_iommu->max_counters = amd_iommu_pc_get_max_counters(idx);
+
+	if (!perf_iommu->iommu ||
+	    !perf_iommu->max_banks ||
+	    !perf_iommu->max_counters) {
+		kfree(perf_iommu);
+		return -EINVAL;
+	}
+
+	snprintf(perf_iommu->name, IOMMU_NAME_SIZE, "amd_iommu_%u", idx);
+
+	ret = perf_pmu_register(&perf_iommu->pmu, perf_iommu->name, -1);
+	if (!ret) {
+		pr_info("Detected AMD IOMMU #%d (%d banks, %d counters/bank).\n",
+			idx, perf_iommu->max_banks, perf_iommu->max_counters);
+		list_add_tail(&perf_iommu->list, &perf_amd_iommu_list);
+	} else {
+		pr_warn("Error initializing IOMMU %d.\n", idx);
+		kfree(perf_iommu);
+	}
+	return ret;
+}
+
+static __init int amd_iommu_pc_init(void)
+{
+	unsigned int i, cnt = 0;
+	int ret;
+
+	/* Make sure the IOMMU PC resource is available */
+	if (!amd_iommu_pc_supported())
+		return -ENODEV;
+
+	ret = _init_events_attrs();
+	if (ret)
+		return ret;
+
+	/*
+	 * An IOMMU PMU is specific to an IOMMU, and can function independently.
+	 * So we go through all IOMMUs and ignore the one that fails init
+	 * unless all IOMMU are failing.
+	 */
+	for (i = 0; i < amd_iommu_get_num_iommus(); i++) {
+		ret = init_one_iommu(i);
+		if (!ret)
+			cnt++;
+	}
+
+	if (!cnt) {
+		kfree(amd_iommu_events_group.attrs);
+		return -ENODEV;
+	}
+
+	/* Init cpumask attributes to only core 0 */
+	cpumask_set_cpu(0, &iommu_cpumask);
+	return 0;
+}
+
+device_initcall(amd_iommu_pc_init);
diff --git a/arch/x86/events/amd/iommu.h b/arch/x86/events/amd/iommu.h
new file mode 100644
index 000000000..0e5c036fd
--- /dev/null
+++ b/arch/x86/events/amd/iommu.h
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2013 Advanced Micro Devices, Inc.
+ *
+ * Author: Steven Kinney <Steven.Kinney@amd.com>
+ * Author: Suravee Suthikulpanit <Suraveee.Suthikulpanit@amd.com>
+ */
+
+#ifndef _PERF_EVENT_AMD_IOMMU_H_
+#define _PERF_EVENT_AMD_IOMMU_H_
+
+/* iommu pc mmio region register indexes */
+#define IOMMU_PC_COUNTER_REG			0x00
+#define IOMMU_PC_COUNTER_SRC_REG		0x08
+#define IOMMU_PC_PASID_MATCH_REG		0x10
+#define IOMMU_PC_DOMID_MATCH_REG		0x18
+#define IOMMU_PC_DEVID_MATCH_REG		0x20
+#define IOMMU_PC_COUNTER_REPORT_REG		0x28
+
+/* maximun specified bank/counters */
+#define PC_MAX_SPEC_BNKS			64
+#define PC_MAX_SPEC_CNTRS			16
+
+struct amd_iommu;
+
+/* amd_iommu_init.c external support functions */
+extern int amd_iommu_get_num_iommus(void);
+
+extern bool amd_iommu_pc_supported(void);
+
+extern u8 amd_iommu_pc_get_max_banks(unsigned int idx);
+
+extern u8 amd_iommu_pc_get_max_counters(unsigned int idx);
+
+extern int amd_iommu_pc_set_reg(struct amd_iommu *iommu, u8 bank, u8 cntr,
+				u8 fxn, u64 *value);
+
+extern int amd_iommu_pc_get_reg(struct amd_iommu *iommu, u8 bank, u8 cntr,
+				u8 fxn, u64 *value);
+
+extern struct amd_iommu *get_amd_iommu(int idx);
+
+#endif /*_PERF_EVENT_AMD_IOMMU_H_*/
diff --git a/arch/x86/events/amd/power.c b/arch/x86/events/amd/power.c
new file mode 100644
index 000000000..37d5b3805
--- /dev/null
+++ b/arch/x86/events/amd/power.c
@@ -0,0 +1,305 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Performance events - AMD Processor Power Reporting Mechanism
+ *
+ * Copyright (C) 2016 Advanced Micro Devices, Inc.
+ *
+ * Author: Huang Rui <ray.huang@amd.com>
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/perf_event.h>
+#include <asm/cpu_device_id.h>
+#include "../perf_event.h"
+
+/* Event code: LSB 8 bits, passed in attr->config any other bit is reserved. */
+#define AMD_POWER_EVENT_MASK		0xFFULL
+
+/*
+ * Accumulated power status counters.
+ */
+#define AMD_POWER_EVENTSEL_PKG		1
+
+/*
+ * The ratio of compute unit power accumulator sample period to the
+ * PTSC period.
+ */
+static unsigned int cpu_pwr_sample_ratio;
+
+/* Maximum accumulated power of a compute unit. */
+static u64 max_cu_acc_power;
+
+static struct pmu pmu_class;
+
+/*
+ * Accumulated power represents the sum of each compute unit's (CU) power
+ * consumption. On any core of each CU we read the total accumulated power from
+ * MSR_F15H_CU_PWR_ACCUMULATOR. cpu_mask represents CPU bit map of all cores
+ * which are picked to measure the power for the CUs they belong to.
+ */
+static cpumask_t cpu_mask;
+
+static void event_update(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 prev_pwr_acc, new_pwr_acc, prev_ptsc, new_ptsc;
+	u64 delta, tdelta;
+
+	prev_pwr_acc = hwc->pwr_acc;
+	prev_ptsc = hwc->ptsc;
+	rdmsrl(MSR_F15H_CU_PWR_ACCUMULATOR, new_pwr_acc);
+	rdmsrl(MSR_F15H_PTSC, new_ptsc);
+
+	/*
+	 * Calculate the CU power consumption over a time period, the unit of
+	 * final value (delta) is micro-Watts. Then add it to the event count.
+	 */
+	if (new_pwr_acc < prev_pwr_acc) {
+		delta = max_cu_acc_power + new_pwr_acc;
+		delta -= prev_pwr_acc;
+	} else
+		delta = new_pwr_acc - prev_pwr_acc;
+
+	delta *= cpu_pwr_sample_ratio * 1000;
+	tdelta = new_ptsc - prev_ptsc;
+
+	do_div(delta, tdelta);
+	local64_add(delta, &event->count);
+}
+
+static void __pmu_event_start(struct perf_event *event)
+{
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	event->hw.state = 0;
+
+	rdmsrl(MSR_F15H_PTSC, event->hw.ptsc);
+	rdmsrl(MSR_F15H_CU_PWR_ACCUMULATOR, event->hw.pwr_acc);
+}
+
+static void pmu_event_start(struct perf_event *event, int mode)
+{
+	__pmu_event_start(event);
+}
+
+static void pmu_event_stop(struct perf_event *event, int mode)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/* Mark event as deactivated and stopped. */
+	if (!(hwc->state & PERF_HES_STOPPED))
+		hwc->state |= PERF_HES_STOPPED;
+
+	/* Check if software counter update is necessary. */
+	if ((mode & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of an event
+		 * that we are disabling:
+		 */
+		event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+}
+
+static int pmu_event_add(struct perf_event *event, int mode)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	if (mode & PERF_EF_START)
+		__pmu_event_start(event);
+
+	return 0;
+}
+
+static void pmu_event_del(struct perf_event *event, int flags)
+{
+	pmu_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int pmu_event_init(struct perf_event *event)
+{
+	u64 cfg = event->attr.config & AMD_POWER_EVENT_MASK;
+
+	/* Only look at AMD power events. */
+	if (event->attr.type != pmu_class.type)
+		return -ENOENT;
+
+	/* Unsupported modes and filters. */
+	if (event->attr.sample_period)
+		return -EINVAL;
+
+	if (cfg != AMD_POWER_EVENTSEL_PKG)
+		return -EINVAL;
+
+	return 0;
+}
+
+static void pmu_event_read(struct perf_event *event)
+{
+	event_update(event);
+}
+
+static ssize_t
+get_attr_cpumask(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	return cpumap_print_to_pagebuf(true, buf, &cpu_mask);
+}
+
+static DEVICE_ATTR(cpumask, S_IRUGO, get_attr_cpumask, NULL);
+
+static struct attribute *pmu_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group pmu_attr_group = {
+	.attrs = pmu_attrs,
+};
+
+/*
+ * Currently it only supports to report the power of each
+ * processor/package.
+ */
+EVENT_ATTR_STR(power-pkg, power_pkg, "event=0x01");
+
+EVENT_ATTR_STR(power-pkg.unit, power_pkg_unit, "mWatts");
+
+/* Convert the count from micro-Watts to milli-Watts. */
+EVENT_ATTR_STR(power-pkg.scale, power_pkg_scale, "1.000000e-3");
+
+static struct attribute *events_attr[] = {
+	EVENT_PTR(power_pkg),
+	EVENT_PTR(power_pkg_unit),
+	EVENT_PTR(power_pkg_scale),
+	NULL,
+};
+
+static struct attribute_group pmu_events_group = {
+	.name	= "events",
+	.attrs	= events_attr,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-7");
+
+static struct attribute *formats_attr[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+
+static struct attribute_group pmu_format_group = {
+	.name	= "format",
+	.attrs	= formats_attr,
+};
+
+static const struct attribute_group *attr_groups[] = {
+	&pmu_attr_group,
+	&pmu_format_group,
+	&pmu_events_group,
+	NULL,
+};
+
+static struct pmu pmu_class = {
+	.attr_groups	= attr_groups,
+	/* system-wide only */
+	.task_ctx_nr	= perf_invalid_context,
+	.event_init	= pmu_event_init,
+	.add		= pmu_event_add,
+	.del		= pmu_event_del,
+	.start		= pmu_event_start,
+	.stop		= pmu_event_stop,
+	.read		= pmu_event_read,
+	.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+	.module		= THIS_MODULE,
+};
+
+static int power_cpu_exit(unsigned int cpu)
+{
+	int target;
+
+	if (!cpumask_test_and_clear_cpu(cpu, &cpu_mask))
+		return 0;
+
+	/*
+	 * Find a new CPU on the same compute unit, if was set in cpumask
+	 * and still some CPUs on compute unit. Then migrate event and
+	 * context to new CPU.
+	 */
+	target = cpumask_any_but(topology_sibling_cpumask(cpu), cpu);
+	if (target < nr_cpumask_bits) {
+		cpumask_set_cpu(target, &cpu_mask);
+		perf_pmu_migrate_context(&pmu_class, cpu, target);
+	}
+	return 0;
+}
+
+static int power_cpu_init(unsigned int cpu)
+{
+	int target;
+
+	/*
+	 * 1) If any CPU is set at cpu_mask in the same compute unit, do
+	 * nothing.
+	 * 2) If no CPU is set at cpu_mask in the same compute unit,
+	 * set current ONLINE CPU.
+	 *
+	 * Note: if there is a CPU aside of the new one already in the
+	 * sibling mask, then it is also in cpu_mask.
+	 */
+	target = cpumask_any_but(topology_sibling_cpumask(cpu), cpu);
+	if (target >= nr_cpumask_bits)
+		cpumask_set_cpu(cpu, &cpu_mask);
+	return 0;
+}
+
+static const struct x86_cpu_id cpu_match[] = {
+	X86_MATCH_VENDOR_FAM(AMD, 0x15, NULL),
+	{},
+};
+
+static int __init amd_power_pmu_init(void)
+{
+	int ret;
+
+	if (!x86_match_cpu(cpu_match))
+		return -ENODEV;
+
+	if (!boot_cpu_has(X86_FEATURE_ACC_POWER))
+		return -ENODEV;
+
+	cpu_pwr_sample_ratio = cpuid_ecx(0x80000007);
+
+	if (rdmsrl_safe(MSR_F15H_CU_MAX_PWR_ACCUMULATOR, &max_cu_acc_power)) {
+		pr_err("Failed to read max compute unit power accumulator MSR\n");
+		return -ENODEV;
+	}
+
+
+	cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_POWER_ONLINE,
+			  "perf/x86/amd/power:online",
+			  power_cpu_init, power_cpu_exit);
+
+	ret = perf_pmu_register(&pmu_class, "power", -1);
+	if (WARN_ON(ret)) {
+		pr_warn("AMD Power PMU registration failed\n");
+		return ret;
+	}
+
+	pr_info("AMD Power PMU detected\n");
+	return ret;
+}
+module_init(amd_power_pmu_init);
+
+static void __exit amd_power_pmu_exit(void)
+{
+	cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_AMD_POWER_ONLINE);
+	perf_pmu_unregister(&pmu_class);
+}
+module_exit(amd_power_pmu_exit);
+
+MODULE_AUTHOR("Huang Rui <ray.huang@amd.com>");
+MODULE_DESCRIPTION("AMD Processor Power Reporting Mechanism");
+MODULE_LICENSE("GPL v2");
diff --git a/arch/x86/events/amd/uncore.c b/arch/x86/events/amd/uncore.c
new file mode 100644
index 000000000..582c0ffb5
--- /dev/null
+++ b/arch/x86/events/amd/uncore.c
@@ -0,0 +1,670 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2013 Advanced Micro Devices, Inc.
+ *
+ * Author: Jacob Shin <jacob.shin@amd.com>
+ */
+
+#include <linux/perf_event.h>
+#include <linux/percpu.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
+
+#include <asm/cpufeature.h>
+#include <asm/perf_event.h>
+#include <asm/msr.h>
+#include <asm/smp.h>
+
+#define NUM_COUNTERS_NB		4
+#define NUM_COUNTERS_L2		4
+#define NUM_COUNTERS_L3		6
+#define MAX_COUNTERS		6
+
+#define RDPMC_BASE_NB		6
+#define RDPMC_BASE_LLC		10
+
+#define COUNTER_SHIFT		16
+
+#undef pr_fmt
+#define pr_fmt(fmt)	"amd_uncore: " fmt
+
+static int num_counters_llc;
+static int num_counters_nb;
+static bool l3_mask;
+
+static HLIST_HEAD(uncore_unused_list);
+
+struct amd_uncore {
+	int id;
+	int refcnt;
+	int cpu;
+	int num_counters;
+	int rdpmc_base;
+	u32 msr_base;
+	cpumask_t *active_mask;
+	struct pmu *pmu;
+	struct perf_event *events[MAX_COUNTERS];
+	struct hlist_node node;
+};
+
+static struct amd_uncore * __percpu *amd_uncore_nb;
+static struct amd_uncore * __percpu *amd_uncore_llc;
+
+static struct pmu amd_nb_pmu;
+static struct pmu amd_llc_pmu;
+
+static cpumask_t amd_nb_active_mask;
+static cpumask_t amd_llc_active_mask;
+
+static bool is_nb_event(struct perf_event *event)
+{
+	return event->pmu->type == amd_nb_pmu.type;
+}
+
+static bool is_llc_event(struct perf_event *event)
+{
+	return event->pmu->type == amd_llc_pmu.type;
+}
+
+static struct amd_uncore *event_to_amd_uncore(struct perf_event *event)
+{
+	if (is_nb_event(event) && amd_uncore_nb)
+		return *per_cpu_ptr(amd_uncore_nb, event->cpu);
+	else if (is_llc_event(event) && amd_uncore_llc)
+		return *per_cpu_ptr(amd_uncore_llc, event->cpu);
+
+	return NULL;
+}
+
+static void amd_uncore_read(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 prev, new;
+	s64 delta;
+
+	/*
+	 * since we do not enable counter overflow interrupts,
+	 * we do not have to worry about prev_count changing on us
+	 */
+
+	prev = local64_read(&hwc->prev_count);
+	rdpmcl(hwc->event_base_rdpmc, new);
+	local64_set(&hwc->prev_count, new);
+	delta = (new << COUNTER_SHIFT) - (prev << COUNTER_SHIFT);
+	delta >>= COUNTER_SHIFT;
+	local64_add(delta, &event->count);
+}
+
+static void amd_uncore_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (flags & PERF_EF_RELOAD)
+		wrmsrl(hwc->event_base, (u64)local64_read(&hwc->prev_count));
+
+	hwc->state = 0;
+	wrmsrl(hwc->config_base, (hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE));
+	perf_event_update_userpage(event);
+}
+
+static void amd_uncore_stop(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	wrmsrl(hwc->config_base, hwc->config);
+	hwc->state |= PERF_HES_STOPPED;
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		amd_uncore_read(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+}
+
+static int amd_uncore_add(struct perf_event *event, int flags)
+{
+	int i;
+	struct amd_uncore *uncore = event_to_amd_uncore(event);
+	struct hw_perf_event *hwc = &event->hw;
+
+	/* are we already assigned? */
+	if (hwc->idx != -1 && uncore->events[hwc->idx] == event)
+		goto out;
+
+	for (i = 0; i < uncore->num_counters; i++) {
+		if (uncore->events[i] == event) {
+			hwc->idx = i;
+			goto out;
+		}
+	}
+
+	/* if not, take the first available counter */
+	hwc->idx = -1;
+	for (i = 0; i < uncore->num_counters; i++) {
+		if (cmpxchg(&uncore->events[i], NULL, event) == NULL) {
+			hwc->idx = i;
+			break;
+		}
+	}
+
+out:
+	if (hwc->idx == -1)
+		return -EBUSY;
+
+	hwc->config_base = uncore->msr_base + (2 * hwc->idx);
+	hwc->event_base = uncore->msr_base + 1 + (2 * hwc->idx);
+	hwc->event_base_rdpmc = uncore->rdpmc_base + hwc->idx;
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	if (flags & PERF_EF_START)
+		amd_uncore_start(event, PERF_EF_RELOAD);
+
+	return 0;
+}
+
+static void amd_uncore_del(struct perf_event *event, int flags)
+{
+	int i;
+	struct amd_uncore *uncore = event_to_amd_uncore(event);
+	struct hw_perf_event *hwc = &event->hw;
+
+	amd_uncore_stop(event, PERF_EF_UPDATE);
+
+	for (i = 0; i < uncore->num_counters; i++) {
+		if (cmpxchg(&uncore->events[i], event, NULL) == event)
+			break;
+	}
+
+	hwc->idx = -1;
+}
+
+/*
+ * Return a full thread and slice mask unless user
+ * has provided them
+ */
+static u64 l3_thread_slice_mask(u64 config)
+{
+	if (boot_cpu_data.x86 <= 0x18)
+		return ((config & AMD64_L3_SLICE_MASK) ? : AMD64_L3_SLICE_MASK) |
+		       ((config & AMD64_L3_THREAD_MASK) ? : AMD64_L3_THREAD_MASK);
+
+	/*
+	 * If the user doesn't specify a threadmask, they're not trying to
+	 * count core 0, so we enable all cores & threads.
+	 * We'll also assume that they want to count slice 0 if they specify
+	 * a threadmask and leave sliceid and enallslices unpopulated.
+	 */
+	if (!(config & AMD64_L3_F19H_THREAD_MASK))
+		return AMD64_L3_F19H_THREAD_MASK | AMD64_L3_EN_ALL_SLICES |
+		       AMD64_L3_EN_ALL_CORES;
+
+	return config & (AMD64_L3_F19H_THREAD_MASK | AMD64_L3_SLICEID_MASK |
+			 AMD64_L3_EN_ALL_CORES | AMD64_L3_EN_ALL_SLICES |
+			 AMD64_L3_COREID_MASK);
+}
+
+static int amd_uncore_event_init(struct perf_event *event)
+{
+	struct amd_uncore *uncore;
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/*
+	 * NB and Last level cache counters (MSRs) are shared across all cores
+	 * that share the same NB / Last level cache.  On family 16h and below,
+	 * Interrupts can be directed to a single target core, however, event
+	 * counts generated by processes running on other cores cannot be masked
+	 * out. So we do not support sampling and per-thread events via
+	 * CAP_NO_INTERRUPT, and we do not enable counter overflow interrupts:
+	 */
+	hwc->config = event->attr.config & AMD64_RAW_EVENT_MASK_NB;
+	hwc->idx = -1;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	/*
+	 * SliceMask and ThreadMask need to be set for certain L3 events.
+	 * For other events, the two fields do not affect the count.
+	 */
+	if (l3_mask && is_llc_event(event))
+		hwc->config |= l3_thread_slice_mask(event->attr.config);
+
+	uncore = event_to_amd_uncore(event);
+	if (!uncore)
+		return -ENODEV;
+
+	/*
+	 * since request can come in to any of the shared cores, we will remap
+	 * to a single common cpu.
+	 */
+	event->cpu = uncore->cpu;
+
+	return 0;
+}
+
+static ssize_t amd_uncore_attr_show_cpumask(struct device *dev,
+					    struct device_attribute *attr,
+					    char *buf)
+{
+	cpumask_t *active_mask;
+	struct pmu *pmu = dev_get_drvdata(dev);
+
+	if (pmu->type == amd_nb_pmu.type)
+		active_mask = &amd_nb_active_mask;
+	else if (pmu->type == amd_llc_pmu.type)
+		active_mask = &amd_llc_active_mask;
+	else
+		return 0;
+
+	return cpumap_print_to_pagebuf(true, buf, active_mask);
+}
+static DEVICE_ATTR(cpumask, S_IRUGO, amd_uncore_attr_show_cpumask, NULL);
+
+static struct attribute *amd_uncore_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group amd_uncore_attr_group = {
+	.attrs = amd_uncore_attrs,
+};
+
+#define DEFINE_UNCORE_FORMAT_ATTR(_var, _name, _format)			\
+static ssize_t __uncore_##_var##_show(struct device *dev,		\
+				struct device_attribute *attr,		\
+				char *page)				\
+{									\
+	BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE);			\
+	return sprintf(page, _format "\n");				\
+}									\
+static struct device_attribute format_attr_##_var =			\
+	__ATTR(_name, 0444, __uncore_##_var##_show, NULL)
+
+DEFINE_UNCORE_FORMAT_ATTR(event12,	event,		"config:0-7,32-35");
+DEFINE_UNCORE_FORMAT_ATTR(event14,	event,		"config:0-7,32-35,59-60"); /* F17h+ DF */
+DEFINE_UNCORE_FORMAT_ATTR(event8,	event,		"config:0-7");		   /* F17h+ L3 */
+DEFINE_UNCORE_FORMAT_ATTR(umask,	umask,		"config:8-15");
+DEFINE_UNCORE_FORMAT_ATTR(coreid,	coreid,		"config:42-44");	   /* F19h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(slicemask,	slicemask,	"config:48-51");	   /* F17h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(threadmask8,	threadmask,	"config:56-63");	   /* F17h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(threadmask2,	threadmask,	"config:56-57");	   /* F19h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(enallslices,	enallslices,	"config:46");		   /* F19h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(enallcores,	enallcores,	"config:47");		   /* F19h L3 */
+DEFINE_UNCORE_FORMAT_ATTR(sliceid,	sliceid,	"config:48-50");	   /* F19h L3 */
+
+static struct attribute *amd_uncore_df_format_attr[] = {
+	&format_attr_event12.attr, /* event14 if F17h+ */
+	&format_attr_umask.attr,
+	NULL,
+};
+
+static struct attribute *amd_uncore_l3_format_attr[] = {
+	&format_attr_event12.attr, /* event8 if F17h+ */
+	&format_attr_umask.attr,
+	NULL, /* slicemask if F17h,	coreid if F19h */
+	NULL, /* threadmask8 if F17h,	enallslices if F19h */
+	NULL, /*			enallcores if F19h */
+	NULL, /*			sliceid if F19h */
+	NULL, /*			threadmask2 if F19h */
+	NULL,
+};
+
+static struct attribute_group amd_uncore_df_format_group = {
+	.name = "format",
+	.attrs = amd_uncore_df_format_attr,
+};
+
+static struct attribute_group amd_uncore_l3_format_group = {
+	.name = "format",
+	.attrs = amd_uncore_l3_format_attr,
+};
+
+static const struct attribute_group *amd_uncore_df_attr_groups[] = {
+	&amd_uncore_attr_group,
+	&amd_uncore_df_format_group,
+	NULL,
+};
+
+static const struct attribute_group *amd_uncore_l3_attr_groups[] = {
+	&amd_uncore_attr_group,
+	&amd_uncore_l3_format_group,
+	NULL,
+};
+
+static struct pmu amd_nb_pmu = {
+	.task_ctx_nr	= perf_invalid_context,
+	.attr_groups	= amd_uncore_df_attr_groups,
+	.name		= "amd_nb",
+	.event_init	= amd_uncore_event_init,
+	.add		= amd_uncore_add,
+	.del		= amd_uncore_del,
+	.start		= amd_uncore_start,
+	.stop		= amd_uncore_stop,
+	.read		= amd_uncore_read,
+	.capabilities	= PERF_PMU_CAP_NO_EXCLUDE | PERF_PMU_CAP_NO_INTERRUPT,
+};
+
+static struct pmu amd_llc_pmu = {
+	.task_ctx_nr	= perf_invalid_context,
+	.attr_groups	= amd_uncore_l3_attr_groups,
+	.name		= "amd_l2",
+	.event_init	= amd_uncore_event_init,
+	.add		= amd_uncore_add,
+	.del		= amd_uncore_del,
+	.start		= amd_uncore_start,
+	.stop		= amd_uncore_stop,
+	.read		= amd_uncore_read,
+	.capabilities	= PERF_PMU_CAP_NO_EXCLUDE | PERF_PMU_CAP_NO_INTERRUPT,
+};
+
+static struct amd_uncore *amd_uncore_alloc(unsigned int cpu)
+{
+	return kzalloc_node(sizeof(struct amd_uncore), GFP_KERNEL,
+			cpu_to_node(cpu));
+}
+
+static int amd_uncore_cpu_up_prepare(unsigned int cpu)
+{
+	struct amd_uncore *uncore_nb = NULL, *uncore_llc;
+
+	if (amd_uncore_nb) {
+		uncore_nb = amd_uncore_alloc(cpu);
+		if (!uncore_nb)
+			goto fail;
+		uncore_nb->cpu = cpu;
+		uncore_nb->num_counters = num_counters_nb;
+		uncore_nb->rdpmc_base = RDPMC_BASE_NB;
+		uncore_nb->msr_base = MSR_F15H_NB_PERF_CTL;
+		uncore_nb->active_mask = &amd_nb_active_mask;
+		uncore_nb->pmu = &amd_nb_pmu;
+		uncore_nb->id = -1;
+		*per_cpu_ptr(amd_uncore_nb, cpu) = uncore_nb;
+	}
+
+	if (amd_uncore_llc) {
+		uncore_llc = amd_uncore_alloc(cpu);
+		if (!uncore_llc)
+			goto fail;
+		uncore_llc->cpu = cpu;
+		uncore_llc->num_counters = num_counters_llc;
+		uncore_llc->rdpmc_base = RDPMC_BASE_LLC;
+		uncore_llc->msr_base = MSR_F16H_L2I_PERF_CTL;
+		uncore_llc->active_mask = &amd_llc_active_mask;
+		uncore_llc->pmu = &amd_llc_pmu;
+		uncore_llc->id = -1;
+		*per_cpu_ptr(amd_uncore_llc, cpu) = uncore_llc;
+	}
+
+	return 0;
+
+fail:
+	if (amd_uncore_nb)
+		*per_cpu_ptr(amd_uncore_nb, cpu) = NULL;
+	kfree(uncore_nb);
+	return -ENOMEM;
+}
+
+static struct amd_uncore *
+amd_uncore_find_online_sibling(struct amd_uncore *this,
+			       struct amd_uncore * __percpu *uncores)
+{
+	unsigned int cpu;
+	struct amd_uncore *that;
+
+	for_each_online_cpu(cpu) {
+		that = *per_cpu_ptr(uncores, cpu);
+
+		if (!that)
+			continue;
+
+		if (this == that)
+			continue;
+
+		if (this->id == that->id) {
+			hlist_add_head(&this->node, &uncore_unused_list);
+			this = that;
+			break;
+		}
+	}
+
+	this->refcnt++;
+	return this;
+}
+
+static int amd_uncore_cpu_starting(unsigned int cpu)
+{
+	unsigned int eax, ebx, ecx, edx;
+	struct amd_uncore *uncore;
+
+	if (amd_uncore_nb) {
+		uncore = *per_cpu_ptr(amd_uncore_nb, cpu);
+		cpuid(0x8000001e, &eax, &ebx, &ecx, &edx);
+		uncore->id = ecx & 0xff;
+
+		uncore = amd_uncore_find_online_sibling(uncore, amd_uncore_nb);
+		*per_cpu_ptr(amd_uncore_nb, cpu) = uncore;
+	}
+
+	if (amd_uncore_llc) {
+		uncore = *per_cpu_ptr(amd_uncore_llc, cpu);
+		uncore->id = per_cpu(cpu_llc_id, cpu);
+
+		uncore = amd_uncore_find_online_sibling(uncore, amd_uncore_llc);
+		*per_cpu_ptr(amd_uncore_llc, cpu) = uncore;
+	}
+
+	return 0;
+}
+
+static void uncore_clean_online(void)
+{
+	struct amd_uncore *uncore;
+	struct hlist_node *n;
+
+	hlist_for_each_entry_safe(uncore, n, &uncore_unused_list, node) {
+		hlist_del(&uncore->node);
+		kfree(uncore);
+	}
+}
+
+static void uncore_online(unsigned int cpu,
+			  struct amd_uncore * __percpu *uncores)
+{
+	struct amd_uncore *uncore = *per_cpu_ptr(uncores, cpu);
+
+	uncore_clean_online();
+
+	if (cpu == uncore->cpu)
+		cpumask_set_cpu(cpu, uncore->active_mask);
+}
+
+static int amd_uncore_cpu_online(unsigned int cpu)
+{
+	if (amd_uncore_nb)
+		uncore_online(cpu, amd_uncore_nb);
+
+	if (amd_uncore_llc)
+		uncore_online(cpu, amd_uncore_llc);
+
+	return 0;
+}
+
+static void uncore_down_prepare(unsigned int cpu,
+				struct amd_uncore * __percpu *uncores)
+{
+	unsigned int i;
+	struct amd_uncore *this = *per_cpu_ptr(uncores, cpu);
+
+	if (this->cpu != cpu)
+		return;
+
+	/* this cpu is going down, migrate to a shared sibling if possible */
+	for_each_online_cpu(i) {
+		struct amd_uncore *that = *per_cpu_ptr(uncores, i);
+
+		if (cpu == i)
+			continue;
+
+		if (this == that) {
+			perf_pmu_migrate_context(this->pmu, cpu, i);
+			cpumask_clear_cpu(cpu, that->active_mask);
+			cpumask_set_cpu(i, that->active_mask);
+			that->cpu = i;
+			break;
+		}
+	}
+}
+
+static int amd_uncore_cpu_down_prepare(unsigned int cpu)
+{
+	if (amd_uncore_nb)
+		uncore_down_prepare(cpu, amd_uncore_nb);
+
+	if (amd_uncore_llc)
+		uncore_down_prepare(cpu, amd_uncore_llc);
+
+	return 0;
+}
+
+static void uncore_dead(unsigned int cpu, struct amd_uncore * __percpu *uncores)
+{
+	struct amd_uncore *uncore = *per_cpu_ptr(uncores, cpu);
+
+	if (cpu == uncore->cpu)
+		cpumask_clear_cpu(cpu, uncore->active_mask);
+
+	if (!--uncore->refcnt)
+		kfree(uncore);
+	*per_cpu_ptr(uncores, cpu) = NULL;
+}
+
+static int amd_uncore_cpu_dead(unsigned int cpu)
+{
+	if (amd_uncore_nb)
+		uncore_dead(cpu, amd_uncore_nb);
+
+	if (amd_uncore_llc)
+		uncore_dead(cpu, amd_uncore_llc);
+
+	return 0;
+}
+
+static int __init amd_uncore_init(void)
+{
+	struct attribute **df_attr = amd_uncore_df_format_attr;
+	struct attribute **l3_attr = amd_uncore_l3_format_attr;
+	int ret = -ENODEV;
+
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD &&
+	    boot_cpu_data.x86_vendor != X86_VENDOR_HYGON)
+		return -ENODEV;
+
+	if (!boot_cpu_has(X86_FEATURE_TOPOEXT))
+		return -ENODEV;
+
+	num_counters_nb	= NUM_COUNTERS_NB;
+	num_counters_llc = NUM_COUNTERS_L2;
+	if (boot_cpu_data.x86 >= 0x17) {
+		/*
+		 * For F17h and above, the Northbridge counters are
+		 * repurposed as Data Fabric counters. Also, L3
+		 * counters are supported too. The PMUs are exported
+		 * based on family as either L2 or L3 and NB or DF.
+		 */
+		num_counters_llc	  = NUM_COUNTERS_L3;
+		amd_nb_pmu.name		  = "amd_df";
+		amd_llc_pmu.name	  = "amd_l3";
+		l3_mask			  = true;
+	}
+
+	if (boot_cpu_has(X86_FEATURE_PERFCTR_NB)) {
+		if (boot_cpu_data.x86 >= 0x17)
+			*df_attr = &format_attr_event14.attr;
+
+		amd_uncore_nb = alloc_percpu(struct amd_uncore *);
+		if (!amd_uncore_nb) {
+			ret = -ENOMEM;
+			goto fail_nb;
+		}
+		ret = perf_pmu_register(&amd_nb_pmu, amd_nb_pmu.name, -1);
+		if (ret)
+			goto fail_nb;
+
+		pr_info("%d %s %s counters detected\n", num_counters_nb,
+			boot_cpu_data.x86_vendor == X86_VENDOR_HYGON ?  "HYGON" : "",
+			amd_nb_pmu.name);
+
+		ret = 0;
+	}
+
+	if (boot_cpu_has(X86_FEATURE_PERFCTR_LLC)) {
+		if (boot_cpu_data.x86 >= 0x19) {
+			*l3_attr++ = &format_attr_event8.attr;
+			*l3_attr++ = &format_attr_umask.attr;
+			*l3_attr++ = &format_attr_coreid.attr;
+			*l3_attr++ = &format_attr_enallslices.attr;
+			*l3_attr++ = &format_attr_enallcores.attr;
+			*l3_attr++ = &format_attr_sliceid.attr;
+			*l3_attr++ = &format_attr_threadmask2.attr;
+		} else if (boot_cpu_data.x86 >= 0x17) {
+			*l3_attr++ = &format_attr_event8.attr;
+			*l3_attr++ = &format_attr_umask.attr;
+			*l3_attr++ = &format_attr_slicemask.attr;
+			*l3_attr++ = &format_attr_threadmask8.attr;
+		}
+
+		amd_uncore_llc = alloc_percpu(struct amd_uncore *);
+		if (!amd_uncore_llc) {
+			ret = -ENOMEM;
+			goto fail_llc;
+		}
+		ret = perf_pmu_register(&amd_llc_pmu, amd_llc_pmu.name, -1);
+		if (ret)
+			goto fail_llc;
+
+		pr_info("%d %s %s counters detected\n", num_counters_llc,
+			boot_cpu_data.x86_vendor == X86_VENDOR_HYGON ?  "HYGON" : "",
+			amd_llc_pmu.name);
+		ret = 0;
+	}
+
+	/*
+	 * Install callbacks. Core will call them for each online cpu.
+	 */
+	if (cpuhp_setup_state(CPUHP_PERF_X86_AMD_UNCORE_PREP,
+			      "perf/x86/amd/uncore:prepare",
+			      amd_uncore_cpu_up_prepare, amd_uncore_cpu_dead))
+		goto fail_llc;
+
+	if (cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_UNCORE_STARTING,
+			      "perf/x86/amd/uncore:starting",
+			      amd_uncore_cpu_starting, NULL))
+		goto fail_prep;
+	if (cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_UNCORE_ONLINE,
+			      "perf/x86/amd/uncore:online",
+			      amd_uncore_cpu_online,
+			      amd_uncore_cpu_down_prepare))
+		goto fail_start;
+	return 0;
+
+fail_start:
+	cpuhp_remove_state(CPUHP_AP_PERF_X86_AMD_UNCORE_STARTING);
+fail_prep:
+	cpuhp_remove_state(CPUHP_PERF_X86_AMD_UNCORE_PREP);
+fail_llc:
+	if (boot_cpu_has(X86_FEATURE_PERFCTR_NB))
+		perf_pmu_unregister(&amd_nb_pmu);
+	if (amd_uncore_llc)
+		free_percpu(amd_uncore_llc);
+fail_nb:
+	if (amd_uncore_nb)
+		free_percpu(amd_uncore_nb);
+
+	return ret;
+}
+device_initcall(amd_uncore_init);
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
new file mode 100644
index 000000000..b79b9f21c
--- /dev/null
+++ b/arch/x86/events/core.c
@@ -0,0 +1,2774 @@
+/*
+ * Performance events x86 architecture code
+ *
+ *  Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
+ *  Copyright (C) 2009 Jaswinder Singh Rajput
+ *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra
+ *  Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
+ *  Copyright (C) 2009 Google, Inc., Stephane Eranian
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+#include <linux/capability.h>
+#include <linux/notifier.h>
+#include <linux/hardirq.h>
+#include <linux/kprobes.h>
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/kdebug.h>
+#include <linux/sched/mm.h>
+#include <linux/sched/clock.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/cpu.h>
+#include <linux/bitops.h>
+#include <linux/device.h>
+#include <linux/nospec.h>
+#include <linux/static_call.h>
+
+#include <asm/apic.h>
+#include <asm/stacktrace.h>
+#include <asm/nmi.h>
+#include <asm/smp.h>
+#include <asm/alternative.h>
+#include <asm/mmu_context.h>
+#include <asm/tlbflush.h>
+#include <asm/timer.h>
+#include <asm/desc.h>
+#include <asm/ldt.h>
+#include <asm/unwind.h>
+
+#include "perf_event.h"
+
+struct x86_pmu x86_pmu __read_mostly;
+static struct pmu pmu;
+
+DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
+	.enabled = 1,
+	.pmu = &pmu,
+};
+
+DEFINE_STATIC_KEY_FALSE(rdpmc_never_available_key);
+DEFINE_STATIC_KEY_FALSE(rdpmc_always_available_key);
+
+/*
+ * This here uses DEFINE_STATIC_CALL_NULL() to get a static_call defined
+ * from just a typename, as opposed to an actual function.
+ */
+DEFINE_STATIC_CALL_NULL(x86_pmu_handle_irq,  *x86_pmu.handle_irq);
+DEFINE_STATIC_CALL_NULL(x86_pmu_disable_all, *x86_pmu.disable_all);
+DEFINE_STATIC_CALL_NULL(x86_pmu_enable_all,  *x86_pmu.enable_all);
+DEFINE_STATIC_CALL_NULL(x86_pmu_enable,	     *x86_pmu.enable);
+DEFINE_STATIC_CALL_NULL(x86_pmu_disable,     *x86_pmu.disable);
+
+DEFINE_STATIC_CALL_NULL(x86_pmu_add,  *x86_pmu.add);
+DEFINE_STATIC_CALL_NULL(x86_pmu_del,  *x86_pmu.del);
+DEFINE_STATIC_CALL_NULL(x86_pmu_read, *x86_pmu.read);
+
+DEFINE_STATIC_CALL_NULL(x86_pmu_schedule_events,       *x86_pmu.schedule_events);
+DEFINE_STATIC_CALL_NULL(x86_pmu_get_event_constraints, *x86_pmu.get_event_constraints);
+DEFINE_STATIC_CALL_NULL(x86_pmu_put_event_constraints, *x86_pmu.put_event_constraints);
+
+DEFINE_STATIC_CALL_NULL(x86_pmu_start_scheduling,  *x86_pmu.start_scheduling);
+DEFINE_STATIC_CALL_NULL(x86_pmu_commit_scheduling, *x86_pmu.commit_scheduling);
+DEFINE_STATIC_CALL_NULL(x86_pmu_stop_scheduling,   *x86_pmu.stop_scheduling);
+
+DEFINE_STATIC_CALL_NULL(x86_pmu_sched_task,    *x86_pmu.sched_task);
+DEFINE_STATIC_CALL_NULL(x86_pmu_swap_task_ctx, *x86_pmu.swap_task_ctx);
+
+DEFINE_STATIC_CALL_NULL(x86_pmu_drain_pebs,   *x86_pmu.drain_pebs);
+DEFINE_STATIC_CALL_NULL(x86_pmu_pebs_aliases, *x86_pmu.pebs_aliases);
+
+u64 __read_mostly hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+u64 __read_mostly hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+
+/*
+ * Propagate event elapsed time into the generic event.
+ * Can only be executed on the CPU where the event is active.
+ * Returns the delta events processed.
+ */
+u64 x86_perf_event_update(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int shift = 64 - x86_pmu.cntval_bits;
+	u64 prev_raw_count, new_raw_count;
+	u64 delta;
+
+	if (unlikely(!hwc->event_base))
+		return 0;
+
+	if (unlikely(is_topdown_count(event)) && x86_pmu.update_topdown_event)
+		return x86_pmu.update_topdown_event(event);
+
+	/*
+	 * Careful: an NMI might modify the previous event value.
+	 *
+	 * Our tactic to handle this is to first atomically read and
+	 * exchange a new raw count - then add that new-prev delta
+	 * count to the generic event atomically:
+	 */
+again:
+	prev_raw_count = local64_read(&hwc->prev_count);
+	rdpmcl(hwc->event_base_rdpmc, new_raw_count);
+
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+					new_raw_count) != prev_raw_count)
+		goto again;
+
+	/*
+	 * Now we have the new raw value and have updated the prev
+	 * timestamp already. We can now calculate the elapsed delta
+	 * (event-)time and add that to the generic event.
+	 *
+	 * Careful, not all hw sign-extends above the physical width
+	 * of the count.
+	 */
+	delta = (new_raw_count << shift) - (prev_raw_count << shift);
+	delta >>= shift;
+
+	local64_add(delta, &event->count);
+	local64_sub(delta, &hwc->period_left);
+
+	return new_raw_count;
+}
+
+/*
+ * Find and validate any extra registers to set up.
+ */
+static int x86_pmu_extra_regs(u64 config, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg;
+	struct extra_reg *er;
+
+	reg = &event->hw.extra_reg;
+
+	if (!x86_pmu.extra_regs)
+		return 0;
+
+	for (er = x86_pmu.extra_regs; er->msr; er++) {
+		if (er->event != (config & er->config_mask))
+			continue;
+		if (event->attr.config1 & ~er->valid_mask)
+			return -EINVAL;
+		/* Check if the extra msrs can be safely accessed*/
+		if (!er->extra_msr_access)
+			return -ENXIO;
+
+		reg->idx = er->idx;
+		reg->config = event->attr.config1;
+		reg->reg = er->msr;
+		break;
+	}
+	return 0;
+}
+
+static atomic_t active_events;
+static atomic_t pmc_refcount;
+static DEFINE_MUTEX(pmc_reserve_mutex);
+
+#ifdef CONFIG_X86_LOCAL_APIC
+
+static bool reserve_pmc_hardware(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		if (!reserve_perfctr_nmi(x86_pmu_event_addr(i)))
+			goto perfctr_fail;
+	}
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		if (!reserve_evntsel_nmi(x86_pmu_config_addr(i)))
+			goto eventsel_fail;
+	}
+
+	return true;
+
+eventsel_fail:
+	for (i--; i >= 0; i--)
+		release_evntsel_nmi(x86_pmu_config_addr(i));
+
+	i = x86_pmu.num_counters;
+
+perfctr_fail:
+	for (i--; i >= 0; i--)
+		release_perfctr_nmi(x86_pmu_event_addr(i));
+
+	return false;
+}
+
+static void release_pmc_hardware(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		release_perfctr_nmi(x86_pmu_event_addr(i));
+		release_evntsel_nmi(x86_pmu_config_addr(i));
+	}
+}
+
+#else
+
+static bool reserve_pmc_hardware(void) { return true; }
+static void release_pmc_hardware(void) {}
+
+#endif
+
+static bool check_hw_exists(void)
+{
+	u64 val, val_fail = -1, val_new= ~0;
+	int i, reg, reg_fail = -1, ret = 0;
+	int bios_fail = 0;
+	int reg_safe = -1;
+
+	/*
+	 * Check to see if the BIOS enabled any of the counters, if so
+	 * complain and bail.
+	 */
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		reg = x86_pmu_config_addr(i);
+		ret = rdmsrl_safe(reg, &val);
+		if (ret)
+			goto msr_fail;
+		if (val & ARCH_PERFMON_EVENTSEL_ENABLE) {
+			bios_fail = 1;
+			val_fail = val;
+			reg_fail = reg;
+		} else {
+			reg_safe = i;
+		}
+	}
+
+	if (x86_pmu.num_counters_fixed) {
+		reg = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
+		ret = rdmsrl_safe(reg, &val);
+		if (ret)
+			goto msr_fail;
+		for (i = 0; i < x86_pmu.num_counters_fixed; i++) {
+			if (val & (0x03 << i*4)) {
+				bios_fail = 1;
+				val_fail = val;
+				reg_fail = reg;
+			}
+		}
+	}
+
+	/*
+	 * If all the counters are enabled, the below test will always
+	 * fail.  The tools will also become useless in this scenario.
+	 * Just fail and disable the hardware counters.
+	 */
+
+	if (reg_safe == -1) {
+		reg = reg_safe;
+		goto msr_fail;
+	}
+
+	/*
+	 * 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.
+	 */
+	reg = x86_pmu_event_addr(reg_safe);
+	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)
+		goto msr_fail;
+
+	/*
+	 * We still allow the PMU driver to operate:
+	 */
+	if (bios_fail) {
+		pr_cont("Broken BIOS detected, complain to your hardware vendor.\n");
+		pr_err(FW_BUG "the BIOS has corrupted hw-PMU resources (MSR %x is %Lx)\n",
+			      reg_fail, val_fail);
+	}
+
+	return true;
+
+msr_fail:
+	if (boot_cpu_has(X86_FEATURE_HYPERVISOR)) {
+		pr_cont("PMU not available due to virtualization, using software events only.\n");
+	} else {
+		pr_cont("Broken PMU hardware detected, using software events only.\n");
+		pr_err("Failed to access perfctr msr (MSR %x is %Lx)\n",
+		       reg, val_new);
+	}
+
+	return false;
+}
+
+static void hw_perf_event_destroy(struct perf_event *event)
+{
+	x86_release_hardware();
+	atomic_dec(&active_events);
+}
+
+void hw_perf_lbr_event_destroy(struct perf_event *event)
+{
+	hw_perf_event_destroy(event);
+
+	/* undo the lbr/bts event accounting */
+	x86_del_exclusive(x86_lbr_exclusive_lbr);
+}
+
+static inline int x86_pmu_initialized(void)
+{
+	return x86_pmu.handle_irq != NULL;
+}
+
+static inline int
+set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event)
+{
+	struct perf_event_attr *attr = &event->attr;
+	unsigned int cache_type, cache_op, cache_result;
+	u64 config, val;
+
+	config = attr->config;
+
+	cache_type = (config >> 0) & 0xff;
+	if (cache_type >= PERF_COUNT_HW_CACHE_MAX)
+		return -EINVAL;
+	cache_type = array_index_nospec(cache_type, PERF_COUNT_HW_CACHE_MAX);
+
+	cache_op = (config >>  8) & 0xff;
+	if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX)
+		return -EINVAL;
+	cache_op = array_index_nospec(cache_op, PERF_COUNT_HW_CACHE_OP_MAX);
+
+	cache_result = (config >> 16) & 0xff;
+	if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX)
+		return -EINVAL;
+	cache_result = array_index_nospec(cache_result, PERF_COUNT_HW_CACHE_RESULT_MAX);
+
+	val = hw_cache_event_ids[cache_type][cache_op][cache_result];
+
+	if (val == 0)
+		return -ENOENT;
+
+	if (val == -1)
+		return -EINVAL;
+
+	hwc->config |= val;
+	attr->config1 = hw_cache_extra_regs[cache_type][cache_op][cache_result];
+	return x86_pmu_extra_regs(val, event);
+}
+
+int x86_reserve_hardware(void)
+{
+	int err = 0;
+
+	if (!atomic_inc_not_zero(&pmc_refcount)) {
+		mutex_lock(&pmc_reserve_mutex);
+		if (atomic_read(&pmc_refcount) == 0) {
+			if (!reserve_pmc_hardware()) {
+				err = -EBUSY;
+			} else {
+				reserve_ds_buffers();
+				reserve_lbr_buffers();
+			}
+		}
+		if (!err)
+			atomic_inc(&pmc_refcount);
+		mutex_unlock(&pmc_reserve_mutex);
+	}
+
+	return err;
+}
+
+void x86_release_hardware(void)
+{
+	if (atomic_dec_and_mutex_lock(&pmc_refcount, &pmc_reserve_mutex)) {
+		release_pmc_hardware();
+		release_ds_buffers();
+		release_lbr_buffers();
+		mutex_unlock(&pmc_reserve_mutex);
+	}
+}
+
+/*
+ * Check if we can create event of a certain type (that no conflicting events
+ * are present).
+ */
+int x86_add_exclusive(unsigned int what)
+{
+	int i;
+
+	/*
+	 * When lbr_pt_coexist we allow PT to coexist with either LBR or BTS.
+	 * LBR and BTS are still mutually exclusive.
+	 */
+	if (x86_pmu.lbr_pt_coexist && what == x86_lbr_exclusive_pt)
+		goto out;
+
+	if (!atomic_inc_not_zero(&x86_pmu.lbr_exclusive[what])) {
+		mutex_lock(&pmc_reserve_mutex);
+		for (i = 0; i < ARRAY_SIZE(x86_pmu.lbr_exclusive); i++) {
+			if (i != what && atomic_read(&x86_pmu.lbr_exclusive[i]))
+				goto fail_unlock;
+		}
+		atomic_inc(&x86_pmu.lbr_exclusive[what]);
+		mutex_unlock(&pmc_reserve_mutex);
+	}
+
+out:
+	atomic_inc(&active_events);
+	return 0;
+
+fail_unlock:
+	mutex_unlock(&pmc_reserve_mutex);
+	return -EBUSY;
+}
+
+void x86_del_exclusive(unsigned int what)
+{
+	atomic_dec(&active_events);
+
+	/*
+	 * See the comment in x86_add_exclusive().
+	 */
+	if (x86_pmu.lbr_pt_coexist && what == x86_lbr_exclusive_pt)
+		return;
+
+	atomic_dec(&x86_pmu.lbr_exclusive[what]);
+}
+
+int x86_setup_perfctr(struct perf_event *event)
+{
+	struct perf_event_attr *attr = &event->attr;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 config;
+
+	if (!is_sampling_event(event)) {
+		hwc->sample_period = x86_pmu.max_period;
+		hwc->last_period = hwc->sample_period;
+		local64_set(&hwc->period_left, hwc->sample_period);
+	}
+
+	if (attr->type == PERF_TYPE_RAW)
+		return x86_pmu_extra_regs(event->attr.config, event);
+
+	if (attr->type == PERF_TYPE_HW_CACHE)
+		return set_ext_hw_attr(hwc, event);
+
+	if (attr->config >= x86_pmu.max_events)
+		return -EINVAL;
+
+	attr->config = array_index_nospec((unsigned long)attr->config, x86_pmu.max_events);
+
+	/*
+	 * The generic map:
+	 */
+	config = x86_pmu.event_map(attr->config);
+
+	if (config == 0)
+		return -ENOENT;
+
+	if (config == -1LL)
+		return -EINVAL;
+
+	hwc->config |= config;
+
+	return 0;
+}
+
+/*
+ * check that branch_sample_type is compatible with
+ * settings needed for precise_ip > 1 which implies
+ * using the LBR to capture ALL taken branches at the
+ * priv levels of the measurement
+ */
+static inline int precise_br_compat(struct perf_event *event)
+{
+	u64 m = event->attr.branch_sample_type;
+	u64 b = 0;
+
+	/* must capture all branches */
+	if (!(m & PERF_SAMPLE_BRANCH_ANY))
+		return 0;
+
+	m &= PERF_SAMPLE_BRANCH_KERNEL | PERF_SAMPLE_BRANCH_USER;
+
+	if (!event->attr.exclude_user)
+		b |= PERF_SAMPLE_BRANCH_USER;
+
+	if (!event->attr.exclude_kernel)
+		b |= PERF_SAMPLE_BRANCH_KERNEL;
+
+	/*
+	 * ignore PERF_SAMPLE_BRANCH_HV, not supported on x86
+	 */
+
+	return m == b;
+}
+
+int x86_pmu_max_precise(void)
+{
+	int precise = 0;
+
+	/* Support for constant skid */
+	if (x86_pmu.pebs_active && !x86_pmu.pebs_broken) {
+		precise++;
+
+		/* Support for IP fixup */
+		if (x86_pmu.lbr_nr || x86_pmu.intel_cap.pebs_format >= 2)
+			precise++;
+
+		if (x86_pmu.pebs_prec_dist)
+			precise++;
+	}
+	return precise;
+}
+
+int x86_pmu_hw_config(struct perf_event *event)
+{
+	if (event->attr.precise_ip) {
+		int precise = x86_pmu_max_precise();
+
+		if (event->attr.precise_ip > precise)
+			return -EOPNOTSUPP;
+
+		/* There's no sense in having PEBS for non sampling events: */
+		if (!is_sampling_event(event))
+			return -EINVAL;
+	}
+	/*
+	 * check that PEBS LBR correction does not conflict with
+	 * whatever the user is asking with attr->branch_sample_type
+	 */
+	if (event->attr.precise_ip > 1 && x86_pmu.intel_cap.pebs_format < 2) {
+		u64 *br_type = &event->attr.branch_sample_type;
+
+		if (has_branch_stack(event)) {
+			if (!precise_br_compat(event))
+				return -EOPNOTSUPP;
+
+			/* branch_sample_type is compatible */
+
+		} else {
+			/*
+			 * user did not specify  branch_sample_type
+			 *
+			 * For PEBS fixups, we capture all
+			 * the branches at the priv level of the
+			 * event.
+			 */
+			*br_type = PERF_SAMPLE_BRANCH_ANY;
+
+			if (!event->attr.exclude_user)
+				*br_type |= PERF_SAMPLE_BRANCH_USER;
+
+			if (!event->attr.exclude_kernel)
+				*br_type |= PERF_SAMPLE_BRANCH_KERNEL;
+		}
+	}
+
+	if (event->attr.branch_sample_type & PERF_SAMPLE_BRANCH_CALL_STACK)
+		event->attach_state |= PERF_ATTACH_TASK_DATA;
+
+	/*
+	 * Generate PMC IRQs:
+	 * (keep 'enabled' bit clear for now)
+	 */
+	event->hw.config = ARCH_PERFMON_EVENTSEL_INT;
+
+	/*
+	 * Count user and OS events unless requested not to
+	 */
+	if (!event->attr.exclude_user)
+		event->hw.config |= ARCH_PERFMON_EVENTSEL_USR;
+	if (!event->attr.exclude_kernel)
+		event->hw.config |= ARCH_PERFMON_EVENTSEL_OS;
+
+	if (event->attr.type == PERF_TYPE_RAW)
+		event->hw.config |= event->attr.config & X86_RAW_EVENT_MASK;
+
+	if (event->attr.sample_period && x86_pmu.limit_period) {
+		if (x86_pmu.limit_period(event, event->attr.sample_period) >
+				event->attr.sample_period)
+			return -EINVAL;
+	}
+
+	/* sample_regs_user never support XMM registers */
+	if (unlikely(event->attr.sample_regs_user & PERF_REG_EXTENDED_MASK))
+		return -EINVAL;
+	/*
+	 * Besides the general purpose registers, XMM registers may
+	 * be collected in PEBS on some platforms, e.g. Icelake
+	 */
+	if (unlikely(event->attr.sample_regs_intr & PERF_REG_EXTENDED_MASK)) {
+		if (!(event->pmu->capabilities & PERF_PMU_CAP_EXTENDED_REGS))
+			return -EINVAL;
+
+		if (!event->attr.precise_ip)
+			return -EINVAL;
+	}
+
+	return x86_setup_perfctr(event);
+}
+
+/*
+ * Setup the hardware configuration for a given attr_type
+ */
+static int __x86_pmu_event_init(struct perf_event *event)
+{
+	int err;
+
+	if (!x86_pmu_initialized())
+		return -ENODEV;
+
+	err = x86_reserve_hardware();
+	if (err)
+		return err;
+
+	atomic_inc(&active_events);
+	event->destroy = hw_perf_event_destroy;
+
+	event->hw.idx = -1;
+	event->hw.last_cpu = -1;
+	event->hw.last_tag = ~0ULL;
+
+	/* mark unused */
+	event->hw.extra_reg.idx = EXTRA_REG_NONE;
+	event->hw.branch_reg.idx = EXTRA_REG_NONE;
+
+	return x86_pmu.hw_config(event);
+}
+
+void x86_pmu_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+		u64 val;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		rdmsrl(x86_pmu_config_addr(idx), val);
+		if (!(val & ARCH_PERFMON_EVENTSEL_ENABLE))
+			continue;
+		val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+		wrmsrl(x86_pmu_config_addr(idx), val);
+		if (is_counter_pair(hwc))
+			wrmsrl(x86_pmu_config_addr(idx + 1), 0);
+	}
+}
+
+/*
+ * There may be PMI landing after enabled=0. The PMI hitting could be before or
+ * after disable_all.
+ *
+ * If PMI hits before disable_all, the PMU will be disabled in the NMI handler.
+ * It will not be re-enabled in the NMI handler again, because enabled=0. After
+ * handling the NMI, disable_all will be called, which will not change the
+ * state either. If PMI hits after disable_all, the PMU is already disabled
+ * before entering NMI handler. The NMI handler will not change the state
+ * either.
+ *
+ * So either situation is harmless.
+ */
+static void x86_pmu_disable(struct pmu *pmu)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!x86_pmu_initialized())
+		return;
+
+	if (!cpuc->enabled)
+		return;
+
+	cpuc->n_added = 0;
+	cpuc->enabled = 0;
+	barrier();
+
+	static_call(x86_pmu_disable_all)();
+}
+
+void x86_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+	}
+}
+
+static inline int is_x86_event(struct perf_event *event)
+{
+	return event->pmu == &pmu;
+}
+
+struct pmu *x86_get_pmu(unsigned int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+
+	/*
+	 * All CPUs of the hybrid type have been offline.
+	 * The x86_get_pmu() should not be invoked.
+	 */
+	if (WARN_ON_ONCE(!cpuc->pmu))
+		return &pmu;
+
+	return cpuc->pmu;
+}
+/*
+ * Event scheduler state:
+ *
+ * Assign events iterating over all events and counters, beginning
+ * with events with least weights first. Keep the current iterator
+ * state in struct sched_state.
+ */
+struct sched_state {
+	int	weight;
+	int	event;		/* event index */
+	int	counter;	/* counter index */
+	int	unassigned;	/* number of events to be assigned left */
+	int	nr_gp;		/* number of GP counters used */
+	u64	used;
+};
+
+/* Total max is X86_PMC_IDX_MAX, but we are O(n!) limited */
+#define	SCHED_STATES_MAX	2
+
+struct perf_sched {
+	int			max_weight;
+	int			max_events;
+	int			max_gp;
+	int			saved_states;
+	struct event_constraint	**constraints;
+	struct sched_state	state;
+	struct sched_state	saved[SCHED_STATES_MAX];
+};
+
+/*
+ * Initialize interator that runs through all events and counters.
+ */
+static void perf_sched_init(struct perf_sched *sched, struct event_constraint **constraints,
+			    int num, int wmin, int wmax, int gpmax)
+{
+	int idx;
+
+	memset(sched, 0, sizeof(*sched));
+	sched->max_events	= num;
+	sched->max_weight	= wmax;
+	sched->max_gp		= gpmax;
+	sched->constraints	= constraints;
+
+	for (idx = 0; idx < num; idx++) {
+		if (constraints[idx]->weight == wmin)
+			break;
+	}
+
+	sched->state.event	= idx;		/* start with min weight */
+	sched->state.weight	= wmin;
+	sched->state.unassigned	= num;
+}
+
+static void perf_sched_save_state(struct perf_sched *sched)
+{
+	if (WARN_ON_ONCE(sched->saved_states >= SCHED_STATES_MAX))
+		return;
+
+	sched->saved[sched->saved_states] = sched->state;
+	sched->saved_states++;
+}
+
+static bool perf_sched_restore_state(struct perf_sched *sched)
+{
+	if (!sched->saved_states)
+		return false;
+
+	sched->saved_states--;
+	sched->state = sched->saved[sched->saved_states];
+
+	/* this assignment didn't work out */
+	/* XXX broken vs EVENT_PAIR */
+	sched->state.used &= ~BIT_ULL(sched->state.counter);
+
+	/* try the next one */
+	sched->state.counter++;
+
+	return true;
+}
+
+/*
+ * Select a counter for the current event to schedule. Return true on
+ * success.
+ */
+static bool __perf_sched_find_counter(struct perf_sched *sched)
+{
+	struct event_constraint *c;
+	int idx;
+
+	if (!sched->state.unassigned)
+		return false;
+
+	if (sched->state.event >= sched->max_events)
+		return false;
+
+	c = sched->constraints[sched->state.event];
+	/* Prefer fixed purpose counters */
+	if (c->idxmsk64 & (~0ULL << INTEL_PMC_IDX_FIXED)) {
+		idx = INTEL_PMC_IDX_FIXED;
+		for_each_set_bit_from(idx, c->idxmsk, X86_PMC_IDX_MAX) {
+			u64 mask = BIT_ULL(idx);
+
+			if (sched->state.used & mask)
+				continue;
+
+			sched->state.used |= mask;
+			goto done;
+		}
+	}
+
+	/* Grab the first unused counter starting with idx */
+	idx = sched->state.counter;
+	for_each_set_bit_from(idx, c->idxmsk, INTEL_PMC_IDX_FIXED) {
+		u64 mask = BIT_ULL(idx);
+
+		if (c->flags & PERF_X86_EVENT_PAIR)
+			mask |= mask << 1;
+
+		if (sched->state.used & mask)
+			continue;
+
+		if (sched->state.nr_gp++ >= sched->max_gp)
+			return false;
+
+		sched->state.used |= mask;
+		goto done;
+	}
+
+	return false;
+
+done:
+	sched->state.counter = idx;
+
+	if (c->overlap)
+		perf_sched_save_state(sched);
+
+	return true;
+}
+
+static bool perf_sched_find_counter(struct perf_sched *sched)
+{
+	while (!__perf_sched_find_counter(sched)) {
+		if (!perf_sched_restore_state(sched))
+			return false;
+	}
+
+	return true;
+}
+
+/*
+ * Go through all unassigned events and find the next one to schedule.
+ * Take events with the least weight first. Return true on success.
+ */
+static bool perf_sched_next_event(struct perf_sched *sched)
+{
+	struct event_constraint *c;
+
+	if (!sched->state.unassigned || !--sched->state.unassigned)
+		return false;
+
+	do {
+		/* next event */
+		sched->state.event++;
+		if (sched->state.event >= sched->max_events) {
+			/* next weight */
+			sched->state.event = 0;
+			sched->state.weight++;
+			if (sched->state.weight > sched->max_weight)
+				return false;
+		}
+		c = sched->constraints[sched->state.event];
+	} while (c->weight != sched->state.weight);
+
+	sched->state.counter = 0;	/* start with first counter */
+
+	return true;
+}
+
+/*
+ * Assign a counter for each event.
+ */
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int gpmax, int *assign)
+{
+	struct perf_sched sched;
+
+	perf_sched_init(&sched, constraints, n, wmin, wmax, gpmax);
+
+	do {
+		if (!perf_sched_find_counter(&sched))
+			break;	/* failed */
+		if (assign)
+			assign[sched.state.event] = sched.state.counter;
+	} while (perf_sched_next_event(&sched));
+
+	return sched.state.unassigned;
+}
+EXPORT_SYMBOL_GPL(perf_assign_events);
+
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+{
+	struct event_constraint *c;
+	struct perf_event *e;
+	int n0, i, wmin, wmax, unsched = 0;
+	struct hw_perf_event *hwc;
+	u64 used_mask = 0;
+
+	/*
+	 * Compute the number of events already present; see x86_pmu_add(),
+	 * validate_group() and x86_pmu_commit_txn(). For the former two
+	 * cpuc->n_events hasn't been updated yet, while for the latter
+	 * cpuc->n_txn contains the number of events added in the current
+	 * transaction.
+	 */
+	n0 = cpuc->n_events;
+	if (cpuc->txn_flags & PERF_PMU_TXN_ADD)
+		n0 -= cpuc->n_txn;
+
+	static_call_cond(x86_pmu_start_scheduling)(cpuc);
+
+	for (i = 0, wmin = X86_PMC_IDX_MAX, wmax = 0; i < n; i++) {
+		c = cpuc->event_constraint[i];
+
+		/*
+		 * Previously scheduled events should have a cached constraint,
+		 * while new events should not have one.
+		 */
+		WARN_ON_ONCE((c && i >= n0) || (!c && i < n0));
+
+		/*
+		 * Request constraints for new events; or for those events that
+		 * have a dynamic constraint -- for those the constraint can
+		 * change due to external factors (sibling state, allow_tfa).
+		 */
+		if (!c || (c->flags & PERF_X86_EVENT_DYNAMIC)) {
+			c = static_call(x86_pmu_get_event_constraints)(cpuc, i, cpuc->event_list[i]);
+			cpuc->event_constraint[i] = c;
+		}
+
+		wmin = min(wmin, c->weight);
+		wmax = max(wmax, c->weight);
+	}
+
+	/*
+	 * fastpath, try to reuse previous register
+	 */
+	for (i = 0; i < n; i++) {
+		u64 mask;
+
+		hwc = &cpuc->event_list[i]->hw;
+		c = cpuc->event_constraint[i];
+
+		/* never assigned */
+		if (hwc->idx == -1)
+			break;
+
+		/* constraint still honored */
+		if (!test_bit(hwc->idx, c->idxmsk))
+			break;
+
+		mask = BIT_ULL(hwc->idx);
+		if (is_counter_pair(hwc))
+			mask |= mask << 1;
+
+		/* not already used */
+		if (used_mask & mask)
+			break;
+
+		used_mask |= mask;
+
+		if (assign)
+			assign[i] = hwc->idx;
+	}
+
+	/* slow path */
+	if (i != n) {
+		int gpmax = x86_pmu.num_counters;
+
+		/*
+		 * Do not allow scheduling of more than half the available
+		 * generic counters.
+		 *
+		 * This helps avoid counter starvation of sibling thread by
+		 * ensuring at most half the counters cannot be in exclusive
+		 * mode. There is no designated counters for the limits. Any
+		 * N/2 counters can be used. This helps with events with
+		 * specific counter constraints.
+		 */
+		if (is_ht_workaround_enabled() && !cpuc->is_fake &&
+		    READ_ONCE(cpuc->excl_cntrs->exclusive_present))
+			gpmax /= 2;
+
+		/*
+		 * Reduce the amount of available counters to allow fitting
+		 * the extra Merge events needed by large increment events.
+		 */
+		if (x86_pmu.flags & PMU_FL_PAIR) {
+			gpmax = x86_pmu.num_counters - cpuc->n_pair;
+			WARN_ON(gpmax <= 0);
+		}
+
+		unsched = perf_assign_events(cpuc->event_constraint, n, wmin,
+					     wmax, gpmax, assign);
+	}
+
+	/*
+	 * In case of success (unsched = 0), mark events as committed,
+	 * so we do not put_constraint() in case new events are added
+	 * and fail to be scheduled
+	 *
+	 * We invoke the lower level commit callback to lock the resource
+	 *
+	 * We do not need to do all of this in case we are called to
+	 * validate an event group (assign == NULL)
+	 */
+	if (!unsched && assign) {
+		for (i = 0; i < n; i++) {
+			e = cpuc->event_list[i];
+			static_call_cond(x86_pmu_commit_scheduling)(cpuc, i, assign[i]);
+		}
+	} else {
+		for (i = n0; i < n; i++) {
+			e = cpuc->event_list[i];
+
+			/*
+			 * release events that failed scheduling
+			 */
+			static_call_cond(x86_pmu_put_event_constraints)(cpuc, e);
+
+			cpuc->event_constraint[i] = NULL;
+		}
+	}
+
+	static_call_cond(x86_pmu_stop_scheduling)(cpuc);
+
+	return unsched ? -EINVAL : 0;
+}
+
+static int add_nr_metric_event(struct cpu_hw_events *cpuc,
+			       struct perf_event *event)
+{
+	if (is_metric_event(event)) {
+		if (cpuc->n_metric == INTEL_TD_METRIC_NUM)
+			return -EINVAL;
+		cpuc->n_metric++;
+		cpuc->n_txn_metric++;
+	}
+
+	return 0;
+}
+
+static void del_nr_metric_event(struct cpu_hw_events *cpuc,
+				struct perf_event *event)
+{
+	if (is_metric_event(event))
+		cpuc->n_metric--;
+}
+
+static int collect_event(struct cpu_hw_events *cpuc, struct perf_event *event,
+			 int max_count, int n)
+{
+
+	if (x86_pmu.intel_cap.perf_metrics && add_nr_metric_event(cpuc, event))
+		return -EINVAL;
+
+	if (n >= max_count + cpuc->n_metric)
+		return -EINVAL;
+
+	cpuc->event_list[n] = event;
+	if (is_counter_pair(&event->hw)) {
+		cpuc->n_pair++;
+		cpuc->n_txn_pair++;
+	}
+
+	return 0;
+}
+
+/*
+ * dogrp: true if must collect siblings events (group)
+ * returns total number of events and error code
+ */
+static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, bool dogrp)
+{
+	struct perf_event *event;
+	int n, max_count;
+
+	max_count = x86_pmu.num_counters + x86_pmu.num_counters_fixed;
+
+	/* current number of events already accepted */
+	n = cpuc->n_events;
+	if (!cpuc->n_events)
+		cpuc->pebs_output = 0;
+
+	if (!cpuc->is_fake && leader->attr.precise_ip) {
+		/*
+		 * For PEBS->PT, if !aux_event, the group leader (PT) went
+		 * away, the group was broken down and this singleton event
+		 * can't schedule any more.
+		 */
+		if (is_pebs_pt(leader) && !leader->aux_event)
+			return -EINVAL;
+
+		/*
+		 * pebs_output: 0: no PEBS so far, 1: PT, 2: DS
+		 */
+		if (cpuc->pebs_output &&
+		    cpuc->pebs_output != is_pebs_pt(leader) + 1)
+			return -EINVAL;
+
+		cpuc->pebs_output = is_pebs_pt(leader) + 1;
+	}
+
+	if (is_x86_event(leader)) {
+		if (collect_event(cpuc, leader, max_count, n))
+			return -EINVAL;
+		n++;
+	}
+
+	if (!dogrp)
+		return n;
+
+	for_each_sibling_event(event, leader) {
+		if (!is_x86_event(event) || event->state <= PERF_EVENT_STATE_OFF)
+			continue;
+
+		if (collect_event(cpuc, event, max_count, n))
+			return -EINVAL;
+
+		n++;
+	}
+	return n;
+}
+
+static inline void x86_assign_hw_event(struct perf_event *event,
+				struct cpu_hw_events *cpuc, int i)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx;
+
+	idx = hwc->idx = cpuc->assign[i];
+	hwc->last_cpu = smp_processor_id();
+	hwc->last_tag = ++cpuc->tags[i];
+
+	switch (hwc->idx) {
+	case INTEL_PMC_IDX_FIXED_BTS:
+	case INTEL_PMC_IDX_FIXED_VLBR:
+		hwc->config_base = 0;
+		hwc->event_base	= 0;
+		break;
+
+	case INTEL_PMC_IDX_METRIC_BASE ... INTEL_PMC_IDX_METRIC_END:
+		/* All the metric events are mapped onto the fixed counter 3. */
+		idx = INTEL_PMC_IDX_FIXED_SLOTS;
+		/* fall through */
+	case INTEL_PMC_IDX_FIXED ... INTEL_PMC_IDX_FIXED_BTS-1:
+		hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
+		hwc->event_base = MSR_ARCH_PERFMON_FIXED_CTR0 +
+				(idx - INTEL_PMC_IDX_FIXED);
+		hwc->event_base_rdpmc = (idx - INTEL_PMC_IDX_FIXED) |
+					INTEL_PMC_FIXED_RDPMC_BASE;
+		break;
+
+	default:
+		hwc->config_base = x86_pmu_config_addr(hwc->idx);
+		hwc->event_base  = x86_pmu_event_addr(hwc->idx);
+		hwc->event_base_rdpmc = x86_pmu_rdpmc_index(hwc->idx);
+		break;
+	}
+}
+
+/**
+ * x86_perf_rdpmc_index - Return PMC counter used for event
+ * @event: the perf_event to which the PMC counter was assigned
+ *
+ * The counter assigned to this performance event may change if interrupts
+ * are enabled. This counter should thus never be used while interrupts are
+ * enabled. Before this function is used to obtain the assigned counter the
+ * event should be checked for validity using, for example,
+ * perf_event_read_local(), within the same interrupt disabled section in
+ * which this counter is planned to be used.
+ *
+ * Return: The index of the performance monitoring counter assigned to
+ * @perf_event.
+ */
+int x86_perf_rdpmc_index(struct perf_event *event)
+{
+	lockdep_assert_irqs_disabled();
+
+	return event->hw.event_base_rdpmc;
+}
+
+static inline int match_prev_assignment(struct hw_perf_event *hwc,
+					struct cpu_hw_events *cpuc,
+					int i)
+{
+	return hwc->idx == cpuc->assign[i] &&
+		hwc->last_cpu == smp_processor_id() &&
+		hwc->last_tag == cpuc->tags[i];
+}
+
+static void x86_pmu_start(struct perf_event *event, int flags);
+
+static void x86_pmu_enable(struct pmu *pmu)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_event *event;
+	struct hw_perf_event *hwc;
+	int i, added = cpuc->n_added;
+
+	if (!x86_pmu_initialized())
+		return;
+
+	if (cpuc->enabled)
+		return;
+
+	if (cpuc->n_added) {
+		int n_running = cpuc->n_events - cpuc->n_added;
+		/*
+		 * apply assignment obtained either from
+		 * hw_perf_group_sched_in() or x86_pmu_enable()
+		 *
+		 * step1: save events moving to new counters
+		 */
+		for (i = 0; i < n_running; i++) {
+			event = cpuc->event_list[i];
+			hwc = &event->hw;
+
+			/*
+			 * we can avoid reprogramming counter if:
+			 * - assigned same counter as last time
+			 * - running on same CPU as last time
+			 * - no other event has used the counter since
+			 */
+			if (hwc->idx == -1 ||
+			    match_prev_assignment(hwc, cpuc, i))
+				continue;
+
+			/*
+			 * Ensure we don't accidentally enable a stopped
+			 * counter simply because we rescheduled.
+			 */
+			if (hwc->state & PERF_HES_STOPPED)
+				hwc->state |= PERF_HES_ARCH;
+
+			x86_pmu_stop(event, PERF_EF_UPDATE);
+		}
+
+		/*
+		 * step2: reprogram moved events into new counters
+		 */
+		for (i = 0; i < cpuc->n_events; i++) {
+			event = cpuc->event_list[i];
+			hwc = &event->hw;
+
+			if (!match_prev_assignment(hwc, cpuc, i))
+				x86_assign_hw_event(event, cpuc, i);
+			else if (i < n_running)
+				continue;
+
+			if (hwc->state & PERF_HES_ARCH)
+				continue;
+
+			x86_pmu_start(event, PERF_EF_RELOAD);
+		}
+		cpuc->n_added = 0;
+		perf_events_lapic_init();
+	}
+
+	cpuc->enabled = 1;
+	barrier();
+
+	static_call(x86_pmu_enable_all)(added);
+}
+
+static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left);
+
+/*
+ * Set the next IRQ period, based on the hwc->period_left value.
+ * To be called with the event disabled in hw:
+ */
+int x86_perf_event_set_period(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	s64 left = local64_read(&hwc->period_left);
+	s64 period = hwc->sample_period;
+	int ret = 0, idx = hwc->idx;
+
+	if (unlikely(!hwc->event_base))
+		return 0;
+
+	if (unlikely(is_topdown_count(event)) &&
+	    x86_pmu.set_topdown_event_period)
+		return x86_pmu.set_topdown_event_period(event);
+
+	/*
+	 * If we are way outside a reasonable range then just skip forward:
+	 */
+	if (unlikely(left <= -period)) {
+		left = period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+
+	if (unlikely(left <= 0)) {
+		left += period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+	/*
+	 * Quirk: certain CPUs dont like it if just 1 hw_event is left:
+	 */
+	if (unlikely(left < 2))
+		left = 2;
+
+	if (left > x86_pmu.max_period)
+		left = x86_pmu.max_period;
+
+	if (x86_pmu.limit_period)
+		left = x86_pmu.limit_period(event, left);
+
+	per_cpu(pmc_prev_left[idx], smp_processor_id()) = left;
+
+	/*
+	 * The hw event starts counting from this event offset,
+	 * mark it to be able to extra future deltas:
+	 */
+	local64_set(&hwc->prev_count, (u64)-left);
+
+	wrmsrl(hwc->event_base, (u64)(-left) & x86_pmu.cntval_mask);
+
+	/*
+	 * Sign extend the Merge event counter's upper 16 bits since
+	 * we currently declare a 48-bit counter width
+	 */
+	if (is_counter_pair(hwc))
+		wrmsrl(x86_pmu_event_addr(idx + 1), 0xffff);
+
+	/*
+	 * Due to erratum on certan cpu we need
+	 * a second write to be sure the register
+	 * is updated properly
+	 */
+	if (x86_pmu.perfctr_second_write) {
+		wrmsrl(hwc->event_base,
+			(u64)(-left) & x86_pmu.cntval_mask);
+	}
+
+	perf_event_update_userpage(event);
+
+	return ret;
+}
+
+void x86_pmu_enable_event(struct perf_event *event)
+{
+	if (__this_cpu_read(cpu_hw_events.enabled))
+		__x86_pmu_enable_event(&event->hw,
+				       ARCH_PERFMON_EVENTSEL_ENABLE);
+}
+
+/*
+ * Add a single event to the PMU.
+ *
+ * The event is added to the group of enabled events
+ * but only if it can be scheduled with existing events.
+ */
+static int x86_pmu_add(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc;
+	int assign[X86_PMC_IDX_MAX];
+	int n, n0, ret;
+
+	hwc = &event->hw;
+
+	n0 = cpuc->n_events;
+	ret = n = collect_events(cpuc, event, false);
+	if (ret < 0)
+		goto out;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
+
+	/*
+	 * If group events scheduling transaction was started,
+	 * skip the schedulability test here, it will be performed
+	 * at commit time (->commit_txn) as a whole.
+	 *
+	 * If commit fails, we'll call ->del() on all events
+	 * for which ->add() was called.
+	 */
+	if (cpuc->txn_flags & PERF_PMU_TXN_ADD)
+		goto done_collect;
+
+	ret = static_call(x86_pmu_schedule_events)(cpuc, n, assign);
+	if (ret)
+		goto out;
+	/*
+	 * copy new assignment, now we know it is possible
+	 * will be used by hw_perf_enable()
+	 */
+	memcpy(cpuc->assign, assign, n*sizeof(int));
+
+done_collect:
+	/*
+	 * Commit the collect_events() state. See x86_pmu_del() and
+	 * x86_pmu_*_txn().
+	 */
+	cpuc->n_events = n;
+	cpuc->n_added += n - n0;
+	cpuc->n_txn += n - n0;
+
+	/*
+	 * This is before x86_pmu_enable() will call x86_pmu_start(),
+	 * so we enable LBRs before an event needs them etc..
+	 */
+	static_call_cond(x86_pmu_add)(event);
+
+	ret = 0;
+out:
+	return ret;
+}
+
+static void x86_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx = event->hw.idx;
+
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		x86_perf_event_set_period(event);
+	}
+
+	event->hw.state = 0;
+
+	cpuc->events[idx] = event;
+	__set_bit(idx, cpuc->active_mask);
+	__set_bit(idx, cpuc->running);
+	static_call(x86_pmu_enable)(event);
+	perf_event_update_userpage(event);
+}
+
+void perf_event_print_debug(void)
+{
+	u64 ctrl, status, overflow, pmc_ctrl, pmc_count, prev_left, fixed;
+	u64 pebs, debugctl;
+	struct cpu_hw_events *cpuc;
+	unsigned long flags;
+	int cpu, idx;
+
+	if (!x86_pmu.num_counters)
+		return;
+
+	local_irq_save(flags);
+
+	cpu = smp_processor_id();
+	cpuc = &per_cpu(cpu_hw_events, cpu);
+
+	if (x86_pmu.version >= 2) {
+		rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, ctrl);
+		rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
+		rdmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, overflow);
+		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR_CTRL, fixed);
+
+		pr_info("\n");
+		pr_info("CPU#%d: ctrl:       %016llx\n", cpu, ctrl);
+		pr_info("CPU#%d: status:     %016llx\n", cpu, status);
+		pr_info("CPU#%d: overflow:   %016llx\n", cpu, overflow);
+		pr_info("CPU#%d: fixed:      %016llx\n", cpu, fixed);
+		if (x86_pmu.pebs_constraints) {
+			rdmsrl(MSR_IA32_PEBS_ENABLE, pebs);
+			pr_info("CPU#%d: pebs:       %016llx\n", cpu, pebs);
+		}
+		if (x86_pmu.lbr_nr) {
+			rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+			pr_info("CPU#%d: debugctl:   %016llx\n", cpu, debugctl);
+		}
+	}
+	pr_info("CPU#%d: active:     %016llx\n", cpu, *(u64 *)cpuc->active_mask);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		rdmsrl(x86_pmu_config_addr(idx), pmc_ctrl);
+		rdmsrl(x86_pmu_event_addr(idx), pmc_count);
+
+		prev_left = per_cpu(pmc_prev_left[idx], cpu);
+
+		pr_info("CPU#%d:   gen-PMC%d ctrl:  %016llx\n",
+			cpu, idx, pmc_ctrl);
+		pr_info("CPU#%d:   gen-PMC%d count: %016llx\n",
+			cpu, idx, pmc_count);
+		pr_info("CPU#%d:   gen-PMC%d left:  %016llx\n",
+			cpu, idx, prev_left);
+	}
+	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++) {
+		rdmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, pmc_count);
+
+		pr_info("CPU#%d: fixed-PMC%d count: %016llx\n",
+			cpu, idx, pmc_count);
+	}
+	local_irq_restore(flags);
+}
+
+void x86_pmu_stop(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (test_bit(hwc->idx, cpuc->active_mask)) {
+		static_call(x86_pmu_disable)(event);
+		__clear_bit(hwc->idx, cpuc->active_mask);
+		cpuc->events[hwc->idx] = NULL;
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		x86_perf_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+}
+
+static void x86_pmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int i;
+
+	/*
+	 * If we're called during a txn, we only need to undo x86_pmu.add.
+	 * The events never got scheduled and ->cancel_txn will truncate
+	 * the event_list.
+	 *
+	 * XXX assumes any ->del() called during a TXN will only be on
+	 * an event added during that same TXN.
+	 */
+	if (cpuc->txn_flags & PERF_PMU_TXN_ADD)
+		goto do_del;
+
+	/*
+	 * Not a TXN, therefore cleanup properly.
+	 */
+	x86_pmu_stop(event, PERF_EF_UPDATE);
+
+	for (i = 0; i < cpuc->n_events; i++) {
+		if (event == cpuc->event_list[i])
+			break;
+	}
+
+	if (WARN_ON_ONCE(i == cpuc->n_events)) /* called ->del() without ->add() ? */
+		return;
+
+	/* If we have a newly added event; make sure to decrease n_added. */
+	if (i >= cpuc->n_events - cpuc->n_added)
+		--cpuc->n_added;
+
+	static_call_cond(x86_pmu_put_event_constraints)(cpuc, event);
+
+	/* Delete the array entry. */
+	while (++i < cpuc->n_events) {
+		cpuc->event_list[i-1] = cpuc->event_list[i];
+		cpuc->event_constraint[i-1] = cpuc->event_constraint[i];
+	}
+	cpuc->event_constraint[i-1] = NULL;
+	--cpuc->n_events;
+	if (x86_pmu.intel_cap.perf_metrics)
+		del_nr_metric_event(cpuc, event);
+
+	perf_event_update_userpage(event);
+
+do_del:
+
+	/*
+	 * This is after x86_pmu_stop(); so we disable LBRs after any
+	 * event can need them etc..
+	 */
+	static_call_cond(x86_pmu_del)(event);
+}
+
+int x86_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc;
+	struct perf_event *event;
+	int idx, handled = 0;
+	u64 val;
+
+	cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * Some chipsets need to unmask the LVTPC in a particular spot
+	 * inside the nmi handler.  As a result, the unmasking was pushed
+	 * into all the nmi handlers.
+	 *
+	 * This generic handler doesn't seem to have any issues where the
+	 * unmasking occurs so it was left at the top.
+	 */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		event = cpuc->events[idx];
+
+		val = x86_perf_event_update(event);
+		if (val & (1ULL << (x86_pmu.cntval_bits - 1)))
+			continue;
+
+		/*
+		 * event overflow
+		 */
+		handled++;
+		perf_sample_data_init(&data, 0, event->hw.last_period);
+
+		if (!x86_perf_event_set_period(event))
+			continue;
+
+		if (perf_event_overflow(event, &data, regs))
+			x86_pmu_stop(event, 0);
+	}
+
+	if (handled)
+		inc_irq_stat(apic_perf_irqs);
+
+	return handled;
+}
+
+void perf_events_lapic_init(void)
+{
+	if (!x86_pmu.apic || !x86_pmu_initialized())
+		return;
+
+	/*
+	 * Always use NMI for PMU
+	 */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+}
+
+static int
+perf_event_nmi_handler(unsigned int cmd, struct pt_regs *regs)
+{
+	u64 start_clock;
+	u64 finish_clock;
+	int ret;
+
+	/*
+	 * All PMUs/events that share this PMI handler should make sure to
+	 * increment active_events for their events.
+	 */
+	if (!atomic_read(&active_events))
+		return NMI_DONE;
+
+	start_clock = sched_clock();
+	ret = static_call(x86_pmu_handle_irq)(regs);
+	finish_clock = sched_clock();
+
+	perf_sample_event_took(finish_clock - start_clock);
+
+	return ret;
+}
+NOKPROBE_SYMBOL(perf_event_nmi_handler);
+
+struct event_constraint emptyconstraint;
+struct event_constraint unconstrained;
+
+static int x86_pmu_prepare_cpu(unsigned int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+	int i;
+
+	for (i = 0 ; i < X86_PERF_KFREE_MAX; i++)
+		cpuc->kfree_on_online[i] = NULL;
+	if (x86_pmu.cpu_prepare)
+		return x86_pmu.cpu_prepare(cpu);
+	return 0;
+}
+
+static int x86_pmu_dead_cpu(unsigned int cpu)
+{
+	if (x86_pmu.cpu_dead)
+		x86_pmu.cpu_dead(cpu);
+	return 0;
+}
+
+static int x86_pmu_online_cpu(unsigned int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+	int i;
+
+	for (i = 0 ; i < X86_PERF_KFREE_MAX; i++) {
+		kfree(cpuc->kfree_on_online[i]);
+		cpuc->kfree_on_online[i] = NULL;
+	}
+	return 0;
+}
+
+static int x86_pmu_starting_cpu(unsigned int cpu)
+{
+	if (x86_pmu.cpu_starting)
+		x86_pmu.cpu_starting(cpu);
+	return 0;
+}
+
+static int x86_pmu_dying_cpu(unsigned int cpu)
+{
+	if (x86_pmu.cpu_dying)
+		x86_pmu.cpu_dying(cpu);
+	return 0;
+}
+
+static void __init pmu_check_apic(void)
+{
+	if (boot_cpu_has(X86_FEATURE_APIC))
+		return;
+
+	x86_pmu.apic = 0;
+	pr_info("no APIC, boot with the \"lapic\" boot parameter to force-enable it.\n");
+	pr_info("no hardware sampling interrupt available.\n");
+
+	/*
+	 * If we have a PMU initialized but no APIC
+	 * interrupts, we cannot sample hardware
+	 * events (user-space has to fall back and
+	 * sample via a hrtimer based software event):
+	 */
+	pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
+}
+
+static struct attribute_group x86_pmu_format_group __ro_after_init = {
+	.name = "format",
+	.attrs = NULL,
+};
+
+ssize_t events_sysfs_show(struct device *dev, struct device_attribute *attr, char *page)
+{
+	struct perf_pmu_events_attr *pmu_attr =
+		container_of(attr, struct perf_pmu_events_attr, attr);
+	u64 config = 0;
+
+	if (pmu_attr->id < x86_pmu.max_events)
+		config = x86_pmu.event_map(pmu_attr->id);
+
+	/* string trumps id */
+	if (pmu_attr->event_str)
+		return sprintf(page, "%s", pmu_attr->event_str);
+
+	return x86_pmu.events_sysfs_show(page, config);
+}
+EXPORT_SYMBOL_GPL(events_sysfs_show);
+
+ssize_t events_ht_sysfs_show(struct device *dev, struct device_attribute *attr,
+			  char *page)
+{
+	struct perf_pmu_events_ht_attr *pmu_attr =
+		container_of(attr, struct perf_pmu_events_ht_attr, attr);
+
+	/*
+	 * Report conditional events depending on Hyper-Threading.
+	 *
+	 * This is overly conservative as usually the HT special
+	 * handling is not needed if the other CPU thread is idle.
+	 *
+	 * Note this does not (and cannot) handle the case when thread
+	 * siblings are invisible, for example with virtualization
+	 * if they are owned by some other guest.  The user tool
+	 * has to re-read when a thread sibling gets onlined later.
+	 */
+	return sprintf(page, "%s",
+			topology_max_smt_threads() > 1 ?
+			pmu_attr->event_str_ht :
+			pmu_attr->event_str_noht);
+}
+
+EVENT_ATTR(cpu-cycles,			CPU_CYCLES		);
+EVENT_ATTR(instructions,		INSTRUCTIONS		);
+EVENT_ATTR(cache-references,		CACHE_REFERENCES	);
+EVENT_ATTR(cache-misses, 		CACHE_MISSES		);
+EVENT_ATTR(branch-instructions,		BRANCH_INSTRUCTIONS	);
+EVENT_ATTR(branch-misses,		BRANCH_MISSES		);
+EVENT_ATTR(bus-cycles,			BUS_CYCLES		);
+EVENT_ATTR(stalled-cycles-frontend,	STALLED_CYCLES_FRONTEND	);
+EVENT_ATTR(stalled-cycles-backend,	STALLED_CYCLES_BACKEND	);
+EVENT_ATTR(ref-cycles,			REF_CPU_CYCLES		);
+
+static struct attribute *empty_attrs;
+
+static struct attribute *events_attr[] = {
+	EVENT_PTR(CPU_CYCLES),
+	EVENT_PTR(INSTRUCTIONS),
+	EVENT_PTR(CACHE_REFERENCES),
+	EVENT_PTR(CACHE_MISSES),
+	EVENT_PTR(BRANCH_INSTRUCTIONS),
+	EVENT_PTR(BRANCH_MISSES),
+	EVENT_PTR(BUS_CYCLES),
+	EVENT_PTR(STALLED_CYCLES_FRONTEND),
+	EVENT_PTR(STALLED_CYCLES_BACKEND),
+	EVENT_PTR(REF_CPU_CYCLES),
+	NULL,
+};
+
+/*
+ * Remove all undefined events (x86_pmu.event_map(id) == 0)
+ * out of events_attr attributes.
+ */
+static umode_t
+is_visible(struct kobject *kobj, struct attribute *attr, int idx)
+{
+	struct perf_pmu_events_attr *pmu_attr;
+
+	if (idx >= x86_pmu.max_events)
+		return 0;
+
+	pmu_attr = container_of(attr, struct perf_pmu_events_attr, attr.attr);
+	/* str trumps id */
+	return pmu_attr->event_str || x86_pmu.event_map(idx) ? attr->mode : 0;
+}
+
+static struct attribute_group x86_pmu_events_group __ro_after_init = {
+	.name = "events",
+	.attrs = events_attr,
+	.is_visible = is_visible,
+};
+
+ssize_t x86_event_sysfs_show(char *page, u64 config, u64 event)
+{
+	u64 umask  = (config & ARCH_PERFMON_EVENTSEL_UMASK) >> 8;
+	u64 cmask  = (config & ARCH_PERFMON_EVENTSEL_CMASK) >> 24;
+	bool edge  = (config & ARCH_PERFMON_EVENTSEL_EDGE);
+	bool pc    = (config & ARCH_PERFMON_EVENTSEL_PIN_CONTROL);
+	bool any   = (config & ARCH_PERFMON_EVENTSEL_ANY);
+	bool inv   = (config & ARCH_PERFMON_EVENTSEL_INV);
+	ssize_t ret;
+
+	/*
+	* We have whole page size to spend and just little data
+	* to write, so we can safely use sprintf.
+	*/
+	ret = sprintf(page, "event=0x%02llx", event);
+
+	if (umask)
+		ret += sprintf(page + ret, ",umask=0x%02llx", umask);
+
+	if (edge)
+		ret += sprintf(page + ret, ",edge");
+
+	if (pc)
+		ret += sprintf(page + ret, ",pc");
+
+	if (any)
+		ret += sprintf(page + ret, ",any");
+
+	if (inv)
+		ret += sprintf(page + ret, ",inv");
+
+	if (cmask)
+		ret += sprintf(page + ret, ",cmask=0x%02llx", cmask);
+
+	ret += sprintf(page + ret, "\n");
+
+	return ret;
+}
+
+static struct attribute_group x86_pmu_attr_group;
+static struct attribute_group x86_pmu_caps_group;
+
+static void x86_pmu_static_call_update(void)
+{
+	static_call_update(x86_pmu_handle_irq, x86_pmu.handle_irq);
+	static_call_update(x86_pmu_disable_all, x86_pmu.disable_all);
+	static_call_update(x86_pmu_enable_all, x86_pmu.enable_all);
+	static_call_update(x86_pmu_enable, x86_pmu.enable);
+	static_call_update(x86_pmu_disable, x86_pmu.disable);
+
+	static_call_update(x86_pmu_add, x86_pmu.add);
+	static_call_update(x86_pmu_del, x86_pmu.del);
+	static_call_update(x86_pmu_read, x86_pmu.read);
+
+	static_call_update(x86_pmu_schedule_events, x86_pmu.schedule_events);
+	static_call_update(x86_pmu_get_event_constraints, x86_pmu.get_event_constraints);
+	static_call_update(x86_pmu_put_event_constraints, x86_pmu.put_event_constraints);
+
+	static_call_update(x86_pmu_start_scheduling, x86_pmu.start_scheduling);
+	static_call_update(x86_pmu_commit_scheduling, x86_pmu.commit_scheduling);
+	static_call_update(x86_pmu_stop_scheduling, x86_pmu.stop_scheduling);
+
+	static_call_update(x86_pmu_sched_task, x86_pmu.sched_task);
+	static_call_update(x86_pmu_swap_task_ctx, x86_pmu.swap_task_ctx);
+
+	static_call_update(x86_pmu_drain_pebs, x86_pmu.drain_pebs);
+	static_call_update(x86_pmu_pebs_aliases, x86_pmu.pebs_aliases);
+}
+
+static void _x86_pmu_read(struct perf_event *event)
+{
+	x86_perf_event_update(event);
+}
+
+static int __init init_hw_perf_events(void)
+{
+	struct x86_pmu_quirk *quirk;
+	int err;
+
+	pr_info("Performance Events: ");
+
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_INTEL:
+		err = intel_pmu_init();
+		break;
+	case X86_VENDOR_AMD:
+		err = amd_pmu_init();
+		break;
+	case X86_VENDOR_HYGON:
+		err = amd_pmu_init();
+		x86_pmu.name = "HYGON";
+		break;
+	case X86_VENDOR_ZHAOXIN:
+	case X86_VENDOR_CENTAUR:
+		err = zhaoxin_pmu_init();
+		break;
+	default:
+		err = -ENOTSUPP;
+	}
+	if (err != 0) {
+		pr_cont("no PMU driver, software events only.\n");
+		return 0;
+	}
+
+	pmu_check_apic();
+
+	/* sanity check that the hardware exists or is emulated */
+	if (!check_hw_exists())
+		return 0;
+
+	pr_cont("%s PMU driver.\n", x86_pmu.name);
+
+	x86_pmu.attr_rdpmc = 1; /* enable userspace RDPMC usage by default */
+
+	for (quirk = x86_pmu.quirks; quirk; quirk = quirk->next)
+		quirk->func();
+
+	if (!x86_pmu.intel_ctrl)
+		x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
+
+	perf_events_lapic_init();
+	register_nmi_handler(NMI_LOCAL, perf_event_nmi_handler, 0, "PMI");
+
+	unconstrained = (struct event_constraint)
+		__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1,
+				   0, x86_pmu.num_counters, 0, 0);
+
+	x86_pmu_format_group.attrs = x86_pmu.format_attrs;
+
+	if (!x86_pmu.events_sysfs_show)
+		x86_pmu_events_group.attrs = &empty_attrs;
+
+	pmu.attr_update = x86_pmu.attr_update;
+
+	pr_info("... version:                %d\n",     x86_pmu.version);
+	pr_info("... bit width:              %d\n",     x86_pmu.cntval_bits);
+	pr_info("... generic registers:      %d\n",     x86_pmu.num_counters);
+	pr_info("... value mask:             %016Lx\n", x86_pmu.cntval_mask);
+	pr_info("... max period:             %016Lx\n", x86_pmu.max_period);
+	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_counters_fixed);
+	pr_info("... event mask:             %016Lx\n", x86_pmu.intel_ctrl);
+
+	if (!x86_pmu.read)
+		x86_pmu.read = _x86_pmu_read;
+
+	x86_pmu_static_call_update();
+
+	/*
+	 * Install callbacks. Core will call them for each online
+	 * cpu.
+	 */
+	err = cpuhp_setup_state(CPUHP_PERF_X86_PREPARE, "perf/x86:prepare",
+				x86_pmu_prepare_cpu, x86_pmu_dead_cpu);
+	if (err)
+		return err;
+
+	err = cpuhp_setup_state(CPUHP_AP_PERF_X86_STARTING,
+				"perf/x86:starting", x86_pmu_starting_cpu,
+				x86_pmu_dying_cpu);
+	if (err)
+		goto out;
+
+	err = cpuhp_setup_state(CPUHP_AP_PERF_X86_ONLINE, "perf/x86:online",
+				x86_pmu_online_cpu, NULL);
+	if (err)
+		goto out1;
+
+	err = perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW);
+	if (err)
+		goto out2;
+
+	return 0;
+
+out2:
+	cpuhp_remove_state(CPUHP_AP_PERF_X86_ONLINE);
+out1:
+	cpuhp_remove_state(CPUHP_AP_PERF_X86_STARTING);
+out:
+	cpuhp_remove_state(CPUHP_PERF_X86_PREPARE);
+	return err;
+}
+early_initcall(init_hw_perf_events);
+
+static void x86_pmu_read(struct perf_event *event)
+{
+	static_call(x86_pmu_read)(event);
+}
+
+/*
+ * Start group events scheduling transaction
+ * Set the flag to make pmu::enable() not perform the
+ * schedulability test, it will be performed at commit time
+ *
+ * We only support PERF_PMU_TXN_ADD transactions. Save the
+ * transaction flags but otherwise ignore non-PERF_PMU_TXN_ADD
+ * transactions.
+ */
+static void x86_pmu_start_txn(struct pmu *pmu, unsigned int txn_flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	WARN_ON_ONCE(cpuc->txn_flags);		/* txn already in flight */
+
+	cpuc->txn_flags = txn_flags;
+	if (txn_flags & ~PERF_PMU_TXN_ADD)
+		return;
+
+	perf_pmu_disable(pmu);
+	__this_cpu_write(cpu_hw_events.n_txn, 0);
+	__this_cpu_write(cpu_hw_events.n_txn_pair, 0);
+	__this_cpu_write(cpu_hw_events.n_txn_metric, 0);
+}
+
+/*
+ * Stop group events scheduling transaction
+ * Clear the flag and pmu::enable() will perform the
+ * schedulability test.
+ */
+static void x86_pmu_cancel_txn(struct pmu *pmu)
+{
+	unsigned int txn_flags;
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	WARN_ON_ONCE(!cpuc->txn_flags);	/* no txn in flight */
+
+	txn_flags = cpuc->txn_flags;
+	cpuc->txn_flags = 0;
+	if (txn_flags & ~PERF_PMU_TXN_ADD)
+		return;
+
+	/*
+	 * Truncate collected array by the number of events added in this
+	 * transaction. See x86_pmu_add() and x86_pmu_*_txn().
+	 */
+	__this_cpu_sub(cpu_hw_events.n_added, __this_cpu_read(cpu_hw_events.n_txn));
+	__this_cpu_sub(cpu_hw_events.n_events, __this_cpu_read(cpu_hw_events.n_txn));
+	__this_cpu_sub(cpu_hw_events.n_pair, __this_cpu_read(cpu_hw_events.n_txn_pair));
+	__this_cpu_sub(cpu_hw_events.n_metric, __this_cpu_read(cpu_hw_events.n_txn_metric));
+	perf_pmu_enable(pmu);
+}
+
+/*
+ * Commit group events scheduling transaction
+ * Perform the group schedulability test as a whole
+ * Return 0 if success
+ *
+ * Does not cancel the transaction on failure; expects the caller to do this.
+ */
+static int x86_pmu_commit_txn(struct pmu *pmu)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int assign[X86_PMC_IDX_MAX];
+	int n, ret;
+
+	WARN_ON_ONCE(!cpuc->txn_flags);	/* no txn in flight */
+
+	if (cpuc->txn_flags & ~PERF_PMU_TXN_ADD) {
+		cpuc->txn_flags = 0;
+		return 0;
+	}
+
+	n = cpuc->n_events;
+
+	if (!x86_pmu_initialized())
+		return -EAGAIN;
+
+	ret = static_call(x86_pmu_schedule_events)(cpuc, n, assign);
+	if (ret)
+		return ret;
+
+	/*
+	 * copy new assignment, now we know it is possible
+	 * will be used by hw_perf_enable()
+	 */
+	memcpy(cpuc->assign, assign, n*sizeof(int));
+
+	cpuc->txn_flags = 0;
+	perf_pmu_enable(pmu);
+	return 0;
+}
+/*
+ * a fake_cpuc is used to validate event groups. Due to
+ * the extra reg logic, we need to also allocate a fake
+ * per_core and per_cpu structure. Otherwise, group events
+ * using extra reg may conflict without the kernel being
+ * able to catch this when the last event gets added to
+ * the group.
+ */
+static void free_fake_cpuc(struct cpu_hw_events *cpuc)
+{
+	intel_cpuc_finish(cpuc);
+	kfree(cpuc);
+}
+
+static struct cpu_hw_events *allocate_fake_cpuc(void)
+{
+	struct cpu_hw_events *cpuc;
+	int cpu = raw_smp_processor_id();
+
+	cpuc = kzalloc(sizeof(*cpuc), GFP_KERNEL);
+	if (!cpuc)
+		return ERR_PTR(-ENOMEM);
+	cpuc->is_fake = 1;
+
+	if (intel_cpuc_prepare(cpuc, cpu))
+		goto error;
+
+	return cpuc;
+error:
+	free_fake_cpuc(cpuc);
+	return ERR_PTR(-ENOMEM);
+}
+
+/*
+ * validate that we can schedule this event
+ */
+static int validate_event(struct perf_event *event)
+{
+	struct cpu_hw_events *fake_cpuc;
+	struct event_constraint *c;
+	int ret = 0;
+
+	fake_cpuc = allocate_fake_cpuc();
+	if (IS_ERR(fake_cpuc))
+		return PTR_ERR(fake_cpuc);
+
+	c = x86_pmu.get_event_constraints(fake_cpuc, 0, event);
+
+	if (!c || !c->weight)
+		ret = -EINVAL;
+
+	if (x86_pmu.put_event_constraints)
+		x86_pmu.put_event_constraints(fake_cpuc, event);
+
+	free_fake_cpuc(fake_cpuc);
+
+	return ret;
+}
+
+/*
+ * validate a single event group
+ *
+ * validation include:
+ *	- check events are compatible which each other
+ *	- events do not compete for the same counter
+ *	- number of events <= number of counters
+ *
+ * validation ensures the group can be loaded onto the
+ * PMU if it was the only group available.
+ */
+static int validate_group(struct perf_event *event)
+{
+	struct perf_event *leader = event->group_leader;
+	struct cpu_hw_events *fake_cpuc;
+	int ret = -EINVAL, n;
+
+	fake_cpuc = allocate_fake_cpuc();
+	if (IS_ERR(fake_cpuc))
+		return PTR_ERR(fake_cpuc);
+	/*
+	 * the event is not yet connected with its
+	 * siblings therefore we must first collect
+	 * existing siblings, then add the new event
+	 * before we can simulate the scheduling
+	 */
+	n = collect_events(fake_cpuc, leader, true);
+	if (n < 0)
+		goto out;
+
+	fake_cpuc->n_events = n;
+	n = collect_events(fake_cpuc, event, false);
+	if (n < 0)
+		goto out;
+
+	fake_cpuc->n_events = 0;
+	ret = x86_pmu.schedule_events(fake_cpuc, n, NULL);
+
+out:
+	free_fake_cpuc(fake_cpuc);
+	return ret;
+}
+
+static int x86_pmu_event_init(struct perf_event *event)
+{
+	struct pmu *tmp;
+	int err;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
+	err = __x86_pmu_event_init(event);
+	if (!err) {
+		/*
+		 * we temporarily connect event to its pmu
+		 * such that validate_group() can classify
+		 * it as an x86 event using is_x86_event()
+		 */
+		tmp = event->pmu;
+		event->pmu = &pmu;
+
+		if (event->group_leader != event)
+			err = validate_group(event);
+		else
+			err = validate_event(event);
+
+		event->pmu = tmp;
+	}
+	if (err) {
+		if (event->destroy)
+			event->destroy(event);
+		event->destroy = NULL;
+	}
+
+	if (READ_ONCE(x86_pmu.attr_rdpmc) &&
+	    !(event->hw.flags & PERF_X86_EVENT_LARGE_PEBS))
+		event->hw.flags |= PERF_X86_EVENT_RDPMC_ALLOWED;
+
+	return err;
+}
+
+static void x86_pmu_event_mapped(struct perf_event *event, struct mm_struct *mm)
+{
+	if (!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+		return;
+
+	/*
+	 * This function relies on not being called concurrently in two
+	 * tasks in the same mm.  Otherwise one task could observe
+	 * perf_rdpmc_allowed > 1 and return all the way back to
+	 * userspace with CR4.PCE clear while another task is still
+	 * doing on_each_cpu_mask() to propagate CR4.PCE.
+	 *
+	 * For now, this can't happen because all callers hold mmap_lock
+	 * for write.  If this changes, we'll need a different solution.
+	 */
+	mmap_assert_write_locked(mm);
+
+	if (atomic_inc_return(&mm->context.perf_rdpmc_allowed) == 1)
+		on_each_cpu_mask(mm_cpumask(mm), cr4_update_pce, NULL, 1);
+}
+
+static void x86_pmu_event_unmapped(struct perf_event *event, struct mm_struct *mm)
+{
+
+	if (!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+		return;
+
+	if (atomic_dec_and_test(&mm->context.perf_rdpmc_allowed))
+		on_each_cpu_mask(mm_cpumask(mm), cr4_update_pce, NULL, 1);
+}
+
+static int x86_pmu_event_idx(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!(hwc->flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+		return 0;
+
+	if (is_metric_idx(hwc->idx))
+		return INTEL_PMC_FIXED_RDPMC_METRICS + 1;
+	else
+		return hwc->event_base_rdpmc + 1;
+}
+
+static ssize_t get_attr_rdpmc(struct device *cdev,
+			      struct device_attribute *attr,
+			      char *buf)
+{
+	return snprintf(buf, 40, "%d\n", x86_pmu.attr_rdpmc);
+}
+
+static ssize_t set_attr_rdpmc(struct device *cdev,
+			      struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	unsigned long val;
+	ssize_t ret;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val > 2)
+		return -EINVAL;
+
+	if (x86_pmu.attr_rdpmc_broken)
+		return -ENOTSUPP;
+
+	if (val != x86_pmu.attr_rdpmc) {
+		/*
+		 * Changing into or out of never available or always available,
+		 * aka perf-event-bypassing mode. This path is extremely slow,
+		 * but only root can trigger it, so it's okay.
+		 */
+		if (val == 0)
+			static_branch_inc(&rdpmc_never_available_key);
+		else if (x86_pmu.attr_rdpmc == 0)
+			static_branch_dec(&rdpmc_never_available_key);
+
+		if (val == 2)
+			static_branch_inc(&rdpmc_always_available_key);
+		else if (x86_pmu.attr_rdpmc == 2)
+			static_branch_dec(&rdpmc_always_available_key);
+
+		on_each_cpu(cr4_update_pce, NULL, 1);
+		x86_pmu.attr_rdpmc = val;
+	}
+
+	return count;
+}
+
+static DEVICE_ATTR(rdpmc, S_IRUSR | S_IWUSR, get_attr_rdpmc, set_attr_rdpmc);
+
+static struct attribute *x86_pmu_attrs[] = {
+	&dev_attr_rdpmc.attr,
+	NULL,
+};
+
+static struct attribute_group x86_pmu_attr_group __ro_after_init = {
+	.attrs = x86_pmu_attrs,
+};
+
+static ssize_t max_precise_show(struct device *cdev,
+				  struct device_attribute *attr,
+				  char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", x86_pmu_max_precise());
+}
+
+static DEVICE_ATTR_RO(max_precise);
+
+static struct attribute *x86_pmu_caps_attrs[] = {
+	&dev_attr_max_precise.attr,
+	NULL
+};
+
+static struct attribute_group x86_pmu_caps_group __ro_after_init = {
+	.name = "caps",
+	.attrs = x86_pmu_caps_attrs,
+};
+
+static const struct attribute_group *x86_pmu_attr_groups[] = {
+	&x86_pmu_attr_group,
+	&x86_pmu_format_group,
+	&x86_pmu_events_group,
+	&x86_pmu_caps_group,
+	NULL,
+};
+
+static void x86_pmu_sched_task(struct perf_event_context *ctx, bool sched_in)
+{
+	static_call_cond(x86_pmu_sched_task)(ctx, sched_in);
+}
+
+static void x86_pmu_swap_task_ctx(struct perf_event_context *prev,
+				  struct perf_event_context *next)
+{
+	static_call_cond(x86_pmu_swap_task_ctx)(prev, next);
+}
+
+void perf_check_microcode(void)
+{
+	if (x86_pmu.check_microcode)
+		x86_pmu.check_microcode();
+}
+
+static int x86_pmu_check_period(struct perf_event *event, u64 value)
+{
+	if (x86_pmu.check_period && x86_pmu.check_period(event, value))
+		return -EINVAL;
+
+	if (value && x86_pmu.limit_period) {
+		if (x86_pmu.limit_period(event, value) > value)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int x86_pmu_aux_output_match(struct perf_event *event)
+{
+	if (!(pmu.capabilities & PERF_PMU_CAP_AUX_OUTPUT))
+		return 0;
+
+	if (x86_pmu.aux_output_match)
+		return x86_pmu.aux_output_match(event);
+
+	return 0;
+}
+
+static struct pmu pmu = {
+	.pmu_enable		= x86_pmu_enable,
+	.pmu_disable		= x86_pmu_disable,
+
+	.attr_groups		= x86_pmu_attr_groups,
+
+	.event_init		= x86_pmu_event_init,
+
+	.event_mapped		= x86_pmu_event_mapped,
+	.event_unmapped		= x86_pmu_event_unmapped,
+
+	.add			= x86_pmu_add,
+	.del			= x86_pmu_del,
+	.start			= x86_pmu_start,
+	.stop			= x86_pmu_stop,
+	.read			= x86_pmu_read,
+
+	.start_txn		= x86_pmu_start_txn,
+	.cancel_txn		= x86_pmu_cancel_txn,
+	.commit_txn		= x86_pmu_commit_txn,
+
+	.event_idx		= x86_pmu_event_idx,
+	.sched_task		= x86_pmu_sched_task,
+	.swap_task_ctx		= x86_pmu_swap_task_ctx,
+	.check_period		= x86_pmu_check_period,
+
+	.aux_output_match	= x86_pmu_aux_output_match,
+};
+
+void arch_perf_update_userpage(struct perf_event *event,
+			       struct perf_event_mmap_page *userpg, u64 now)
+{
+	struct cyc2ns_data data;
+	u64 offset;
+
+	userpg->cap_user_time = 0;
+	userpg->cap_user_time_zero = 0;
+	userpg->cap_user_rdpmc =
+		!!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED);
+	userpg->pmc_width = x86_pmu.cntval_bits;
+
+	if (!using_native_sched_clock() || !sched_clock_stable())
+		return;
+
+	cyc2ns_read_begin(&data);
+
+	offset = data.cyc2ns_offset + __sched_clock_offset;
+
+	/*
+	 * Internal timekeeping for enabled/running/stopped times
+	 * is always in the local_clock domain.
+	 */
+	userpg->cap_user_time = 1;
+	userpg->time_mult = data.cyc2ns_mul;
+	userpg->time_shift = data.cyc2ns_shift;
+	userpg->time_offset = offset - now;
+
+	/*
+	 * cap_user_time_zero doesn't make sense when we're using a different
+	 * time base for the records.
+	 */
+	if (!event->attr.use_clockid) {
+		userpg->cap_user_time_zero = 1;
+		userpg->time_zero = offset;
+	}
+
+	cyc2ns_read_end();
+}
+
+/*
+ * Determine whether the regs were taken from an irq/exception handler rather
+ * than from perf_arch_fetch_caller_regs().
+ */
+static bool perf_hw_regs(struct pt_regs *regs)
+{
+	return regs->flags & X86_EFLAGS_FIXED;
+}
+
+void
+perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
+{
+	struct perf_guest_info_callbacks *guest_cbs = perf_get_guest_cbs();
+	struct unwind_state state;
+	unsigned long addr;
+
+	if (guest_cbs && guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
+	if (perf_callchain_store(entry, regs->ip))
+		return;
+
+	if (perf_hw_regs(regs))
+		unwind_start(&state, current, regs, NULL);
+	else
+		unwind_start(&state, current, NULL, (void *)regs->sp);
+
+	for (; !unwind_done(&state); unwind_next_frame(&state)) {
+		addr = unwind_get_return_address(&state);
+		if (!addr || perf_callchain_store(entry, addr))
+			return;
+	}
+}
+
+static inline int
+valid_user_frame(const void __user *fp, unsigned long size)
+{
+	return (__range_not_ok(fp, size, TASK_SIZE) == 0);
+}
+
+static unsigned long get_segment_base(unsigned int segment)
+{
+	struct desc_struct *desc;
+	unsigned int idx = segment >> 3;
+
+	if ((segment & SEGMENT_TI_MASK) == SEGMENT_LDT) {
+#ifdef CONFIG_MODIFY_LDT_SYSCALL
+		struct ldt_struct *ldt;
+
+		/* IRQs are off, so this synchronizes with smp_store_release */
+		ldt = READ_ONCE(current->active_mm->context.ldt);
+		if (!ldt || idx >= ldt->nr_entries)
+			return 0;
+
+		desc = &ldt->entries[idx];
+#else
+		return 0;
+#endif
+	} else {
+		if (idx >= GDT_ENTRIES)
+			return 0;
+
+		desc = raw_cpu_ptr(gdt_page.gdt) + idx;
+	}
+
+	return get_desc_base(desc);
+}
+
+#ifdef CONFIG_IA32_EMULATION
+
+#include <linux/compat.h>
+
+static inline int
+perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry_ctx *entry)
+{
+	/* 32-bit process in 64-bit kernel. */
+	unsigned long ss_base, cs_base;
+	struct stack_frame_ia32 frame;
+	const struct stack_frame_ia32 __user *fp;
+
+	if (!test_thread_flag(TIF_IA32))
+		return 0;
+
+	cs_base = get_segment_base(regs->cs);
+	ss_base = get_segment_base(regs->ss);
+
+	fp = compat_ptr(ss_base + regs->bp);
+	pagefault_disable();
+	while (entry->nr < entry->max_stack) {
+		if (!valid_user_frame(fp, sizeof(frame)))
+			break;
+
+		if (__get_user(frame.next_frame, &fp->next_frame))
+			break;
+		if (__get_user(frame.return_address, &fp->return_address))
+			break;
+
+		perf_callchain_store(entry, cs_base + frame.return_address);
+		fp = compat_ptr(ss_base + frame.next_frame);
+	}
+	pagefault_enable();
+	return 1;
+}
+#else
+static inline int
+perf_callchain_user32(struct pt_regs *regs, struct perf_callchain_entry_ctx *entry)
+{
+    return 0;
+}
+#endif
+
+void
+perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
+{
+	struct perf_guest_info_callbacks *guest_cbs = perf_get_guest_cbs();
+	struct stack_frame frame;
+	const struct stack_frame __user *fp;
+
+	if (guest_cbs && guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
+	/*
+	 * We don't know what to do with VM86 stacks.. ignore them for now.
+	 */
+	if (regs->flags & (X86_VM_MASK | PERF_EFLAGS_VM))
+		return;
+
+	fp = (void __user *)regs->bp;
+
+	perf_callchain_store(entry, regs->ip);
+
+	if (!nmi_uaccess_okay())
+		return;
+
+	if (perf_callchain_user32(regs, entry))
+		return;
+
+	pagefault_disable();
+	while (entry->nr < entry->max_stack) {
+		if (!valid_user_frame(fp, sizeof(frame)))
+			break;
+
+		if (__get_user(frame.next_frame, &fp->next_frame))
+			break;
+		if (__get_user(frame.return_address, &fp->return_address))
+			break;
+
+		perf_callchain_store(entry, frame.return_address);
+		fp = (void __user *)frame.next_frame;
+	}
+	pagefault_enable();
+}
+
+/*
+ * Deal with code segment offsets for the various execution modes:
+ *
+ *   VM86 - the good olde 16 bit days, where the linear address is
+ *          20 bits and we use regs->ip + 0x10 * regs->cs.
+ *
+ *   IA32 - Where we need to look at GDT/LDT segment descriptor tables
+ *          to figure out what the 32bit base address is.
+ *
+ *    X32 - has TIF_X32 set, but is running in x86_64
+ *
+ * X86_64 - CS,DS,SS,ES are all zero based.
+ */
+static unsigned long code_segment_base(struct pt_regs *regs)
+{
+	/*
+	 * For IA32 we look at the GDT/LDT segment base to convert the
+	 * effective IP to a linear address.
+	 */
+
+#ifdef CONFIG_X86_32
+	/*
+	 * If we are in VM86 mode, add the segment offset to convert to a
+	 * linear address.
+	 */
+	if (regs->flags & X86_VM_MASK)
+		return 0x10 * regs->cs;
+
+	if (user_mode(regs) && regs->cs != __USER_CS)
+		return get_segment_base(regs->cs);
+#else
+	if (user_mode(regs) && !user_64bit_mode(regs) &&
+	    regs->cs != __USER32_CS)
+		return get_segment_base(regs->cs);
+#endif
+	return 0;
+}
+
+unsigned long perf_instruction_pointer(struct pt_regs *regs)
+{
+	struct perf_guest_info_callbacks *guest_cbs = perf_get_guest_cbs();
+
+	if (guest_cbs && guest_cbs->is_in_guest())
+		return guest_cbs->get_guest_ip();
+
+	return regs->ip + code_segment_base(regs);
+}
+
+unsigned long perf_misc_flags(struct pt_regs *regs)
+{
+	struct perf_guest_info_callbacks *guest_cbs = perf_get_guest_cbs();
+	int misc = 0;
+
+	if (guest_cbs && guest_cbs->is_in_guest()) {
+		if (guest_cbs->is_user_mode())
+			misc |= PERF_RECORD_MISC_GUEST_USER;
+		else
+			misc |= PERF_RECORD_MISC_GUEST_KERNEL;
+	} else {
+		if (user_mode(regs))
+			misc |= PERF_RECORD_MISC_USER;
+		else
+			misc |= PERF_RECORD_MISC_KERNEL;
+	}
+
+	if (regs->flags & PERF_EFLAGS_EXACT)
+		misc |= PERF_RECORD_MISC_EXACT_IP;
+
+	return misc;
+}
+
+void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap)
+{
+	cap->version		= x86_pmu.version;
+	cap->num_counters_gp	= x86_pmu.num_counters;
+	cap->num_counters_fixed	= x86_pmu.num_counters_fixed;
+	cap->bit_width_gp	= x86_pmu.cntval_bits;
+	cap->bit_width_fixed	= x86_pmu.cntval_bits;
+	cap->events_mask	= (unsigned int)x86_pmu.events_maskl;
+	cap->events_mask_len	= x86_pmu.events_mask_len;
+}
+EXPORT_SYMBOL_GPL(perf_get_x86_pmu_capability);
diff --git a/arch/x86/events/intel/Makefile b/arch/x86/events/intel/Makefile
new file mode 100644
index 000000000..e67a58863
--- /dev/null
+++ b/arch/x86/events/intel/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_CPU_SUP_INTEL)		+= core.o bts.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= ds.o knc.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= lbr.o p4.o p6.o pt.o
+obj-$(CONFIG_PERF_EVENTS_INTEL_UNCORE)	+= intel-uncore.o
+intel-uncore-objs			:= uncore.o uncore_nhmex.o uncore_snb.o uncore_snbep.o
+obj-$(CONFIG_PERF_EVENTS_INTEL_CSTATE)	+= intel-cstate.o
+intel-cstate-objs			:= cstate.o
diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c
new file mode 100644
index 000000000..731dd8d0d
--- /dev/null
+++ b/arch/x86/events/intel/bts.c
@@ -0,0 +1,619 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * BTS PMU driver for perf
+ * Copyright (c) 2013-2014, Intel Corporation.
+ */
+
+#undef DEBUG
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/bitops.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/debugfs.h>
+#include <linux/device.h>
+#include <linux/coredump.h>
+
+#include <linux/sizes.h>
+#include <asm/perf_event.h>
+
+#include "../perf_event.h"
+
+struct bts_ctx {
+	struct perf_output_handle	handle;
+	struct debug_store		ds_back;
+	int				state;
+};
+
+/* BTS context states: */
+enum {
+	/* no ongoing AUX transactions */
+	BTS_STATE_STOPPED = 0,
+	/* AUX transaction is on, BTS tracing is disabled */
+	BTS_STATE_INACTIVE,
+	/* AUX transaction is on, BTS tracing is running */
+	BTS_STATE_ACTIVE,
+};
+
+static DEFINE_PER_CPU(struct bts_ctx, bts_ctx);
+
+#define BTS_RECORD_SIZE		24
+#define BTS_SAFETY_MARGIN	4080
+
+struct bts_phys {
+	struct page	*page;
+	unsigned long	size;
+	unsigned long	offset;
+	unsigned long	displacement;
+};
+
+struct bts_buffer {
+	size_t		real_size;	/* multiple of BTS_RECORD_SIZE */
+	unsigned int	nr_pages;
+	unsigned int	nr_bufs;
+	unsigned int	cur_buf;
+	bool		snapshot;
+	local_t		data_size;
+	local_t		head;
+	unsigned long	end;
+	void		**data_pages;
+	struct bts_phys	buf[];
+};
+
+static struct pmu bts_pmu;
+
+static int buf_nr_pages(struct page *page)
+{
+	if (!PagePrivate(page))
+		return 1;
+
+	return 1 << page_private(page);
+}
+
+static size_t buf_size(struct page *page)
+{
+	return buf_nr_pages(page) * PAGE_SIZE;
+}
+
+static void *
+bts_buffer_setup_aux(struct perf_event *event, void **pages,
+		     int nr_pages, bool overwrite)
+{
+	struct bts_buffer *buf;
+	struct page *page;
+	int cpu = event->cpu;
+	int node = (cpu == -1) ? cpu : cpu_to_node(cpu);
+	unsigned long offset;
+	size_t size = nr_pages << PAGE_SHIFT;
+	int pg, nbuf, pad;
+
+	/* count all the high order buffers */
+	for (pg = 0, nbuf = 0; pg < nr_pages;) {
+		page = virt_to_page(pages[pg]);
+		pg += buf_nr_pages(page);
+		nbuf++;
+	}
+
+	/*
+	 * to avoid interrupts in overwrite mode, only allow one physical
+	 */
+	if (overwrite && nbuf > 1)
+		return NULL;
+
+	buf = kzalloc_node(offsetof(struct bts_buffer, buf[nbuf]), GFP_KERNEL, node);
+	if (!buf)
+		return NULL;
+
+	buf->nr_pages = nr_pages;
+	buf->nr_bufs = nbuf;
+	buf->snapshot = overwrite;
+	buf->data_pages = pages;
+	buf->real_size = size - size % BTS_RECORD_SIZE;
+
+	for (pg = 0, nbuf = 0, offset = 0, pad = 0; nbuf < buf->nr_bufs; nbuf++) {
+		unsigned int __nr_pages;
+
+		page = virt_to_page(pages[pg]);
+		__nr_pages = buf_nr_pages(page);
+		buf->buf[nbuf].page = page;
+		buf->buf[nbuf].offset = offset;
+		buf->buf[nbuf].displacement = (pad ? BTS_RECORD_SIZE - pad : 0);
+		buf->buf[nbuf].size = buf_size(page) - buf->buf[nbuf].displacement;
+		pad = buf->buf[nbuf].size % BTS_RECORD_SIZE;
+		buf->buf[nbuf].size -= pad;
+
+		pg += __nr_pages;
+		offset += __nr_pages << PAGE_SHIFT;
+	}
+
+	return buf;
+}
+
+static void bts_buffer_free_aux(void *data)
+{
+	kfree(data);
+}
+
+static unsigned long bts_buffer_offset(struct bts_buffer *buf, unsigned int idx)
+{
+	return buf->buf[idx].offset + buf->buf[idx].displacement;
+}
+
+static void
+bts_config_buffer(struct bts_buffer *buf)
+{
+	int cpu = raw_smp_processor_id();
+	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
+	struct bts_phys *phys = &buf->buf[buf->cur_buf];
+	unsigned long index, thresh = 0, end = phys->size;
+	struct page *page = phys->page;
+
+	index = local_read(&buf->head);
+
+	if (!buf->snapshot) {
+		if (buf->end < phys->offset + buf_size(page))
+			end = buf->end - phys->offset - phys->displacement;
+
+		index -= phys->offset + phys->displacement;
+
+		if (end - index > BTS_SAFETY_MARGIN)
+			thresh = end - BTS_SAFETY_MARGIN;
+		else if (end - index > BTS_RECORD_SIZE)
+			thresh = end - BTS_RECORD_SIZE;
+		else
+			thresh = end;
+	}
+
+	ds->bts_buffer_base = (u64)(long)page_address(page) + phys->displacement;
+	ds->bts_index = ds->bts_buffer_base + index;
+	ds->bts_absolute_maximum = ds->bts_buffer_base + end;
+	ds->bts_interrupt_threshold = !buf->snapshot
+		? ds->bts_buffer_base + thresh
+		: ds->bts_absolute_maximum + BTS_RECORD_SIZE;
+}
+
+static void bts_buffer_pad_out(struct bts_phys *phys, unsigned long head)
+{
+	unsigned long index = head - phys->offset;
+
+	memset(page_address(phys->page) + index, 0, phys->size - index);
+}
+
+static void bts_update(struct bts_ctx *bts)
+{
+	int cpu = raw_smp_processor_id();
+	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
+	struct bts_buffer *buf = perf_get_aux(&bts->handle);
+	unsigned long index = ds->bts_index - ds->bts_buffer_base, old, head;
+
+	if (!buf)
+		return;
+
+	head = index + bts_buffer_offset(buf, buf->cur_buf);
+	old = local_xchg(&buf->head, head);
+
+	if (!buf->snapshot) {
+		if (old == head)
+			return;
+
+		if (ds->bts_index >= ds->bts_absolute_maximum)
+			perf_aux_output_flag(&bts->handle,
+			                     PERF_AUX_FLAG_TRUNCATED);
+
+		/*
+		 * old and head are always in the same physical buffer, so we
+		 * can subtract them to get the data size.
+		 */
+		local_add(head - old, &buf->data_size);
+	} else {
+		local_set(&buf->data_size, head);
+	}
+}
+
+static int
+bts_buffer_reset(struct bts_buffer *buf, struct perf_output_handle *handle);
+
+/*
+ * Ordering PMU callbacks wrt themselves and the PMI is done by means
+ * of bts::state, which:
+ *  - is set when bts::handle::event is valid, that is, between
+ *    perf_aux_output_begin() and perf_aux_output_end();
+ *  - is zero otherwise;
+ *  - is ordered against bts::handle::event with a compiler barrier.
+ */
+
+static void __bts_event_start(struct perf_event *event)
+{
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	struct bts_buffer *buf = perf_get_aux(&bts->handle);
+	u64 config = 0;
+
+	if (!buf->snapshot)
+		config |= ARCH_PERFMON_EVENTSEL_INT;
+	if (!event->attr.exclude_kernel)
+		config |= ARCH_PERFMON_EVENTSEL_OS;
+	if (!event->attr.exclude_user)
+		config |= ARCH_PERFMON_EVENTSEL_USR;
+
+	bts_config_buffer(buf);
+
+	/*
+	 * local barrier to make sure that ds configuration made it
+	 * before we enable BTS and bts::state goes ACTIVE
+	 */
+	wmb();
+
+	/* INACTIVE/STOPPED -> ACTIVE */
+	WRITE_ONCE(bts->state, BTS_STATE_ACTIVE);
+
+	intel_pmu_enable_bts(config);
+
+}
+
+static void bts_event_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	struct bts_buffer *buf;
+
+	buf = perf_aux_output_begin(&bts->handle, event);
+	if (!buf)
+		goto fail_stop;
+
+	if (bts_buffer_reset(buf, &bts->handle))
+		goto fail_end_stop;
+
+	bts->ds_back.bts_buffer_base = cpuc->ds->bts_buffer_base;
+	bts->ds_back.bts_absolute_maximum = cpuc->ds->bts_absolute_maximum;
+	bts->ds_back.bts_interrupt_threshold = cpuc->ds->bts_interrupt_threshold;
+
+	perf_event_itrace_started(event);
+	event->hw.state = 0;
+
+	__bts_event_start(event);
+
+	return;
+
+fail_end_stop:
+	perf_aux_output_end(&bts->handle, 0);
+
+fail_stop:
+	event->hw.state = PERF_HES_STOPPED;
+}
+
+static void __bts_event_stop(struct perf_event *event, int state)
+{
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+
+	/* ACTIVE -> INACTIVE(PMI)/STOPPED(->stop()) */
+	WRITE_ONCE(bts->state, state);
+
+	/*
+	 * No extra synchronization is mandated by the documentation to have
+	 * BTS data stores globally visible.
+	 */
+	intel_pmu_disable_bts();
+}
+
+static void bts_event_stop(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	struct bts_buffer *buf = NULL;
+	int state = READ_ONCE(bts->state);
+
+	if (state == BTS_STATE_ACTIVE)
+		__bts_event_stop(event, BTS_STATE_STOPPED);
+
+	if (state != BTS_STATE_STOPPED)
+		buf = perf_get_aux(&bts->handle);
+
+	event->hw.state |= PERF_HES_STOPPED;
+
+	if (flags & PERF_EF_UPDATE) {
+		bts_update(bts);
+
+		if (buf) {
+			if (buf->snapshot)
+				bts->handle.head =
+					local_xchg(&buf->data_size,
+						   buf->nr_pages << PAGE_SHIFT);
+			perf_aux_output_end(&bts->handle,
+			                    local_xchg(&buf->data_size, 0));
+		}
+
+		cpuc->ds->bts_index = bts->ds_back.bts_buffer_base;
+		cpuc->ds->bts_buffer_base = bts->ds_back.bts_buffer_base;
+		cpuc->ds->bts_absolute_maximum = bts->ds_back.bts_absolute_maximum;
+		cpuc->ds->bts_interrupt_threshold = bts->ds_back.bts_interrupt_threshold;
+	}
+}
+
+void intel_bts_enable_local(void)
+{
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	int state = READ_ONCE(bts->state);
+
+	/*
+	 * Here we transition from INACTIVE to ACTIVE;
+	 * if we instead are STOPPED from the interrupt handler,
+	 * stay that way. Can't be ACTIVE here though.
+	 */
+	if (WARN_ON_ONCE(state == BTS_STATE_ACTIVE))
+		return;
+
+	if (state == BTS_STATE_STOPPED)
+		return;
+
+	if (bts->handle.event)
+		__bts_event_start(bts->handle.event);
+}
+
+void intel_bts_disable_local(void)
+{
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+
+	/*
+	 * Here we transition from ACTIVE to INACTIVE;
+	 * do nothing for STOPPED or INACTIVE.
+	 */
+	if (READ_ONCE(bts->state) != BTS_STATE_ACTIVE)
+		return;
+
+	if (bts->handle.event)
+		__bts_event_stop(bts->handle.event, BTS_STATE_INACTIVE);
+}
+
+static int
+bts_buffer_reset(struct bts_buffer *buf, struct perf_output_handle *handle)
+{
+	unsigned long head, space, next_space, pad, gap, skip, wakeup;
+	unsigned int next_buf;
+	struct bts_phys *phys, *next_phys;
+	int ret;
+
+	if (buf->snapshot)
+		return 0;
+
+	head = handle->head & ((buf->nr_pages << PAGE_SHIFT) - 1);
+
+	phys = &buf->buf[buf->cur_buf];
+	space = phys->offset + phys->displacement + phys->size - head;
+	pad = space;
+	if (space > handle->size) {
+		space = handle->size;
+		space -= space % BTS_RECORD_SIZE;
+	}
+	if (space <= BTS_SAFETY_MARGIN) {
+		/* See if next phys buffer has more space */
+		next_buf = buf->cur_buf + 1;
+		if (next_buf >= buf->nr_bufs)
+			next_buf = 0;
+		next_phys = &buf->buf[next_buf];
+		gap = buf_size(phys->page) - phys->displacement - phys->size +
+		      next_phys->displacement;
+		skip = pad + gap;
+		if (handle->size >= skip) {
+			next_space = next_phys->size;
+			if (next_space + skip > handle->size) {
+				next_space = handle->size - skip;
+				next_space -= next_space % BTS_RECORD_SIZE;
+			}
+			if (next_space > space || !space) {
+				if (pad)
+					bts_buffer_pad_out(phys, head);
+				ret = perf_aux_output_skip(handle, skip);
+				if (ret)
+					return ret;
+				/* Advance to next phys buffer */
+				phys = next_phys;
+				space = next_space;
+				head = phys->offset + phys->displacement;
+				/*
+				 * After this, cur_buf and head won't match ds
+				 * anymore, so we must not be racing with
+				 * bts_update().
+				 */
+				buf->cur_buf = next_buf;
+				local_set(&buf->head, head);
+			}
+		}
+	}
+
+	/* Don't go far beyond wakeup watermark */
+	wakeup = BTS_SAFETY_MARGIN + BTS_RECORD_SIZE + handle->wakeup -
+		 handle->head;
+	if (space > wakeup) {
+		space = wakeup;
+		space -= space % BTS_RECORD_SIZE;
+	}
+
+	buf->end = head + space;
+
+	/*
+	 * If we have no space, the lost notification would have been sent when
+	 * we hit absolute_maximum - see bts_update()
+	 */
+	if (!space)
+		return -ENOSPC;
+
+	return 0;
+}
+
+int intel_bts_interrupt(void)
+{
+	struct debug_store *ds = this_cpu_ptr(&cpu_hw_events)->ds;
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	struct perf_event *event = bts->handle.event;
+	struct bts_buffer *buf;
+	s64 old_head;
+	int err = -ENOSPC, handled = 0;
+
+	/*
+	 * The only surefire way of knowing if this NMI is ours is by checking
+	 * the write ptr against the PMI threshold.
+	 */
+	if (ds && (ds->bts_index >= ds->bts_interrupt_threshold))
+		handled = 1;
+
+	/*
+	 * this is wrapped in intel_bts_enable_local/intel_bts_disable_local,
+	 * so we can only be INACTIVE or STOPPED
+	 */
+	if (READ_ONCE(bts->state) == BTS_STATE_STOPPED)
+		return handled;
+
+	buf = perf_get_aux(&bts->handle);
+	if (!buf)
+		return handled;
+
+	/*
+	 * Skip snapshot counters: they don't use the interrupt, but
+	 * there's no other way of telling, because the pointer will
+	 * keep moving
+	 */
+	if (buf->snapshot)
+		return 0;
+
+	old_head = local_read(&buf->head);
+	bts_update(bts);
+
+	/* no new data */
+	if (old_head == local_read(&buf->head))
+		return handled;
+
+	perf_aux_output_end(&bts->handle, local_xchg(&buf->data_size, 0));
+
+	buf = perf_aux_output_begin(&bts->handle, event);
+	if (buf)
+		err = bts_buffer_reset(buf, &bts->handle);
+
+	if (err) {
+		WRITE_ONCE(bts->state, BTS_STATE_STOPPED);
+
+		if (buf) {
+			/*
+			 * BTS_STATE_STOPPED should be visible before
+			 * cleared handle::event
+			 */
+			barrier();
+			perf_aux_output_end(&bts->handle, 0);
+		}
+	}
+
+	return 1;
+}
+
+static void bts_event_del(struct perf_event *event, int mode)
+{
+	bts_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int bts_event_add(struct perf_event *event, int mode)
+{
+	struct bts_ctx *bts = this_cpu_ptr(&bts_ctx);
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+
+	event->hw.state = PERF_HES_STOPPED;
+
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask))
+		return -EBUSY;
+
+	if (bts->handle.event)
+		return -EBUSY;
+
+	if (mode & PERF_EF_START) {
+		bts_event_start(event, 0);
+		if (hwc->state & PERF_HES_STOPPED)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void bts_event_destroy(struct perf_event *event)
+{
+	x86_release_hardware();
+	x86_del_exclusive(x86_lbr_exclusive_bts);
+}
+
+static int bts_event_init(struct perf_event *event)
+{
+	int ret;
+
+	if (event->attr.type != bts_pmu.type)
+		return -ENOENT;
+
+	/*
+	 * BTS leaks kernel addresses even when CPL0 tracing is
+	 * disabled, so disallow intel_bts driver for unprivileged
+	 * users on paranoid systems since it provides trace data
+	 * to the user in a zero-copy fashion.
+	 *
+	 * Note that the default paranoia setting permits unprivileged
+	 * users to profile the kernel.
+	 */
+	if (event->attr.exclude_kernel) {
+		ret = perf_allow_kernel(&event->attr);
+		if (ret)
+			return ret;
+	}
+
+	if (x86_add_exclusive(x86_lbr_exclusive_bts))
+		return -EBUSY;
+
+	ret = x86_reserve_hardware();
+	if (ret) {
+		x86_del_exclusive(x86_lbr_exclusive_bts);
+		return ret;
+	}
+
+	event->destroy = bts_event_destroy;
+
+	return 0;
+}
+
+static void bts_event_read(struct perf_event *event)
+{
+}
+
+static __init int bts_init(void)
+{
+	if (!boot_cpu_has(X86_FEATURE_DTES64) || !x86_pmu.bts)
+		return -ENODEV;
+
+	if (boot_cpu_has(X86_FEATURE_PTI)) {
+		/*
+		 * BTS hardware writes through a virtual memory map we must
+		 * either use the kernel physical map, or the user mapping of
+		 * the AUX buffer.
+		 *
+		 * However, since this driver supports per-CPU and per-task inherit
+		 * we cannot use the user mapping since it will not be available
+		 * if we're not running the owning process.
+		 *
+		 * With PTI we can't use the kernal map either, because its not
+		 * there when we run userspace.
+		 *
+		 * For now, disable this driver when using PTI.
+		 */
+		return -ENODEV;
+	}
+
+	bts_pmu.capabilities	= PERF_PMU_CAP_AUX_NO_SG | PERF_PMU_CAP_ITRACE |
+				  PERF_PMU_CAP_EXCLUSIVE;
+	bts_pmu.task_ctx_nr	= perf_sw_context;
+	bts_pmu.event_init	= bts_event_init;
+	bts_pmu.add		= bts_event_add;
+	bts_pmu.del		= bts_event_del;
+	bts_pmu.start		= bts_event_start;
+	bts_pmu.stop		= bts_event_stop;
+	bts_pmu.read		= bts_event_read;
+	bts_pmu.setup_aux	= bts_buffer_setup_aux;
+	bts_pmu.free_aux	= bts_buffer_free_aux;
+
+	return perf_pmu_register(&bts_pmu, "intel_bts", -1);
+}
+arch_initcall(bts_init);
diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
new file mode 100644
index 000000000..990d5543e
--- /dev/null
+++ b/arch/x86/events/intel/core.c
@@ -0,0 +1,5670 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Per core/cpu state
+ *
+ * Used to coordinate shared registers between HT threads or
+ * among events on a single PMU.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <linux/nmi.h>
+
+#include <asm/cpufeature.h>
+#include <asm/hardirq.h>
+#include <asm/intel-family.h>
+#include <asm/intel_pt.h>
+#include <asm/apic.h>
+#include <asm/cpu_device_id.h>
+
+#include "../perf_event.h"
+
+/*
+ * Intel PerfMon, used on Core and later.
+ */
+static u64 intel_perfmon_event_map[PERF_COUNT_HW_MAX] __read_mostly =
+{
+	[PERF_COUNT_HW_CPU_CYCLES]		= 0x003c,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= 0x4f2e,
+	[PERF_COUNT_HW_CACHE_MISSES]		= 0x412e,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
+	[PERF_COUNT_HW_BUS_CYCLES]		= 0x013c,
+	[PERF_COUNT_HW_REF_CPU_CYCLES]		= 0x0300, /* pseudo-encoding */
+};
+
+static struct event_constraint intel_core_event_constraints[] __read_mostly =
+{
+	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 */
+	INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */
+	INTEL_EVENT_CONSTRAINT(0xc1, 0x1), /* FP_COMP_INSTR_RET */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_core2_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */
+	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 */
+	INTEL_EVENT_CONSTRAINT(0x18, 0x1), /* IDLE_DURING_DIV */
+	INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */
+	INTEL_EVENT_CONSTRAINT(0xa1, 0x1), /* RS_UOPS_DISPATCH_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0xc9, 0x1), /* ITLB_MISS_RETIRED (T30-9) */
+	INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_nehalem_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_EVENT_CONSTRAINT(0x40, 0x3), /* L1D_CACHE_LD */
+	INTEL_EVENT_CONSTRAINT(0x41, 0x3), /* L1D_CACHE_ST */
+	INTEL_EVENT_CONSTRAINT(0x42, 0x3), /* L1D_CACHE_LOCK */
+	INTEL_EVENT_CONSTRAINT(0x43, 0x3), /* L1D_ALL_REF */
+	INTEL_EVENT_CONSTRAINT(0x48, 0x3), /* L1D_PEND_MISS */
+	INTEL_EVENT_CONSTRAINT(0x4e, 0x3), /* L1D_PREFETCH */
+	INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */
+	INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg intel_nehalem_extra_regs[] __read_mostly =
+{
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0xffff, RSP_0),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x100b),
+	EVENT_EXTRA_END
+};
+
+static struct event_constraint intel_westmere_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */
+	INTEL_EVENT_CONSTRAINT(0x60, 0x1), /* OFFCORE_REQUESTS_OUTSTANDING */
+	INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */
+	INTEL_EVENT_CONSTRAINT(0xb3, 0x1), /* SNOOPQ_REQUEST_OUTSTANDING */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_snb_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_NO_DISPATCH */
+	INTEL_UEVENT_CONSTRAINT(0x05a3, 0xf), /* CYCLE_ACTIVITY.STALLS_L2_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x02a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x06a3, 0x4), /* CYCLE_ACTIVITY.STALLS_L1D_PENDING */
+	INTEL_EVENT_CONSTRAINT(0x48, 0x4), /* L1D_PEND_MISS.PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */
+	INTEL_EVENT_CONSTRAINT(0xcd, 0x8), /* MEM_TRANS_RETIRED.LOAD_LATENCY */
+	INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_NO_DISPATCH */
+	INTEL_UEVENT_CONSTRAINT(0x02a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
+
+	/*
+	 * When HT is off these events can only run on the bottom 4 counters
+	 * When HT is on, they are impacted by the HT bug and require EXCL access
+	 */
+	INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_ivb_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_UEVENT_CONSTRAINT(0x0148, 0x4), /* L1D_PEND_MISS.PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x0279, 0xf), /* IDQ.EMTPY */
+	INTEL_UEVENT_CONSTRAINT(0x019c, 0xf), /* IDQ_UOPS_NOT_DELIVERED.CORE */
+	INTEL_UEVENT_CONSTRAINT(0x02a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_LDM_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_NO_EXECUTE */
+	INTEL_UEVENT_CONSTRAINT(0x05a3, 0xf), /* CYCLE_ACTIVITY.STALLS_L2_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x06a3, 0xf), /* CYCLE_ACTIVITY.STALLS_LDM_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x08a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4), /* CYCLE_ACTIVITY.STALLS_L1D_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */
+
+	/*
+	 * When HT is off these events can only run on the bottom 4 counters
+	 * When HT is on, they are impacted by the HT bug and require EXCL access
+	 */
+	INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg intel_westmere_extra_regs[] __read_mostly =
+{
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0xffff, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0xffff, RSP_1),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x100b),
+	EVENT_EXTRA_END
+};
+
+static struct event_constraint intel_v1_event_constraints[] __read_mostly =
+{
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_gen_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_slm_event_constraints[] __read_mostly =
+{
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* pseudo CPU_CLK_UNHALTED.REF */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_skl_event_constraints[] = {
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0),	/* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1),	/* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2),	/* CPU_CLK_UNHALTED.REF */
+	INTEL_UEVENT_CONSTRAINT(0x1c0, 0x2),	/* INST_RETIRED.PREC_DIST */
+
+	/*
+	 * when HT is off, these can only run on the bottom 4 counters
+	 */
+	INTEL_EVENT_CONSTRAINT(0xd0, 0xf),	/* MEM_INST_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xd1, 0xf),	/* MEM_LOAD_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xd2, 0xf),	/* MEM_LOAD_L3_HIT_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xcd, 0xf),	/* MEM_TRANS_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xc6, 0xf),	/* FRONTEND_RETIRED.* */
+
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg intel_knl_extra_regs[] __read_mostly = {
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x799ffbb6e7ull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0x399ffbffe7ull, RSP_1),
+	EVENT_EXTRA_END
+};
+
+static struct extra_reg intel_snb_extra_regs[] __read_mostly = {
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3f807f8fffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3f807f8fffull, RSP_1),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd),
+	EVENT_EXTRA_END
+};
+
+static struct extra_reg intel_snbep_extra_regs[] __read_mostly = {
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd),
+	EVENT_EXTRA_END
+};
+
+static struct extra_reg intel_skl_extra_regs[] __read_mostly = {
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd),
+	/*
+	 * Note the low 8 bits eventsel code is not a continuous field, containing
+	 * some #GPing bits. These are masked out.
+	 */
+	INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x7fff17, FE),
+	EVENT_EXTRA_END
+};
+
+static struct event_constraint intel_icl_event_constraints[] = {
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0),	/* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x01c0, 0),	/* INST_RETIRED.PREC_DIST */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1),	/* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2),	/* CPU_CLK_UNHALTED.REF */
+	FIXED_EVENT_CONSTRAINT(0x0400, 3),	/* SLOTS */
+	METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_RETIRING, 0),
+	METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_BAD_SPEC, 1),
+	METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_FE_BOUND, 2),
+	METRIC_EVENT_CONSTRAINT(INTEL_TD_METRIC_BE_BOUND, 3),
+	INTEL_EVENT_CONSTRAINT_RANGE(0x03, 0x0a, 0xf),
+	INTEL_EVENT_CONSTRAINT_RANGE(0x1f, 0x28, 0xf),
+	INTEL_EVENT_CONSTRAINT(0x32, 0xf),	/* SW_PREFETCH_ACCESS.* */
+	INTEL_EVENT_CONSTRAINT_RANGE(0x48, 0x56, 0xf),
+	INTEL_EVENT_CONSTRAINT_RANGE(0x60, 0x8b, 0xf),
+	INTEL_UEVENT_CONSTRAINT(0x04a3, 0xff),  /* CYCLE_ACTIVITY.STALLS_TOTAL */
+	INTEL_UEVENT_CONSTRAINT(0x10a3, 0xff),  /* CYCLE_ACTIVITY.CYCLES_MEM_ANY */
+	INTEL_UEVENT_CONSTRAINT(0x14a3, 0xff),  /* CYCLE_ACTIVITY.STALLS_MEM_ANY */
+	INTEL_EVENT_CONSTRAINT(0xa3, 0xf),      /* CYCLE_ACTIVITY.* */
+	INTEL_EVENT_CONSTRAINT_RANGE(0xa8, 0xb0, 0xf),
+	INTEL_EVENT_CONSTRAINT_RANGE(0xb7, 0xbd, 0xf),
+	INTEL_EVENT_CONSTRAINT_RANGE(0xd0, 0xe6, 0xf),
+	INTEL_EVENT_CONSTRAINT(0xef, 0xf),
+	INTEL_EVENT_CONSTRAINT_RANGE(0xf0, 0xf4, 0xf),
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg intel_icl_extra_regs[] __read_mostly = {
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffffbfffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffffbfffull, RSP_1),
+	INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd),
+	INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x7fff17, FE),
+	EVENT_EXTRA_END
+};
+
+EVENT_ATTR_STR(mem-loads,	mem_ld_nhm,	"event=0x0b,umask=0x10,ldlat=3");
+EVENT_ATTR_STR(mem-loads,	mem_ld_snb,	"event=0xcd,umask=0x1,ldlat=3");
+EVENT_ATTR_STR(mem-stores,	mem_st_snb,	"event=0xcd,umask=0x2");
+
+static struct attribute *nhm_mem_events_attrs[] = {
+	EVENT_PTR(mem_ld_nhm),
+	NULL,
+};
+
+/*
+ * topdown events for Intel Core CPUs.
+ *
+ * The events are all in slots, which is a free slot in a 4 wide
+ * pipeline. Some events are already reported in slots, for cycle
+ * events we multiply by the pipeline width (4).
+ *
+ * With Hyper Threading on, topdown metrics are either summed or averaged
+ * between the threads of a core: (count_t0 + count_t1).
+ *
+ * For the average case the metric is always scaled to pipeline width,
+ * so we use factor 2 ((count_t0 + count_t1) / 2 * 4)
+ */
+
+EVENT_ATTR_STR_HT(topdown-total-slots, td_total_slots,
+	"event=0x3c,umask=0x0",			/* cpu_clk_unhalted.thread */
+	"event=0x3c,umask=0x0,any=1");		/* cpu_clk_unhalted.thread_any */
+EVENT_ATTR_STR_HT(topdown-total-slots.scale, td_total_slots_scale, "4", "2");
+EVENT_ATTR_STR(topdown-slots-issued, td_slots_issued,
+	"event=0xe,umask=0x1");			/* uops_issued.any */
+EVENT_ATTR_STR(topdown-slots-retired, td_slots_retired,
+	"event=0xc2,umask=0x2");		/* uops_retired.retire_slots */
+EVENT_ATTR_STR(topdown-fetch-bubbles, td_fetch_bubbles,
+	"event=0x9c,umask=0x1");		/* idq_uops_not_delivered_core */
+EVENT_ATTR_STR_HT(topdown-recovery-bubbles, td_recovery_bubbles,
+	"event=0xd,umask=0x3,cmask=1",		/* int_misc.recovery_cycles */
+	"event=0xd,umask=0x3,cmask=1,any=1");	/* int_misc.recovery_cycles_any */
+EVENT_ATTR_STR_HT(topdown-recovery-bubbles.scale, td_recovery_bubbles_scale,
+	"4", "2");
+
+EVENT_ATTR_STR(slots,			slots,		"event=0x00,umask=0x4");
+EVENT_ATTR_STR(topdown-retiring,	td_retiring,	"event=0x00,umask=0x80");
+EVENT_ATTR_STR(topdown-bad-spec,	td_bad_spec,	"event=0x00,umask=0x81");
+EVENT_ATTR_STR(topdown-fe-bound,	td_fe_bound,	"event=0x00,umask=0x82");
+EVENT_ATTR_STR(topdown-be-bound,	td_be_bound,	"event=0x00,umask=0x83");
+
+static struct attribute *snb_events_attrs[] = {
+	EVENT_PTR(td_slots_issued),
+	EVENT_PTR(td_slots_retired),
+	EVENT_PTR(td_fetch_bubbles),
+	EVENT_PTR(td_total_slots),
+	EVENT_PTR(td_total_slots_scale),
+	EVENT_PTR(td_recovery_bubbles),
+	EVENT_PTR(td_recovery_bubbles_scale),
+	NULL,
+};
+
+static struct attribute *snb_mem_events_attrs[] = {
+	EVENT_PTR(mem_ld_snb),
+	EVENT_PTR(mem_st_snb),
+	NULL,
+};
+
+static struct event_constraint intel_hsw_event_constraints[] = {
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */
+	INTEL_UEVENT_CONSTRAINT(0x148, 0x4),	/* L1D_PEND_MISS.PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */
+	INTEL_EVENT_CONSTRAINT(0xcd, 0x8), /* MEM_TRANS_RETIRED.LOAD_LATENCY */
+	/* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x08a3, 0x4),
+	/* CYCLE_ACTIVITY.STALLS_L1D_PENDING */
+	INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4),
+	/* CYCLE_ACTIVITY.CYCLES_NO_EXECUTE */
+	INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf),
+
+	/*
+	 * When HT is off these events can only run on the bottom 4 counters
+	 * When HT is on, they are impacted by the HT bug and require EXCL access
+	 */
+	INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint intel_bdw_event_constraints[] = {
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0),	/* INST_RETIRED.ANY */
+	FIXED_EVENT_CONSTRAINT(0x003c, 1),	/* CPU_CLK_UNHALTED.CORE */
+	FIXED_EVENT_CONSTRAINT(0x0300, 2),	/* CPU_CLK_UNHALTED.REF */
+	INTEL_UEVENT_CONSTRAINT(0x148, 0x4),	/* L1D_PEND_MISS.PENDING */
+	INTEL_UBIT_EVENT_CONSTRAINT(0x8a3, 0x4),	/* CYCLE_ACTIVITY.CYCLES_L1D_MISS */
+	/*
+	 * when HT is off, these can only run on the bottom 4 counters
+	 */
+	INTEL_EVENT_CONSTRAINT(0xd0, 0xf),	/* MEM_INST_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xd1, 0xf),	/* MEM_LOAD_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xd2, 0xf),	/* MEM_LOAD_L3_HIT_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xcd, 0xf),	/* MEM_TRANS_RETIRED.* */
+	EVENT_CONSTRAINT_END
+};
+
+static u64 intel_pmu_event_map(int hw_event)
+{
+	return intel_perfmon_event_map[hw_event];
+}
+
+/*
+ * Notes on the events:
+ * - data reads do not include code reads (comparable to earlier tables)
+ * - data counts include speculative execution (except L1 write, dtlb, bpu)
+ * - remote node access includes remote memory, remote cache, remote mmio.
+ * - prefetches are not included in the counts.
+ * - icache miss does not include decoded icache
+ */
+
+#define SKL_DEMAND_DATA_RD		BIT_ULL(0)
+#define SKL_DEMAND_RFO			BIT_ULL(1)
+#define SKL_ANY_RESPONSE		BIT_ULL(16)
+#define SKL_SUPPLIER_NONE		BIT_ULL(17)
+#define SKL_L3_MISS_LOCAL_DRAM		BIT_ULL(26)
+#define SKL_L3_MISS_REMOTE_HOP0_DRAM	BIT_ULL(27)
+#define SKL_L3_MISS_REMOTE_HOP1_DRAM	BIT_ULL(28)
+#define SKL_L3_MISS_REMOTE_HOP2P_DRAM	BIT_ULL(29)
+#define SKL_L3_MISS			(SKL_L3_MISS_LOCAL_DRAM| \
+					 SKL_L3_MISS_REMOTE_HOP0_DRAM| \
+					 SKL_L3_MISS_REMOTE_HOP1_DRAM| \
+					 SKL_L3_MISS_REMOTE_HOP2P_DRAM)
+#define SKL_SPL_HIT			BIT_ULL(30)
+#define SKL_SNOOP_NONE			BIT_ULL(31)
+#define SKL_SNOOP_NOT_NEEDED		BIT_ULL(32)
+#define SKL_SNOOP_MISS			BIT_ULL(33)
+#define SKL_SNOOP_HIT_NO_FWD		BIT_ULL(34)
+#define SKL_SNOOP_HIT_WITH_FWD		BIT_ULL(35)
+#define SKL_SNOOP_HITM			BIT_ULL(36)
+#define SKL_SNOOP_NON_DRAM		BIT_ULL(37)
+#define SKL_ANY_SNOOP			(SKL_SPL_HIT|SKL_SNOOP_NONE| \
+					 SKL_SNOOP_NOT_NEEDED|SKL_SNOOP_MISS| \
+					 SKL_SNOOP_HIT_NO_FWD|SKL_SNOOP_HIT_WITH_FWD| \
+					 SKL_SNOOP_HITM|SKL_SNOOP_NON_DRAM)
+#define SKL_DEMAND_READ			SKL_DEMAND_DATA_RD
+#define SKL_SNOOP_DRAM			(SKL_SNOOP_NONE| \
+					 SKL_SNOOP_NOT_NEEDED|SKL_SNOOP_MISS| \
+					 SKL_SNOOP_HIT_NO_FWD|SKL_SNOOP_HIT_WITH_FWD| \
+					 SKL_SNOOP_HITM|SKL_SPL_HIT)
+#define SKL_DEMAND_WRITE		SKL_DEMAND_RFO
+#define SKL_LLC_ACCESS			SKL_ANY_RESPONSE
+#define SKL_L3_MISS_REMOTE		(SKL_L3_MISS_REMOTE_HOP0_DRAM| \
+					 SKL_L3_MISS_REMOTE_HOP1_DRAM| \
+					 SKL_L3_MISS_REMOTE_HOP2P_DRAM)
+
+static __initconst const u64 skl_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) ] = 0x81d0,	/* MEM_INST_RETIRED.ALL_LOADS */
+		[ C(RESULT_MISS)   ] = 0x151,	/* L1D.REPLACEMENT */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x82d0,	/* MEM_INST_RETIRED.ALL_STORES */
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x283,	/* ICACHE_64B.MISS */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x81d0,	/* MEM_INST_RETIRED.ALL_LOADS */
+		[ C(RESULT_MISS)   ] = 0xe08,	/* DTLB_LOAD_MISSES.WALK_COMPLETED */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x82d0,	/* MEM_INST_RETIRED.ALL_STORES */
+		[ C(RESULT_MISS)   ] = 0xe49,	/* DTLB_STORE_MISSES.WALK_COMPLETED */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2085,	/* ITLB_MISSES.STLB_HIT */
+		[ C(RESULT_MISS)   ] = 0xe85,	/* ITLB_MISSES.WALK_COMPLETED */
+	},
+	[ 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) ] = 0xc4,	/* BR_INST_RETIRED.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0xc5,	/* BR_MISP_RETIRED.ALL_BRANCHES */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+};
+
+static __initconst const u64 skl_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = SKL_DEMAND_READ|
+				       SKL_LLC_ACCESS|SKL_ANY_SNOOP,
+		[ C(RESULT_MISS)   ] = SKL_DEMAND_READ|
+				       SKL_L3_MISS|SKL_ANY_SNOOP|
+				       SKL_SUPPLIER_NONE,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = SKL_DEMAND_WRITE|
+				       SKL_LLC_ACCESS|SKL_ANY_SNOOP,
+		[ C(RESULT_MISS)   ] = SKL_DEMAND_WRITE|
+				       SKL_L3_MISS|SKL_ANY_SNOOP|
+				       SKL_SUPPLIER_NONE,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = SKL_DEMAND_READ|
+				       SKL_L3_MISS_LOCAL_DRAM|SKL_SNOOP_DRAM,
+		[ C(RESULT_MISS)   ] = SKL_DEMAND_READ|
+				       SKL_L3_MISS_REMOTE|SKL_SNOOP_DRAM,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = SKL_DEMAND_WRITE|
+				       SKL_L3_MISS_LOCAL_DRAM|SKL_SNOOP_DRAM,
+		[ C(RESULT_MISS)   ] = SKL_DEMAND_WRITE|
+				       SKL_L3_MISS_REMOTE|SKL_SNOOP_DRAM,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+};
+
+#define SNB_DMND_DATA_RD	(1ULL << 0)
+#define SNB_DMND_RFO		(1ULL << 1)
+#define SNB_DMND_IFETCH		(1ULL << 2)
+#define SNB_DMND_WB		(1ULL << 3)
+#define SNB_PF_DATA_RD		(1ULL << 4)
+#define SNB_PF_RFO		(1ULL << 5)
+#define SNB_PF_IFETCH		(1ULL << 6)
+#define SNB_LLC_DATA_RD		(1ULL << 7)
+#define SNB_LLC_RFO		(1ULL << 8)
+#define SNB_LLC_IFETCH		(1ULL << 9)
+#define SNB_BUS_LOCKS		(1ULL << 10)
+#define SNB_STRM_ST		(1ULL << 11)
+#define SNB_OTHER		(1ULL << 15)
+#define SNB_RESP_ANY		(1ULL << 16)
+#define SNB_NO_SUPP		(1ULL << 17)
+#define SNB_LLC_HITM		(1ULL << 18)
+#define SNB_LLC_HITE		(1ULL << 19)
+#define SNB_LLC_HITS		(1ULL << 20)
+#define SNB_LLC_HITF		(1ULL << 21)
+#define SNB_LOCAL		(1ULL << 22)
+#define SNB_REMOTE		(0xffULL << 23)
+#define SNB_SNP_NONE		(1ULL << 31)
+#define SNB_SNP_NOT_NEEDED	(1ULL << 32)
+#define SNB_SNP_MISS		(1ULL << 33)
+#define SNB_NO_FWD		(1ULL << 34)
+#define SNB_SNP_FWD		(1ULL << 35)
+#define SNB_HITM		(1ULL << 36)
+#define SNB_NON_DRAM		(1ULL << 37)
+
+#define SNB_DMND_READ		(SNB_DMND_DATA_RD|SNB_LLC_DATA_RD)
+#define SNB_DMND_WRITE		(SNB_DMND_RFO|SNB_LLC_RFO)
+#define SNB_DMND_PREFETCH	(SNB_PF_DATA_RD|SNB_PF_RFO)
+
+#define SNB_SNP_ANY		(SNB_SNP_NONE|SNB_SNP_NOT_NEEDED| \
+				 SNB_SNP_MISS|SNB_NO_FWD|SNB_SNP_FWD| \
+				 SNB_HITM)
+
+#define SNB_DRAM_ANY		(SNB_LOCAL|SNB_REMOTE|SNB_SNP_ANY)
+#define SNB_DRAM_REMOTE		(SNB_REMOTE|SNB_SNP_ANY)
+
+#define SNB_L3_ACCESS		SNB_RESP_ANY
+#define SNB_L3_MISS		(SNB_DRAM_ANY|SNB_NON_DRAM)
+
+static __initconst const u64 snb_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_READ|SNB_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = SNB_DMND_READ|SNB_L3_MISS,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_WRITE|SNB_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = SNB_DMND_WRITE|SNB_L3_MISS,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_PREFETCH|SNB_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = SNB_DMND_PREFETCH|SNB_L3_MISS,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_READ|SNB_DRAM_ANY,
+		[ C(RESULT_MISS)   ] = SNB_DMND_READ|SNB_DRAM_REMOTE,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_WRITE|SNB_DRAM_ANY,
+		[ C(RESULT_MISS)   ] = SNB_DMND_WRITE|SNB_DRAM_REMOTE,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = SNB_DMND_PREFETCH|SNB_DRAM_ANY,
+		[ C(RESULT_MISS)   ] = SNB_DMND_PREFETCH|SNB_DRAM_REMOTE,
+	},
+ },
+};
+
+static __initconst const u64 snb_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) ] = 0xf1d0, /* MEM_UOP_RETIRED.LOADS        */
+		[ C(RESULT_MISS)   ] = 0x0151, /* L1D.REPLACEMENT              */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0xf2d0, /* MEM_UOP_RETIRED.STORES       */
+		[ C(RESULT_MISS)   ] = 0x0851, /* L1D.ALL_M_REPLACEMENT        */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x024e, /* HW_PRE_REQ.DL1_MISS          */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0280, /* ICACHE.MISSES */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		/* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_DATA.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_WRITE) ] = {
+		/* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		/* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x81d0, /* MEM_UOP_RETIRED.ALL_LOADS */
+		[ C(RESULT_MISS)   ] = 0x0108, /* DTLB_LOAD_MISSES.CAUSES_A_WALK */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x82d0, /* MEM_UOP_RETIRED.ALL_STORES */
+		[ C(RESULT_MISS)   ] = 0x0149, /* DTLB_STORE_MISSES.MISS_CAUSES_A_WALK */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1085, /* ITLB_MISSES.STLB_HIT         */
+		[ C(RESULT_MISS)   ] = 0x0185, /* ITLB_MISSES.CAUSES_A_WALK    */
+	},
+	[ 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.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BR_MISP_RETIRED.ALL_BRANCHES */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+
+};
+
+/*
+ * Notes on the events:
+ * - data reads do not include code reads (comparable to earlier tables)
+ * - data counts include speculative execution (except L1 write, dtlb, bpu)
+ * - remote node access includes remote memory, remote cache, remote mmio.
+ * - prefetches are not included in the counts because they are not
+ *   reliably counted.
+ */
+
+#define HSW_DEMAND_DATA_RD		BIT_ULL(0)
+#define HSW_DEMAND_RFO			BIT_ULL(1)
+#define HSW_ANY_RESPONSE		BIT_ULL(16)
+#define HSW_SUPPLIER_NONE		BIT_ULL(17)
+#define HSW_L3_MISS_LOCAL_DRAM		BIT_ULL(22)
+#define HSW_L3_MISS_REMOTE_HOP0		BIT_ULL(27)
+#define HSW_L3_MISS_REMOTE_HOP1		BIT_ULL(28)
+#define HSW_L3_MISS_REMOTE_HOP2P	BIT_ULL(29)
+#define HSW_L3_MISS			(HSW_L3_MISS_LOCAL_DRAM| \
+					 HSW_L3_MISS_REMOTE_HOP0|HSW_L3_MISS_REMOTE_HOP1| \
+					 HSW_L3_MISS_REMOTE_HOP2P)
+#define HSW_SNOOP_NONE			BIT_ULL(31)
+#define HSW_SNOOP_NOT_NEEDED		BIT_ULL(32)
+#define HSW_SNOOP_MISS			BIT_ULL(33)
+#define HSW_SNOOP_HIT_NO_FWD		BIT_ULL(34)
+#define HSW_SNOOP_HIT_WITH_FWD		BIT_ULL(35)
+#define HSW_SNOOP_HITM			BIT_ULL(36)
+#define HSW_SNOOP_NON_DRAM		BIT_ULL(37)
+#define HSW_ANY_SNOOP			(HSW_SNOOP_NONE| \
+					 HSW_SNOOP_NOT_NEEDED|HSW_SNOOP_MISS| \
+					 HSW_SNOOP_HIT_NO_FWD|HSW_SNOOP_HIT_WITH_FWD| \
+					 HSW_SNOOP_HITM|HSW_SNOOP_NON_DRAM)
+#define HSW_SNOOP_DRAM			(HSW_ANY_SNOOP & ~HSW_SNOOP_NON_DRAM)
+#define HSW_DEMAND_READ			HSW_DEMAND_DATA_RD
+#define HSW_DEMAND_WRITE		HSW_DEMAND_RFO
+#define HSW_L3_MISS_REMOTE		(HSW_L3_MISS_REMOTE_HOP0|\
+					 HSW_L3_MISS_REMOTE_HOP1|HSW_L3_MISS_REMOTE_HOP2P)
+#define HSW_LLC_ACCESS			HSW_ANY_RESPONSE
+
+#define BDW_L3_MISS_LOCAL		BIT(26)
+#define BDW_L3_MISS			(BDW_L3_MISS_LOCAL| \
+					 HSW_L3_MISS_REMOTE_HOP0|HSW_L3_MISS_REMOTE_HOP1| \
+					 HSW_L3_MISS_REMOTE_HOP2P)
+
+
+static __initconst const u64 hsw_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) ] = 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+		[ C(RESULT_MISS)   ] = 0x151,	/* L1D.REPLACEMENT */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x280,	/* ICACHE.MISSES */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+		[ C(RESULT_MISS)   ] = 0x108,	/* DTLB_LOAD_MISSES.MISS_CAUSES_A_WALK */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+		[ C(RESULT_MISS)   ] = 0x149,	/* DTLB_STORE_MISSES.MISS_CAUSES_A_WALK */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x6085,	/* ITLB_MISSES.STLB_HIT */
+		[ C(RESULT_MISS)   ] = 0x185,	/* ITLB_MISSES.MISS_CAUSES_A_WALK */
+	},
+	[ 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) ] = 0xc4,	/* BR_INST_RETIRED.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0xc5,	/* BR_MISP_RETIRED.ALL_BRANCHES */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x1b7,	/* OFFCORE_RESPONSE */
+		[ C(RESULT_MISS)   ] = 0x1b7,	/* OFFCORE_RESPONSE */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+};
+
+static __initconst const u64 hsw_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = HSW_DEMAND_READ|
+				       HSW_LLC_ACCESS,
+		[ C(RESULT_MISS)   ] = HSW_DEMAND_READ|
+				       HSW_L3_MISS|HSW_ANY_SNOOP,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = HSW_DEMAND_WRITE|
+				       HSW_LLC_ACCESS,
+		[ C(RESULT_MISS)   ] = HSW_DEMAND_WRITE|
+				       HSW_L3_MISS|HSW_ANY_SNOOP,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = HSW_DEMAND_READ|
+				       HSW_L3_MISS_LOCAL_DRAM|
+				       HSW_SNOOP_DRAM,
+		[ C(RESULT_MISS)   ] = HSW_DEMAND_READ|
+				       HSW_L3_MISS_REMOTE|
+				       HSW_SNOOP_DRAM,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = HSW_DEMAND_WRITE|
+				       HSW_L3_MISS_LOCAL_DRAM|
+				       HSW_SNOOP_DRAM,
+		[ C(RESULT_MISS)   ] = HSW_DEMAND_WRITE|
+				       HSW_L3_MISS_REMOTE|
+				       HSW_SNOOP_DRAM,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+};
+
+static __initconst const u64 westmere_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) ] = 0x010b, /* MEM_INST_RETIRED.LOADS       */
+		[ C(RESULT_MISS)   ] = 0x0151, /* L1D.REPL                     */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES      */
+		[ C(RESULT_MISS)   ] = 0x0251, /* L1D.M_REPL                   */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS        */
+		[ C(RESULT_MISS)   ] = 0x024e, /* L1D_PREFETCH.MISS            */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                    */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		/* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_DATA.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	/*
+	 * Use RFO, not WRITEBACK, because a write miss would typically occur
+	 * on RFO.
+	 */
+	[ C(OP_WRITE) ] = {
+		/* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		/* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x010b, /* MEM_INST_RETIRED.LOADS       */
+		[ C(RESULT_MISS)   ] = 0x0108, /* DTLB_LOAD_MISSES.ANY         */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES      */
+		[ C(RESULT_MISS)   ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS  */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P           */
+		[ C(RESULT_MISS)   ] = 0x0185, /* ITLB_MISSES.ANY              */
+	},
+	[ 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.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x03e8, /* BPU_CLEARS.ANY               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+};
+
+/*
+ * Nehalem/Westmere MSR_OFFCORE_RESPONSE bits;
+ * See IA32 SDM Vol 3B 30.6.1.3
+ */
+
+#define NHM_DMND_DATA_RD	(1 << 0)
+#define NHM_DMND_RFO		(1 << 1)
+#define NHM_DMND_IFETCH		(1 << 2)
+#define NHM_DMND_WB		(1 << 3)
+#define NHM_PF_DATA_RD		(1 << 4)
+#define NHM_PF_DATA_RFO		(1 << 5)
+#define NHM_PF_IFETCH		(1 << 6)
+#define NHM_OFFCORE_OTHER	(1 << 7)
+#define NHM_UNCORE_HIT		(1 << 8)
+#define NHM_OTHER_CORE_HIT_SNP	(1 << 9)
+#define NHM_OTHER_CORE_HITM	(1 << 10)
+        			/* reserved */
+#define NHM_REMOTE_CACHE_FWD	(1 << 12)
+#define NHM_REMOTE_DRAM		(1 << 13)
+#define NHM_LOCAL_DRAM		(1 << 14)
+#define NHM_NON_DRAM		(1 << 15)
+
+#define NHM_LOCAL		(NHM_LOCAL_DRAM|NHM_REMOTE_CACHE_FWD)
+#define NHM_REMOTE		(NHM_REMOTE_DRAM)
+
+#define NHM_DMND_READ		(NHM_DMND_DATA_RD)
+#define NHM_DMND_WRITE		(NHM_DMND_RFO|NHM_DMND_WB)
+#define NHM_DMND_PREFETCH	(NHM_PF_DATA_RD|NHM_PF_DATA_RFO)
+
+#define NHM_L3_HIT	(NHM_UNCORE_HIT|NHM_OTHER_CORE_HIT_SNP|NHM_OTHER_CORE_HITM)
+#define NHM_L3_MISS	(NHM_NON_DRAM|NHM_LOCAL_DRAM|NHM_REMOTE_DRAM|NHM_REMOTE_CACHE_FWD)
+#define NHM_L3_ACCESS	(NHM_L3_HIT|NHM_L3_MISS)
+
+static __initconst const u64 nehalem_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_READ|NHM_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = NHM_DMND_READ|NHM_L3_MISS,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_WRITE|NHM_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = NHM_DMND_WRITE|NHM_L3_MISS,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_PREFETCH|NHM_L3_ACCESS,
+		[ C(RESULT_MISS)   ] = NHM_DMND_PREFETCH|NHM_L3_MISS,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_READ|NHM_LOCAL|NHM_REMOTE,
+		[ C(RESULT_MISS)   ] = NHM_DMND_READ|NHM_REMOTE,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_WRITE|NHM_LOCAL|NHM_REMOTE,
+		[ C(RESULT_MISS)   ] = NHM_DMND_WRITE|NHM_REMOTE,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = NHM_DMND_PREFETCH|NHM_LOCAL|NHM_REMOTE,
+		[ C(RESULT_MISS)   ] = NHM_DMND_PREFETCH|NHM_REMOTE,
+	},
+ },
+};
+
+static __initconst const u64 nehalem_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) ] = 0x010b, /* MEM_INST_RETIRED.LOADS       */
+		[ C(RESULT_MISS)   ] = 0x0151, /* L1D.REPL                     */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x020b, /* MEM_INST_RETURED.STORES      */
+		[ C(RESULT_MISS)   ] = 0x0251, /* L1D.M_REPL                   */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x014e, /* L1D_PREFETCH.REQUESTS        */
+		[ C(RESULT_MISS)   ] = 0x024e, /* L1D_PREFETCH.MISS            */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                    */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                   */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		/* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_DATA.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	/*
+	 * Use RFO, not WRITEBACK, because a write miss would typically occur
+	 * on RFO.
+	 */
+	[ C(OP_WRITE) ] = {
+		/* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		/* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI   (alias)  */
+		[ C(RESULT_MISS)   ] = 0x0108, /* DTLB_LOAD_MISSES.ANY         */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI   (alias)  */
+		[ C(RESULT_MISS)   ] = 0x010c, /* MEM_STORE_RETIRED.DTLB_MISS  */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01c0, /* INST_RETIRED.ANY_P           */
+		[ C(RESULT_MISS)   ] = 0x20c8, /* ITLB_MISS_RETIRED            */
+	},
+	[ 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.ALL_BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x03e8, /* BPU_CLEARS.ANY               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+};
+
+static __initconst const u64 core2_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) ] = 0x0f40, /* L1D_CACHE_LD.MESI          */
+		[ C(RESULT_MISS)   ] = 0x0140, /* L1D_CACHE_LD.I_STATE       */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI          */
+		[ C(RESULT_MISS)   ] = 0x0141, /* L1D_CACHE_ST.I_STATE       */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x104e, /* L1D_PREFETCH.REQUESTS      */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0080, /* L1I.READS                  */
+		[ C(RESULT_MISS)   ] = 0x0081, /* L1I.MISSES                 */
+	},
+	[ 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) ] = 0x4f29, /* L2_LD.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x4129, /* L2_LD.ISTATE               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x412A, /* L2_ST.ISTATE               */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f40, /* L1D_CACHE_LD.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0208, /* DTLB_MISSES.MISS_LD        */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0f41, /* L1D_CACHE_ST.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0808, /* DTLB_MISSES.MISS_ST        */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P         */
+		[ C(RESULT_MISS)   ] = 0x1282, /* ITLBMISSES                 */
+	},
+	[ 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.ANY        */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BP_INST_RETIRED.MISPRED    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+static __initconst const u64 atom_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) ] = 0x2140, /* L1D_CACHE.LD               */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE.ST               */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* L1I.READS                  */
+		[ C(RESULT_MISS)   ] = 0x0280, /* L1I.MISSES                 */
+	},
+	[ 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) ] = 0x4f29, /* L2_LD.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x4129, /* L2_LD.ISTATE               */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x4f2A, /* L2_ST.MESI                 */
+		[ C(RESULT_MISS)   ] = 0x412A, /* L2_ST.ISTATE               */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2140, /* L1D_CACHE_LD.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0508, /* DTLB_MISSES.MISS_LD        */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x2240, /* L1D_CACHE_ST.MESI  (alias) */
+		[ C(RESULT_MISS)   ] = 0x0608, /* DTLB_MISSES.MISS_ST        */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x00c0, /* INST_RETIRED.ANY_P         */
+		[ C(RESULT_MISS)   ] = 0x0282, /* ITLB.MISSES                */
+	},
+	[ 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.ANY        */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BP_INST_RETIRED.MISPRED    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+EVENT_ATTR_STR(topdown-total-slots, td_total_slots_slm, "event=0x3c");
+EVENT_ATTR_STR(topdown-total-slots.scale, td_total_slots_scale_slm, "2");
+/* no_alloc_cycles.not_delivered */
+EVENT_ATTR_STR(topdown-fetch-bubbles, td_fetch_bubbles_slm,
+	       "event=0xca,umask=0x50");
+EVENT_ATTR_STR(topdown-fetch-bubbles.scale, td_fetch_bubbles_scale_slm, "2");
+/* uops_retired.all */
+EVENT_ATTR_STR(topdown-slots-issued, td_slots_issued_slm,
+	       "event=0xc2,umask=0x10");
+/* uops_retired.all */
+EVENT_ATTR_STR(topdown-slots-retired, td_slots_retired_slm,
+	       "event=0xc2,umask=0x10");
+
+static struct attribute *slm_events_attrs[] = {
+	EVENT_PTR(td_total_slots_slm),
+	EVENT_PTR(td_total_slots_scale_slm),
+	EVENT_PTR(td_fetch_bubbles_slm),
+	EVENT_PTR(td_fetch_bubbles_scale_slm),
+	EVENT_PTR(td_slots_issued_slm),
+	EVENT_PTR(td_slots_retired_slm),
+	NULL
+};
+
+static struct extra_reg intel_slm_extra_regs[] __read_mostly =
+{
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x768005ffffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0x368005ffffull, RSP_1),
+	EVENT_EXTRA_END
+};
+
+#define SLM_DMND_READ		SNB_DMND_DATA_RD
+#define SLM_DMND_WRITE		SNB_DMND_RFO
+#define SLM_DMND_PREFETCH	(SNB_PF_DATA_RD|SNB_PF_RFO)
+
+#define SLM_SNP_ANY		(SNB_SNP_NONE|SNB_SNP_MISS|SNB_NO_FWD|SNB_HITM)
+#define SLM_LLC_ACCESS		SNB_RESP_ANY
+#define SLM_LLC_MISS		(SLM_SNP_ANY|SNB_NON_DRAM)
+
+static __initconst const u64 slm_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = SLM_DMND_READ|SLM_LLC_ACCESS,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = SLM_DMND_WRITE|SLM_LLC_ACCESS,
+		[ C(RESULT_MISS)   ] = SLM_DMND_WRITE|SLM_LLC_MISS,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = SLM_DMND_PREFETCH|SLM_LLC_ACCESS,
+		[ C(RESULT_MISS)   ] = SLM_DMND_PREFETCH|SLM_LLC_MISS,
+	},
+ },
+};
+
+static __initconst const u64 slm_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) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x0104, /* LD_DCU_MISS */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0380, /* ICACHE.ACCESSES */
+		[ C(RESULT_MISS)   ] = 0x0280, /* ICACGE.MISSES */
+	},
+	[ 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) ] = {
+		/* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_WRITE) ] = {
+		/* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+	[ C(OP_PREFETCH) ] = {
+		/* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */
+		[ C(RESULT_ACCESS) ] = 0x01b7,
+		/* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */
+		[ C(RESULT_MISS)   ] = 0x01b7,
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x0804, /* LD_DTLB_MISS */
+	},
+	[ 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) ] = 0x00c0, /* INST_RETIRED.ANY_P */
+		[ C(RESULT_MISS)   ] = 0x40205, /* PAGE_WALKS.I_SIDE_WALKS */
+	},
+	[ 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.ANY */
+		[ C(RESULT_MISS)   ] = 0x00c5, /* BP_INST_RETIRED.MISPRED */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+EVENT_ATTR_STR(topdown-total-slots, td_total_slots_glm, "event=0x3c");
+EVENT_ATTR_STR(topdown-total-slots.scale, td_total_slots_scale_glm, "3");
+/* UOPS_NOT_DELIVERED.ANY */
+EVENT_ATTR_STR(topdown-fetch-bubbles, td_fetch_bubbles_glm, "event=0x9c");
+/* ISSUE_SLOTS_NOT_CONSUMED.RECOVERY */
+EVENT_ATTR_STR(topdown-recovery-bubbles, td_recovery_bubbles_glm, "event=0xca,umask=0x02");
+/* UOPS_RETIRED.ANY */
+EVENT_ATTR_STR(topdown-slots-retired, td_slots_retired_glm, "event=0xc2");
+/* UOPS_ISSUED.ANY */
+EVENT_ATTR_STR(topdown-slots-issued, td_slots_issued_glm, "event=0x0e");
+
+static struct attribute *glm_events_attrs[] = {
+	EVENT_PTR(td_total_slots_glm),
+	EVENT_PTR(td_total_slots_scale_glm),
+	EVENT_PTR(td_fetch_bubbles_glm),
+	EVENT_PTR(td_recovery_bubbles_glm),
+	EVENT_PTR(td_slots_issued_glm),
+	EVENT_PTR(td_slots_retired_glm),
+	NULL
+};
+
+static struct extra_reg intel_glm_extra_regs[] __read_mostly = {
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x760005ffbfull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0x360005ffbfull, RSP_1),
+	EVENT_EXTRA_END
+};
+
+#define GLM_DEMAND_DATA_RD		BIT_ULL(0)
+#define GLM_DEMAND_RFO			BIT_ULL(1)
+#define GLM_ANY_RESPONSE		BIT_ULL(16)
+#define GLM_SNP_NONE_OR_MISS		BIT_ULL(33)
+#define GLM_DEMAND_READ			GLM_DEMAND_DATA_RD
+#define GLM_DEMAND_WRITE		GLM_DEMAND_RFO
+#define GLM_DEMAND_PREFETCH		(SNB_PF_DATA_RD|SNB_PF_RFO)
+#define GLM_LLC_ACCESS			GLM_ANY_RESPONSE
+#define GLM_SNP_ANY			(GLM_SNP_NONE_OR_MISS|SNB_NO_FWD|SNB_HITM)
+#define GLM_LLC_MISS			(GLM_SNP_ANY|SNB_NON_DRAM)
+
+static __initconst const u64 glm_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)]	= 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(L1I)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x0380,	/* ICACHE.ACCESSES */
+			[C(RESULT_MISS)]	= 0x0280,	/* ICACHE.MISSES */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+			[C(RESULT_MISS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+			[C(RESULT_MISS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+			[C(RESULT_MISS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+		},
+	},
+	[C(DTLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(ITLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x00c0,	/* INST_RETIRED.ANY_P */
+			[C(RESULT_MISS)]	= 0x0481,	/* 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.ALL_BRANCHES */
+			[C(RESULT_MISS)]	= 0x00c5,	/* BR_MISP_RETIRED.ALL_BRANCHES */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+	},
+};
+
+static __initconst const u64 glm_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= GLM_DEMAND_READ|
+						  GLM_LLC_ACCESS,
+			[C(RESULT_MISS)]	= GLM_DEMAND_READ|
+						  GLM_LLC_MISS,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= GLM_DEMAND_WRITE|
+						  GLM_LLC_ACCESS,
+			[C(RESULT_MISS)]	= GLM_DEMAND_WRITE|
+						  GLM_LLC_MISS,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= GLM_DEMAND_PREFETCH|
+						  GLM_LLC_ACCESS,
+			[C(RESULT_MISS)]	= GLM_DEMAND_PREFETCH|
+						  GLM_LLC_MISS,
+		},
+	},
+};
+
+static __initconst const u64 glp_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)]	= 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+			[C(RESULT_MISS)]	= 0x0,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(L1I)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x0380,	/* ICACHE.ACCESSES */
+			[C(RESULT_MISS)]	= 0x0280,	/* ICACHE.MISSES */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+			[C(RESULT_MISS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+			[C(RESULT_MISS)]	= 0x1b7,	/* OFFCORE_RESPONSE */
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(DTLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x81d0,	/* MEM_UOPS_RETIRED.ALL_LOADS */
+			[C(RESULT_MISS)]	= 0xe08,	/* DTLB_LOAD_MISSES.WALK_COMPLETED */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= 0x82d0,	/* MEM_UOPS_RETIRED.ALL_STORES */
+			[C(RESULT_MISS)]	= 0xe49,	/* DTLB_STORE_MISSES.WALK_COMPLETED */
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+	[C(ITLB)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= 0x00c0,	/* INST_RETIRED.ANY_P */
+			[C(RESULT_MISS)]	= 0x0481,	/* 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.ALL_BRANCHES */
+			[C(RESULT_MISS)]	= 0x00c5,	/* BR_MISP_RETIRED.ALL_BRANCHES */
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= -1,
+			[C(RESULT_MISS)]	= -1,
+		},
+	},
+};
+
+static __initconst const u64 glp_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= GLM_DEMAND_READ|
+						  GLM_LLC_ACCESS,
+			[C(RESULT_MISS)]	= GLM_DEMAND_READ|
+						  GLM_LLC_MISS,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= GLM_DEMAND_WRITE|
+						  GLM_LLC_ACCESS,
+			[C(RESULT_MISS)]	= GLM_DEMAND_WRITE|
+						  GLM_LLC_MISS,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+};
+
+#define TNT_LOCAL_DRAM			BIT_ULL(26)
+#define TNT_DEMAND_READ			GLM_DEMAND_DATA_RD
+#define TNT_DEMAND_WRITE		GLM_DEMAND_RFO
+#define TNT_LLC_ACCESS			GLM_ANY_RESPONSE
+#define TNT_SNP_ANY			(SNB_SNP_NOT_NEEDED|SNB_SNP_MISS| \
+					 SNB_NO_FWD|SNB_SNP_FWD|SNB_HITM)
+#define TNT_LLC_MISS			(TNT_SNP_ANY|SNB_NON_DRAM|TNT_LOCAL_DRAM)
+
+static __initconst const u64 tnt_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)]	= TNT_DEMAND_READ|
+						  TNT_LLC_ACCESS,
+			[C(RESULT_MISS)]	= TNT_DEMAND_READ|
+						  TNT_LLC_MISS,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)]	= TNT_DEMAND_WRITE|
+						  TNT_LLC_ACCESS,
+			[C(RESULT_MISS)]	= TNT_DEMAND_WRITE|
+						  TNT_LLC_MISS,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)]	= 0x0,
+			[C(RESULT_MISS)]	= 0x0,
+		},
+	},
+};
+
+static struct extra_reg intel_tnt_extra_regs[] __read_mostly = {
+	/* must define OFFCORE_RSP_X first, see intel_fixup_er() */
+	INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x800ff0ffffff9fffull, RSP_0),
+	INTEL_UEVENT_EXTRA_REG(0x02b7, MSR_OFFCORE_RSP_1, 0xff0ffffff9fffull, RSP_1),
+	EVENT_EXTRA_END
+};
+
+#define KNL_OT_L2_HITE		BIT_ULL(19) /* Other Tile L2 Hit */
+#define KNL_OT_L2_HITF		BIT_ULL(20) /* Other Tile L2 Hit */
+#define KNL_MCDRAM_LOCAL	BIT_ULL(21)
+#define KNL_MCDRAM_FAR		BIT_ULL(22)
+#define KNL_DDR_LOCAL		BIT_ULL(23)
+#define KNL_DDR_FAR		BIT_ULL(24)
+#define KNL_DRAM_ANY		(KNL_MCDRAM_LOCAL | KNL_MCDRAM_FAR | \
+				    KNL_DDR_LOCAL | KNL_DDR_FAR)
+#define KNL_L2_READ		SLM_DMND_READ
+#define KNL_L2_WRITE		SLM_DMND_WRITE
+#define KNL_L2_PREFETCH		SLM_DMND_PREFETCH
+#define KNL_L2_ACCESS		SLM_LLC_ACCESS
+#define KNL_L2_MISS		(KNL_OT_L2_HITE | KNL_OT_L2_HITF | \
+				   KNL_DRAM_ANY | SNB_SNP_ANY | \
+						  SNB_NON_DRAM)
+
+static __initconst const u64 knl_hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+	[C(LL)] = {
+		[C(OP_READ)] = {
+			[C(RESULT_ACCESS)] = KNL_L2_READ | KNL_L2_ACCESS,
+			[C(RESULT_MISS)]   = 0,
+		},
+		[C(OP_WRITE)] = {
+			[C(RESULT_ACCESS)] = KNL_L2_WRITE | KNL_L2_ACCESS,
+			[C(RESULT_MISS)]   = KNL_L2_WRITE | KNL_L2_MISS,
+		},
+		[C(OP_PREFETCH)] = {
+			[C(RESULT_ACCESS)] = KNL_L2_PREFETCH | KNL_L2_ACCESS,
+			[C(RESULT_MISS)]   = KNL_L2_PREFETCH | KNL_L2_MISS,
+		},
+	},
+};
+
+/*
+ * Used from PMIs where the LBRs are already disabled.
+ *
+ * This function could be called consecutively. It is required to remain in
+ * disabled state if called consecutively.
+ *
+ * During consecutive calls, the same disable value will be written to related
+ * registers, so the PMU state remains unchanged.
+ *
+ * intel_bts events don't coexist with intel PMU's BTS events because of
+ * x86_add_exclusive(x86_lbr_exclusive_lbr); there's no need to keep them
+ * disabled around intel PMU's event batching etc, only inside the PMI handler.
+ *
+ * Avoid PEBS_ENABLE MSR access in PMIs.
+ * The GLOBAL_CTRL has been disabled. All the counters do not count anymore.
+ * It doesn't matter if the PEBS is enabled or not.
+ * Usually, the PEBS status are not changed in PMIs. It's unnecessary to
+ * access PEBS_ENABLE MSR in disable_all()/enable_all().
+ * However, there are some cases which may change PEBS status, e.g. PMI
+ * throttle. The PEBS_ENABLE should be updated where the status changes.
+ */
+static void __intel_pmu_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask))
+		intel_pmu_disable_bts();
+}
+
+static void intel_pmu_disable_all(void)
+{
+	__intel_pmu_disable_all();
+	intel_pmu_pebs_disable_all();
+	intel_pmu_lbr_disable_all();
+}
+
+static void __intel_pmu_enable_all(int added, bool pmi)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	intel_pmu_lbr_enable_all(pmi);
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL,
+			x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask);
+
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
+		struct perf_event *event =
+			cpuc->events[INTEL_PMC_IDX_FIXED_BTS];
+
+		if (WARN_ON_ONCE(!event))
+			return;
+
+		intel_pmu_enable_bts(event->hw.config);
+	}
+}
+
+static void intel_pmu_enable_all(int added)
+{
+	intel_pmu_pebs_enable_all();
+	__intel_pmu_enable_all(added, false);
+}
+
+/*
+ * Workaround for:
+ *   Intel Errata AAK100 (model 26)
+ *   Intel Errata AAP53  (model 30)
+ *   Intel Errata BD53   (model 44)
+ *
+ * The official story:
+ *   These chips need to be 'reset' when adding counters by programming the
+ *   magic three (non-counting) events 0x4300B5, 0x4300D2, and 0x4300B1 either
+ *   in sequence on the same PMC or on different PMCs.
+ *
+ * In practise it appears some of these events do in fact count, and
+ * we need to program all 4 events.
+ */
+static void intel_pmu_nhm_workaround(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	static const unsigned long nhm_magic[4] = {
+		0x4300B5,
+		0x4300D2,
+		0x4300B1,
+		0x4300B1
+	};
+	struct perf_event *event;
+	int i;
+
+	/*
+	 * The Errata requires below steps:
+	 * 1) Clear MSR_IA32_PEBS_ENABLE and MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 2) Configure 4 PERFEVTSELx with the magic events and clear
+	 *    the corresponding PMCx;
+	 * 3) set bit0~bit3 of MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 4) Clear MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 5) Clear 4 pairs of ERFEVTSELx and PMCx;
+	 */
+
+	/*
+	 * The real steps we choose are a little different from above.
+	 * A) To reduce MSR operations, we don't run step 1) as they
+	 *    are already cleared before this function is called;
+	 * B) Call x86_perf_event_update to save PMCx before configuring
+	 *    PERFEVTSELx with magic number;
+	 * C) With step 5), we do clear only when the PERFEVTSELx is
+	 *    not used currently.
+	 * D) Call x86_perf_event_set_period to restore PMCx;
+	 */
+
+	/* We always operate 4 pairs of PERF Counters */
+	for (i = 0; i < 4; i++) {
+		event = cpuc->events[i];
+		if (event)
+			x86_perf_event_update(event);
+	}
+
+	for (i = 0; i < 4; i++) {
+		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, nhm_magic[i]);
+		wrmsrl(MSR_ARCH_PERFMON_PERFCTR0 + i, 0x0);
+	}
+
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0xf);
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0);
+
+	for (i = 0; i < 4; i++) {
+		event = cpuc->events[i];
+
+		if (event) {
+			x86_perf_event_set_period(event);
+			__x86_pmu_enable_event(&event->hw,
+					ARCH_PERFMON_EVENTSEL_ENABLE);
+		} else
+			wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, 0x0);
+	}
+}
+
+static void intel_pmu_nhm_enable_all(int added)
+{
+	if (added)
+		intel_pmu_nhm_workaround();
+	intel_pmu_enable_all(added);
+}
+
+static void intel_set_tfa(struct cpu_hw_events *cpuc, bool on)
+{
+	u64 val = on ? MSR_TFA_RTM_FORCE_ABORT : 0;
+
+	if (cpuc->tfa_shadow != val) {
+		cpuc->tfa_shadow = val;
+		wrmsrl(MSR_TSX_FORCE_ABORT, val);
+	}
+}
+
+static void intel_tfa_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr)
+{
+	/*
+	 * We're going to use PMC3, make sure TFA is set before we touch it.
+	 */
+	if (cntr == 3)
+		intel_set_tfa(cpuc, true);
+}
+
+static void intel_tfa_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * If we find PMC3 is no longer used when we enable the PMU, we can
+	 * clear TFA.
+	 */
+	if (!test_bit(3, cpuc->active_mask))
+		intel_set_tfa(cpuc, false);
+
+	intel_pmu_enable_all(added);
+}
+
+static void enable_counter_freeze(void)
+{
+	update_debugctlmsr(get_debugctlmsr() |
+			DEBUGCTLMSR_FREEZE_PERFMON_ON_PMI);
+}
+
+static void disable_counter_freeze(void)
+{
+	update_debugctlmsr(get_debugctlmsr() &
+			~DEBUGCTLMSR_FREEZE_PERFMON_ON_PMI);
+}
+
+static inline u64 intel_pmu_get_status(void)
+{
+	u64 status;
+
+	rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
+
+	return status;
+}
+
+static inline void intel_pmu_ack_status(u64 ack)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack);
+}
+
+static inline bool event_is_checkpointed(struct perf_event *event)
+{
+	return unlikely(event->hw.config & HSW_IN_TX_CHECKPOINTED) != 0;
+}
+
+static inline void intel_set_masks(struct perf_event *event, int idx)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (event->attr.exclude_host)
+		__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_guest_mask);
+	if (event->attr.exclude_guest)
+		__set_bit(idx, (unsigned long *)&cpuc->intel_ctrl_host_mask);
+	if (event_is_checkpointed(event))
+		__set_bit(idx, (unsigned long *)&cpuc->intel_cp_status);
+}
+
+static inline void intel_clear_masks(struct perf_event *event, int idx)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	__clear_bit(idx, (unsigned long *)&cpuc->intel_ctrl_guest_mask);
+	__clear_bit(idx, (unsigned long *)&cpuc->intel_ctrl_host_mask);
+	__clear_bit(idx, (unsigned long *)&cpuc->intel_cp_status);
+}
+
+static void intel_pmu_disable_fixed(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 ctrl_val, mask;
+	int idx = hwc->idx;
+
+	if (is_topdown_idx(idx)) {
+		struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+		/*
+		 * When there are other active TopDown events,
+		 * don't disable the fixed counter 3.
+		 */
+		if (*(u64 *)cpuc->active_mask & INTEL_PMC_OTHER_TOPDOWN_BITS(idx))
+			return;
+		idx = INTEL_PMC_IDX_FIXED_SLOTS;
+	}
+
+	intel_clear_masks(event, idx);
+
+	mask = 0xfULL << ((idx - INTEL_PMC_IDX_FIXED) * 4);
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static void intel_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	switch (idx) {
+	case 0 ... INTEL_PMC_IDX_FIXED - 1:
+		intel_clear_masks(event, idx);
+		x86_pmu_disable_event(event);
+		break;
+	case INTEL_PMC_IDX_FIXED ... INTEL_PMC_IDX_FIXED_BTS - 1:
+	case INTEL_PMC_IDX_METRIC_BASE ... INTEL_PMC_IDX_METRIC_END:
+		intel_pmu_disable_fixed(event);
+		break;
+	case INTEL_PMC_IDX_FIXED_BTS:
+		intel_pmu_disable_bts();
+		intel_pmu_drain_bts_buffer();
+		return;
+	case INTEL_PMC_IDX_FIXED_VLBR:
+		intel_clear_masks(event, idx);
+		break;
+	default:
+		intel_clear_masks(event, idx);
+		pr_warn("Failed to disable the event with invalid index %d\n",
+			idx);
+		return;
+	}
+
+	/*
+	 * Needs to be called after x86_pmu_disable_event,
+	 * so we don't trigger the event without PEBS bit set.
+	 */
+	if (unlikely(event->attr.precise_ip))
+		intel_pmu_pebs_disable(event);
+}
+
+static void intel_pmu_del_event(struct perf_event *event)
+{
+	if (needs_branch_stack(event))
+		intel_pmu_lbr_del(event);
+	if (event->attr.precise_ip)
+		intel_pmu_pebs_del(event);
+}
+
+static int icl_set_topdown_event_period(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	s64 left = local64_read(&hwc->period_left);
+
+	/*
+	 * The values in PERF_METRICS MSR are derived from fixed counter 3.
+	 * Software should start both registers, PERF_METRICS and fixed
+	 * counter 3, from zero.
+	 * Clear PERF_METRICS and Fixed counter 3 in initialization.
+	 * After that, both MSRs will be cleared for each read.
+	 * Don't need to clear them again.
+	 */
+	if (left == x86_pmu.max_period) {
+		wrmsrl(MSR_CORE_PERF_FIXED_CTR3, 0);
+		wrmsrl(MSR_PERF_METRICS, 0);
+		hwc->saved_slots = 0;
+		hwc->saved_metric = 0;
+	}
+
+	if ((hwc->saved_slots) && is_slots_event(event)) {
+		wrmsrl(MSR_CORE_PERF_FIXED_CTR3, hwc->saved_slots);
+		wrmsrl(MSR_PERF_METRICS, hwc->saved_metric);
+	}
+
+	perf_event_update_userpage(event);
+
+	return 0;
+}
+
+static inline u64 icl_get_metrics_event_value(u64 metric, u64 slots, int idx)
+{
+	u32 val;
+
+	/*
+	 * The metric is reported as an 8bit integer fraction
+	 * suming up to 0xff.
+	 * slots-in-metric = (Metric / 0xff) * slots
+	 */
+	val = (metric >> ((idx - INTEL_PMC_IDX_METRIC_BASE) * 8)) & 0xff;
+	return  mul_u64_u32_div(slots, val, 0xff);
+}
+
+static u64 icl_get_topdown_value(struct perf_event *event,
+				       u64 slots, u64 metrics)
+{
+	int idx = event->hw.idx;
+	u64 delta;
+
+	if (is_metric_idx(idx))
+		delta = icl_get_metrics_event_value(metrics, slots, idx);
+	else
+		delta = slots;
+
+	return delta;
+}
+
+static void __icl_update_topdown_event(struct perf_event *event,
+				       u64 slots, u64 metrics,
+				       u64 last_slots, u64 last_metrics)
+{
+	u64 delta, last = 0;
+
+	delta = icl_get_topdown_value(event, slots, metrics);
+	if (last_slots)
+		last = icl_get_topdown_value(event, last_slots, last_metrics);
+
+	/*
+	 * The 8bit integer fraction of metric may be not accurate,
+	 * especially when the changes is very small.
+	 * For example, if only a few bad_spec happens, the fraction
+	 * may be reduced from 1 to 0. If so, the bad_spec event value
+	 * will be 0 which is definitely less than the last value.
+	 * Avoid update event->count for this case.
+	 */
+	if (delta > last) {
+		delta -= last;
+		local64_add(delta, &event->count);
+	}
+}
+
+static void update_saved_topdown_regs(struct perf_event *event,
+				      u64 slots, u64 metrics)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_event *other;
+	int idx;
+
+	event->hw.saved_slots = slots;
+	event->hw.saved_metric = metrics;
+
+	for_each_set_bit(idx, cpuc->active_mask, INTEL_PMC_IDX_TD_BE_BOUND + 1) {
+		if (!is_topdown_idx(idx))
+			continue;
+		other = cpuc->events[idx];
+		other->hw.saved_slots = slots;
+		other->hw.saved_metric = metrics;
+	}
+}
+
+/*
+ * Update all active Topdown events.
+ *
+ * The PERF_METRICS and Fixed counter 3 are read separately. The values may be
+ * modify by a NMI. PMU has to be disabled before calling this function.
+ */
+static u64 icl_update_topdown_event(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_event *other;
+	u64 slots, metrics;
+	bool reset = true;
+	int idx;
+
+	/* read Fixed counter 3 */
+	rdpmcl((3 | INTEL_PMC_FIXED_RDPMC_BASE), slots);
+	if (!slots)
+		return 0;
+
+	/* read PERF_METRICS */
+	rdpmcl(INTEL_PMC_FIXED_RDPMC_METRICS, metrics);
+
+	for_each_set_bit(idx, cpuc->active_mask, INTEL_PMC_IDX_TD_BE_BOUND + 1) {
+		if (!is_topdown_idx(idx))
+			continue;
+		other = cpuc->events[idx];
+		__icl_update_topdown_event(other, slots, metrics,
+					   event ? event->hw.saved_slots : 0,
+					   event ? event->hw.saved_metric : 0);
+	}
+
+	/*
+	 * Check and update this event, which may have been cleared
+	 * in active_mask e.g. x86_pmu_stop()
+	 */
+	if (event && !test_bit(event->hw.idx, cpuc->active_mask)) {
+		__icl_update_topdown_event(event, slots, metrics,
+					   event->hw.saved_slots,
+					   event->hw.saved_metric);
+
+		/*
+		 * In x86_pmu_stop(), the event is cleared in active_mask first,
+		 * then drain the delta, which indicates context switch for
+		 * counting.
+		 * Save metric and slots for context switch.
+		 * Don't need to reset the PERF_METRICS and Fixed counter 3.
+		 * Because the values will be restored in next schedule in.
+		 */
+		update_saved_topdown_regs(event, slots, metrics);
+		reset = false;
+	}
+
+	if (reset) {
+		/* The fixed counter 3 has to be written before the PERF_METRICS. */
+		wrmsrl(MSR_CORE_PERF_FIXED_CTR3, 0);
+		wrmsrl(MSR_PERF_METRICS, 0);
+		if (event)
+			update_saved_topdown_regs(event, 0, 0);
+	}
+
+	return slots;
+}
+
+static void intel_pmu_read_topdown_event(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/* Only need to call update_topdown_event() once for group read. */
+	if ((cpuc->txn_flags & PERF_PMU_TXN_READ) &&
+	    !is_slots_event(event))
+		return;
+
+	perf_pmu_disable(event->pmu);
+	x86_pmu.update_topdown_event(event);
+	perf_pmu_enable(event->pmu);
+}
+
+static void intel_pmu_read_event(struct perf_event *event)
+{
+	if (event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD)
+		intel_pmu_auto_reload_read(event);
+	else if (is_topdown_count(event) && x86_pmu.update_topdown_event)
+		intel_pmu_read_topdown_event(event);
+	else
+		x86_perf_event_update(event);
+}
+
+static void intel_pmu_enable_fixed(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 ctrl_val, mask, bits = 0;
+	int idx = hwc->idx;
+
+	if (is_topdown_idx(idx)) {
+		struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+		/*
+		 * When there are other active TopDown events,
+		 * don't enable the fixed counter 3 again.
+		 */
+		if (*(u64 *)cpuc->active_mask & INTEL_PMC_OTHER_TOPDOWN_BITS(idx))
+			return;
+
+		idx = INTEL_PMC_IDX_FIXED_SLOTS;
+	}
+
+	intel_set_masks(event, idx);
+
+	/*
+	 * Enable IRQ generation (0x8), if not PEBS,
+	 * and enable ring-3 counting (0x2) and ring-0 counting (0x1)
+	 * if requested:
+	 */
+	if (!event->attr.precise_ip)
+		bits |= 0x8;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_USR)
+		bits |= 0x2;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_OS)
+		bits |= 0x1;
+
+	/*
+	 * ANY bit is supported in v3 and up
+	 */
+	if (x86_pmu.version > 2 && hwc->config & ARCH_PERFMON_EVENTSEL_ANY)
+		bits |= 0x4;
+
+	idx -= INTEL_PMC_IDX_FIXED;
+	bits <<= (idx * 4);
+	mask = 0xfULL << (idx * 4);
+
+	if (x86_pmu.intel_cap.pebs_baseline && event->attr.precise_ip) {
+		bits |= ICL_FIXED_0_ADAPTIVE << (idx * 4);
+		mask |= ICL_FIXED_0_ADAPTIVE << (idx * 4);
+	}
+
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	ctrl_val |= bits;
+	wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static void intel_pmu_enable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	if (unlikely(event->attr.precise_ip))
+		intel_pmu_pebs_enable(event);
+
+	switch (idx) {
+	case 0 ... INTEL_PMC_IDX_FIXED - 1:
+		intel_set_masks(event, idx);
+		__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+		break;
+	case INTEL_PMC_IDX_FIXED ... INTEL_PMC_IDX_FIXED_BTS - 1:
+	case INTEL_PMC_IDX_METRIC_BASE ... INTEL_PMC_IDX_METRIC_END:
+		intel_pmu_enable_fixed(event);
+		break;
+	case INTEL_PMC_IDX_FIXED_BTS:
+		if (!__this_cpu_read(cpu_hw_events.enabled))
+			return;
+		intel_pmu_enable_bts(hwc->config);
+		break;
+	case INTEL_PMC_IDX_FIXED_VLBR:
+		intel_set_masks(event, idx);
+		break;
+	default:
+		pr_warn("Failed to enable the event with invalid index %d\n",
+			idx);
+	}
+}
+
+static void intel_pmu_add_event(struct perf_event *event)
+{
+	if (event->attr.precise_ip)
+		intel_pmu_pebs_add(event);
+	if (needs_branch_stack(event))
+		intel_pmu_lbr_add(event);
+}
+
+/*
+ * Save and restart an expired event. Called by NMI contexts,
+ * so it has to be careful about preempting normal event ops:
+ */
+int intel_pmu_save_and_restart(struct perf_event *event)
+{
+	x86_perf_event_update(event);
+	/*
+	 * For a checkpointed counter always reset back to 0.  This
+	 * avoids a situation where the counter overflows, aborts the
+	 * transaction and is then set back to shortly before the
+	 * overflow, and overflows and aborts again.
+	 */
+	if (unlikely(event_is_checkpointed(event))) {
+		/* No race with NMIs because the counter should not be armed */
+		wrmsrl(event->hw.event_base, 0);
+		local64_set(&event->hw.prev_count, 0);
+	}
+	return x86_perf_event_set_period(event);
+}
+
+static void intel_pmu_reset(void)
+{
+	struct debug_store *ds = __this_cpu_read(cpu_hw_events.ds);
+	unsigned long flags;
+	int idx;
+
+	if (!x86_pmu.num_counters)
+		return;
+
+	local_irq_save(flags);
+
+	pr_info("clearing PMU state on CPU#%d\n", smp_processor_id());
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		wrmsrl_safe(x86_pmu_config_addr(idx), 0ull);
+		wrmsrl_safe(x86_pmu_event_addr(idx),  0ull);
+	}
+	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++)
+		wrmsrl_safe(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull);
+
+	if (ds)
+		ds->bts_index = ds->bts_buffer_base;
+
+	/* Ack all overflows and disable fixed counters */
+	if (x86_pmu.version >= 2) {
+		intel_pmu_ack_status(intel_pmu_get_status());
+		wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+	}
+
+	/* Reset LBRs and LBR freezing */
+	if (x86_pmu.lbr_nr) {
+		update_debugctlmsr(get_debugctlmsr() &
+			~(DEBUGCTLMSR_FREEZE_LBRS_ON_PMI|DEBUGCTLMSR_LBR));
+	}
+
+	local_irq_restore(flags);
+}
+
+static int handle_pmi_common(struct pt_regs *regs, u64 status)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_guest_info_callbacks *guest_cbs;
+	int bit;
+	int handled = 0;
+
+	inc_irq_stat(apic_perf_irqs);
+
+	/*
+	 * Ignore a range of extra bits in status that do not indicate
+	 * overflow by themselves.
+	 */
+	status &= ~(GLOBAL_STATUS_COND_CHG |
+		    GLOBAL_STATUS_ASIF |
+		    GLOBAL_STATUS_LBRS_FROZEN);
+	if (!status)
+		return 0;
+	/*
+	 * In case multiple PEBS events are sampled at the same time,
+	 * it is possible to have GLOBAL_STATUS bit 62 set indicating
+	 * PEBS buffer overflow and also seeing at most 3 PEBS counters
+	 * having their bits set in the status register. This is a sign
+	 * that there was at least one PEBS record pending at the time
+	 * of the PMU interrupt. PEBS counters must only be processed
+	 * via the drain_pebs() calls and not via the regular sample
+	 * processing loop coming after that the function, otherwise
+	 * phony regular samples may be generated in the sampling buffer
+	 * not marked with the EXACT tag. Another possibility is to have
+	 * one PEBS event and at least one non-PEBS event whic hoverflows
+	 * while PEBS has armed. In this case, bit 62 of GLOBAL_STATUS will
+	 * not be set, yet the overflow status bit for the PEBS counter will
+	 * be on Skylake.
+	 *
+	 * To avoid this problem, we systematically ignore the PEBS-enabled
+	 * counters from the GLOBAL_STATUS mask and we always process PEBS
+	 * events via drain_pebs().
+	 */
+	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
+		status &= ~cpuc->pebs_enabled;
+	else
+		status &= ~(cpuc->pebs_enabled & PEBS_COUNTER_MASK);
+
+	/*
+	 * PEBS overflow sets bit 62 in the global status register
+	 */
+	if (__test_and_clear_bit(GLOBAL_STATUS_BUFFER_OVF_BIT, (unsigned long *)&status)) {
+		u64 pebs_enabled = cpuc->pebs_enabled;
+
+		handled++;
+		x86_pmu.drain_pebs(regs, &data);
+		status &= x86_pmu.intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI;
+
+		/*
+		 * PMI throttle may be triggered, which stops the PEBS event.
+		 * Although cpuc->pebs_enabled is updated accordingly, the
+		 * MSR_IA32_PEBS_ENABLE is not updated. Because the
+		 * cpuc->enabled has been forced to 0 in PMI.
+		 * Update the MSR if pebs_enabled is changed.
+		 */
+		if (pebs_enabled != cpuc->pebs_enabled)
+			wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+	}
+
+	/*
+	 * Intel PT
+	 */
+	if (__test_and_clear_bit(GLOBAL_STATUS_TRACE_TOPAPMI_BIT, (unsigned long *)&status)) {
+		handled++;
+
+		guest_cbs = perf_get_guest_cbs();
+		if (unlikely(guest_cbs && guest_cbs->is_in_guest() &&
+			     guest_cbs->handle_intel_pt_intr))
+			guest_cbs->handle_intel_pt_intr();
+		else
+			intel_pt_interrupt();
+	}
+
+	/*
+	 * Intel Perf mertrics
+	 */
+	if (__test_and_clear_bit(GLOBAL_STATUS_PERF_METRICS_OVF_BIT, (unsigned long *)&status)) {
+		handled++;
+		if (x86_pmu.update_topdown_event)
+			x86_pmu.update_topdown_event(NULL);
+	}
+
+	/*
+	 * Checkpointed counters can lead to 'spurious' PMIs because the
+	 * rollback caused by the PMI will have cleared the overflow status
+	 * bit. Therefore always force probe these counters.
+	 */
+	status |= cpuc->intel_cp_status;
+
+	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 (has_branch_stack(event))
+			data.br_stack = &cpuc->lbr_stack;
+
+		if (perf_event_overflow(event, &data, regs))
+			x86_pmu_stop(event, 0);
+	}
+
+	return handled;
+}
+
+static bool disable_counter_freezing = true;
+static int __init intel_perf_counter_freezing_setup(char *s)
+{
+	bool res;
+
+	if (kstrtobool(s, &res))
+		return -EINVAL;
+
+	disable_counter_freezing = !res;
+	return 1;
+}
+__setup("perf_v4_pmi=", intel_perf_counter_freezing_setup);
+
+/*
+ * Simplified handler for Arch Perfmon v4:
+ * - We rely on counter freezing/unfreezing to enable/disable the PMU.
+ * This is done automatically on PMU ack.
+ * - Ack the PMU only after the APIC.
+ */
+
+static int intel_pmu_handle_irq_v4(struct pt_regs *regs)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int handled = 0;
+	bool bts = false;
+	u64 status;
+	int pmu_enabled = cpuc->enabled;
+	int loops = 0;
+
+	/* PMU has been disabled because of counter freezing */
+	cpuc->enabled = 0;
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
+		bts = true;
+		intel_bts_disable_local();
+		handled = intel_pmu_drain_bts_buffer();
+		handled += intel_bts_interrupt();
+	}
+	status = intel_pmu_get_status();
+	if (!status)
+		goto done;
+again:
+	intel_pmu_lbr_read();
+	if (++loops > 100) {
+		static bool warned;
+
+		if (!warned) {
+			WARN(1, "perfevents: irq loop stuck!\n");
+			perf_event_print_debug();
+			warned = true;
+		}
+		intel_pmu_reset();
+		goto done;
+	}
+
+
+	handled += handle_pmi_common(regs, status);
+done:
+	/* Ack the PMI in the APIC */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+
+	/*
+	 * The counters start counting immediately while ack the status.
+	 * Make it as close as possible to IRET. This avoids bogus
+	 * freezing on Skylake CPUs.
+	 */
+	if (status) {
+		intel_pmu_ack_status(status);
+	} else {
+		/*
+		 * CPU may issues two PMIs very close to each other.
+		 * When the PMI handler services the first one, the
+		 * GLOBAL_STATUS is already updated to reflect both.
+		 * When it IRETs, the second PMI is immediately
+		 * handled and it sees clear status. At the meantime,
+		 * there may be a third PMI, because the freezing bit
+		 * isn't set since the ack in first PMI handlers.
+		 * Double check if there is more work to be done.
+		 */
+		status = intel_pmu_get_status();
+		if (status)
+			goto again;
+	}
+
+	if (bts)
+		intel_bts_enable_local();
+	cpuc->enabled = pmu_enabled;
+	return handled;
+}
+
+/*
+ * This handler is triggered by the local APIC, so the APIC IRQ handling
+ * rules apply:
+ */
+static int intel_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct cpu_hw_events *cpuc;
+	int loops;
+	u64 status;
+	int handled;
+	int pmu_enabled;
+
+	cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * Save the PMU state.
+	 * It needs to be restored when leaving the handler.
+	 */
+	pmu_enabled = cpuc->enabled;
+	/*
+	 * No known reason to not always do late ACK,
+	 * but just in case do it opt-in.
+	 */
+	if (!x86_pmu.late_ack)
+		apic_write(APIC_LVTPC, APIC_DM_NMI);
+	intel_bts_disable_local();
+	cpuc->enabled = 0;
+	__intel_pmu_disable_all();
+	handled = intel_pmu_drain_bts_buffer();
+	handled += intel_bts_interrupt();
+	status = intel_pmu_get_status();
+	if (!status)
+		goto done;
+
+	loops = 0;
+again:
+	intel_pmu_lbr_read();
+	intel_pmu_ack_status(status);
+	if (++loops > 100) {
+		static bool warned;
+
+		if (!warned) {
+			WARN(1, "perfevents: irq loop stuck!\n");
+			perf_event_print_debug();
+			warned = true;
+		}
+		intel_pmu_reset();
+		goto done;
+	}
+
+	handled += handle_pmi_common(regs, status);
+
+	/*
+	 * Repeat if there is more work to be done:
+	 */
+	status = intel_pmu_get_status();
+	if (status)
+		goto again;
+
+done:
+	/* Only restore PMU state when it's active. See x86_pmu_disable(). */
+	cpuc->enabled = pmu_enabled;
+	if (pmu_enabled)
+		__intel_pmu_enable_all(0, true);
+	intel_bts_enable_local();
+
+	/*
+	 * Only unmask the NMI after the overflow counters
+	 * have been reset. This avoids spurious NMIs on
+	 * Haswell CPUs.
+	 */
+	if (x86_pmu.late_ack)
+		apic_write(APIC_LVTPC, APIC_DM_NMI);
+	return handled;
+}
+
+static struct event_constraint *
+intel_bts_constraints(struct perf_event *event)
+{
+	if (unlikely(intel_pmu_has_bts(event)))
+		return &bts_constraint;
+
+	return NULL;
+}
+
+/*
+ * Note: matches a fake event, like Fixed2.
+ */
+static struct event_constraint *
+intel_vlbr_constraints(struct perf_event *event)
+{
+	struct event_constraint *c = &vlbr_constraint;
+
+	if (unlikely(constraint_match(c, event->hw.config))) {
+		event->hw.flags |= c->flags;
+		return c;
+	}
+
+	return NULL;
+}
+
+static int intel_alt_er(int idx, u64 config)
+{
+	int alt_idx = idx;
+
+	if (!(x86_pmu.flags & PMU_FL_HAS_RSP_1))
+		return idx;
+
+	if (idx == EXTRA_REG_RSP_0)
+		alt_idx = EXTRA_REG_RSP_1;
+
+	if (idx == EXTRA_REG_RSP_1)
+		alt_idx = EXTRA_REG_RSP_0;
+
+	if (config & ~x86_pmu.extra_regs[alt_idx].valid_mask)
+		return idx;
+
+	return alt_idx;
+}
+
+static void intel_fixup_er(struct perf_event *event, int idx)
+{
+	event->hw.extra_reg.idx = idx;
+
+	if (idx == EXTRA_REG_RSP_0) {
+		event->hw.config &= ~INTEL_ARCH_EVENT_MASK;
+		event->hw.config |= x86_pmu.extra_regs[EXTRA_REG_RSP_0].event;
+		event->hw.extra_reg.reg = MSR_OFFCORE_RSP_0;
+	} else if (idx == EXTRA_REG_RSP_1) {
+		event->hw.config &= ~INTEL_ARCH_EVENT_MASK;
+		event->hw.config |= x86_pmu.extra_regs[EXTRA_REG_RSP_1].event;
+		event->hw.extra_reg.reg = MSR_OFFCORE_RSP_1;
+	}
+}
+
+/*
+ * manage allocation of shared extra msr for certain events
+ *
+ * sharing can be:
+ * per-cpu: to be shared between the various events on a single PMU
+ * per-core: per-cpu + shared by HT threads
+ */
+static struct event_constraint *
+__intel_shared_reg_get_constraints(struct cpu_hw_events *cpuc,
+				   struct perf_event *event,
+				   struct hw_perf_event_extra *reg)
+{
+	struct event_constraint *c = &emptyconstraint;
+	struct er_account *era;
+	unsigned long flags;
+	int idx = reg->idx;
+
+	/*
+	 * reg->alloc can be set due to existing state, so for fake cpuc we
+	 * need to ignore this, otherwise we might fail to allocate proper fake
+	 * state for this extra reg constraint. Also see the comment below.
+	 */
+	if (reg->alloc && !cpuc->is_fake)
+		return NULL; /* call x86_get_event_constraint() */
+
+again:
+	era = &cpuc->shared_regs->regs[idx];
+	/*
+	 * we use spin_lock_irqsave() to avoid lockdep issues when
+	 * passing a fake cpuc
+	 */
+	raw_spin_lock_irqsave(&era->lock, flags);
+
+	if (!atomic_read(&era->ref) || era->config == reg->config) {
+
+		/*
+		 * If its a fake cpuc -- as per validate_{group,event}() we
+		 * shouldn't touch event state and we can avoid doing so
+		 * since both will only call get_event_constraints() once
+		 * on each event, this avoids the need for reg->alloc.
+		 *
+		 * Not doing the ER fixup will only result in era->reg being
+		 * wrong, but since we won't actually try and program hardware
+		 * this isn't a problem either.
+		 */
+		if (!cpuc->is_fake) {
+			if (idx != reg->idx)
+				intel_fixup_er(event, idx);
+
+			/*
+			 * x86_schedule_events() can call get_event_constraints()
+			 * multiple times on events in the case of incremental
+			 * scheduling(). reg->alloc ensures we only do the ER
+			 * allocation once.
+			 */
+			reg->alloc = 1;
+		}
+
+		/* lock in msr value */
+		era->config = reg->config;
+		era->reg = reg->reg;
+
+		/* one more user */
+		atomic_inc(&era->ref);
+
+		/*
+		 * need to call x86_get_event_constraint()
+		 * to check if associated event has constraints
+		 */
+		c = NULL;
+	} else {
+		idx = intel_alt_er(idx, reg->config);
+		if (idx != reg->idx) {
+			raw_spin_unlock_irqrestore(&era->lock, flags);
+			goto again;
+		}
+	}
+	raw_spin_unlock_irqrestore(&era->lock, flags);
+
+	return c;
+}
+
+static void
+__intel_shared_reg_put_constraints(struct cpu_hw_events *cpuc,
+				   struct hw_perf_event_extra *reg)
+{
+	struct er_account *era;
+
+	/*
+	 * Only put constraint if extra reg was actually allocated. Also takes
+	 * care of event which do not use an extra shared reg.
+	 *
+	 * Also, if this is a fake cpuc we shouldn't touch any event state
+	 * (reg->alloc) and we don't care about leaving inconsistent cpuc state
+	 * either since it'll be thrown out.
+	 */
+	if (!reg->alloc || cpuc->is_fake)
+		return;
+
+	era = &cpuc->shared_regs->regs[reg->idx];
+
+	/* one fewer user */
+	atomic_dec(&era->ref);
+
+	/* allocate again next time */
+	reg->alloc = 0;
+}
+
+static struct event_constraint *
+intel_shared_regs_constraints(struct cpu_hw_events *cpuc,
+			      struct perf_event *event)
+{
+	struct event_constraint *c = NULL, *d;
+	struct hw_perf_event_extra *xreg, *breg;
+
+	xreg = &event->hw.extra_reg;
+	if (xreg->idx != EXTRA_REG_NONE) {
+		c = __intel_shared_reg_get_constraints(cpuc, event, xreg);
+		if (c == &emptyconstraint)
+			return c;
+	}
+	breg = &event->hw.branch_reg;
+	if (breg->idx != EXTRA_REG_NONE) {
+		d = __intel_shared_reg_get_constraints(cpuc, event, breg);
+		if (d == &emptyconstraint) {
+			__intel_shared_reg_put_constraints(cpuc, xreg);
+			c = d;
+		}
+	}
+	return c;
+}
+
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	if (x86_pmu.event_constraints) {
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			if (constraint_match(c, event->hw.config)) {
+				event->hw.flags |= c->flags;
+				return c;
+			}
+		}
+	}
+
+	return &unconstrained;
+}
+
+static struct event_constraint *
+__intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			    struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	c = intel_vlbr_constraints(event);
+	if (c)
+		return c;
+
+	c = intel_bts_constraints(event);
+	if (c)
+		return c;
+
+	c = intel_shared_regs_constraints(cpuc, event);
+	if (c)
+		return c;
+
+	c = intel_pebs_constraints(event);
+	if (c)
+		return c;
+
+	return x86_get_event_constraints(cpuc, idx, event);
+}
+
+static void
+intel_start_scheduling(struct cpu_hw_events *cpuc)
+{
+	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
+	struct intel_excl_states *xl;
+	int tid = cpuc->excl_thread_id;
+
+	/*
+	 * nothing needed if in group validation mode
+	 */
+	if (cpuc->is_fake || !is_ht_workaround_enabled())
+		return;
+
+	/*
+	 * no exclusion needed
+	 */
+	if (WARN_ON_ONCE(!excl_cntrs))
+		return;
+
+	xl = &excl_cntrs->states[tid];
+
+	xl->sched_started = true;
+	/*
+	 * lock shared state until we are done scheduling
+	 * in stop_event_scheduling()
+	 * makes scheduling appear as a transaction
+	 */
+	raw_spin_lock(&excl_cntrs->lock);
+}
+
+static void intel_commit_scheduling(struct cpu_hw_events *cpuc, int idx, int cntr)
+{
+	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
+	struct event_constraint *c = cpuc->event_constraint[idx];
+	struct intel_excl_states *xl;
+	int tid = cpuc->excl_thread_id;
+
+	if (cpuc->is_fake || !is_ht_workaround_enabled())
+		return;
+
+	if (WARN_ON_ONCE(!excl_cntrs))
+		return;
+
+	if (!(c->flags & PERF_X86_EVENT_DYNAMIC))
+		return;
+
+	xl = &excl_cntrs->states[tid];
+
+	lockdep_assert_held(&excl_cntrs->lock);
+
+	if (c->flags & PERF_X86_EVENT_EXCL)
+		xl->state[cntr] = INTEL_EXCL_EXCLUSIVE;
+	else
+		xl->state[cntr] = INTEL_EXCL_SHARED;
+}
+
+static void
+intel_stop_scheduling(struct cpu_hw_events *cpuc)
+{
+	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
+	struct intel_excl_states *xl;
+	int tid = cpuc->excl_thread_id;
+
+	/*
+	 * nothing needed if in group validation mode
+	 */
+	if (cpuc->is_fake || !is_ht_workaround_enabled())
+		return;
+	/*
+	 * no exclusion needed
+	 */
+	if (WARN_ON_ONCE(!excl_cntrs))
+		return;
+
+	xl = &excl_cntrs->states[tid];
+
+	xl->sched_started = false;
+	/*
+	 * release shared state lock (acquired in intel_start_scheduling())
+	 */
+	raw_spin_unlock(&excl_cntrs->lock);
+}
+
+static struct event_constraint *
+dyn_constraint(struct cpu_hw_events *cpuc, struct event_constraint *c, int idx)
+{
+	WARN_ON_ONCE(!cpuc->constraint_list);
+
+	if (!(c->flags & PERF_X86_EVENT_DYNAMIC)) {
+		struct event_constraint *cx;
+
+		/*
+		 * grab pre-allocated constraint entry
+		 */
+		cx = &cpuc->constraint_list[idx];
+
+		/*
+		 * initialize dynamic constraint
+		 * with static constraint
+		 */
+		*cx = *c;
+
+		/*
+		 * mark constraint as dynamic
+		 */
+		cx->flags |= PERF_X86_EVENT_DYNAMIC;
+		c = cx;
+	}
+
+	return c;
+}
+
+static struct event_constraint *
+intel_get_excl_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
+			   int idx, struct event_constraint *c)
+{
+	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
+	struct intel_excl_states *xlo;
+	int tid = cpuc->excl_thread_id;
+	int is_excl, i, w;
+
+	/*
+	 * validating a group does not require
+	 * enforcing cross-thread  exclusion
+	 */
+	if (cpuc->is_fake || !is_ht_workaround_enabled())
+		return c;
+
+	/*
+	 * no exclusion needed
+	 */
+	if (WARN_ON_ONCE(!excl_cntrs))
+		return c;
+
+	/*
+	 * because we modify the constraint, we need
+	 * to make a copy. Static constraints come
+	 * from static const tables.
+	 *
+	 * only needed when constraint has not yet
+	 * been cloned (marked dynamic)
+	 */
+	c = dyn_constraint(cpuc, c, idx);
+
+	/*
+	 * From here on, the constraint is dynamic.
+	 * Either it was just allocated above, or it
+	 * was allocated during a earlier invocation
+	 * of this function
+	 */
+
+	/*
+	 * state of sibling HT
+	 */
+	xlo = &excl_cntrs->states[tid ^ 1];
+
+	/*
+	 * event requires exclusive counter access
+	 * across HT threads
+	 */
+	is_excl = c->flags & PERF_X86_EVENT_EXCL;
+	if (is_excl && !(event->hw.flags & PERF_X86_EVENT_EXCL_ACCT)) {
+		event->hw.flags |= PERF_X86_EVENT_EXCL_ACCT;
+		if (!cpuc->n_excl++)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 1);
+	}
+
+	/*
+	 * Modify static constraint with current dynamic
+	 * state of thread
+	 *
+	 * EXCLUSIVE: sibling counter measuring exclusive event
+	 * SHARED   : sibling counter measuring non-exclusive event
+	 * UNUSED   : sibling counter unused
+	 */
+	w = c->weight;
+	for_each_set_bit(i, c->idxmsk, X86_PMC_IDX_MAX) {
+		/*
+		 * exclusive event in sibling counter
+		 * our corresponding counter cannot be used
+		 * regardless of our event
+		 */
+		if (xlo->state[i] == INTEL_EXCL_EXCLUSIVE) {
+			__clear_bit(i, c->idxmsk);
+			w--;
+			continue;
+		}
+		/*
+		 * if measuring an exclusive event, sibling
+		 * measuring non-exclusive, then counter cannot
+		 * be used
+		 */
+		if (is_excl && xlo->state[i] == INTEL_EXCL_SHARED) {
+			__clear_bit(i, c->idxmsk);
+			w--;
+			continue;
+		}
+	}
+
+	/*
+	 * if we return an empty mask, then switch
+	 * back to static empty constraint to avoid
+	 * the cost of freeing later on
+	 */
+	if (!w)
+		c = &emptyconstraint;
+
+	c->weight = w;
+
+	return c;
+}
+
+static struct event_constraint *
+intel_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			    struct perf_event *event)
+{
+	struct event_constraint *c1, *c2;
+
+	c1 = cpuc->event_constraint[idx];
+
+	/*
+	 * first time only
+	 * - static constraint: no change across incremental scheduling calls
+	 * - dynamic constraint: handled by intel_get_excl_constraints()
+	 */
+	c2 = __intel_get_event_constraints(cpuc, idx, event);
+	if (c1) {
+	        WARN_ON_ONCE(!(c1->flags & PERF_X86_EVENT_DYNAMIC));
+		bitmap_copy(c1->idxmsk, c2->idxmsk, X86_PMC_IDX_MAX);
+		c1->weight = c2->weight;
+		c2 = c1;
+	}
+
+	if (cpuc->excl_cntrs)
+		return intel_get_excl_constraints(cpuc, event, idx, c2);
+
+	return c2;
+}
+
+static void intel_put_excl_constraints(struct cpu_hw_events *cpuc,
+		struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct intel_excl_cntrs *excl_cntrs = cpuc->excl_cntrs;
+	int tid = cpuc->excl_thread_id;
+	struct intel_excl_states *xl;
+
+	/*
+	 * nothing needed if in group validation mode
+	 */
+	if (cpuc->is_fake)
+		return;
+
+	if (WARN_ON_ONCE(!excl_cntrs))
+		return;
+
+	if (hwc->flags & PERF_X86_EVENT_EXCL_ACCT) {
+		hwc->flags &= ~PERF_X86_EVENT_EXCL_ACCT;
+		if (!--cpuc->n_excl)
+			WRITE_ONCE(excl_cntrs->has_exclusive[tid], 0);
+	}
+
+	/*
+	 * If event was actually assigned, then mark the counter state as
+	 * unused now.
+	 */
+	if (hwc->idx >= 0) {
+		xl = &excl_cntrs->states[tid];
+
+		/*
+		 * put_constraint may be called from x86_schedule_events()
+		 * which already has the lock held so here make locking
+		 * conditional.
+		 */
+		if (!xl->sched_started)
+			raw_spin_lock(&excl_cntrs->lock);
+
+		xl->state[hwc->idx] = INTEL_EXCL_UNUSED;
+
+		if (!xl->sched_started)
+			raw_spin_unlock(&excl_cntrs->lock);
+	}
+}
+
+static void
+intel_put_shared_regs_event_constraints(struct cpu_hw_events *cpuc,
+					struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg;
+
+	reg = &event->hw.extra_reg;
+	if (reg->idx != EXTRA_REG_NONE)
+		__intel_shared_reg_put_constraints(cpuc, reg);
+
+	reg = &event->hw.branch_reg;
+	if (reg->idx != EXTRA_REG_NONE)
+		__intel_shared_reg_put_constraints(cpuc, reg);
+}
+
+static void intel_put_event_constraints(struct cpu_hw_events *cpuc,
+					struct perf_event *event)
+{
+	intel_put_shared_regs_event_constraints(cpuc, event);
+
+	/*
+	 * is PMU has exclusive counter restrictions, then
+	 * all events are subject to and must call the
+	 * put_excl_constraints() routine
+	 */
+	if (cpuc->excl_cntrs)
+		intel_put_excl_constraints(cpuc, event);
+}
+
+static void intel_pebs_aliases_core2(struct perf_event *event)
+{
+	if ((event->hw.config & X86_RAW_EVENT_MASK) == 0x003c) {
+		/*
+		 * Use an alternative encoding for CPU_CLK_UNHALTED.THREAD_P
+		 * (0x003c) so that we can use it with PEBS.
+		 *
+		 * The regular CPU_CLK_UNHALTED.THREAD_P event (0x003c) isn't
+		 * PEBS capable. However we can use INST_RETIRED.ANY_P
+		 * (0x00c0), which is a PEBS capable event, to get the same
+		 * count.
+		 *
+		 * INST_RETIRED.ANY_P counts the number of cycles that retires
+		 * CNTMASK instructions. By setting CNTMASK to a value (16)
+		 * larger than the maximum number of instructions that can be
+		 * retired per cycle (4) and then inverting the condition, we
+		 * count all cycles that retire 16 or less instructions, which
+		 * is every cycle.
+		 *
+		 * Thereby we gain a PEBS capable cycle counter.
+		 */
+		u64 alt_config = X86_CONFIG(.event=0xc0, .inv=1, .cmask=16);
+
+		alt_config |= (event->hw.config & ~X86_RAW_EVENT_MASK);
+		event->hw.config = alt_config;
+	}
+}
+
+static void intel_pebs_aliases_snb(struct perf_event *event)
+{
+	if ((event->hw.config & X86_RAW_EVENT_MASK) == 0x003c) {
+		/*
+		 * Use an alternative encoding for CPU_CLK_UNHALTED.THREAD_P
+		 * (0x003c) so that we can use it with PEBS.
+		 *
+		 * The regular CPU_CLK_UNHALTED.THREAD_P event (0x003c) isn't
+		 * PEBS capable. However we can use UOPS_RETIRED.ALL
+		 * (0x01c2), which is a PEBS capable event, to get the same
+		 * count.
+		 *
+		 * UOPS_RETIRED.ALL counts the number of cycles that retires
+		 * CNTMASK micro-ops. By setting CNTMASK to a value (16)
+		 * larger than the maximum number of micro-ops that can be
+		 * retired per cycle (4) and then inverting the condition, we
+		 * count all cycles that retire 16 or less micro-ops, which
+		 * is every cycle.
+		 *
+		 * Thereby we gain a PEBS capable cycle counter.
+		 */
+		u64 alt_config = X86_CONFIG(.event=0xc2, .umask=0x01, .inv=1, .cmask=16);
+
+		alt_config |= (event->hw.config & ~X86_RAW_EVENT_MASK);
+		event->hw.config = alt_config;
+	}
+}
+
+static void intel_pebs_aliases_precdist(struct perf_event *event)
+{
+	if ((event->hw.config & X86_RAW_EVENT_MASK) == 0x003c) {
+		/*
+		 * Use an alternative encoding for CPU_CLK_UNHALTED.THREAD_P
+		 * (0x003c) so that we can use it with PEBS.
+		 *
+		 * The regular CPU_CLK_UNHALTED.THREAD_P event (0x003c) isn't
+		 * PEBS capable. However we can use INST_RETIRED.PREC_DIST
+		 * (0x01c0), which is a PEBS capable event, to get the same
+		 * count.
+		 *
+		 * The PREC_DIST event has special support to minimize sample
+		 * shadowing effects. One drawback is that it can be
+		 * only programmed on counter 1, but that seems like an
+		 * acceptable trade off.
+		 */
+		u64 alt_config = X86_CONFIG(.event=0xc0, .umask=0x01, .inv=1, .cmask=16);
+
+		alt_config |= (event->hw.config & ~X86_RAW_EVENT_MASK);
+		event->hw.config = alt_config;
+	}
+}
+
+static void intel_pebs_aliases_ivb(struct perf_event *event)
+{
+	if (event->attr.precise_ip < 3)
+		return intel_pebs_aliases_snb(event);
+	return intel_pebs_aliases_precdist(event);
+}
+
+static void intel_pebs_aliases_skl(struct perf_event *event)
+{
+	if (event->attr.precise_ip < 3)
+		return intel_pebs_aliases_core2(event);
+	return intel_pebs_aliases_precdist(event);
+}
+
+static unsigned long intel_pmu_large_pebs_flags(struct perf_event *event)
+{
+	unsigned long flags = x86_pmu.large_pebs_flags;
+
+	if (event->attr.use_clockid)
+		flags &= ~PERF_SAMPLE_TIME;
+	if (!event->attr.exclude_kernel)
+		flags &= ~PERF_SAMPLE_REGS_USER;
+	if (event->attr.sample_regs_user & ~PEBS_GP_REGS)
+		flags &= ~(PERF_SAMPLE_REGS_USER | PERF_SAMPLE_REGS_INTR);
+	return flags;
+}
+
+static int intel_pmu_bts_config(struct perf_event *event)
+{
+	struct perf_event_attr *attr = &event->attr;
+
+	if (unlikely(intel_pmu_has_bts(event))) {
+		/* BTS is not supported by this architecture. */
+		if (!x86_pmu.bts_active)
+			return -EOPNOTSUPP;
+
+		/* BTS is currently only allowed for user-mode. */
+		if (!attr->exclude_kernel)
+			return -EOPNOTSUPP;
+
+		/* BTS is not allowed for precise events. */
+		if (attr->precise_ip)
+			return -EOPNOTSUPP;
+
+		/* disallow bts if conflicting events are present */
+		if (x86_add_exclusive(x86_lbr_exclusive_lbr))
+			return -EBUSY;
+
+		event->destroy = hw_perf_lbr_event_destroy;
+	}
+
+	return 0;
+}
+
+static int core_pmu_hw_config(struct perf_event *event)
+{
+	int ret = x86_pmu_hw_config(event);
+
+	if (ret)
+		return ret;
+
+	return intel_pmu_bts_config(event);
+}
+
+static int intel_pmu_hw_config(struct perf_event *event)
+{
+	int ret = x86_pmu_hw_config(event);
+
+	if (ret)
+		return ret;
+
+	ret = intel_pmu_bts_config(event);
+	if (ret)
+		return ret;
+
+	if (event->attr.precise_ip) {
+		if ((event->attr.config & INTEL_ARCH_EVENT_MASK) == INTEL_FIXED_VLBR_EVENT)
+			return -EINVAL;
+
+		if (!(event->attr.freq || (event->attr.wakeup_events && !event->attr.watermark))) {
+			event->hw.flags |= PERF_X86_EVENT_AUTO_RELOAD;
+			if (!(event->attr.sample_type &
+			      ~intel_pmu_large_pebs_flags(event))) {
+				event->hw.flags |= PERF_X86_EVENT_LARGE_PEBS;
+				event->attach_state |= PERF_ATTACH_SCHED_CB;
+			}
+		}
+		if (x86_pmu.pebs_aliases)
+			x86_pmu.pebs_aliases(event);
+
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+			event->attr.sample_type |= __PERF_SAMPLE_CALLCHAIN_EARLY;
+	}
+
+	if (needs_branch_stack(event)) {
+		ret = intel_pmu_setup_lbr_filter(event);
+		if (ret)
+			return ret;
+		event->attach_state |= PERF_ATTACH_SCHED_CB;
+
+		/*
+		 * BTS is set up earlier in this path, so don't account twice
+		 */
+		if (!unlikely(intel_pmu_has_bts(event))) {
+			/* disallow lbr if conflicting events are present */
+			if (x86_add_exclusive(x86_lbr_exclusive_lbr))
+				return -EBUSY;
+
+			event->destroy = hw_perf_lbr_event_destroy;
+		}
+	}
+
+	if (event->attr.aux_output) {
+		if (!event->attr.precise_ip)
+			return -EINVAL;
+
+		event->hw.flags |= PERF_X86_EVENT_PEBS_VIA_PT;
+	}
+
+	if (event->attr.type != PERF_TYPE_RAW)
+		return 0;
+
+	/*
+	 * Config Topdown slots and metric events
+	 *
+	 * The slots event on Fixed Counter 3 can support sampling,
+	 * which will be handled normally in x86_perf_event_update().
+	 *
+	 * Metric events don't support sampling and require being paired
+	 * with a slots event as group leader. When the slots event
+	 * is used in a metrics group, it too cannot support sampling.
+	 */
+	if (x86_pmu.intel_cap.perf_metrics && is_topdown_event(event)) {
+		if (event->attr.config1 || event->attr.config2)
+			return -EINVAL;
+
+		/*
+		 * The TopDown metrics events and slots event don't
+		 * support any filters.
+		 */
+		if (event->attr.config & X86_ALL_EVENT_FLAGS)
+			return -EINVAL;
+
+		if (is_metric_event(event)) {
+			struct perf_event *leader = event->group_leader;
+
+			/* The metric events don't support sampling. */
+			if (is_sampling_event(event))
+				return -EINVAL;
+
+			/* The metric events require a slots group leader. */
+			if (!is_slots_event(leader))
+				return -EINVAL;
+
+			/*
+			 * The leader/SLOTS must not be a sampling event for
+			 * metric use; hardware requires it starts at 0 when used
+			 * in conjunction with MSR_PERF_METRICS.
+			 */
+			if (is_sampling_event(leader))
+				return -EINVAL;
+
+			event->event_caps |= PERF_EV_CAP_SIBLING;
+			/*
+			 * Only once we have a METRICs sibling do we
+			 * need TopDown magic.
+			 */
+			leader->hw.flags |= PERF_X86_EVENT_TOPDOWN;
+			event->hw.flags  |= PERF_X86_EVENT_TOPDOWN;
+		}
+	}
+
+	if (!(event->attr.config & ARCH_PERFMON_EVENTSEL_ANY))
+		return 0;
+
+	if (x86_pmu.version < 3)
+		return -EINVAL;
+
+	ret = perf_allow_cpu(&event->attr);
+	if (ret)
+		return ret;
+
+	event->hw.config |= ARCH_PERFMON_EVENTSEL_ANY;
+
+	return 0;
+}
+
+#ifdef CONFIG_RETPOLINE
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr);
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr);
+#endif
+
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+{
+#ifdef CONFIG_RETPOLINE
+	if (x86_pmu.guest_get_msrs == intel_guest_get_msrs)
+		return intel_guest_get_msrs(nr);
+	else if (x86_pmu.guest_get_msrs == core_guest_get_msrs)
+		return core_guest_get_msrs(nr);
+#endif
+	if (x86_pmu.guest_get_msrs)
+		return x86_pmu.guest_get_msrs(nr);
+	*nr = 0;
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(perf_guest_get_msrs);
+
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+
+	arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
+	arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
+	arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask;
+	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
+		arr[0].guest &= ~cpuc->pebs_enabled;
+	else
+		arr[0].guest &= ~(cpuc->pebs_enabled & PEBS_COUNTER_MASK);
+	*nr = 1;
+
+	if (x86_pmu.pebs && x86_pmu.pebs_no_isolation) {
+		/*
+		 * If PMU counter has PEBS enabled it is not enough to
+		 * disable counter on a guest entry since PEBS memory
+		 * write can overshoot guest entry and corrupt guest
+		 * memory. Disabling PEBS solves the problem.
+		 *
+		 * Don't do this if the CPU already enforces it.
+		 */
+		arr[1].msr = MSR_IA32_PEBS_ENABLE;
+		arr[1].host = cpuc->pebs_enabled;
+		arr[1].guest = 0;
+		*nr = 2;
+	}
+
+	return arr;
+}
+
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++)  {
+		struct perf_event *event = cpuc->events[idx];
+
+		arr[idx].msr = x86_pmu_config_addr(idx);
+		arr[idx].host = arr[idx].guest = 0;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		arr[idx].host = arr[idx].guest =
+			event->hw.config | ARCH_PERFMON_EVENTSEL_ENABLE;
+
+		if (event->attr.exclude_host)
+			arr[idx].host &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+		else if (event->attr.exclude_guest)
+			arr[idx].guest &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+	}
+
+	*nr = x86_pmu.num_counters;
+	return arr;
+}
+
+static void core_pmu_enable_event(struct perf_event *event)
+{
+	if (!event->attr.exclude_host)
+		x86_pmu_enable_event(event);
+}
+
+static void core_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+
+		if (!test_bit(idx, cpuc->active_mask) ||
+				cpuc->events[idx]->attr.exclude_host)
+			continue;
+
+		__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+	}
+}
+
+static int hsw_hw_config(struct perf_event *event)
+{
+	int ret = intel_pmu_hw_config(event);
+
+	if (ret)
+		return ret;
+	if (!boot_cpu_has(X86_FEATURE_RTM) && !boot_cpu_has(X86_FEATURE_HLE))
+		return 0;
+	event->hw.config |= event->attr.config & (HSW_IN_TX|HSW_IN_TX_CHECKPOINTED);
+
+	/*
+	 * IN_TX/IN_TX-CP filters are not supported by the Haswell PMU with
+	 * PEBS or in ANY thread mode. Since the results are non-sensical forbid
+	 * this combination.
+	 */
+	if ((event->hw.config & (HSW_IN_TX|HSW_IN_TX_CHECKPOINTED)) &&
+	     ((event->hw.config & ARCH_PERFMON_EVENTSEL_ANY) ||
+	      event->attr.precise_ip > 0))
+		return -EOPNOTSUPP;
+
+	if (event_is_checkpointed(event)) {
+		/*
+		 * Sampling of checkpointed events can cause situations where
+		 * the CPU constantly aborts because of a overflow, which is
+		 * then checkpointed back and ignored. Forbid checkpointing
+		 * for sampling.
+		 *
+		 * But still allow a long sampling period, so that perf stat
+		 * from KVM works.
+		 */
+		if (event->attr.sample_period > 0 &&
+		    event->attr.sample_period < 0x7fffffff)
+			return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static struct event_constraint counter0_constraint =
+			INTEL_ALL_EVENT_CONSTRAINT(0, 0x1);
+
+static struct event_constraint counter2_constraint =
+			EVENT_CONSTRAINT(0, 0x4, 0);
+
+static struct event_constraint fixed0_constraint =
+			FIXED_EVENT_CONSTRAINT(0x00c0, 0);
+
+static struct event_constraint fixed0_counter0_constraint =
+			INTEL_ALL_EVENT_CONSTRAINT(0, 0x100000001ULL);
+
+static struct event_constraint *
+hsw_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	c = intel_get_event_constraints(cpuc, idx, event);
+
+	/* Handle special quirk on in_tx_checkpointed only in counter 2 */
+	if (event->hw.config & HSW_IN_TX_CHECKPOINTED) {
+		if (c->idxmsk64 & (1U << 2))
+			return &counter2_constraint;
+		return &emptyconstraint;
+	}
+
+	return c;
+}
+
+static struct event_constraint *
+icl_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	/*
+	 * Fixed counter 0 has less skid.
+	 * Force instruction:ppp in Fixed counter 0
+	 */
+	if ((event->attr.precise_ip == 3) &&
+	    constraint_match(&fixed0_constraint, event->hw.config))
+		return &fixed0_constraint;
+
+	return hsw_get_event_constraints(cpuc, idx, event);
+}
+
+static struct event_constraint *
+glp_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	/* :ppp means to do reduced skid PEBS which is PMC0 only. */
+	if (event->attr.precise_ip == 3)
+		return &counter0_constraint;
+
+	c = intel_get_event_constraints(cpuc, idx, event);
+
+	return c;
+}
+
+static struct event_constraint *
+tnt_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	/*
+	 * :ppp means to do reduced skid PEBS,
+	 * which is available on PMC0 and fixed counter 0.
+	 */
+	if (event->attr.precise_ip == 3) {
+		/* Force instruction:ppp on PMC0 and Fixed counter 0 */
+		if (constraint_match(&fixed0_constraint, event->hw.config))
+			return &fixed0_counter0_constraint;
+
+		return &counter0_constraint;
+	}
+
+	c = intel_get_event_constraints(cpuc, idx, event);
+
+	return c;
+}
+
+static bool allow_tsx_force_abort = true;
+
+static struct event_constraint *
+tfa_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event)
+{
+	struct event_constraint *c = hsw_get_event_constraints(cpuc, idx, event);
+
+	/*
+	 * Without TFA we must not use PMC3.
+	 */
+	if (!allow_tsx_force_abort && test_bit(3, c->idxmsk)) {
+		c = dyn_constraint(cpuc, c, idx);
+		c->idxmsk64 &= ~(1ULL << 3);
+		c->weight--;
+	}
+
+	return c;
+}
+
+/*
+ * Broadwell:
+ *
+ * The INST_RETIRED.ALL period always needs to have lowest 6 bits cleared
+ * (BDM55) and it must not use a period smaller than 100 (BDM11). We combine
+ * the two to enforce a minimum period of 128 (the smallest value that has bits
+ * 0-5 cleared and >= 100).
+ *
+ * Because of how the code in x86_perf_event_set_period() works, the truncation
+ * of the lower 6 bits is 'harmless' as we'll occasionally add a longer period
+ * to make up for the 'lost' events due to carrying the 'error' in period_left.
+ *
+ * Therefore the effective (average) period matches the requested period,
+ * despite coarser hardware granularity.
+ */
+static u64 bdw_limit_period(struct perf_event *event, u64 left)
+{
+	if ((event->hw.config & INTEL_ARCH_EVENT_MASK) ==
+			X86_CONFIG(.event=0xc0, .umask=0x01)) {
+		if (left < 128)
+			left = 128;
+		left &= ~0x3fULL;
+	}
+	return left;
+}
+
+static u64 nhm_limit_period(struct perf_event *event, u64 left)
+{
+	return max(left, 32ULL);
+}
+
+PMU_FORMAT_ATTR(event,	"config:0-7"	);
+PMU_FORMAT_ATTR(umask,	"config:8-15"	);
+PMU_FORMAT_ATTR(edge,	"config:18"	);
+PMU_FORMAT_ATTR(pc,	"config:19"	);
+PMU_FORMAT_ATTR(any,	"config:21"	); /* v3 + */
+PMU_FORMAT_ATTR(inv,	"config:23"	);
+PMU_FORMAT_ATTR(cmask,	"config:24-31"	);
+PMU_FORMAT_ATTR(in_tx,  "config:32");
+PMU_FORMAT_ATTR(in_tx_cp, "config:33");
+
+static struct attribute *intel_arch_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_pc.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+ssize_t intel_event_sysfs_show(char *page, u64 config)
+{
+	u64 event = (config & ARCH_PERFMON_EVENTSEL_EVENT);
+
+	return x86_event_sysfs_show(page, config, event);
+}
+
+static struct intel_shared_regs *allocate_shared_regs(int cpu)
+{
+	struct intel_shared_regs *regs;
+	int i;
+
+	regs = kzalloc_node(sizeof(struct intel_shared_regs),
+			    GFP_KERNEL, cpu_to_node(cpu));
+	if (regs) {
+		/*
+		 * initialize the locks to keep lockdep happy
+		 */
+		for (i = 0; i < EXTRA_REG_MAX; i++)
+			raw_spin_lock_init(&regs->regs[i].lock);
+
+		regs->core_id = -1;
+	}
+	return regs;
+}
+
+static struct intel_excl_cntrs *allocate_excl_cntrs(int cpu)
+{
+	struct intel_excl_cntrs *c;
+
+	c = kzalloc_node(sizeof(struct intel_excl_cntrs),
+			 GFP_KERNEL, cpu_to_node(cpu));
+	if (c) {
+		raw_spin_lock_init(&c->lock);
+		c->core_id = -1;
+	}
+	return c;
+}
+
+
+int intel_cpuc_prepare(struct cpu_hw_events *cpuc, int cpu)
+{
+	cpuc->pebs_record_size = x86_pmu.pebs_record_size;
+
+	if (x86_pmu.extra_regs || x86_pmu.lbr_sel_map) {
+		cpuc->shared_regs = allocate_shared_regs(cpu);
+		if (!cpuc->shared_regs)
+			goto err;
+	}
+
+	if (x86_pmu.flags & (PMU_FL_EXCL_CNTRS | PMU_FL_TFA)) {
+		size_t sz = X86_PMC_IDX_MAX * sizeof(struct event_constraint);
+
+		cpuc->constraint_list = kzalloc_node(sz, GFP_KERNEL, cpu_to_node(cpu));
+		if (!cpuc->constraint_list)
+			goto err_shared_regs;
+	}
+
+	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
+		cpuc->excl_cntrs = allocate_excl_cntrs(cpu);
+		if (!cpuc->excl_cntrs)
+			goto err_constraint_list;
+
+		cpuc->excl_thread_id = 0;
+	}
+
+	return 0;
+
+err_constraint_list:
+	kfree(cpuc->constraint_list);
+	cpuc->constraint_list = NULL;
+
+err_shared_regs:
+	kfree(cpuc->shared_regs);
+	cpuc->shared_regs = NULL;
+
+err:
+	return -ENOMEM;
+}
+
+static int intel_pmu_cpu_prepare(int cpu)
+{
+	return intel_cpuc_prepare(&per_cpu(cpu_hw_events, cpu), cpu);
+}
+
+static void flip_smm_bit(void *data)
+{
+	unsigned long set = *(unsigned long *)data;
+
+	if (set > 0) {
+		msr_set_bit(MSR_IA32_DEBUGCTLMSR,
+			    DEBUGCTLMSR_FREEZE_IN_SMM_BIT);
+	} else {
+		msr_clear_bit(MSR_IA32_DEBUGCTLMSR,
+			      DEBUGCTLMSR_FREEZE_IN_SMM_BIT);
+	}
+}
+
+static void intel_pmu_cpu_starting(int cpu)
+{
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+	int core_id = topology_core_id(cpu);
+	int i;
+
+	init_debug_store_on_cpu(cpu);
+	/*
+	 * Deal with CPUs that don't clear their LBRs on power-up.
+	 */
+	intel_pmu_lbr_reset();
+
+	cpuc->lbr_sel = NULL;
+
+	if (x86_pmu.flags & PMU_FL_TFA) {
+		WARN_ON_ONCE(cpuc->tfa_shadow);
+		cpuc->tfa_shadow = ~0ULL;
+		intel_set_tfa(cpuc, false);
+	}
+
+	if (x86_pmu.version > 1)
+		flip_smm_bit(&x86_pmu.attr_freeze_on_smi);
+
+	if (x86_pmu.counter_freezing)
+		enable_counter_freeze();
+
+	/* Disable perf metrics if any added CPU doesn't support it. */
+	if (x86_pmu.intel_cap.perf_metrics) {
+		union perf_capabilities perf_cap;
+
+		rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_cap.capabilities);
+		if (!perf_cap.perf_metrics) {
+			x86_pmu.intel_cap.perf_metrics = 0;
+			x86_pmu.intel_ctrl &= ~(1ULL << GLOBAL_CTRL_EN_PERF_METRICS);
+		}
+	}
+
+	if (!cpuc->shared_regs)
+		return;
+
+	if (!(x86_pmu.flags & PMU_FL_NO_HT_SHARING)) {
+		for_each_cpu(i, topology_sibling_cpumask(cpu)) {
+			struct intel_shared_regs *pc;
+
+			pc = per_cpu(cpu_hw_events, i).shared_regs;
+			if (pc && pc->core_id == core_id) {
+				cpuc->kfree_on_online[0] = cpuc->shared_regs;
+				cpuc->shared_regs = pc;
+				break;
+			}
+		}
+		cpuc->shared_regs->core_id = core_id;
+		cpuc->shared_regs->refcnt++;
+	}
+
+	if (x86_pmu.lbr_sel_map)
+		cpuc->lbr_sel = &cpuc->shared_regs->regs[EXTRA_REG_LBR];
+
+	if (x86_pmu.flags & PMU_FL_EXCL_CNTRS) {
+		for_each_cpu(i, topology_sibling_cpumask(cpu)) {
+			struct cpu_hw_events *sibling;
+			struct intel_excl_cntrs *c;
+
+			sibling = &per_cpu(cpu_hw_events, i);
+			c = sibling->excl_cntrs;
+			if (c && c->core_id == core_id) {
+				cpuc->kfree_on_online[1] = cpuc->excl_cntrs;
+				cpuc->excl_cntrs = c;
+				if (!sibling->excl_thread_id)
+					cpuc->excl_thread_id = 1;
+				break;
+			}
+		}
+		cpuc->excl_cntrs->core_id = core_id;
+		cpuc->excl_cntrs->refcnt++;
+	}
+}
+
+static void free_excl_cntrs(struct cpu_hw_events *cpuc)
+{
+	struct intel_excl_cntrs *c;
+
+	c = cpuc->excl_cntrs;
+	if (c) {
+		if (c->core_id == -1 || --c->refcnt == 0)
+			kfree(c);
+		cpuc->excl_cntrs = NULL;
+	}
+
+	kfree(cpuc->constraint_list);
+	cpuc->constraint_list = NULL;
+}
+
+static void intel_pmu_cpu_dying(int cpu)
+{
+	fini_debug_store_on_cpu(cpu);
+
+	if (x86_pmu.counter_freezing)
+		disable_counter_freeze();
+}
+
+void intel_cpuc_finish(struct cpu_hw_events *cpuc)
+{
+	struct intel_shared_regs *pc;
+
+	pc = cpuc->shared_regs;
+	if (pc) {
+		if (pc->core_id == -1 || --pc->refcnt == 0)
+			kfree(pc);
+		cpuc->shared_regs = NULL;
+	}
+
+	free_excl_cntrs(cpuc);
+}
+
+static void intel_pmu_cpu_dead(int cpu)
+{
+	intel_cpuc_finish(&per_cpu(cpu_hw_events, cpu));
+}
+
+static void intel_pmu_sched_task(struct perf_event_context *ctx,
+				 bool sched_in)
+{
+	intel_pmu_pebs_sched_task(ctx, sched_in);
+	intel_pmu_lbr_sched_task(ctx, sched_in);
+}
+
+static void intel_pmu_swap_task_ctx(struct perf_event_context *prev,
+				    struct perf_event_context *next)
+{
+	intel_pmu_lbr_swap_task_ctx(prev, next);
+}
+
+static int intel_pmu_check_period(struct perf_event *event, u64 value)
+{
+	return intel_pmu_has_bts_period(event, value) ? -EINVAL : 0;
+}
+
+static int intel_pmu_aux_output_match(struct perf_event *event)
+{
+	if (!x86_pmu.intel_cap.pebs_output_pt_available)
+		return 0;
+
+	return is_intel_pt_event(event);
+}
+
+PMU_FORMAT_ATTR(offcore_rsp, "config1:0-63");
+
+PMU_FORMAT_ATTR(ldlat, "config1:0-15");
+
+PMU_FORMAT_ATTR(frontend, "config1:0-23");
+
+static struct attribute *intel_arch3_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_pc.attr,
+	&format_attr_any.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+static struct attribute *hsw_format_attr[] = {
+	&format_attr_in_tx.attr,
+	&format_attr_in_tx_cp.attr,
+	&format_attr_offcore_rsp.attr,
+	&format_attr_ldlat.attr,
+	NULL
+};
+
+static struct attribute *nhm_format_attr[] = {
+	&format_attr_offcore_rsp.attr,
+	&format_attr_ldlat.attr,
+	NULL
+};
+
+static struct attribute *slm_format_attr[] = {
+	&format_attr_offcore_rsp.attr,
+	NULL
+};
+
+static struct attribute *skl_format_attr[] = {
+	&format_attr_frontend.attr,
+	NULL,
+};
+
+static __initconst const struct x86_pmu core_pmu = {
+	.name			= "core",
+	.handle_irq		= x86_pmu_handle_irq,
+	.disable_all		= x86_pmu_disable_all,
+	.enable_all		= core_pmu_enable_all,
+	.enable			= core_pmu_enable_event,
+	.disable		= x86_pmu_disable_event,
+	.hw_config		= core_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
+	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
+	.event_map		= intel_pmu_event_map,
+	.max_events		= ARRAY_SIZE(intel_perfmon_event_map),
+	.apic			= 1,
+	.large_pebs_flags	= LARGE_PEBS_FLAGS,
+
+	/*
+	 * Intel PMCs cannot be accessed sanely above 32-bit width,
+	 * so we install an artificial 1<<31 period regardless of
+	 * the generic event period:
+	 */
+	.max_period		= (1ULL<<31) - 1,
+	.get_event_constraints	= intel_get_event_constraints,
+	.put_event_constraints	= intel_put_event_constraints,
+	.event_constraints	= intel_core_event_constraints,
+	.guest_get_msrs		= core_guest_get_msrs,
+	.format_attrs		= intel_arch_formats_attr,
+	.events_sysfs_show	= intel_event_sysfs_show,
+
+	/*
+	 * Virtual (or funny metal) CPU can define x86_pmu.extra_regs
+	 * together with PMU version 1 and thus be using core_pmu with
+	 * shared_regs. We need following callbacks here to allocate
+	 * it properly.
+	 */
+	.cpu_prepare		= intel_pmu_cpu_prepare,
+	.cpu_starting		= intel_pmu_cpu_starting,
+	.cpu_dying		= intel_pmu_cpu_dying,
+	.cpu_dead		= intel_pmu_cpu_dead,
+
+	.check_period		= intel_pmu_check_period,
+
+	.lbr_reset		= intel_pmu_lbr_reset_64,
+	.lbr_read		= intel_pmu_lbr_read_64,
+	.lbr_save		= intel_pmu_lbr_save,
+	.lbr_restore		= intel_pmu_lbr_restore,
+};
+
+static __initconst const struct x86_pmu intel_pmu = {
+	.name			= "Intel",
+	.handle_irq		= intel_pmu_handle_irq,
+	.disable_all		= intel_pmu_disable_all,
+	.enable_all		= intel_pmu_enable_all,
+	.enable			= intel_pmu_enable_event,
+	.disable		= intel_pmu_disable_event,
+	.add			= intel_pmu_add_event,
+	.del			= intel_pmu_del_event,
+	.read			= intel_pmu_read_event,
+	.hw_config		= intel_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
+	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
+	.event_map		= intel_pmu_event_map,
+	.max_events		= ARRAY_SIZE(intel_perfmon_event_map),
+	.apic			= 1,
+	.large_pebs_flags	= LARGE_PEBS_FLAGS,
+	/*
+	 * Intel PMCs cannot be accessed sanely above 32 bit width,
+	 * so we install an artificial 1<<31 period regardless of
+	 * the generic event period:
+	 */
+	.max_period		= (1ULL << 31) - 1,
+	.get_event_constraints	= intel_get_event_constraints,
+	.put_event_constraints	= intel_put_event_constraints,
+	.pebs_aliases		= intel_pebs_aliases_core2,
+
+	.format_attrs		= intel_arch3_formats_attr,
+	.events_sysfs_show	= intel_event_sysfs_show,
+
+	.cpu_prepare		= intel_pmu_cpu_prepare,
+	.cpu_starting		= intel_pmu_cpu_starting,
+	.cpu_dying		= intel_pmu_cpu_dying,
+	.cpu_dead		= intel_pmu_cpu_dead,
+
+	.guest_get_msrs		= intel_guest_get_msrs,
+	.sched_task		= intel_pmu_sched_task,
+	.swap_task_ctx		= intel_pmu_swap_task_ctx,
+
+	.check_period		= intel_pmu_check_period,
+
+	.aux_output_match	= intel_pmu_aux_output_match,
+
+	.lbr_reset		= intel_pmu_lbr_reset_64,
+	.lbr_read		= intel_pmu_lbr_read_64,
+	.lbr_save		= intel_pmu_lbr_save,
+	.lbr_restore		= intel_pmu_lbr_restore,
+
+	/*
+	 * SMM has access to all 4 rings and while traditionally SMM code only
+	 * ran in CPL0, 2021-era firmware is starting to make use of CPL3 in SMM.
+	 *
+	 * Since the EVENTSEL.{USR,OS} CPL filtering makes no distinction
+	 * between SMM or not, this results in what should be pure userspace
+	 * counters including SMM data.
+	 *
+	 * This is a clear privilege issue, therefore globally disable
+	 * counting SMM by default.
+	 */
+	.attr_freeze_on_smi	= 1,
+};
+
+static __init void intel_clovertown_quirk(void)
+{
+	/*
+	 * PEBS is unreliable due to:
+	 *
+	 *   AJ67  - PEBS may experience CPL leaks
+	 *   AJ68  - PEBS PMI may be delayed by one event
+	 *   AJ69  - GLOBAL_STATUS[62] will only be set when DEBUGCTL[12]
+	 *   AJ106 - FREEZE_LBRS_ON_PMI doesn't work in combination with PEBS
+	 *
+	 * AJ67 could be worked around by restricting the OS/USR flags.
+	 * AJ69 could be worked around by setting PMU_FREEZE_ON_PMI.
+	 *
+	 * AJ106 could possibly be worked around by not allowing LBR
+	 *       usage from PEBS, including the fixup.
+	 * AJ68  could possibly be worked around by always programming
+	 *	 a pebs_event_reset[0] value and coping with the lost events.
+	 *
+	 * But taken together it might just make sense to not enable PEBS on
+	 * these chips.
+	 */
+	pr_warn("PEBS disabled due to CPU errata\n");
+	x86_pmu.pebs = 0;
+	x86_pmu.pebs_constraints = NULL;
+}
+
+static const struct x86_cpu_desc isolation_ucodes[] = {
+	INTEL_CPU_DESC(INTEL_FAM6_HASWELL,		 3, 0x0000001f),
+	INTEL_CPU_DESC(INTEL_FAM6_HASWELL_L,		 1, 0x0000001e),
+	INTEL_CPU_DESC(INTEL_FAM6_HASWELL_G,		 1, 0x00000015),
+	INTEL_CPU_DESC(INTEL_FAM6_HASWELL_X,		 2, 0x00000037),
+	INTEL_CPU_DESC(INTEL_FAM6_HASWELL_X,		 4, 0x0000000a),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL,		 4, 0x00000023),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_G,		 1, 0x00000014),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_D,		 2, 0x00000010),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_D,		 3, 0x07000009),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_D,		 4, 0x0f000009),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_D,		 5, 0x0e000002),
+	INTEL_CPU_DESC(INTEL_FAM6_BROADWELL_X,		 1, 0x0b000014),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		 3, 0x00000021),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		 4, 0x00000000),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		 5, 0x00000000),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		 6, 0x00000000),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		 7, 0x00000000),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_X,		11, 0x00000000),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE_L,		 3, 0x0000007c),
+	INTEL_CPU_DESC(INTEL_FAM6_SKYLAKE,		 3, 0x0000007c),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE,		 9, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE_L,		 9, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE_L,		10, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE_L,		11, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE_L,		12, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE,		10, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE,		11, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE,		12, 0x0000004e),
+	INTEL_CPU_DESC(INTEL_FAM6_KABYLAKE,		13, 0x0000004e),
+	{}
+};
+
+static void intel_check_pebs_isolation(void)
+{
+	x86_pmu.pebs_no_isolation = !x86_cpu_has_min_microcode_rev(isolation_ucodes);
+}
+
+static __init void intel_pebs_isolation_quirk(void)
+{
+	WARN_ON_ONCE(x86_pmu.check_microcode);
+	x86_pmu.check_microcode = intel_check_pebs_isolation;
+	intel_check_pebs_isolation();
+}
+
+static const struct x86_cpu_desc pebs_ucodes[] = {
+	INTEL_CPU_DESC(INTEL_FAM6_SANDYBRIDGE,		7, 0x00000028),
+	INTEL_CPU_DESC(INTEL_FAM6_SANDYBRIDGE_X,	6, 0x00000618),
+	INTEL_CPU_DESC(INTEL_FAM6_SANDYBRIDGE_X,	7, 0x0000070c),
+	{}
+};
+
+static bool intel_snb_pebs_broken(void)
+{
+	return !x86_cpu_has_min_microcode_rev(pebs_ucodes);
+}
+
+static void intel_snb_check_microcode(void)
+{
+	if (intel_snb_pebs_broken() == x86_pmu.pebs_broken)
+		return;
+
+	/*
+	 * Serialized by the microcode lock..
+	 */
+	if (x86_pmu.pebs_broken) {
+		pr_info("PEBS enabled due to microcode update\n");
+		x86_pmu.pebs_broken = 0;
+	} else {
+		pr_info("PEBS disabled due to CPU errata, please upgrade microcode\n");
+		x86_pmu.pebs_broken = 1;
+	}
+}
+
+static bool is_lbr_from(unsigned long msr)
+{
+	unsigned long lbr_from_nr = x86_pmu.lbr_from + x86_pmu.lbr_nr;
+
+	return x86_pmu.lbr_from <= msr && msr < lbr_from_nr;
+}
+
+/*
+ * Under certain circumstances, access certain MSR may cause #GP.
+ * The function tests if the input MSR can be safely accessed.
+ */
+static bool check_msr(unsigned long msr, u64 mask)
+{
+	u64 val_old, val_new, val_tmp;
+
+	/*
+	 * Disable the check for real HW, so we don't
+	 * mess with potentionaly enabled registers:
+	 */
+	if (!boot_cpu_has(X86_FEATURE_HYPERVISOR))
+		return true;
+
+	/*
+	 * 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.
+	 */
+	if (rdmsrl_safe(msr, &val_old))
+		return false;
+
+	/*
+	 * Only change the bits which can be updated by wrmsrl.
+	 */
+	val_tmp = val_old ^ mask;
+
+	if (is_lbr_from(msr))
+		val_tmp = lbr_from_signext_quirk_wr(val_tmp);
+
+	if (wrmsrl_safe(msr, val_tmp) ||
+	    rdmsrl_safe(msr, &val_new))
+		return false;
+
+	/*
+	 * Quirk only affects validation in wrmsr(), so wrmsrl()'s value
+	 * should equal rdmsrl()'s even with the quirk.
+	 */
+	if (val_new != val_tmp)
+		return false;
+
+	if (is_lbr_from(msr))
+		val_old = lbr_from_signext_quirk_wr(val_old);
+
+	/* Here it's sure that the MSR can be safely accessed.
+	 * Restore the old value and return.
+	 */
+	wrmsrl(msr, val_old);
+
+	return true;
+}
+
+static __init void intel_sandybridge_quirk(void)
+{
+	x86_pmu.check_microcode = intel_snb_check_microcode;
+	cpus_read_lock();
+	intel_snb_check_microcode();
+	cpus_read_unlock();
+}
+
+static const struct { int id; char *name; } intel_arch_events_map[] __initconst = {
+	{ PERF_COUNT_HW_CPU_CYCLES, "cpu cycles" },
+	{ PERF_COUNT_HW_INSTRUCTIONS, "instructions" },
+	{ PERF_COUNT_HW_BUS_CYCLES, "bus cycles" },
+	{ PERF_COUNT_HW_CACHE_REFERENCES, "cache references" },
+	{ PERF_COUNT_HW_CACHE_MISSES, "cache misses" },
+	{ PERF_COUNT_HW_BRANCH_INSTRUCTIONS, "branch instructions" },
+	{ PERF_COUNT_HW_BRANCH_MISSES, "branch misses" },
+};
+
+static __init void intel_arch_events_quirk(void)
+{
+	int bit;
+
+	/* disable event that reported as not presend by cpuid */
+	for_each_set_bit(bit, x86_pmu.events_mask, ARRAY_SIZE(intel_arch_events_map)) {
+		intel_perfmon_event_map[intel_arch_events_map[bit].id] = 0;
+		pr_warn("CPUID marked event: \'%s\' unavailable\n",
+			intel_arch_events_map[bit].name);
+	}
+}
+
+static __init void intel_nehalem_quirk(void)
+{
+	union cpuid10_ebx ebx;
+
+	ebx.full = x86_pmu.events_maskl;
+	if (ebx.split.no_branch_misses_retired) {
+		/*
+		 * Erratum AAJ80 detected, we work it around by using
+		 * the BR_MISP_EXEC.ANY event. This will over-count
+		 * branch-misses, but it's still much better than the
+		 * architectural event which is often completely bogus:
+		 */
+		intel_perfmon_event_map[PERF_COUNT_HW_BRANCH_MISSES] = 0x7f89;
+		ebx.split.no_branch_misses_retired = 0;
+		x86_pmu.events_maskl = ebx.full;
+		pr_info("CPU erratum AAJ80 worked around\n");
+	}
+}
+
+static const struct x86_cpu_desc counter_freezing_ucodes[] = {
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT,	 2, 0x0000000e),
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT,	 9, 0x0000002e),
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT,	10, 0x00000008),
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT_D,	 1, 0x00000028),
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT_PLUS,	 1, 0x00000028),
+	INTEL_CPU_DESC(INTEL_FAM6_ATOM_GOLDMONT_PLUS,	 8, 0x00000006),
+	{}
+};
+
+static bool intel_counter_freezing_broken(void)
+{
+	return !x86_cpu_has_min_microcode_rev(counter_freezing_ucodes);
+}
+
+static __init void intel_counter_freezing_quirk(void)
+{
+	/* Check if it's already disabled */
+	if (disable_counter_freezing)
+		return;
+
+	/*
+	 * If the system starts with the wrong ucode, leave the
+	 * counter-freezing feature permanently disabled.
+	 */
+	if (intel_counter_freezing_broken()) {
+		pr_info("PMU counter freezing disabled due to CPU errata,"
+			"please upgrade microcode\n");
+		x86_pmu.counter_freezing = false;
+		x86_pmu.handle_irq = intel_pmu_handle_irq;
+	}
+}
+
+/*
+ * enable software workaround for errata:
+ * SNB: BJ122
+ * IVB: BV98
+ * HSW: HSD29
+ *
+ * Only needed when HT is enabled. However detecting
+ * if HT is enabled is difficult (model specific). So instead,
+ * we enable the workaround in the early boot, and verify if
+ * it is needed in a later initcall phase once we have valid
+ * topology information to check if HT is actually enabled
+ */
+static __init void intel_ht_bug(void)
+{
+	x86_pmu.flags |= PMU_FL_EXCL_CNTRS | PMU_FL_EXCL_ENABLED;
+
+	x86_pmu.start_scheduling = intel_start_scheduling;
+	x86_pmu.commit_scheduling = intel_commit_scheduling;
+	x86_pmu.stop_scheduling = intel_stop_scheduling;
+}
+
+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")
+
+/* Haswell special events */
+EVENT_ATTR_STR(tx-start,	tx_start,	"event=0xc9,umask=0x1");
+EVENT_ATTR_STR(tx-commit,	tx_commit,	"event=0xc9,umask=0x2");
+EVENT_ATTR_STR(tx-abort,	tx_abort,	"event=0xc9,umask=0x4");
+EVENT_ATTR_STR(tx-capacity,	tx_capacity,	"event=0x54,umask=0x2");
+EVENT_ATTR_STR(tx-conflict,	tx_conflict,	"event=0x54,umask=0x1");
+EVENT_ATTR_STR(el-start,	el_start,	"event=0xc8,umask=0x1");
+EVENT_ATTR_STR(el-commit,	el_commit,	"event=0xc8,umask=0x2");
+EVENT_ATTR_STR(el-abort,	el_abort,	"event=0xc8,umask=0x4");
+EVENT_ATTR_STR(el-capacity,	el_capacity,	"event=0x54,umask=0x2");
+EVENT_ATTR_STR(el-conflict,	el_conflict,	"event=0x54,umask=0x1");
+EVENT_ATTR_STR(cycles-t,	cycles_t,	"event=0x3c,in_tx=1");
+EVENT_ATTR_STR(cycles-ct,	cycles_ct,	"event=0x3c,in_tx=1,in_tx_cp=1");
+
+static struct attribute *hsw_events_attrs[] = {
+	EVENT_PTR(td_slots_issued),
+	EVENT_PTR(td_slots_retired),
+	EVENT_PTR(td_fetch_bubbles),
+	EVENT_PTR(td_total_slots),
+	EVENT_PTR(td_total_slots_scale),
+	EVENT_PTR(td_recovery_bubbles),
+	EVENT_PTR(td_recovery_bubbles_scale),
+	NULL
+};
+
+static struct attribute *hsw_mem_events_attrs[] = {
+	EVENT_PTR(mem_ld_hsw),
+	EVENT_PTR(mem_st_hsw),
+	NULL,
+};
+
+static struct attribute *hsw_tsx_events_attrs[] = {
+	EVENT_PTR(tx_start),
+	EVENT_PTR(tx_commit),
+	EVENT_PTR(tx_abort),
+	EVENT_PTR(tx_capacity),
+	EVENT_PTR(tx_conflict),
+	EVENT_PTR(el_start),
+	EVENT_PTR(el_commit),
+	EVENT_PTR(el_abort),
+	EVENT_PTR(el_capacity),
+	EVENT_PTR(el_conflict),
+	EVENT_PTR(cycles_t),
+	EVENT_PTR(cycles_ct),
+	NULL
+};
+
+EVENT_ATTR_STR(tx-capacity-read,  tx_capacity_read,  "event=0x54,umask=0x80");
+EVENT_ATTR_STR(tx-capacity-write, tx_capacity_write, "event=0x54,umask=0x2");
+EVENT_ATTR_STR(el-capacity-read,  el_capacity_read,  "event=0x54,umask=0x80");
+EVENT_ATTR_STR(el-capacity-write, el_capacity_write, "event=0x54,umask=0x2");
+
+static struct attribute *icl_events_attrs[] = {
+	EVENT_PTR(mem_ld_hsw),
+	EVENT_PTR(mem_st_hsw),
+	NULL,
+};
+
+static struct attribute *icl_td_events_attrs[] = {
+	EVENT_PTR(slots),
+	EVENT_PTR(td_retiring),
+	EVENT_PTR(td_bad_spec),
+	EVENT_PTR(td_fe_bound),
+	EVENT_PTR(td_be_bound),
+	NULL,
+};
+
+static struct attribute *icl_tsx_events_attrs[] = {
+	EVENT_PTR(tx_start),
+	EVENT_PTR(tx_abort),
+	EVENT_PTR(tx_commit),
+	EVENT_PTR(tx_capacity_read),
+	EVENT_PTR(tx_capacity_write),
+	EVENT_PTR(tx_conflict),
+	EVENT_PTR(el_start),
+	EVENT_PTR(el_abort),
+	EVENT_PTR(el_commit),
+	EVENT_PTR(el_capacity_read),
+	EVENT_PTR(el_capacity_write),
+	EVENT_PTR(el_conflict),
+	EVENT_PTR(cycles_t),
+	EVENT_PTR(cycles_ct),
+	NULL,
+};
+
+static ssize_t freeze_on_smi_show(struct device *cdev,
+				  struct device_attribute *attr,
+				  char *buf)
+{
+	return sprintf(buf, "%lu\n", x86_pmu.attr_freeze_on_smi);
+}
+
+static DEFINE_MUTEX(freeze_on_smi_mutex);
+
+static ssize_t freeze_on_smi_store(struct device *cdev,
+				   struct device_attribute *attr,
+				   const char *buf, size_t count)
+{
+	unsigned long val;
+	ssize_t ret;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val > 1)
+		return -EINVAL;
+
+	mutex_lock(&freeze_on_smi_mutex);
+
+	if (x86_pmu.attr_freeze_on_smi == val)
+		goto done;
+
+	x86_pmu.attr_freeze_on_smi = val;
+
+	get_online_cpus();
+	on_each_cpu(flip_smm_bit, &val, 1);
+	put_online_cpus();
+done:
+	mutex_unlock(&freeze_on_smi_mutex);
+
+	return count;
+}
+
+static void update_tfa_sched(void *ignored)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * check if PMC3 is used
+	 * and if so force schedule out for all event types all contexts
+	 */
+	if (test_bit(3, cpuc->active_mask))
+		perf_pmu_resched(x86_get_pmu(smp_processor_id()));
+}
+
+static ssize_t show_sysctl_tfa(struct device *cdev,
+			      struct device_attribute *attr,
+			      char *buf)
+{
+	return snprintf(buf, 40, "%d\n", allow_tsx_force_abort);
+}
+
+static ssize_t set_sysctl_tfa(struct device *cdev,
+			      struct device_attribute *attr,
+			      const char *buf, size_t count)
+{
+	bool val;
+	ssize_t ret;
+
+	ret = kstrtobool(buf, &val);
+	if (ret)
+		return ret;
+
+	/* no change */
+	if (val == allow_tsx_force_abort)
+		return count;
+
+	allow_tsx_force_abort = val;
+
+	get_online_cpus();
+	on_each_cpu(update_tfa_sched, NULL, 1);
+	put_online_cpus();
+
+	return count;
+}
+
+
+static DEVICE_ATTR_RW(freeze_on_smi);
+
+static ssize_t branches_show(struct device *cdev,
+			     struct device_attribute *attr,
+			     char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", x86_pmu.lbr_nr);
+}
+
+static DEVICE_ATTR_RO(branches);
+
+static struct attribute *lbr_attrs[] = {
+	&dev_attr_branches.attr,
+	NULL
+};
+
+static char pmu_name_str[30];
+
+static ssize_t pmu_name_show(struct device *cdev,
+			     struct device_attribute *attr,
+			     char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%s\n", pmu_name_str);
+}
+
+static DEVICE_ATTR_RO(pmu_name);
+
+static struct attribute *intel_pmu_caps_attrs[] = {
+       &dev_attr_pmu_name.attr,
+       NULL
+};
+
+static DEVICE_ATTR(allow_tsx_force_abort, 0644,
+		   show_sysctl_tfa,
+		   set_sysctl_tfa);
+
+static struct attribute *intel_pmu_attrs[] = {
+	&dev_attr_freeze_on_smi.attr,
+	&dev_attr_allow_tsx_force_abort.attr,
+	NULL,
+};
+
+static umode_t
+tsx_is_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return boot_cpu_has(X86_FEATURE_RTM) ? attr->mode : 0;
+}
+
+static umode_t
+pebs_is_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return x86_pmu.pebs ? attr->mode : 0;
+}
+
+static umode_t
+lbr_is_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return x86_pmu.lbr_nr ? attr->mode : 0;
+}
+
+static umode_t
+exra_is_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return x86_pmu.version >= 2 ? attr->mode : 0;
+}
+
+static umode_t
+default_is_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	if (attr == &dev_attr_allow_tsx_force_abort.attr)
+		return x86_pmu.flags & PMU_FL_TFA ? attr->mode : 0;
+
+	return attr->mode;
+}
+
+static struct attribute_group group_events_td  = {
+	.name = "events",
+};
+
+static struct attribute_group group_events_mem = {
+	.name       = "events",
+	.is_visible = pebs_is_visible,
+};
+
+static struct attribute_group group_events_tsx = {
+	.name       = "events",
+	.is_visible = tsx_is_visible,
+};
+
+static struct attribute_group group_caps_gen = {
+	.name  = "caps",
+	.attrs = intel_pmu_caps_attrs,
+};
+
+static struct attribute_group group_caps_lbr = {
+	.name       = "caps",
+	.attrs	    = lbr_attrs,
+	.is_visible = lbr_is_visible,
+};
+
+static struct attribute_group group_format_extra = {
+	.name       = "format",
+	.is_visible = exra_is_visible,
+};
+
+static struct attribute_group group_format_extra_skl = {
+	.name       = "format",
+	.is_visible = exra_is_visible,
+};
+
+static struct attribute_group group_default = {
+	.attrs      = intel_pmu_attrs,
+	.is_visible = default_is_visible,
+};
+
+static const struct attribute_group *attr_update[] = {
+	&group_events_td,
+	&group_events_mem,
+	&group_events_tsx,
+	&group_caps_gen,
+	&group_caps_lbr,
+	&group_format_extra,
+	&group_format_extra_skl,
+	&group_default,
+	NULL,
+};
+
+static struct attribute *empty_attrs;
+
+__init int intel_pmu_init(void)
+{
+	struct attribute **extra_skl_attr = &empty_attrs;
+	struct attribute **extra_attr = &empty_attrs;
+	struct attribute **td_attr    = &empty_attrs;
+	struct attribute **mem_attr   = &empty_attrs;
+	struct attribute **tsx_attr   = &empty_attrs;
+	union cpuid10_edx edx;
+	union cpuid10_eax eax;
+	union cpuid10_ebx ebx;
+	struct event_constraint *c;
+	unsigned int unused;
+	struct extra_reg *er;
+	bool pmem = false;
+	int version, i;
+	char *name;
+
+	if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+		switch (boot_cpu_data.x86) {
+		case 0x6:
+			return p6_pmu_init();
+		case 0xb:
+			return knc_pmu_init();
+		case 0xf:
+			return p4_pmu_init();
+		}
+		return -ENODEV;
+	}
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Branch Misses Retired hw_event or not.
+	 */
+	cpuid(10, &eax.full, &ebx.full, &unused, &edx.full);
+	if (eax.split.mask_length < ARCH_PERFMON_EVENTS_COUNT)
+		return -ENODEV;
+
+	version = eax.split.version_id;
+	if (version < 2)
+		x86_pmu = core_pmu;
+	else
+		x86_pmu = intel_pmu;
+
+	x86_pmu.version			= version;
+	x86_pmu.num_counters		= eax.split.num_counters;
+	x86_pmu.cntval_bits		= eax.split.bit_width;
+	x86_pmu.cntval_mask		= (1ULL << eax.split.bit_width) - 1;
+
+	x86_pmu.events_maskl		= ebx.full;
+	x86_pmu.events_mask_len		= eax.split.mask_length;
+
+	x86_pmu.max_pebs_events		= min_t(unsigned, MAX_PEBS_EVENTS, x86_pmu.num_counters);
+
+	/*
+	 * Quirk: v2 perfmon does not report fixed-purpose events, so
+	 * assume at least 3 events, when not running in a hypervisor:
+	 */
+	if (version > 1) {
+		int assume = 3 * !boot_cpu_has(X86_FEATURE_HYPERVISOR);
+
+		x86_pmu.num_counters_fixed =
+			max((int)edx.split.num_counters_fixed, assume);
+	}
+
+	if (version >= 4)
+		x86_pmu.counter_freezing = !disable_counter_freezing;
+
+	if (boot_cpu_has(X86_FEATURE_PDCM)) {
+		u64 capabilities;
+
+		rdmsrl(MSR_IA32_PERF_CAPABILITIES, capabilities);
+		x86_pmu.intel_cap.capabilities = capabilities;
+	}
+
+	if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_32) {
+		x86_pmu.lbr_reset = intel_pmu_lbr_reset_32;
+		x86_pmu.lbr_read = intel_pmu_lbr_read_32;
+	}
+
+	if (boot_cpu_has(X86_FEATURE_ARCH_LBR))
+		intel_pmu_arch_lbr_init();
+
+	intel_ds_init();
+
+	x86_add_quirk(intel_arch_events_quirk); /* Install first, so it runs last */
+
+	if (version >= 5) {
+		x86_pmu.intel_cap.anythread_deprecated = edx.split.anythread_deprecated;
+		if (x86_pmu.intel_cap.anythread_deprecated)
+			pr_cont(" AnyThread deprecated, ");
+	}
+
+	/*
+	 * Install the hw-cache-events table:
+	 */
+	switch (boot_cpu_data.x86_model) {
+	case INTEL_FAM6_CORE_YONAH:
+		pr_cont("Core events, ");
+		name = "core";
+		break;
+
+	case INTEL_FAM6_CORE2_MEROM:
+		x86_add_quirk(intel_clovertown_quirk);
+		fallthrough;
+
+	case INTEL_FAM6_CORE2_MEROM_L:
+	case INTEL_FAM6_CORE2_PENRYN:
+	case INTEL_FAM6_CORE2_DUNNINGTON:
+		memcpy(hw_cache_event_ids, core2_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		intel_pmu_lbr_init_core();
+
+		x86_pmu.event_constraints = intel_core2_event_constraints;
+		x86_pmu.pebs_constraints = intel_core2_pebs_event_constraints;
+		pr_cont("Core2 events, ");
+		name = "core2";
+		break;
+
+	case INTEL_FAM6_NEHALEM:
+	case INTEL_FAM6_NEHALEM_EP:
+	case INTEL_FAM6_NEHALEM_EX:
+		memcpy(hw_cache_event_ids, nehalem_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, nehalem_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_nhm();
+
+		x86_pmu.event_constraints = intel_nehalem_event_constraints;
+		x86_pmu.pebs_constraints = intel_nehalem_pebs_event_constraints;
+		x86_pmu.enable_all = intel_pmu_nhm_enable_all;
+		x86_pmu.extra_regs = intel_nehalem_extra_regs;
+		x86_pmu.limit_period = nhm_limit_period;
+
+		mem_attr = nhm_mem_events_attrs;
+
+		/* UOPS_ISSUED.STALLED_CYCLES */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
+		/* UOPS_EXECUTED.CORE_ACTIVE_CYCLES,c=1,i=1 */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] =
+			X86_CONFIG(.event=0xb1, .umask=0x3f, .inv=1, .cmask=1);
+
+		intel_pmu_pebs_data_source_nhm();
+		x86_add_quirk(intel_nehalem_quirk);
+		x86_pmu.pebs_no_tlb = 1;
+		extra_attr = nhm_format_attr;
+
+		pr_cont("Nehalem events, ");
+		name = "nehalem";
+		break;
+
+	case INTEL_FAM6_ATOM_BONNELL:
+	case INTEL_FAM6_ATOM_BONNELL_MID:
+	case INTEL_FAM6_ATOM_SALTWELL:
+	case INTEL_FAM6_ATOM_SALTWELL_MID:
+	case INTEL_FAM6_ATOM_SALTWELL_TABLET:
+		memcpy(hw_cache_event_ids, atom_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+
+		intel_pmu_lbr_init_atom();
+
+		x86_pmu.event_constraints = intel_gen_event_constraints;
+		x86_pmu.pebs_constraints = intel_atom_pebs_event_constraints;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_core2;
+		pr_cont("Atom events, ");
+		name = "bonnell";
+		break;
+
+	case INTEL_FAM6_ATOM_SILVERMONT:
+	case INTEL_FAM6_ATOM_SILVERMONT_D:
+	case INTEL_FAM6_ATOM_SILVERMONT_MID:
+	case INTEL_FAM6_ATOM_AIRMONT:
+	case INTEL_FAM6_ATOM_AIRMONT_MID:
+		memcpy(hw_cache_event_ids, slm_hw_cache_event_ids,
+			sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, slm_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_slm();
+
+		x86_pmu.event_constraints = intel_slm_event_constraints;
+		x86_pmu.pebs_constraints = intel_slm_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_slm_extra_regs;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		td_attr = slm_events_attrs;
+		extra_attr = slm_format_attr;
+		pr_cont("Silvermont events, ");
+		name = "silvermont";
+		break;
+
+	case INTEL_FAM6_ATOM_GOLDMONT:
+	case INTEL_FAM6_ATOM_GOLDMONT_D:
+		x86_add_quirk(intel_counter_freezing_quirk);
+		memcpy(hw_cache_event_ids, glm_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, glm_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_skl();
+
+		x86_pmu.event_constraints = intel_slm_event_constraints;
+		x86_pmu.pebs_constraints = intel_glm_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_glm_extra_regs;
+		/*
+		 * It's recommended to use CPU_CLK_UNHALTED.CORE_P + NPEBS
+		 * for precise cycles.
+		 * :pp is identical to :ppp
+		 */
+		x86_pmu.pebs_aliases = NULL;
+		x86_pmu.pebs_prec_dist = true;
+		x86_pmu.lbr_pt_coexist = true;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		td_attr = glm_events_attrs;
+		extra_attr = slm_format_attr;
+		pr_cont("Goldmont events, ");
+		name = "goldmont";
+		break;
+
+	case INTEL_FAM6_ATOM_GOLDMONT_PLUS:
+		x86_add_quirk(intel_counter_freezing_quirk);
+		memcpy(hw_cache_event_ids, glp_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, glp_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_skl();
+
+		x86_pmu.event_constraints = intel_slm_event_constraints;
+		x86_pmu.extra_regs = intel_glm_extra_regs;
+		/*
+		 * It's recommended to use CPU_CLK_UNHALTED.CORE_P + NPEBS
+		 * for precise cycles.
+		 */
+		x86_pmu.pebs_aliases = NULL;
+		x86_pmu.pebs_prec_dist = true;
+		x86_pmu.lbr_pt_coexist = true;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_PEBS_ALL;
+		x86_pmu.get_event_constraints = glp_get_event_constraints;
+		td_attr = glm_events_attrs;
+		/* Goldmont Plus has 4-wide pipeline */
+		event_attr_td_total_slots_scale_glm.event_str = "4";
+		extra_attr = slm_format_attr;
+		pr_cont("Goldmont plus events, ");
+		name = "goldmont_plus";
+		break;
+
+	case INTEL_FAM6_ATOM_TREMONT_D:
+	case INTEL_FAM6_ATOM_TREMONT:
+	case INTEL_FAM6_ATOM_TREMONT_L:
+		x86_pmu.late_ack = true;
+		memcpy(hw_cache_event_ids, glp_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, tnt_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+		hw_cache_event_ids[C(ITLB)][C(OP_READ)][C(RESULT_ACCESS)] = -1;
+
+		intel_pmu_lbr_init_skl();
+
+		x86_pmu.event_constraints = intel_slm_event_constraints;
+		x86_pmu.extra_regs = intel_tnt_extra_regs;
+		/*
+		 * It's recommended to use CPU_CLK_UNHALTED.CORE_P + NPEBS
+		 * for precise cycles.
+		 */
+		x86_pmu.pebs_aliases = NULL;
+		x86_pmu.pebs_prec_dist = true;
+		x86_pmu.lbr_pt_coexist = true;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.get_event_constraints = tnt_get_event_constraints;
+		extra_attr = slm_format_attr;
+		pr_cont("Tremont events, ");
+		name = "Tremont";
+		break;
+
+	case INTEL_FAM6_WESTMERE:
+	case INTEL_FAM6_WESTMERE_EP:
+	case INTEL_FAM6_WESTMERE_EX:
+		memcpy(hw_cache_event_ids, westmere_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, nehalem_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_nhm();
+
+		x86_pmu.event_constraints = intel_westmere_event_constraints;
+		x86_pmu.enable_all = intel_pmu_nhm_enable_all;
+		x86_pmu.pebs_constraints = intel_westmere_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_westmere_extra_regs;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+
+		mem_attr = nhm_mem_events_attrs;
+
+		/* UOPS_ISSUED.STALLED_CYCLES */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
+		/* UOPS_EXECUTED.CORE_ACTIVE_CYCLES,c=1,i=1 */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] =
+			X86_CONFIG(.event=0xb1, .umask=0x3f, .inv=1, .cmask=1);
+
+		intel_pmu_pebs_data_source_nhm();
+		extra_attr = nhm_format_attr;
+		pr_cont("Westmere events, ");
+		name = "westmere";
+		break;
+
+	case INTEL_FAM6_SANDYBRIDGE:
+	case INTEL_FAM6_SANDYBRIDGE_X:
+		x86_add_quirk(intel_sandybridge_quirk);
+		x86_add_quirk(intel_ht_bug);
+		memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_snb();
+
+		x86_pmu.event_constraints = intel_snb_event_constraints;
+		x86_pmu.pebs_constraints = intel_snb_pebs_event_constraints;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_snb;
+		if (boot_cpu_data.x86_model == INTEL_FAM6_SANDYBRIDGE_X)
+			x86_pmu.extra_regs = intel_snbep_extra_regs;
+		else
+			x86_pmu.extra_regs = intel_snb_extra_regs;
+
+
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		td_attr  = snb_events_attrs;
+		mem_attr = snb_mem_events_attrs;
+
+		/* UOPS_ISSUED.ANY,c=1,i=1 to count stall cycles */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
+		/* UOPS_DISPATCHED.THREAD,c=1,i=1 to count stall cycles*/
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] =
+			X86_CONFIG(.event=0xb1, .umask=0x01, .inv=1, .cmask=1);
+
+		extra_attr = nhm_format_attr;
+
+		pr_cont("SandyBridge events, ");
+		name = "sandybridge";
+		break;
+
+	case INTEL_FAM6_IVYBRIDGE:
+	case INTEL_FAM6_IVYBRIDGE_X:
+		x86_add_quirk(intel_ht_bug);
+		memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		/* dTLB-load-misses on IVB is different than SNB */
+		hw_cache_event_ids[C(DTLB)][C(OP_READ)][C(RESULT_MISS)] = 0x8108; /* DTLB_LOAD_MISSES.DEMAND_LD_MISS_CAUSES_A_WALK */
+
+		memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_snb();
+
+		x86_pmu.event_constraints = intel_ivb_event_constraints;
+		x86_pmu.pebs_constraints = intel_ivb_pebs_event_constraints;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_ivb;
+		x86_pmu.pebs_prec_dist = true;
+		if (boot_cpu_data.x86_model == INTEL_FAM6_IVYBRIDGE_X)
+			x86_pmu.extra_regs = intel_snbep_extra_regs;
+		else
+			x86_pmu.extra_regs = intel_snb_extra_regs;
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		td_attr  = snb_events_attrs;
+		mem_attr = snb_mem_events_attrs;
+
+		/* UOPS_ISSUED.ANY,c=1,i=1 to count stall cycles */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
+
+		extra_attr = nhm_format_attr;
+
+		pr_cont("IvyBridge events, ");
+		name = "ivybridge";
+		break;
+
+
+	case INTEL_FAM6_HASWELL:
+	case INTEL_FAM6_HASWELL_X:
+	case INTEL_FAM6_HASWELL_L:
+	case INTEL_FAM6_HASWELL_G:
+		x86_add_quirk(intel_ht_bug);
+		x86_add_quirk(intel_pebs_isolation_quirk);
+		x86_pmu.late_ack = true;
+		memcpy(hw_cache_event_ids, hsw_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, hsw_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_hsw();
+
+		x86_pmu.event_constraints = intel_hsw_event_constraints;
+		x86_pmu.pebs_constraints = intel_hsw_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_snbep_extra_regs;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_ivb;
+		x86_pmu.pebs_prec_dist = true;
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		x86_pmu.hw_config = hsw_hw_config;
+		x86_pmu.get_event_constraints = hsw_get_event_constraints;
+		x86_pmu.lbr_double_abort = true;
+		extra_attr = boot_cpu_has(X86_FEATURE_RTM) ?
+			hsw_format_attr : nhm_format_attr;
+		td_attr  = hsw_events_attrs;
+		mem_attr = hsw_mem_events_attrs;
+		tsx_attr = hsw_tsx_events_attrs;
+		pr_cont("Haswell events, ");
+		name = "haswell";
+		break;
+
+	case INTEL_FAM6_BROADWELL:
+	case INTEL_FAM6_BROADWELL_D:
+	case INTEL_FAM6_BROADWELL_G:
+	case INTEL_FAM6_BROADWELL_X:
+		x86_add_quirk(intel_pebs_isolation_quirk);
+		x86_pmu.late_ack = true;
+		memcpy(hw_cache_event_ids, hsw_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, hsw_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
+
+		/* L3_MISS_LOCAL_DRAM is BIT(26) in Broadwell */
+		hw_cache_extra_regs[C(LL)][C(OP_READ)][C(RESULT_MISS)] = HSW_DEMAND_READ |
+									 BDW_L3_MISS|HSW_SNOOP_DRAM;
+		hw_cache_extra_regs[C(LL)][C(OP_WRITE)][C(RESULT_MISS)] = HSW_DEMAND_WRITE|BDW_L3_MISS|
+									  HSW_SNOOP_DRAM;
+		hw_cache_extra_regs[C(NODE)][C(OP_READ)][C(RESULT_ACCESS)] = HSW_DEMAND_READ|
+									     BDW_L3_MISS_LOCAL|HSW_SNOOP_DRAM;
+		hw_cache_extra_regs[C(NODE)][C(OP_WRITE)][C(RESULT_ACCESS)] = HSW_DEMAND_WRITE|
+									      BDW_L3_MISS_LOCAL|HSW_SNOOP_DRAM;
+
+		intel_pmu_lbr_init_hsw();
+
+		x86_pmu.event_constraints = intel_bdw_event_constraints;
+		x86_pmu.pebs_constraints = intel_bdw_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_snbep_extra_regs;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_ivb;
+		x86_pmu.pebs_prec_dist = true;
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		x86_pmu.hw_config = hsw_hw_config;
+		x86_pmu.get_event_constraints = hsw_get_event_constraints;
+		x86_pmu.limit_period = bdw_limit_period;
+		extra_attr = boot_cpu_has(X86_FEATURE_RTM) ?
+			hsw_format_attr : nhm_format_attr;
+		td_attr  = hsw_events_attrs;
+		mem_attr = hsw_mem_events_attrs;
+		tsx_attr = hsw_tsx_events_attrs;
+		pr_cont("Broadwell events, ");
+		name = "broadwell";
+		break;
+
+	case INTEL_FAM6_XEON_PHI_KNL:
+	case INTEL_FAM6_XEON_PHI_KNM:
+		memcpy(hw_cache_event_ids,
+		       slm_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs,
+		       knl_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
+		intel_pmu_lbr_init_knl();
+
+		x86_pmu.event_constraints = intel_slm_event_constraints;
+		x86_pmu.pebs_constraints = intel_slm_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_knl_extra_regs;
+
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+		extra_attr = slm_format_attr;
+		pr_cont("Knights Landing/Mill events, ");
+		name = "knights-landing";
+		break;
+
+	case INTEL_FAM6_SKYLAKE_X:
+		pmem = true;
+		fallthrough;
+	case INTEL_FAM6_SKYLAKE_L:
+	case INTEL_FAM6_SKYLAKE:
+	case INTEL_FAM6_KABYLAKE_L:
+	case INTEL_FAM6_KABYLAKE:
+	case INTEL_FAM6_COMETLAKE_L:
+	case INTEL_FAM6_COMETLAKE:
+		x86_add_quirk(intel_pebs_isolation_quirk);
+		x86_pmu.late_ack = true;
+		memcpy(hw_cache_event_ids, skl_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, skl_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
+		intel_pmu_lbr_init_skl();
+
+		/* INT_MISC.RECOVERY_CYCLES has umask 1 in Skylake */
+		event_attr_td_recovery_bubbles.event_str_noht =
+			"event=0xd,umask=0x1,cmask=1";
+		event_attr_td_recovery_bubbles.event_str_ht =
+			"event=0xd,umask=0x1,cmask=1,any=1";
+
+		x86_pmu.event_constraints = intel_skl_event_constraints;
+		x86_pmu.pebs_constraints = intel_skl_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_skl_extra_regs;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_skl;
+		x86_pmu.pebs_prec_dist = true;
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		x86_pmu.hw_config = hsw_hw_config;
+		x86_pmu.get_event_constraints = hsw_get_event_constraints;
+		extra_attr = boot_cpu_has(X86_FEATURE_RTM) ?
+			hsw_format_attr : nhm_format_attr;
+		extra_skl_attr = skl_format_attr;
+		td_attr  = hsw_events_attrs;
+		mem_attr = hsw_mem_events_attrs;
+		tsx_attr = hsw_tsx_events_attrs;
+		intel_pmu_pebs_data_source_skl(pmem);
+
+		if (boot_cpu_has(X86_FEATURE_TSX_FORCE_ABORT)) {
+			x86_pmu.flags |= PMU_FL_TFA;
+			x86_pmu.get_event_constraints = tfa_get_event_constraints;
+			x86_pmu.enable_all = intel_tfa_pmu_enable_all;
+			x86_pmu.commit_scheduling = intel_tfa_commit_scheduling;
+		}
+
+		pr_cont("Skylake events, ");
+		name = "skylake";
+		break;
+
+	case INTEL_FAM6_ICELAKE_X:
+	case INTEL_FAM6_ICELAKE_D:
+		pmem = true;
+		fallthrough;
+	case INTEL_FAM6_ICELAKE_L:
+	case INTEL_FAM6_ICELAKE:
+	case INTEL_FAM6_TIGERLAKE_L:
+	case INTEL_FAM6_TIGERLAKE:
+		x86_pmu.late_ack = true;
+		memcpy(hw_cache_event_ids, skl_hw_cache_event_ids, sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, skl_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
+		hw_cache_event_ids[C(ITLB)][C(OP_READ)][C(RESULT_ACCESS)] = -1;
+		intel_pmu_lbr_init_skl();
+
+		x86_pmu.event_constraints = intel_icl_event_constraints;
+		x86_pmu.pebs_constraints = intel_icl_pebs_event_constraints;
+		x86_pmu.extra_regs = intel_icl_extra_regs;
+		x86_pmu.pebs_aliases = NULL;
+		x86_pmu.pebs_prec_dist = true;
+		x86_pmu.flags |= PMU_FL_HAS_RSP_1;
+		x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
+
+		x86_pmu.hw_config = hsw_hw_config;
+		x86_pmu.get_event_constraints = icl_get_event_constraints;
+		extra_attr = boot_cpu_has(X86_FEATURE_RTM) ?
+			hsw_format_attr : nhm_format_attr;
+		extra_skl_attr = skl_format_attr;
+		mem_attr = icl_events_attrs;
+		td_attr = icl_td_events_attrs;
+		tsx_attr = icl_tsx_events_attrs;
+		x86_pmu.rtm_abort_event = X86_CONFIG(.event=0xc9, .umask=0x04);
+		x86_pmu.lbr_pt_coexist = true;
+		intel_pmu_pebs_data_source_skl(pmem);
+		x86_pmu.update_topdown_event = icl_update_topdown_event;
+		x86_pmu.set_topdown_event_period = icl_set_topdown_event_period;
+		pr_cont("Icelake events, ");
+		name = "icelake";
+		break;
+
+	default:
+		switch (x86_pmu.version) {
+		case 1:
+			x86_pmu.event_constraints = intel_v1_event_constraints;
+			pr_cont("generic architected perfmon v1, ");
+			name = "generic_arch_v1";
+			break;
+		default:
+			/*
+			 * default constraints for v2 and up
+			 */
+			x86_pmu.event_constraints = intel_gen_event_constraints;
+			pr_cont("generic architected perfmon, ");
+			name = "generic_arch_v2+";
+			break;
+		}
+	}
+
+	snprintf(pmu_name_str, sizeof(pmu_name_str), "%s", name);
+
+
+	group_events_td.attrs  = td_attr;
+	group_events_mem.attrs = mem_attr;
+	group_events_tsx.attrs = tsx_attr;
+	group_format_extra.attrs = extra_attr;
+	group_format_extra_skl.attrs = extra_skl_attr;
+
+	x86_pmu.attr_update = attr_update;
+
+	if (x86_pmu.num_counters > INTEL_PMC_MAX_GENERIC) {
+		WARN(1, KERN_ERR "hw perf events %d > max(%d), clipping!",
+		     x86_pmu.num_counters, INTEL_PMC_MAX_GENERIC);
+		x86_pmu.num_counters = INTEL_PMC_MAX_GENERIC;
+	}
+	x86_pmu.intel_ctrl = (1ULL << x86_pmu.num_counters) - 1;
+
+	if (x86_pmu.num_counters_fixed > INTEL_PMC_MAX_FIXED) {
+		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
+		     x86_pmu.num_counters_fixed, INTEL_PMC_MAX_FIXED);
+		x86_pmu.num_counters_fixed = INTEL_PMC_MAX_FIXED;
+	}
+
+	x86_pmu.intel_ctrl |=
+		((1LL << x86_pmu.num_counters_fixed)-1) << INTEL_PMC_IDX_FIXED;
+
+	/* AnyThread may be deprecated on arch perfmon v5 or later */
+	if (x86_pmu.intel_cap.anythread_deprecated)
+		x86_pmu.format_attrs = intel_arch_formats_attr;
+
+	if (x86_pmu.event_constraints) {
+		/*
+		 * event on fixed counter2 (REF_CYCLES) only works on this
+		 * counter, so do not extend mask to generic counters
+		 */
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			/*
+			 * Don't extend the topdown slots and metrics
+			 * events to the generic counters.
+			 */
+			if (c->idxmsk64 & INTEL_PMC_MSK_TOPDOWN) {
+				c->weight = hweight64(c->idxmsk64);
+				continue;
+			}
+
+			if (c->cmask == FIXED_EVENT_FLAGS
+			    && c->idxmsk64 != INTEL_PMC_MSK_FIXED_REF_CYCLES) {
+				c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
+			}
+			c->idxmsk64 &=
+				~(~0ULL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed));
+			c->weight = hweight64(c->idxmsk64);
+		}
+	}
+
+	/*
+	 * Access LBR MSR may cause #GP under certain circumstances.
+	 * E.g. KVM doesn't support LBR MSR
+	 * Check all LBT MSR here.
+	 * Disable LBR access if any LBR MSRs can not be accessed.
+	 */
+	if (x86_pmu.lbr_tos && !check_msr(x86_pmu.lbr_tos, 0x3UL))
+		x86_pmu.lbr_nr = 0;
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		if (!(check_msr(x86_pmu.lbr_from + i, 0xffffUL) &&
+		      check_msr(x86_pmu.lbr_to + i, 0xffffUL)))
+			x86_pmu.lbr_nr = 0;
+	}
+
+	if (x86_pmu.lbr_nr)
+		pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr);
+
+	/*
+	 * Access extra MSR may cause #GP under certain circumstances.
+	 * E.g. KVM doesn't support offcore event
+	 * Check all extra_regs here.
+	 */
+	if (x86_pmu.extra_regs) {
+		for (er = x86_pmu.extra_regs; er->msr; er++) {
+			er->extra_msr_access = check_msr(er->msr, 0x11UL);
+			/* Disable LBR select mapping */
+			if ((er->idx == EXTRA_REG_LBR) && !er->extra_msr_access)
+				x86_pmu.lbr_sel_map = NULL;
+		}
+	}
+
+	/* Support full width counters using alternative MSR range */
+	if (x86_pmu.intel_cap.full_width_write) {
+		x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
+		x86_pmu.perfctr = MSR_IA32_PMC0;
+		pr_cont("full-width counters, ");
+	}
+
+	/*
+	 * For arch perfmon 4 use counter freezing to avoid
+	 * several MSR accesses in the PMI.
+	 */
+	if (x86_pmu.counter_freezing)
+		x86_pmu.handle_irq = intel_pmu_handle_irq_v4;
+
+	if (x86_pmu.intel_cap.perf_metrics)
+		x86_pmu.intel_ctrl |= 1ULL << GLOBAL_CTRL_EN_PERF_METRICS;
+
+	return 0;
+}
+
+/*
+ * HT bug: phase 2 init
+ * Called once we have valid topology information to check
+ * whether or not HT is enabled
+ * If HT is off, then we disable the workaround
+ */
+static __init int fixup_ht_bug(void)
+{
+	int c;
+	/*
+	 * problem not present on this CPU model, nothing to do
+	 */
+	if (!(x86_pmu.flags & PMU_FL_EXCL_ENABLED))
+		return 0;
+
+	if (topology_max_smt_threads() > 1) {
+		pr_info("PMU erratum BJ122, BV98, HSD29 worked around, HT is on\n");
+		return 0;
+	}
+
+	cpus_read_lock();
+
+	hardlockup_detector_perf_stop();
+
+	x86_pmu.flags &= ~(PMU_FL_EXCL_CNTRS | PMU_FL_EXCL_ENABLED);
+
+	x86_pmu.start_scheduling = NULL;
+	x86_pmu.commit_scheduling = NULL;
+	x86_pmu.stop_scheduling = NULL;
+
+	hardlockup_detector_perf_restart();
+
+	for_each_online_cpu(c)
+		free_excl_cntrs(&per_cpu(cpu_hw_events, c));
+
+	cpus_read_unlock();
+	pr_info("PMU erratum BJ122, BV98, HSD29 workaround disabled, HT off\n");
+	return 0;
+}
+subsys_initcall(fixup_ht_bug)
diff --git a/arch/x86/events/intel/cstate.c b/arch/x86/events/intel/cstate.c
new file mode 100644
index 000000000..4eb7ee5fe
--- /dev/null
+++ b/arch/x86/events/intel/cstate.c
@@ -0,0 +1,752 @@
+/*
+ * Support cstate residency counters
+ *
+ * Copyright (C) 2015, Intel Corp.
+ * Author: Kan Liang (kan.liang@intel.com)
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Library General Public
+ * License as published by the Free Software Foundation; either
+ * version 2 of the License, or (at your option) any later version.
+ *
+ * This library is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Library General Public License for more details.
+ *
+ */
+
+/*
+ * This file export cstate related free running (read-only) counters
+ * for perf. These counters may be use simultaneously by other tools,
+ * such as turbostat. However, it still make sense to implement them
+ * in perf. Because we can conveniently collect them together with
+ * other events, and allow to use them from tools without special MSR
+ * access code.
+ *
+ * The events only support system-wide mode counting. There is no
+ * sampling support because it is not supported by the hardware.
+ *
+ * According to counters' scope and category, two PMUs are registered
+ * with the perf_event core subsystem.
+ *  - 'cstate_core': The counter is available for each physical core.
+ *    The counters include CORE_C*_RESIDENCY.
+ *  - 'cstate_pkg': The counter is available for each physical package.
+ *    The counters include PKG_C*_RESIDENCY.
+ *
+ * All of these counters are specified in the Intel® 64 and IA-32
+ * Architectures Software Developer.s Manual Vol3b.
+ *
+ * Model specific counters:
+ *	MSR_CORE_C1_RES: CORE C1 Residency Counter
+ *			 perf code: 0x00
+ *			 Available model: SLM,AMT,GLM,CNL,TNT
+ *			 Scope: Core (each processor core has a MSR)
+ *	MSR_CORE_C3_RESIDENCY: CORE C3 Residency Counter
+ *			       perf code: 0x01
+ *			       Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL,GLM,
+ *						CNL,KBL,CML,TNT
+ *			       Scope: Core
+ *	MSR_CORE_C6_RESIDENCY: CORE C6 Residency Counter
+ *			       perf code: 0x02
+ *			       Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,
+ *						SKL,KNL,GLM,CNL,KBL,CML,ICL,TGL,
+ *						TNT
+ *			       Scope: Core
+ *	MSR_CORE_C7_RESIDENCY: CORE C7 Residency Counter
+ *			       perf code: 0x03
+ *			       Available model: SNB,IVB,HSW,BDW,SKL,CNL,KBL,CML,
+ *						ICL,TGL
+ *			       Scope: Core
+ *	MSR_PKG_C2_RESIDENCY:  Package C2 Residency Counter.
+ *			       perf code: 0x00
+ *			       Available model: SNB,IVB,HSW,BDW,SKL,KNL,GLM,CNL,
+ *						KBL,CML,ICL,TGL,TNT
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C3_RESIDENCY:  Package C3 Residency Counter.
+ *			       perf code: 0x01
+ *			       Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL,KNL,
+ *						GLM,CNL,KBL,CML,ICL,TGL,TNT
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C6_RESIDENCY:  Package C6 Residency Counter.
+ *			       perf code: 0x02
+ *			       Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,
+ *						SKL,KNL,GLM,CNL,KBL,CML,ICL,TGL,
+ *						TNT
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C7_RESIDENCY:  Package C7 Residency Counter.
+ *			       perf code: 0x03
+ *			       Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL,CNL,
+ *						KBL,CML,ICL,TGL
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C8_RESIDENCY:  Package C8 Residency Counter.
+ *			       perf code: 0x04
+ *			       Available model: HSW ULT,KBL,CNL,CML,ICL,TGL
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C9_RESIDENCY:  Package C9 Residency Counter.
+ *			       perf code: 0x05
+ *			       Available model: HSW ULT,KBL,CNL,CML,ICL,TGL
+ *			       Scope: Package (physical package)
+ *	MSR_PKG_C10_RESIDENCY: Package C10 Residency Counter.
+ *			       perf code: 0x06
+ *			       Available model: HSW ULT,KBL,GLM,CNL,CML,ICL,TGL,
+ *						TNT
+ *			       Scope: Package (physical package)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/perf_event.h>
+#include <linux/nospec.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include "../perf_event.h"
+#include "../probe.h"
+
+MODULE_LICENSE("GPL");
+
+#define DEFINE_CSTATE_FORMAT_ATTR(_var, _name, _format)		\
+static ssize_t __cstate_##_var##_show(struct device *dev,	\
+				struct device_attribute *attr,	\
+				char *page)			\
+{								\
+	BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE);		\
+	return sprintf(page, _format "\n");			\
+}								\
+static struct device_attribute format_attr_##_var =		\
+	__ATTR(_name, 0444, __cstate_##_var##_show, NULL)
+
+static ssize_t cstate_get_attr_cpumask(struct device *dev,
+				       struct device_attribute *attr,
+				       char *buf);
+
+/* Model -> events mapping */
+struct cstate_model {
+	unsigned long		core_events;
+	unsigned long		pkg_events;
+	unsigned long		quirks;
+};
+
+/* Quirk flags */
+#define SLM_PKG_C6_USE_C7_MSR	(1UL << 0)
+#define KNL_CORE_C6_MSR		(1UL << 1)
+
+struct perf_cstate_msr {
+	u64	msr;
+	struct	perf_pmu_events_attr *attr;
+};
+
+
+/* cstate_core PMU */
+static struct pmu cstate_core_pmu;
+static bool has_cstate_core;
+
+enum perf_cstate_core_events {
+	PERF_CSTATE_CORE_C1_RES = 0,
+	PERF_CSTATE_CORE_C3_RES,
+	PERF_CSTATE_CORE_C6_RES,
+	PERF_CSTATE_CORE_C7_RES,
+
+	PERF_CSTATE_CORE_EVENT_MAX,
+};
+
+PMU_EVENT_ATTR_STRING(c1-residency, attr_cstate_core_c1, "event=0x00");
+PMU_EVENT_ATTR_STRING(c3-residency, attr_cstate_core_c3, "event=0x01");
+PMU_EVENT_ATTR_STRING(c6-residency, attr_cstate_core_c6, "event=0x02");
+PMU_EVENT_ATTR_STRING(c7-residency, attr_cstate_core_c7, "event=0x03");
+
+static unsigned long core_msr_mask;
+
+PMU_EVENT_GROUP(events, cstate_core_c1);
+PMU_EVENT_GROUP(events, cstate_core_c3);
+PMU_EVENT_GROUP(events, cstate_core_c6);
+PMU_EVENT_GROUP(events, cstate_core_c7);
+
+static bool test_msr(int idx, void *data)
+{
+	return test_bit(idx, (unsigned long *) data);
+}
+
+static struct perf_msr core_msr[] = {
+	[PERF_CSTATE_CORE_C1_RES] = { MSR_CORE_C1_RES,		&group_cstate_core_c1,	test_msr },
+	[PERF_CSTATE_CORE_C3_RES] = { MSR_CORE_C3_RESIDENCY,	&group_cstate_core_c3,	test_msr },
+	[PERF_CSTATE_CORE_C6_RES] = { MSR_CORE_C6_RESIDENCY,	&group_cstate_core_c6,	test_msr },
+	[PERF_CSTATE_CORE_C7_RES] = { MSR_CORE_C7_RESIDENCY,	&group_cstate_core_c7,	test_msr },
+};
+
+static struct attribute *attrs_empty[] = {
+	NULL,
+};
+
+/*
+ * There are no default events, but we need to create
+ * "events" group (with empty attrs) before updating
+ * it with detected events.
+ */
+static struct attribute_group core_events_attr_group = {
+	.name = "events",
+	.attrs = attrs_empty,
+};
+
+DEFINE_CSTATE_FORMAT_ATTR(core_event, event, "config:0-63");
+static struct attribute *core_format_attrs[] = {
+	&format_attr_core_event.attr,
+	NULL,
+};
+
+static struct attribute_group core_format_attr_group = {
+	.name = "format",
+	.attrs = core_format_attrs,
+};
+
+static cpumask_t cstate_core_cpu_mask;
+static DEVICE_ATTR(cpumask, S_IRUGO, cstate_get_attr_cpumask, NULL);
+
+static struct attribute *cstate_cpumask_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group cpumask_attr_group = {
+	.attrs = cstate_cpumask_attrs,
+};
+
+static const struct attribute_group *core_attr_groups[] = {
+	&core_events_attr_group,
+	&core_format_attr_group,
+	&cpumask_attr_group,
+	NULL,
+};
+
+/* cstate_pkg PMU */
+static struct pmu cstate_pkg_pmu;
+static bool has_cstate_pkg;
+
+enum perf_cstate_pkg_events {
+	PERF_CSTATE_PKG_C2_RES = 0,
+	PERF_CSTATE_PKG_C3_RES,
+	PERF_CSTATE_PKG_C6_RES,
+	PERF_CSTATE_PKG_C7_RES,
+	PERF_CSTATE_PKG_C8_RES,
+	PERF_CSTATE_PKG_C9_RES,
+	PERF_CSTATE_PKG_C10_RES,
+
+	PERF_CSTATE_PKG_EVENT_MAX,
+};
+
+PMU_EVENT_ATTR_STRING(c2-residency,  attr_cstate_pkg_c2,  "event=0x00");
+PMU_EVENT_ATTR_STRING(c3-residency,  attr_cstate_pkg_c3,  "event=0x01");
+PMU_EVENT_ATTR_STRING(c6-residency,  attr_cstate_pkg_c6,  "event=0x02");
+PMU_EVENT_ATTR_STRING(c7-residency,  attr_cstate_pkg_c7,  "event=0x03");
+PMU_EVENT_ATTR_STRING(c8-residency,  attr_cstate_pkg_c8,  "event=0x04");
+PMU_EVENT_ATTR_STRING(c9-residency,  attr_cstate_pkg_c9,  "event=0x05");
+PMU_EVENT_ATTR_STRING(c10-residency, attr_cstate_pkg_c10, "event=0x06");
+
+static unsigned long pkg_msr_mask;
+
+PMU_EVENT_GROUP(events, cstate_pkg_c2);
+PMU_EVENT_GROUP(events, cstate_pkg_c3);
+PMU_EVENT_GROUP(events, cstate_pkg_c6);
+PMU_EVENT_GROUP(events, cstate_pkg_c7);
+PMU_EVENT_GROUP(events, cstate_pkg_c8);
+PMU_EVENT_GROUP(events, cstate_pkg_c9);
+PMU_EVENT_GROUP(events, cstate_pkg_c10);
+
+static struct perf_msr pkg_msr[] = {
+	[PERF_CSTATE_PKG_C2_RES]  = { MSR_PKG_C2_RESIDENCY,	&group_cstate_pkg_c2,	test_msr },
+	[PERF_CSTATE_PKG_C3_RES]  = { MSR_PKG_C3_RESIDENCY,	&group_cstate_pkg_c3,	test_msr },
+	[PERF_CSTATE_PKG_C6_RES]  = { MSR_PKG_C6_RESIDENCY,	&group_cstate_pkg_c6,	test_msr },
+	[PERF_CSTATE_PKG_C7_RES]  = { MSR_PKG_C7_RESIDENCY,	&group_cstate_pkg_c7,	test_msr },
+	[PERF_CSTATE_PKG_C8_RES]  = { MSR_PKG_C8_RESIDENCY,	&group_cstate_pkg_c8,	test_msr },
+	[PERF_CSTATE_PKG_C9_RES]  = { MSR_PKG_C9_RESIDENCY,	&group_cstate_pkg_c9,	test_msr },
+	[PERF_CSTATE_PKG_C10_RES] = { MSR_PKG_C10_RESIDENCY,	&group_cstate_pkg_c10,	test_msr },
+};
+
+static struct attribute_group pkg_events_attr_group = {
+	.name = "events",
+	.attrs = attrs_empty,
+};
+
+DEFINE_CSTATE_FORMAT_ATTR(pkg_event, event, "config:0-63");
+static struct attribute *pkg_format_attrs[] = {
+	&format_attr_pkg_event.attr,
+	NULL,
+};
+static struct attribute_group pkg_format_attr_group = {
+	.name = "format",
+	.attrs = pkg_format_attrs,
+};
+
+static cpumask_t cstate_pkg_cpu_mask;
+
+static const struct attribute_group *pkg_attr_groups[] = {
+	&pkg_events_attr_group,
+	&pkg_format_attr_group,
+	&cpumask_attr_group,
+	NULL,
+};
+
+static ssize_t cstate_get_attr_cpumask(struct device *dev,
+				       struct device_attribute *attr,
+				       char *buf)
+{
+	struct pmu *pmu = dev_get_drvdata(dev);
+
+	if (pmu == &cstate_core_pmu)
+		return cpumap_print_to_pagebuf(true, buf, &cstate_core_cpu_mask);
+	else if (pmu == &cstate_pkg_pmu)
+		return cpumap_print_to_pagebuf(true, buf, &cstate_pkg_cpu_mask);
+	else
+		return 0;
+}
+
+static int cstate_pmu_event_init(struct perf_event *event)
+{
+	u64 cfg = event->attr.config;
+	int cpu;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/* unsupported modes and filters */
+	if (event->attr.sample_period) /* no sampling */
+		return -EINVAL;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	if (event->pmu == &cstate_core_pmu) {
+		if (cfg >= PERF_CSTATE_CORE_EVENT_MAX)
+			return -EINVAL;
+		cfg = array_index_nospec((unsigned long)cfg, PERF_CSTATE_CORE_EVENT_MAX);
+		if (!(core_msr_mask & (1 << cfg)))
+			return -EINVAL;
+		event->hw.event_base = core_msr[cfg].msr;
+		cpu = cpumask_any_and(&cstate_core_cpu_mask,
+				      topology_sibling_cpumask(event->cpu));
+	} else if (event->pmu == &cstate_pkg_pmu) {
+		if (cfg >= PERF_CSTATE_PKG_EVENT_MAX)
+			return -EINVAL;
+		cfg = array_index_nospec((unsigned long)cfg, PERF_CSTATE_PKG_EVENT_MAX);
+		if (!(pkg_msr_mask & (1 << cfg)))
+			return -EINVAL;
+		event->hw.event_base = pkg_msr[cfg].msr;
+		cpu = cpumask_any_and(&cstate_pkg_cpu_mask,
+				      topology_die_cpumask(event->cpu));
+	} else {
+		return -ENOENT;
+	}
+
+	if (cpu >= nr_cpu_ids)
+		return -ENODEV;
+
+	event->cpu = cpu;
+	event->hw.config = cfg;
+	event->hw.idx = -1;
+	return 0;
+}
+
+static inline u64 cstate_pmu_read_counter(struct perf_event *event)
+{
+	u64 val;
+
+	rdmsrl(event->hw.event_base, val);
+	return val;
+}
+
+static void cstate_pmu_event_update(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 prev_raw_count, new_raw_count;
+
+again:
+	prev_raw_count = local64_read(&hwc->prev_count);
+	new_raw_count = cstate_pmu_read_counter(event);
+
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+			    new_raw_count) != prev_raw_count)
+		goto again;
+
+	local64_add(new_raw_count - prev_raw_count, &event->count);
+}
+
+static void cstate_pmu_event_start(struct perf_event *event, int mode)
+{
+	local64_set(&event->hw.prev_count, cstate_pmu_read_counter(event));
+}
+
+static void cstate_pmu_event_stop(struct perf_event *event, int mode)
+{
+	cstate_pmu_event_update(event);
+}
+
+static void cstate_pmu_event_del(struct perf_event *event, int mode)
+{
+	cstate_pmu_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int cstate_pmu_event_add(struct perf_event *event, int mode)
+{
+	if (mode & PERF_EF_START)
+		cstate_pmu_event_start(event, mode);
+
+	return 0;
+}
+
+/*
+ * Check if exiting cpu is the designated reader. If so migrate the
+ * events when there is a valid target available
+ */
+static int cstate_cpu_exit(unsigned int cpu)
+{
+	unsigned int target;
+
+	if (has_cstate_core &&
+	    cpumask_test_and_clear_cpu(cpu, &cstate_core_cpu_mask)) {
+
+		target = cpumask_any_but(topology_sibling_cpumask(cpu), cpu);
+		/* Migrate events if there is a valid target */
+		if (target < nr_cpu_ids) {
+			cpumask_set_cpu(target, &cstate_core_cpu_mask);
+			perf_pmu_migrate_context(&cstate_core_pmu, cpu, target);
+		}
+	}
+
+	if (has_cstate_pkg &&
+	    cpumask_test_and_clear_cpu(cpu, &cstate_pkg_cpu_mask)) {
+
+		target = cpumask_any_but(topology_die_cpumask(cpu), cpu);
+		/* Migrate events if there is a valid target */
+		if (target < nr_cpu_ids) {
+			cpumask_set_cpu(target, &cstate_pkg_cpu_mask);
+			perf_pmu_migrate_context(&cstate_pkg_pmu, cpu, target);
+		}
+	}
+	return 0;
+}
+
+static int cstate_cpu_init(unsigned int cpu)
+{
+	unsigned int target;
+
+	/*
+	 * If this is the first online thread of that core, set it in
+	 * the core cpu mask as the designated reader.
+	 */
+	target = cpumask_any_and(&cstate_core_cpu_mask,
+				 topology_sibling_cpumask(cpu));
+
+	if (has_cstate_core && target >= nr_cpu_ids)
+		cpumask_set_cpu(cpu, &cstate_core_cpu_mask);
+
+	/*
+	 * If this is the first online thread of that package, set it
+	 * in the package cpu mask as the designated reader.
+	 */
+	target = cpumask_any_and(&cstate_pkg_cpu_mask,
+				 topology_die_cpumask(cpu));
+	if (has_cstate_pkg && target >= nr_cpu_ids)
+		cpumask_set_cpu(cpu, &cstate_pkg_cpu_mask);
+
+	return 0;
+}
+
+static const struct attribute_group *core_attr_update[] = {
+	&group_cstate_core_c1,
+	&group_cstate_core_c3,
+	&group_cstate_core_c6,
+	&group_cstate_core_c7,
+	NULL,
+};
+
+static const struct attribute_group *pkg_attr_update[] = {
+	&group_cstate_pkg_c2,
+	&group_cstate_pkg_c3,
+	&group_cstate_pkg_c6,
+	&group_cstate_pkg_c7,
+	&group_cstate_pkg_c8,
+	&group_cstate_pkg_c9,
+	&group_cstate_pkg_c10,
+	NULL,
+};
+
+static struct pmu cstate_core_pmu = {
+	.attr_groups	= core_attr_groups,
+	.attr_update	= core_attr_update,
+	.name		= "cstate_core",
+	.task_ctx_nr	= perf_invalid_context,
+	.event_init	= cstate_pmu_event_init,
+	.add		= cstate_pmu_event_add,
+	.del		= cstate_pmu_event_del,
+	.start		= cstate_pmu_event_start,
+	.stop		= cstate_pmu_event_stop,
+	.read		= cstate_pmu_event_update,
+	.capabilities	= PERF_PMU_CAP_NO_INTERRUPT | PERF_PMU_CAP_NO_EXCLUDE,
+	.module		= THIS_MODULE,
+};
+
+static struct pmu cstate_pkg_pmu = {
+	.attr_groups	= pkg_attr_groups,
+	.attr_update	= pkg_attr_update,
+	.name		= "cstate_pkg",
+	.task_ctx_nr	= perf_invalid_context,
+	.event_init	= cstate_pmu_event_init,
+	.add		= cstate_pmu_event_add,
+	.del		= cstate_pmu_event_del,
+	.start		= cstate_pmu_event_start,
+	.stop		= cstate_pmu_event_stop,
+	.read		= cstate_pmu_event_update,
+	.capabilities	= PERF_PMU_CAP_NO_INTERRUPT | PERF_PMU_CAP_NO_EXCLUDE,
+	.module		= THIS_MODULE,
+};
+
+static const struct cstate_model nhm_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C3_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C7_RES),
+};
+
+static const struct cstate_model snb_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C3_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES) |
+				  BIT(PERF_CSTATE_CORE_C7_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C7_RES),
+};
+
+static const struct cstate_model hswult_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C3_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES) |
+				  BIT(PERF_CSTATE_CORE_C7_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C7_RES) |
+				  BIT(PERF_CSTATE_PKG_C8_RES) |
+				  BIT(PERF_CSTATE_PKG_C9_RES) |
+				  BIT(PERF_CSTATE_PKG_C10_RES),
+};
+
+static const struct cstate_model cnl_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C1_RES) |
+				  BIT(PERF_CSTATE_CORE_C3_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES) |
+				  BIT(PERF_CSTATE_CORE_C7_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C7_RES) |
+				  BIT(PERF_CSTATE_PKG_C8_RES) |
+				  BIT(PERF_CSTATE_PKG_C9_RES) |
+				  BIT(PERF_CSTATE_PKG_C10_RES),
+};
+
+static const struct cstate_model icl_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C6_RES) |
+				  BIT(PERF_CSTATE_CORE_C7_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C7_RES) |
+				  BIT(PERF_CSTATE_PKG_C8_RES) |
+				  BIT(PERF_CSTATE_PKG_C9_RES) |
+				  BIT(PERF_CSTATE_PKG_C10_RES),
+};
+
+static const struct cstate_model slm_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C1_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C6_RES),
+	.quirks			= SLM_PKG_C6_USE_C7_MSR,
+};
+
+
+static const struct cstate_model knl_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C6_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES),
+	.quirks			= KNL_CORE_C6_MSR,
+};
+
+
+static const struct cstate_model glm_cstates __initconst = {
+	.core_events		= BIT(PERF_CSTATE_CORE_C1_RES) |
+				  BIT(PERF_CSTATE_CORE_C3_RES) |
+				  BIT(PERF_CSTATE_CORE_C6_RES),
+
+	.pkg_events		= BIT(PERF_CSTATE_PKG_C2_RES) |
+				  BIT(PERF_CSTATE_PKG_C3_RES) |
+				  BIT(PERF_CSTATE_PKG_C6_RES) |
+				  BIT(PERF_CSTATE_PKG_C10_RES),
+};
+
+
+static const struct x86_cpu_id intel_cstates_match[] __initconst = {
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM,		&nhm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EP,		&nhm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EX,		&nhm_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE,		&nhm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EP,		&nhm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EX,		&nhm_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE_X,	&snb_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE_X,		&snb_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_X,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_G,		&snb_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_L,		&hswult_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT,	&slm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D,	&slm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_AIRMONT,	&slm_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_D,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_G,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_X,		&snb_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE,		&snb_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X,		&snb_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L,		&hswult_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE,		&hswult_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE_L,		&hswult_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE,		&hswult_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(CANNONLAKE_L,	&cnl_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL,	&knl_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM,	&knl_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT,	&glm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D,	&glm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_PLUS,	&glm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_D,	&glm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT,	&glm_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_L,	&glm_cstates),
+
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L,		&icl_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE,		&icl_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE_L,		&icl_cstates),
+	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE,		&icl_cstates),
+	{ },
+};
+MODULE_DEVICE_TABLE(x86cpu, intel_cstates_match);
+
+static int __init cstate_probe(const struct cstate_model *cm)
+{
+	/* SLM has different MSR for PKG C6 */
+	if (cm->quirks & SLM_PKG_C6_USE_C7_MSR)
+		pkg_msr[PERF_CSTATE_PKG_C6_RES].msr = MSR_PKG_C7_RESIDENCY;
+
+	/* KNL has different MSR for CORE C6 */
+	if (cm->quirks & KNL_CORE_C6_MSR)
+		pkg_msr[PERF_CSTATE_CORE_C6_RES].msr = MSR_KNL_CORE_C6_RESIDENCY;
+
+
+	core_msr_mask = perf_msr_probe(core_msr, PERF_CSTATE_CORE_EVENT_MAX,
+				       true, (void *) &cm->core_events);
+
+	pkg_msr_mask = perf_msr_probe(pkg_msr, PERF_CSTATE_PKG_EVENT_MAX,
+				      true, (void *) &cm->pkg_events);
+
+	has_cstate_core = !!core_msr_mask;
+	has_cstate_pkg  = !!pkg_msr_mask;
+
+	return (has_cstate_core || has_cstate_pkg) ? 0 : -ENODEV;
+}
+
+static inline void cstate_cleanup(void)
+{
+	cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_CSTATE_ONLINE);
+	cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_CSTATE_STARTING);
+
+	if (has_cstate_core)
+		perf_pmu_unregister(&cstate_core_pmu);
+
+	if (has_cstate_pkg)
+		perf_pmu_unregister(&cstate_pkg_pmu);
+}
+
+static int __init cstate_init(void)
+{
+	int err;
+
+	cpuhp_setup_state(CPUHP_AP_PERF_X86_CSTATE_STARTING,
+			  "perf/x86/cstate:starting", cstate_cpu_init, NULL);
+	cpuhp_setup_state(CPUHP_AP_PERF_X86_CSTATE_ONLINE,
+			  "perf/x86/cstate:online", NULL, cstate_cpu_exit);
+
+	if (has_cstate_core) {
+		err = perf_pmu_register(&cstate_core_pmu, cstate_core_pmu.name, -1);
+		if (err) {
+			has_cstate_core = false;
+			pr_info("Failed to register cstate core pmu\n");
+			cstate_cleanup();
+			return err;
+		}
+	}
+
+	if (has_cstate_pkg) {
+		if (topology_max_die_per_package() > 1) {
+			err = perf_pmu_register(&cstate_pkg_pmu,
+						"cstate_die", -1);
+		} else {
+			err = perf_pmu_register(&cstate_pkg_pmu,
+						cstate_pkg_pmu.name, -1);
+		}
+		if (err) {
+			has_cstate_pkg = false;
+			pr_info("Failed to register cstate pkg pmu\n");
+			cstate_cleanup();
+			return err;
+		}
+	}
+	return 0;
+}
+
+static int __init cstate_pmu_init(void)
+{
+	const struct x86_cpu_id *id;
+	int err;
+
+	if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
+		return -ENODEV;
+
+	id = x86_match_cpu(intel_cstates_match);
+	if (!id)
+		return -ENODEV;
+
+	err = cstate_probe((const struct cstate_model *) id->driver_data);
+	if (err)
+		return err;
+
+	return cstate_init();
+}
+module_init(cstate_pmu_init);
+
+static void __exit cstate_pmu_exit(void)
+{
+	cstate_cleanup();
+}
+module_exit(cstate_pmu_exit);
diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
new file mode 100644
index 000000000..48f30ffef
--- /dev/null
+++ b/arch/x86/events/intel/ds.c
@@ -0,0 +1,2119 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bitops.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+
+#include <asm/cpu_entry_area.h>
+#include <asm/perf_event.h>
+#include <asm/tlbflush.h>
+#include <asm/insn.h>
+#include <asm/io.h>
+
+#include "../perf_event.h"
+
+/* Waste a full page so it can be mapped into the cpu_entry_area */
+DEFINE_PER_CPU_PAGE_ALIGNED(struct debug_store, cpu_debug_store);
+
+/* The size of a BTS record in bytes: */
+#define BTS_RECORD_SIZE		24
+
+#define PEBS_FIXUP_SIZE		PAGE_SIZE
+
+/*
+ * pebs_record_32 for p4 and core not supported
+
+struct pebs_record_32 {
+	u32 flags, ip;
+	u32 ax, bc, cx, dx;
+	u32 si, di, bp, sp;
+};
+
+ */
+
+union intel_x86_pebs_dse {
+	u64 val;
+	struct {
+		unsigned int ld_dse:4;
+		unsigned int ld_stlb_miss:1;
+		unsigned int ld_locked:1;
+		unsigned int ld_reserved:26;
+	};
+	struct {
+		unsigned int st_l1d_hit:1;
+		unsigned int st_reserved1:3;
+		unsigned int st_stlb_miss:1;
+		unsigned int st_locked:1;
+		unsigned int st_reserved2:26;
+	};
+};
+
+
+/*
+ * Map PEBS Load Latency Data Source encodings to generic
+ * memory data source information
+ */
+#define P(a, b) PERF_MEM_S(a, b)
+#define OP_LH (P(OP, LOAD) | P(LVL, HIT))
+#define LEVEL(x) P(LVLNUM, x)
+#define REM P(REMOTE, REMOTE)
+#define SNOOP_NONE_MISS (P(SNOOP, NONE) | P(SNOOP, MISS))
+
+/* Version for Sandy Bridge and later */
+static u64 pebs_data_source[] = {
+	P(OP, LOAD) | P(LVL, MISS) | LEVEL(L3) | P(SNOOP, NA),/* 0x00:ukn L3 */
+	OP_LH | P(LVL, L1)  | LEVEL(L1) | P(SNOOP, NONE),  /* 0x01: L1 local */
+	OP_LH | P(LVL, LFB) | LEVEL(LFB) | P(SNOOP, NONE), /* 0x02: LFB hit */
+	OP_LH | P(LVL, L2)  | LEVEL(L2) | P(SNOOP, NONE),  /* 0x03: L2 hit */
+	OP_LH | P(LVL, L3)  | LEVEL(L3) | P(SNOOP, NONE),  /* 0x04: L3 hit */
+	OP_LH | P(LVL, L3)  | LEVEL(L3) | P(SNOOP, MISS),  /* 0x05: L3 hit, snoop miss */
+	OP_LH | P(LVL, L3)  | LEVEL(L3) | P(SNOOP, HIT),   /* 0x06: L3 hit, snoop hit */
+	OP_LH | P(LVL, L3)  | LEVEL(L3) | P(SNOOP, HITM),  /* 0x07: L3 hit, snoop hitm */
+	OP_LH | P(LVL, REM_CCE1) | REM | LEVEL(L3) | P(SNOOP, HIT),  /* 0x08: L3 miss snoop hit */
+	OP_LH | P(LVL, REM_CCE1) | REM | LEVEL(L3) | P(SNOOP, HITM), /* 0x09: L3 miss snoop hitm*/
+	OP_LH | P(LVL, LOC_RAM)  | LEVEL(RAM) | P(SNOOP, HIT),       /* 0x0a: L3 miss, shared */
+	OP_LH | P(LVL, REM_RAM1) | REM | LEVEL(L3) | P(SNOOP, HIT),  /* 0x0b: L3 miss, shared */
+	OP_LH | P(LVL, LOC_RAM)  | LEVEL(RAM) | SNOOP_NONE_MISS,     /* 0x0c: L3 miss, excl */
+	OP_LH | P(LVL, REM_RAM1) | LEVEL(RAM) | REM | SNOOP_NONE_MISS, /* 0x0d: L3 miss, excl */
+	OP_LH | P(LVL, IO)  | LEVEL(NA) | P(SNOOP, NONE), /* 0x0e: I/O */
+	OP_LH | P(LVL, UNC) | LEVEL(NA) | P(SNOOP, NONE), /* 0x0f: uncached */
+};
+
+/* Patch up minor differences in the bits */
+void __init intel_pmu_pebs_data_source_nhm(void)
+{
+	pebs_data_source[0x05] = OP_LH | P(LVL, L3) | LEVEL(L3) | P(SNOOP, HIT);
+	pebs_data_source[0x06] = OP_LH | P(LVL, L3) | LEVEL(L3) | P(SNOOP, HITM);
+	pebs_data_source[0x07] = OP_LH | P(LVL, L3) | LEVEL(L3) | P(SNOOP, HITM);
+}
+
+void __init intel_pmu_pebs_data_source_skl(bool pmem)
+{
+	u64 pmem_or_l4 = pmem ? LEVEL(PMEM) : LEVEL(L4);
+
+	pebs_data_source[0x08] = OP_LH | pmem_or_l4 | P(SNOOP, HIT);
+	pebs_data_source[0x09] = OP_LH | pmem_or_l4 | REM | P(SNOOP, HIT);
+	pebs_data_source[0x0b] = OP_LH | LEVEL(RAM) | REM | P(SNOOP, NONE);
+	pebs_data_source[0x0c] = OP_LH | LEVEL(ANY_CACHE) | REM | P(SNOOPX, FWD);
+	pebs_data_source[0x0d] = OP_LH | LEVEL(ANY_CACHE) | REM | P(SNOOP, HITM);
+}
+
+static u64 precise_store_data(u64 status)
+{
+	union intel_x86_pebs_dse dse;
+	u64 val = P(OP, STORE) | P(SNOOP, NA) | P(LVL, L1) | P(TLB, L2);
+
+	dse.val = status;
+
+	/*
+	 * bit 4: TLB access
+	 * 1 = stored missed 2nd level TLB
+	 *
+	 * so it either hit the walker or the OS
+	 * otherwise hit 2nd level TLB
+	 */
+	if (dse.st_stlb_miss)
+		val |= P(TLB, MISS);
+	else
+		val |= P(TLB, HIT);
+
+	/*
+	 * bit 0: hit L1 data cache
+	 * if not set, then all we know is that
+	 * it missed L1D
+	 */
+	if (dse.st_l1d_hit)
+		val |= P(LVL, HIT);
+	else
+		val |= P(LVL, MISS);
+
+	/*
+	 * bit 5: Locked prefix
+	 */
+	if (dse.st_locked)
+		val |= P(LOCK, LOCKED);
+
+	return val;
+}
+
+static u64 precise_datala_hsw(struct perf_event *event, u64 status)
+{
+	union perf_mem_data_src dse;
+
+	dse.val = PERF_MEM_NA;
+
+	if (event->hw.flags & PERF_X86_EVENT_PEBS_ST_HSW)
+		dse.mem_op = PERF_MEM_OP_STORE;
+	else if (event->hw.flags & PERF_X86_EVENT_PEBS_LD_HSW)
+		dse.mem_op = PERF_MEM_OP_LOAD;
+
+	/*
+	 * L1 info only valid for following events:
+	 *
+	 * MEM_UOPS_RETIRED.STLB_MISS_STORES
+	 * MEM_UOPS_RETIRED.LOCK_STORES
+	 * MEM_UOPS_RETIRED.SPLIT_STORES
+	 * MEM_UOPS_RETIRED.ALL_STORES
+	 */
+	if (event->hw.flags & PERF_X86_EVENT_PEBS_ST_HSW) {
+		if (status & 1)
+			dse.mem_lvl = PERF_MEM_LVL_L1 | PERF_MEM_LVL_HIT;
+		else
+			dse.mem_lvl = PERF_MEM_LVL_L1 | PERF_MEM_LVL_MISS;
+	}
+	return dse.val;
+}
+
+static u64 load_latency_data(u64 status)
+{
+	union intel_x86_pebs_dse dse;
+	u64 val;
+
+	dse.val = status;
+
+	/*
+	 * use the mapping table for bit 0-3
+	 */
+	val = pebs_data_source[dse.ld_dse];
+
+	/*
+	 * Nehalem models do not support TLB, Lock infos
+	 */
+	if (x86_pmu.pebs_no_tlb) {
+		val |= P(TLB, NA) | P(LOCK, NA);
+		return val;
+	}
+	/*
+	 * bit 4: TLB access
+	 * 0 = did not miss 2nd level TLB
+	 * 1 = missed 2nd level TLB
+	 */
+	if (dse.ld_stlb_miss)
+		val |= P(TLB, MISS) | P(TLB, L2);
+	else
+		val |= P(TLB, HIT) | P(TLB, L1) | P(TLB, L2);
+
+	/*
+	 * bit 5: locked prefix
+	 */
+	if (dse.ld_locked)
+		val |= P(LOCK, LOCKED);
+
+	return val;
+}
+
+struct pebs_record_core {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+};
+
+struct pebs_record_nhm {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+	u64 status, dla, dse, lat;
+};
+
+/*
+ * Same as pebs_record_nhm, with two additional fields.
+ */
+struct pebs_record_hsw {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+	u64 status, dla, dse, lat;
+	u64 real_ip, tsx_tuning;
+};
+
+union hsw_tsx_tuning {
+	struct {
+		u32 cycles_last_block     : 32,
+		    hle_abort		  : 1,
+		    rtm_abort		  : 1,
+		    instruction_abort     : 1,
+		    non_instruction_abort : 1,
+		    retry		  : 1,
+		    data_conflict	  : 1,
+		    capacity_writes	  : 1,
+		    capacity_reads	  : 1;
+	};
+	u64	    value;
+};
+
+#define PEBS_HSW_TSX_FLAGS	0xff00000000ULL
+
+/* Same as HSW, plus TSC */
+
+struct pebs_record_skl {
+	u64 flags, ip;
+	u64 ax, bx, cx, dx;
+	u64 si, di, bp, sp;
+	u64 r8,  r9,  r10, r11;
+	u64 r12, r13, r14, r15;
+	u64 status, dla, dse, lat;
+	u64 real_ip, tsx_tuning;
+	u64 tsc;
+};
+
+void init_debug_store_on_cpu(int cpu)
+{
+	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
+
+	if (!ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA,
+		     (u32)((u64)(unsigned long)ds),
+		     (u32)((u64)(unsigned long)ds >> 32));
+}
+
+void fini_debug_store_on_cpu(int cpu)
+{
+	if (!per_cpu(cpu_hw_events, cpu).ds)
+		return;
+
+	wrmsr_on_cpu(cpu, MSR_IA32_DS_AREA, 0, 0);
+}
+
+static DEFINE_PER_CPU(void *, insn_buffer);
+
+static void ds_update_cea(void *cea, void *addr, size_t size, pgprot_t prot)
+{
+	unsigned long start = (unsigned long)cea;
+	phys_addr_t pa;
+	size_t msz = 0;
+
+	pa = virt_to_phys(addr);
+
+	preempt_disable();
+	for (; msz < size; msz += PAGE_SIZE, pa += PAGE_SIZE, cea += PAGE_SIZE)
+		cea_set_pte(cea, pa, prot);
+
+	/*
+	 * This is a cross-CPU update of the cpu_entry_area, we must shoot down
+	 * all TLB entries for it.
+	 */
+	flush_tlb_kernel_range(start, start + size);
+	preempt_enable();
+}
+
+static void ds_clear_cea(void *cea, size_t size)
+{
+	unsigned long start = (unsigned long)cea;
+	size_t msz = 0;
+
+	preempt_disable();
+	for (; msz < size; msz += PAGE_SIZE, cea += PAGE_SIZE)
+		cea_set_pte(cea, 0, PAGE_NONE);
+
+	flush_tlb_kernel_range(start, start + size);
+	preempt_enable();
+}
+
+static void *dsalloc_pages(size_t size, gfp_t flags, int cpu)
+{
+	unsigned int order = get_order(size);
+	int node = cpu_to_node(cpu);
+	struct page *page;
+
+	page = __alloc_pages_node(node, flags | __GFP_ZERO, order);
+	return page ? page_address(page) : NULL;
+}
+
+static void dsfree_pages(const void *buffer, size_t size)
+{
+	if (buffer)
+		free_pages((unsigned long)buffer, get_order(size));
+}
+
+static int alloc_pebs_buffer(int cpu)
+{
+	struct cpu_hw_events *hwev = per_cpu_ptr(&cpu_hw_events, cpu);
+	struct debug_store *ds = hwev->ds;
+	size_t bsiz = x86_pmu.pebs_buffer_size;
+	int max, node = cpu_to_node(cpu);
+	void *buffer, *insn_buff, *cea;
+
+	if (!x86_pmu.pebs)
+		return 0;
+
+	buffer = dsalloc_pages(bsiz, GFP_KERNEL, cpu);
+	if (unlikely(!buffer))
+		return -ENOMEM;
+
+	/*
+	 * HSW+ already provides us the eventing ip; no need to allocate this
+	 * buffer then.
+	 */
+	if (x86_pmu.intel_cap.pebs_format < 2) {
+		insn_buff = kzalloc_node(PEBS_FIXUP_SIZE, GFP_KERNEL, node);
+		if (!insn_buff) {
+			dsfree_pages(buffer, bsiz);
+			return -ENOMEM;
+		}
+		per_cpu(insn_buffer, cpu) = insn_buff;
+	}
+	hwev->ds_pebs_vaddr = buffer;
+	/* Update the cpu entry area mapping */
+	cea = &get_cpu_entry_area(cpu)->cpu_debug_buffers.pebs_buffer;
+	ds->pebs_buffer_base = (unsigned long) cea;
+	ds_update_cea(cea, buffer, bsiz, PAGE_KERNEL);
+	ds->pebs_index = ds->pebs_buffer_base;
+	max = x86_pmu.pebs_record_size * (bsiz / x86_pmu.pebs_record_size);
+	ds->pebs_absolute_maximum = ds->pebs_buffer_base + max;
+	return 0;
+}
+
+static void release_pebs_buffer(int cpu)
+{
+	struct cpu_hw_events *hwev = per_cpu_ptr(&cpu_hw_events, cpu);
+	void *cea;
+
+	if (!x86_pmu.pebs)
+		return;
+
+	kfree(per_cpu(insn_buffer, cpu));
+	per_cpu(insn_buffer, cpu) = NULL;
+
+	/* Clear the fixmap */
+	cea = &get_cpu_entry_area(cpu)->cpu_debug_buffers.pebs_buffer;
+	ds_clear_cea(cea, x86_pmu.pebs_buffer_size);
+	dsfree_pages(hwev->ds_pebs_vaddr, x86_pmu.pebs_buffer_size);
+	hwev->ds_pebs_vaddr = NULL;
+}
+
+static int alloc_bts_buffer(int cpu)
+{
+	struct cpu_hw_events *hwev = per_cpu_ptr(&cpu_hw_events, cpu);
+	struct debug_store *ds = hwev->ds;
+	void *buffer, *cea;
+	int max;
+
+	if (!x86_pmu.bts)
+		return 0;
+
+	buffer = dsalloc_pages(BTS_BUFFER_SIZE, GFP_KERNEL | __GFP_NOWARN, cpu);
+	if (unlikely(!buffer)) {
+		WARN_ONCE(1, "%s: BTS buffer allocation failure\n", __func__);
+		return -ENOMEM;
+	}
+	hwev->ds_bts_vaddr = buffer;
+	/* Update the fixmap */
+	cea = &get_cpu_entry_area(cpu)->cpu_debug_buffers.bts_buffer;
+	ds->bts_buffer_base = (unsigned long) cea;
+	ds_update_cea(cea, buffer, BTS_BUFFER_SIZE, PAGE_KERNEL);
+	ds->bts_index = ds->bts_buffer_base;
+	max = BTS_BUFFER_SIZE / BTS_RECORD_SIZE;
+	ds->bts_absolute_maximum = ds->bts_buffer_base +
+					max * BTS_RECORD_SIZE;
+	ds->bts_interrupt_threshold = ds->bts_absolute_maximum -
+					(max / 16) * BTS_RECORD_SIZE;
+	return 0;
+}
+
+static void release_bts_buffer(int cpu)
+{
+	struct cpu_hw_events *hwev = per_cpu_ptr(&cpu_hw_events, cpu);
+	void *cea;
+
+	if (!x86_pmu.bts)
+		return;
+
+	/* Clear the fixmap */
+	cea = &get_cpu_entry_area(cpu)->cpu_debug_buffers.bts_buffer;
+	ds_clear_cea(cea, BTS_BUFFER_SIZE);
+	dsfree_pages(hwev->ds_bts_vaddr, BTS_BUFFER_SIZE);
+	hwev->ds_bts_vaddr = NULL;
+}
+
+static int alloc_ds_buffer(int cpu)
+{
+	struct debug_store *ds = &get_cpu_entry_area(cpu)->cpu_debug_store;
+
+	memset(ds, 0, sizeof(*ds));
+	per_cpu(cpu_hw_events, cpu).ds = ds;
+	return 0;
+}
+
+static void release_ds_buffer(int cpu)
+{
+	per_cpu(cpu_hw_events, cpu).ds = NULL;
+}
+
+void release_ds_buffers(void)
+{
+	int cpu;
+
+	if (!x86_pmu.bts && !x86_pmu.pebs)
+		return;
+
+	for_each_possible_cpu(cpu)
+		release_ds_buffer(cpu);
+
+	for_each_possible_cpu(cpu) {
+		/*
+		 * Again, ignore errors from offline CPUs, they will no longer
+		 * observe cpu_hw_events.ds and not program the DS_AREA when
+		 * they come up.
+		 */
+		fini_debug_store_on_cpu(cpu);
+	}
+
+	for_each_possible_cpu(cpu) {
+		release_pebs_buffer(cpu);
+		release_bts_buffer(cpu);
+	}
+}
+
+void reserve_ds_buffers(void)
+{
+	int bts_err = 0, pebs_err = 0;
+	int cpu;
+
+	x86_pmu.bts_active = 0;
+	x86_pmu.pebs_active = 0;
+
+	if (!x86_pmu.bts && !x86_pmu.pebs)
+		return;
+
+	if (!x86_pmu.bts)
+		bts_err = 1;
+
+	if (!x86_pmu.pebs)
+		pebs_err = 1;
+
+	for_each_possible_cpu(cpu) {
+		if (alloc_ds_buffer(cpu)) {
+			bts_err = 1;
+			pebs_err = 1;
+		}
+
+		if (!bts_err && alloc_bts_buffer(cpu))
+			bts_err = 1;
+
+		if (!pebs_err && alloc_pebs_buffer(cpu))
+			pebs_err = 1;
+
+		if (bts_err && pebs_err)
+			break;
+	}
+
+	if (bts_err) {
+		for_each_possible_cpu(cpu)
+			release_bts_buffer(cpu);
+	}
+
+	if (pebs_err) {
+		for_each_possible_cpu(cpu)
+			release_pebs_buffer(cpu);
+	}
+
+	if (bts_err && pebs_err) {
+		for_each_possible_cpu(cpu)
+			release_ds_buffer(cpu);
+	} else {
+		if (x86_pmu.bts && !bts_err)
+			x86_pmu.bts_active = 1;
+
+		if (x86_pmu.pebs && !pebs_err)
+			x86_pmu.pebs_active = 1;
+
+		for_each_possible_cpu(cpu) {
+			/*
+			 * Ignores wrmsr_on_cpu() errors for offline CPUs they
+			 * will get this call through intel_pmu_cpu_starting().
+			 */
+			init_debug_store_on_cpu(cpu);
+		}
+	}
+}
+
+/*
+ * BTS
+ */
+
+struct event_constraint bts_constraint =
+	EVENT_CONSTRAINT(0, 1ULL << INTEL_PMC_IDX_FIXED_BTS, 0);
+
+void intel_pmu_enable_bts(u64 config)
+{
+	unsigned long debugctlmsr;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr |= DEBUGCTLMSR_TR;
+	debugctlmsr |= DEBUGCTLMSR_BTS;
+	if (config & ARCH_PERFMON_EVENTSEL_INT)
+		debugctlmsr |= DEBUGCTLMSR_BTINT;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_OS))
+		debugctlmsr |= DEBUGCTLMSR_BTS_OFF_OS;
+
+	if (!(config & ARCH_PERFMON_EVENTSEL_USR))
+		debugctlmsr |= DEBUGCTLMSR_BTS_OFF_USR;
+
+	update_debugctlmsr(debugctlmsr);
+}
+
+void intel_pmu_disable_bts(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	unsigned long debugctlmsr;
+
+	if (!cpuc->ds)
+		return;
+
+	debugctlmsr = get_debugctlmsr();
+
+	debugctlmsr &=
+		~(DEBUGCTLMSR_TR | DEBUGCTLMSR_BTS | DEBUGCTLMSR_BTINT |
+		  DEBUGCTLMSR_BTS_OFF_OS | DEBUGCTLMSR_BTS_OFF_USR);
+
+	update_debugctlmsr(debugctlmsr);
+}
+
+int intel_pmu_drain_bts_buffer(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct bts_record {
+		u64	from;
+		u64	to;
+		u64	flags;
+	};
+	struct perf_event *event = cpuc->events[INTEL_PMC_IDX_FIXED_BTS];
+	struct bts_record *at, *base, *top;
+	struct perf_output_handle handle;
+	struct perf_event_header header;
+	struct perf_sample_data data;
+	unsigned long skip = 0;
+	struct pt_regs regs;
+
+	if (!event)
+		return 0;
+
+	if (!x86_pmu.bts_active)
+		return 0;
+
+	base = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
+	top  = (struct bts_record *)(unsigned long)ds->bts_index;
+
+	if (top <= base)
+		return 0;
+
+	memset(&regs, 0, sizeof(regs));
+
+	ds->bts_index = ds->bts_buffer_base;
+
+	perf_sample_data_init(&data, 0, event->hw.last_period);
+
+	/*
+	 * BTS leaks kernel addresses in branches across the cpl boundary,
+	 * such as traps or system calls, so unless the user is asking for
+	 * kernel tracing (and right now it's not possible), we'd need to
+	 * filter them out. But first we need to count how many of those we
+	 * have in the current batch. This is an extra O(n) pass, however,
+	 * it's much faster than the other one especially considering that
+	 * n <= 2560 (BTS_BUFFER_SIZE / BTS_RECORD_SIZE * 15/16; see the
+	 * alloc_bts_buffer()).
+	 */
+	for (at = base; at < top; at++) {
+		/*
+		 * Note that right now *this* BTS code only works if
+		 * attr::exclude_kernel is set, but let's keep this extra
+		 * check here in case that changes.
+		 */
+		if (event->attr.exclude_kernel &&
+		    (kernel_ip(at->from) || kernel_ip(at->to)))
+			skip++;
+	}
+
+	/*
+	 * Prepare a generic sample, i.e. fill in the invariant fields.
+	 * We will overwrite the from and to address before we output
+	 * the sample.
+	 */
+	rcu_read_lock();
+	perf_prepare_sample(&header, &data, event, &regs);
+
+	if (perf_output_begin(&handle, &data, event,
+			      header.size * (top - base - skip)))
+		goto unlock;
+
+	for (at = base; at < top; at++) {
+		/* Filter out any records that contain kernel addresses. */
+		if (event->attr.exclude_kernel &&
+		    (kernel_ip(at->from) || kernel_ip(at->to)))
+			continue;
+
+		data.ip		= at->from;
+		data.addr	= at->to;
+
+		perf_output_sample(&handle, &header, &data, event);
+	}
+
+	perf_output_end(&handle);
+
+	/* There's new data available. */
+	event->hw.interrupts++;
+	event->pending_kill = POLL_IN;
+unlock:
+	rcu_read_unlock();
+	return 1;
+}
+
+static inline void intel_pmu_drain_pebs_buffer(void)
+{
+	struct perf_sample_data data;
+
+	x86_pmu.drain_pebs(NULL, &data);
+}
+
+/*
+ * PEBS
+ */
+struct event_constraint intel_core2_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x1fc7, 0x1), /* SIMD_INST_RETURED.ANY */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xcb, 0x1),    /* MEM_LOAD_RETIRED.* */
+	/* INST_RETIRED.ANY_P, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x01),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_atom_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x00c5, 0x1), /* MISPREDICTED_BRANCH_RETIRED */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xcb, 0x1),    /* MEM_LOAD_RETIRED.* */
+	/* INST_RETIRED.ANY_P, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x01),
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_slm_pebs_event_constraints[] = {
+	/* INST_RETIRED.ANY_P, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x1),
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_glm_pebs_event_constraints[] = {
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_nehalem_pebs_event_constraints[] = {
+	INTEL_PLD_CONSTRAINT(0x100b, 0xf),      /* MEM_INST_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0x0f, 0xf),    /* MEM_UNCORE_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc0, 0xf),    /* INST_RETIRED.ANY */
+	INTEL_EVENT_CONSTRAINT(0xc2, 0xf),    /* UOPS_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc4, 0xf),    /* BR_INST_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x02c5, 0xf), /* BR_MISP_RETIRED.NEAR_CALL */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc7, 0xf),    /* SSEX_UOPS_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x20c8, 0xf), /* ITLB_MISS_RETIRED */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xcb, 0xf),    /* MEM_LOAD_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xf7, 0xf),    /* FP_ASSIST.* */
+	/* INST_RETIRED.ANY_P, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x0f),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_westmere_pebs_event_constraints[] = {
+	INTEL_PLD_CONSTRAINT(0x100b, 0xf),      /* MEM_INST_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0x0f, 0xf),    /* MEM_UNCORE_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc0, 0xf),    /* INSTR_RETIRED.* */
+	INTEL_EVENT_CONSTRAINT(0xc2, 0xf),    /* UOPS_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc4, 0xf),    /* BR_INST_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc5, 0xf),    /* BR_MISP_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xc7, 0xf),    /* SSEX_UOPS_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x20c8, 0xf), /* ITLB_MISS_RETIRED */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xcb, 0xf),    /* MEM_LOAD_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xf7, 0xf),    /* FP_ASSIST.* */
+	/* INST_RETIRED.ANY_P, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x0f),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_snb_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
+	INTEL_PLD_CONSTRAINT(0x01cd, 0x8),    /* MEM_TRANS_RETIRED.LAT_ABOVE_THR */
+	INTEL_PST_CONSTRAINT(0x02cd, 0x8),    /* MEM_TRANS_RETIRED.PRECISE_STORES */
+	/* UOPS_RETIRED.ALL, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c2, 0xf),
+        INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf),    /* MEM_UOP_RETIRED.* */
+        INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf),    /* MEM_LOAD_UOPS_RETIRED.* */
+        INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf),    /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+        INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf),    /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0xf),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_ivb_pebs_event_constraints[] = {
+        INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
+        INTEL_PLD_CONSTRAINT(0x01cd, 0x8),    /* MEM_TRANS_RETIRED.LAT_ABOVE_THR */
+	INTEL_PST_CONSTRAINT(0x02cd, 0x8),    /* MEM_TRANS_RETIRED.PRECISE_STORES */
+	/* UOPS_RETIRED.ALL, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c2, 0xf),
+	/* INST_RETIRED.PREC_DIST, inv=1, cmask=16 (cycles:ppp). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c0, 0x2),
+	INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf),    /* MEM_UOP_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf),    /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf),    /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+	INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf),    /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0xf),
+        EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_hsw_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
+	INTEL_PLD_CONSTRAINT(0x01cd, 0xf),    /* MEM_TRANS_RETIRED.* */
+	/* UOPS_RETIRED.ALL, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c2, 0xf),
+	/* INST_RETIRED.PREC_DIST, inv=1, cmask=16 (cycles:ppp). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c0, 0x2),
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_NA(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XLD(0x11d0, 0xf), /* MEM_UOPS_RETIRED.STLB_MISS_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XLD(0x21d0, 0xf), /* MEM_UOPS_RETIRED.LOCK_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XLD(0x41d0, 0xf), /* MEM_UOPS_RETIRED.SPLIT_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XLD(0x81d0, 0xf), /* MEM_UOPS_RETIRED.ALL_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XST(0x12d0, 0xf), /* MEM_UOPS_RETIRED.STLB_MISS_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XST(0x42d0, 0xf), /* MEM_UOPS_RETIRED.SPLIT_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XST(0x82d0, 0xf), /* MEM_UOPS_RETIRED.ALL_STORES */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_XLD(0xd1, 0xf),    /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_XLD(0xd2, 0xf),    /* MEM_LOAD_UOPS_L3_HIT_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_XLD(0xd3, 0xf),    /* MEM_LOAD_UOPS_L3_MISS_RETIRED.* */
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0xf),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_bdw_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
+	INTEL_PLD_CONSTRAINT(0x01cd, 0xf),    /* MEM_TRANS_RETIRED.* */
+	/* UOPS_RETIRED.ALL, inv=1, cmask=16 (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c2, 0xf),
+	/* INST_RETIRED.PREC_DIST, inv=1, cmask=16 (cycles:ppp). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c0, 0x2),
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_NA(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf), /* MEM_UOPS_RETIRED.STLB_MISS_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x21d0, 0xf), /* MEM_UOPS_RETIRED.LOCK_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x41d0, 0xf), /* MEM_UOPS_RETIRED.SPLIT_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x81d0, 0xf), /* MEM_UOPS_RETIRED.ALL_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf), /* MEM_UOPS_RETIRED.STLB_MISS_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x42d0, 0xf), /* MEM_UOPS_RETIRED.SPLIT_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x82d0, 0xf), /* MEM_UOPS_RETIRED.ALL_STORES */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd1, 0xf),    /* MEM_LOAD_UOPS_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd2, 0xf),    /* MEM_LOAD_UOPS_L3_HIT_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd3, 0xf),    /* MEM_LOAD_UOPS_L3_MISS_RETIRED.* */
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0xf),
+	EVENT_CONSTRAINT_END
+};
+
+
+struct event_constraint intel_skl_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x1c0, 0x2),	/* INST_RETIRED.PREC_DIST */
+	/* INST_RETIRED.PREC_DIST, inv=1, cmask=16 (cycles:ppp). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108001c0, 0x2),
+	/* INST_RETIRED.TOTAL_CYCLES_PS (inv=1, cmask=16) (cycles:p). */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x108000c0, 0x0f),
+	INTEL_PLD_CONSTRAINT(0x1cd, 0xf),		      /* MEM_TRANS_RETIRED.* */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf), /* MEM_INST_RETIRED.STLB_MISS_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf), /* MEM_INST_RETIRED.STLB_MISS_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x21d0, 0xf), /* MEM_INST_RETIRED.LOCK_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x22d0, 0xf), /* MEM_INST_RETIRED.LOCK_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x41d0, 0xf), /* MEM_INST_RETIRED.SPLIT_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x42d0, 0xf), /* MEM_INST_RETIRED.SPLIT_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x81d0, 0xf), /* MEM_INST_RETIRED.ALL_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x82d0, 0xf), /* MEM_INST_RETIRED.ALL_STORES */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd1, 0xf),    /* MEM_LOAD_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd2, 0xf),    /* MEM_LOAD_L3_HIT_RETIRED.* */
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(0xd3, 0xf),    /* MEM_LOAD_L3_MISS_RETIRED.* */
+	/* Allow all events as PEBS with no flags */
+	INTEL_ALL_EVENT_CONSTRAINT(0, 0xf),
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint intel_icl_pebs_event_constraints[] = {
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x1c0, 0x100000000ULL),	/* INST_RETIRED.PREC_DIST */
+	INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL),	/* SLOTS */
+
+	INTEL_PLD_CONSTRAINT(0x1cd, 0xff),			/* MEM_TRANS_RETIRED.LOAD_LATENCY */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x11d0, 0xf),	/* MEM_INST_RETIRED.STLB_MISS_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x12d0, 0xf),	/* MEM_INST_RETIRED.STLB_MISS_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x21d0, 0xf),	/* MEM_INST_RETIRED.LOCK_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x41d0, 0xf),	/* MEM_INST_RETIRED.SPLIT_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x42d0, 0xf),	/* MEM_INST_RETIRED.SPLIT_STORES */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(0x81d0, 0xf),	/* MEM_INST_RETIRED.ALL_LOADS */
+	INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(0x82d0, 0xf),	/* MEM_INST_RETIRED.ALL_STORES */
+
+	INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD_RANGE(0xd1, 0xd4, 0xf), /* MEM_LOAD_*_RETIRED.* */
+
+	INTEL_FLAGS_EVENT_CONSTRAINT(0xd0, 0xf),		/* MEM_INST_RETIRED.* */
+
+	/*
+	 * Everything else is handled by PMU_FL_PEBS_ALL, because we
+	 * need the full constraints from the main table.
+	 */
+
+	EVENT_CONSTRAINT_END
+};
+
+struct event_constraint *intel_pebs_constraints(struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	if (!event->attr.precise_ip)
+		return NULL;
+
+	if (x86_pmu.pebs_constraints) {
+		for_each_event_constraint(c, x86_pmu.pebs_constraints) {
+			if (constraint_match(c, event->hw.config)) {
+				event->hw.flags |= c->flags;
+				return c;
+			}
+		}
+	}
+
+	/*
+	 * Extended PEBS support
+	 * Makes the PEBS code search the normal constraints.
+	 */
+	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
+		return NULL;
+
+	return &emptyconstraint;
+}
+
+/*
+ * We need the sched_task callback even for per-cpu events when we use
+ * the large interrupt threshold, such that we can provide PID and TID
+ * to PEBS samples.
+ */
+static inline bool pebs_needs_sched_cb(struct cpu_hw_events *cpuc)
+{
+	if (cpuc->n_pebs == cpuc->n_pebs_via_pt)
+		return false;
+
+	return cpuc->n_pebs && (cpuc->n_pebs == cpuc->n_large_pebs);
+}
+
+void intel_pmu_pebs_sched_task(struct perf_event_context *ctx, bool sched_in)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!sched_in && pebs_needs_sched_cb(cpuc))
+		intel_pmu_drain_pebs_buffer();
+}
+
+static inline void pebs_update_threshold(struct cpu_hw_events *cpuc)
+{
+	struct debug_store *ds = cpuc->ds;
+	u64 threshold;
+	int reserved;
+
+	if (cpuc->n_pebs_via_pt)
+		return;
+
+	if (x86_pmu.flags & PMU_FL_PEBS_ALL)
+		reserved = x86_pmu.max_pebs_events + x86_pmu.num_counters_fixed;
+	else
+		reserved = x86_pmu.max_pebs_events;
+
+	if (cpuc->n_pebs == cpuc->n_large_pebs) {
+		threshold = ds->pebs_absolute_maximum -
+			reserved * cpuc->pebs_record_size;
+	} else {
+		threshold = ds->pebs_buffer_base + cpuc->pebs_record_size;
+	}
+
+	ds->pebs_interrupt_threshold = threshold;
+}
+
+static void adaptive_pebs_record_size_update(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	u64 pebs_data_cfg = cpuc->pebs_data_cfg;
+	int sz = sizeof(struct pebs_basic);
+
+	if (pebs_data_cfg & PEBS_DATACFG_MEMINFO)
+		sz += sizeof(struct pebs_meminfo);
+	if (pebs_data_cfg & PEBS_DATACFG_GP)
+		sz += sizeof(struct pebs_gprs);
+	if (pebs_data_cfg & PEBS_DATACFG_XMMS)
+		sz += sizeof(struct pebs_xmm);
+	if (pebs_data_cfg & PEBS_DATACFG_LBRS)
+		sz += x86_pmu.lbr_nr * sizeof(struct lbr_entry);
+
+	cpuc->pebs_record_size = sz;
+}
+
+#define PERF_PEBS_MEMINFO_TYPE	(PERF_SAMPLE_ADDR | PERF_SAMPLE_DATA_SRC |   \
+				PERF_SAMPLE_PHYS_ADDR | PERF_SAMPLE_WEIGHT | \
+				PERF_SAMPLE_TRANSACTION)
+
+static u64 pebs_update_adaptive_cfg(struct perf_event *event)
+{
+	struct perf_event_attr *attr = &event->attr;
+	u64 sample_type = attr->sample_type;
+	u64 pebs_data_cfg = 0;
+	bool gprs, tsx_weight;
+
+	if (!(sample_type & ~(PERF_SAMPLE_IP|PERF_SAMPLE_TIME)) &&
+	    attr->precise_ip > 1)
+		return pebs_data_cfg;
+
+	if (sample_type & PERF_PEBS_MEMINFO_TYPE)
+		pebs_data_cfg |= PEBS_DATACFG_MEMINFO;
+
+	/*
+	 * We need GPRs when:
+	 * + user requested them
+	 * + precise_ip < 2 for the non event IP
+	 * + For RTM TSX weight we need GPRs for the abort code.
+	 */
+	gprs = (sample_type & PERF_SAMPLE_REGS_INTR) &&
+	       (attr->sample_regs_intr & PEBS_GP_REGS);
+
+	tsx_weight = (sample_type & PERF_SAMPLE_WEIGHT) &&
+		     ((attr->config & INTEL_ARCH_EVENT_MASK) ==
+		      x86_pmu.rtm_abort_event);
+
+	if (gprs || (attr->precise_ip < 2) || tsx_weight)
+		pebs_data_cfg |= PEBS_DATACFG_GP;
+
+	if ((sample_type & PERF_SAMPLE_REGS_INTR) &&
+	    (attr->sample_regs_intr & PERF_REG_EXTENDED_MASK))
+		pebs_data_cfg |= PEBS_DATACFG_XMMS;
+
+	if (sample_type & PERF_SAMPLE_BRANCH_STACK) {
+		/*
+		 * For now always log all LBRs. Could configure this
+		 * later.
+		 */
+		pebs_data_cfg |= PEBS_DATACFG_LBRS |
+			((x86_pmu.lbr_nr-1) << PEBS_DATACFG_LBR_SHIFT);
+	}
+
+	return pebs_data_cfg;
+}
+
+static void
+pebs_update_state(bool needed_cb, struct cpu_hw_events *cpuc,
+		  struct perf_event *event, bool add)
+{
+	struct pmu *pmu = event->ctx->pmu;
+	/*
+	 * Make sure we get updated with the first PEBS
+	 * event. It will trigger also during removal, but
+	 * that does not hurt:
+	 */
+	bool update = cpuc->n_pebs == 1;
+
+	if (needed_cb != pebs_needs_sched_cb(cpuc)) {
+		if (!needed_cb)
+			perf_sched_cb_inc(pmu);
+		else
+			perf_sched_cb_dec(pmu);
+
+		update = true;
+	}
+
+	/*
+	 * The PEBS record doesn't shrink on pmu::del(). Doing so would require
+	 * iterating all remaining PEBS events to reconstruct the config.
+	 */
+	if (x86_pmu.intel_cap.pebs_baseline && add) {
+		u64 pebs_data_cfg;
+
+		/* Clear pebs_data_cfg and pebs_record_size for first PEBS. */
+		if (cpuc->n_pebs == 1) {
+			cpuc->pebs_data_cfg = 0;
+			cpuc->pebs_record_size = sizeof(struct pebs_basic);
+		}
+
+		pebs_data_cfg = pebs_update_adaptive_cfg(event);
+
+		/* Update pebs_record_size if new event requires more data. */
+		if (pebs_data_cfg & ~cpuc->pebs_data_cfg) {
+			cpuc->pebs_data_cfg |= pebs_data_cfg;
+			adaptive_pebs_record_size_update();
+			update = true;
+		}
+	}
+
+	if (update)
+		pebs_update_threshold(cpuc);
+}
+
+void intel_pmu_pebs_add(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	bool needed_cb = pebs_needs_sched_cb(cpuc);
+
+	cpuc->n_pebs++;
+	if (hwc->flags & PERF_X86_EVENT_LARGE_PEBS)
+		cpuc->n_large_pebs++;
+	if (hwc->flags & PERF_X86_EVENT_PEBS_VIA_PT)
+		cpuc->n_pebs_via_pt++;
+
+	pebs_update_state(needed_cb, cpuc, event, true);
+}
+
+static void intel_pmu_pebs_via_pt_disable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!is_pebs_pt(event))
+		return;
+
+	if (!(cpuc->pebs_enabled & ~PEBS_VIA_PT_MASK))
+		cpuc->pebs_enabled &= ~PEBS_VIA_PT_MASK;
+}
+
+static void intel_pmu_pebs_via_pt_enable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	struct debug_store *ds = cpuc->ds;
+
+	if (!is_pebs_pt(event))
+		return;
+
+	if (!(event->hw.flags & PERF_X86_EVENT_LARGE_PEBS))
+		cpuc->pebs_enabled |= PEBS_PMI_AFTER_EACH_RECORD;
+
+	cpuc->pebs_enabled |= PEBS_OUTPUT_PT;
+
+	wrmsrl(MSR_RELOAD_PMC0 + hwc->idx, ds->pebs_event_reset[hwc->idx]);
+}
+
+void intel_pmu_pebs_enable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	struct debug_store *ds = cpuc->ds;
+
+	hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT;
+
+	cpuc->pebs_enabled |= 1ULL << hwc->idx;
+
+	if ((event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT) && (x86_pmu.version < 5))
+		cpuc->pebs_enabled |= 1ULL << (hwc->idx + 32);
+	else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST)
+		cpuc->pebs_enabled |= 1ULL << 63;
+
+	if (x86_pmu.intel_cap.pebs_baseline) {
+		hwc->config |= ICL_EVENTSEL_ADAPTIVE;
+		if (cpuc->pebs_data_cfg != cpuc->active_pebs_data_cfg) {
+			wrmsrl(MSR_PEBS_DATA_CFG, cpuc->pebs_data_cfg);
+			cpuc->active_pebs_data_cfg = cpuc->pebs_data_cfg;
+		}
+	}
+
+	/*
+	 * Use auto-reload if possible to save a MSR write in the PMI.
+	 * This must be done in pmu::start(), because PERF_EVENT_IOC_PERIOD.
+	 */
+	if (hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) {
+		unsigned int idx = hwc->idx;
+
+		if (idx >= INTEL_PMC_IDX_FIXED)
+			idx = MAX_PEBS_EVENTS + (idx - INTEL_PMC_IDX_FIXED);
+		ds->pebs_event_reset[idx] =
+			(u64)(-hwc->sample_period) & x86_pmu.cntval_mask;
+	} else {
+		ds->pebs_event_reset[hwc->idx] = 0;
+	}
+
+	intel_pmu_pebs_via_pt_enable(event);
+}
+
+void intel_pmu_pebs_del(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	bool needed_cb = pebs_needs_sched_cb(cpuc);
+
+	cpuc->n_pebs--;
+	if (hwc->flags & PERF_X86_EVENT_LARGE_PEBS)
+		cpuc->n_large_pebs--;
+	if (hwc->flags & PERF_X86_EVENT_PEBS_VIA_PT)
+		cpuc->n_pebs_via_pt--;
+
+	pebs_update_state(needed_cb, cpuc, event, false);
+}
+
+void intel_pmu_pebs_disable(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (cpuc->n_pebs == cpuc->n_large_pebs &&
+	    cpuc->n_pebs != cpuc->n_pebs_via_pt)
+		intel_pmu_drain_pebs_buffer();
+
+	cpuc->pebs_enabled &= ~(1ULL << hwc->idx);
+
+	if ((event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT) &&
+	    (x86_pmu.version < 5))
+		cpuc->pebs_enabled &= ~(1ULL << (hwc->idx + 32));
+	else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST)
+		cpuc->pebs_enabled &= ~(1ULL << 63);
+
+	intel_pmu_pebs_via_pt_disable(event);
+
+	if (cpuc->enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+
+	hwc->config |= ARCH_PERFMON_EVENTSEL_INT;
+}
+
+void intel_pmu_pebs_enable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (cpuc->pebs_enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
+}
+
+void intel_pmu_pebs_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (cpuc->pebs_enabled)
+		wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
+}
+
+static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	unsigned long from = cpuc->lbr_entries[0].from;
+	unsigned long old_to, to = cpuc->lbr_entries[0].to;
+	unsigned long ip = regs->ip;
+	int is_64bit = 0;
+	void *kaddr;
+	int size;
+
+	/*
+	 * We don't need to fixup if the PEBS assist is fault like
+	 */
+	if (!x86_pmu.intel_cap.pebs_trap)
+		return 1;
+
+	/*
+	 * No LBR entry, no basic block, no rewinding
+	 */
+	if (!cpuc->lbr_stack.nr || !from || !to)
+		return 0;
+
+	/*
+	 * Basic blocks should never cross user/kernel boundaries
+	 */
+	if (kernel_ip(ip) != kernel_ip(to))
+		return 0;
+
+	/*
+	 * unsigned math, either ip is before the start (impossible) or
+	 * the basic block is larger than 1 page (sanity)
+	 */
+	if ((ip - to) > PEBS_FIXUP_SIZE)
+		return 0;
+
+	/*
+	 * We sampled a branch insn, rewind using the LBR stack
+	 */
+	if (ip == to) {
+		set_linear_ip(regs, from);
+		return 1;
+	}
+
+	size = ip - to;
+	if (!kernel_ip(ip)) {
+		int bytes;
+		u8 *buf = this_cpu_read(insn_buffer);
+
+		/* 'size' must fit our buffer, see above */
+		bytes = copy_from_user_nmi(buf, (void __user *)to, size);
+		if (bytes != 0)
+			return 0;
+
+		kaddr = buf;
+	} else {
+		kaddr = (void *)to;
+	}
+
+	do {
+		struct insn insn;
+
+		old_to = to;
+
+#ifdef CONFIG_X86_64
+		is_64bit = kernel_ip(to) || !test_thread_flag(TIF_IA32);
+#endif
+		insn_init(&insn, kaddr, size, is_64bit);
+		insn_get_length(&insn);
+		/*
+		 * Make sure there was not a problem decoding the
+		 * instruction and getting the length.  This is
+		 * doubly important because we have an infinite
+		 * loop if insn.length=0.
+		 */
+		if (!insn.length)
+			break;
+
+		to += insn.length;
+		kaddr += insn.length;
+		size -= insn.length;
+	} while (to < ip);
+
+	if (to == ip) {
+		set_linear_ip(regs, old_to);
+		return 1;
+	}
+
+	/*
+	 * Even though we decoded the basic block, the instruction stream
+	 * never matched the given IP, either the TO or the IP got corrupted.
+	 */
+	return 0;
+}
+
+static inline u64 intel_get_tsx_weight(u64 tsx_tuning)
+{
+	if (tsx_tuning) {
+		union hsw_tsx_tuning tsx = { .value = tsx_tuning };
+		return tsx.cycles_last_block;
+	}
+	return 0;
+}
+
+static inline u64 intel_get_tsx_transaction(u64 tsx_tuning, u64 ax)
+{
+	u64 txn = (tsx_tuning & PEBS_HSW_TSX_FLAGS) >> 32;
+
+	/* For RTM XABORTs also log the abort code from AX */
+	if ((txn & PERF_TXN_TRANSACTION) && (ax & 1))
+		txn |= ((ax >> 24) & 0xff) << PERF_TXN_ABORT_SHIFT;
+	return txn;
+}
+
+static inline u64 get_pebs_status(void *n)
+{
+	if (x86_pmu.intel_cap.pebs_format < 4)
+		return ((struct pebs_record_nhm *)n)->status;
+	return ((struct pebs_basic *)n)->applicable_counters;
+}
+
+#define PERF_X86_EVENT_PEBS_HSW_PREC \
+		(PERF_X86_EVENT_PEBS_ST_HSW | \
+		 PERF_X86_EVENT_PEBS_LD_HSW | \
+		 PERF_X86_EVENT_PEBS_NA_HSW)
+
+static u64 get_data_src(struct perf_event *event, u64 aux)
+{
+	u64 val = PERF_MEM_NA;
+	int fl = event->hw.flags;
+	bool fst = fl & (PERF_X86_EVENT_PEBS_ST | PERF_X86_EVENT_PEBS_HSW_PREC);
+
+	if (fl & PERF_X86_EVENT_PEBS_LDLAT)
+		val = load_latency_data(aux);
+	else if (fst && (fl & PERF_X86_EVENT_PEBS_HSW_PREC))
+		val = precise_datala_hsw(event, aux);
+	else if (fst)
+		val = precise_store_data(aux);
+	return val;
+}
+
+static void setup_pebs_fixed_sample_data(struct perf_event *event,
+				   struct pt_regs *iregs, void *__pebs,
+				   struct perf_sample_data *data,
+				   struct pt_regs *regs)
+{
+	/*
+	 * We cast to the biggest pebs_record but are careful not to
+	 * unconditionally access the 'extra' entries.
+	 */
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct pebs_record_skl *pebs = __pebs;
+	u64 sample_type;
+	int fll;
+
+	if (pebs == NULL)
+		return;
+
+	sample_type = event->attr.sample_type;
+	fll = event->hw.flags & PERF_X86_EVENT_PEBS_LDLAT;
+
+	perf_sample_data_init(data, 0, event->hw.last_period);
+
+	data->period = event->hw.last_period;
+
+	/*
+	 * Use latency for weight (only avail with PEBS-LL)
+	 */
+	if (fll && (sample_type & PERF_SAMPLE_WEIGHT))
+		data->weight = pebs->lat;
+
+	/*
+	 * data.data_src encodes the data source
+	 */
+	if (sample_type & PERF_SAMPLE_DATA_SRC)
+		data->data_src.val = get_data_src(event, pebs->dse);
+
+	/*
+	 * We must however always use iregs for the unwinder to stay sane; the
+	 * record BP,SP,IP can point into thin air when the record is from a
+	 * previous PMI context or an (I)RET happened between the record and
+	 * PMI.
+	 */
+	if (sample_type & PERF_SAMPLE_CALLCHAIN)
+		data->callchain = perf_callchain(event, iregs);
+
+	/*
+	 * We use the interrupt regs as a base because the PEBS record does not
+	 * contain a full regs set, specifically it seems to lack segment
+	 * descriptors, which get used by things like user_mode().
+	 *
+	 * In the simple case fix up only the IP for PERF_SAMPLE_IP.
+	 */
+	*regs = *iregs;
+
+	/*
+	 * Initialize regs_>flags from PEBS,
+	 * Clear exact bit (which uses x86 EFLAGS Reserved bit 3),
+	 * i.e., do not rely on it being zero:
+	 */
+	regs->flags = pebs->flags & ~PERF_EFLAGS_EXACT;
+
+	if (sample_type & PERF_SAMPLE_REGS_INTR) {
+		regs->ax = pebs->ax;
+		regs->bx = pebs->bx;
+		regs->cx = pebs->cx;
+		regs->dx = pebs->dx;
+		regs->si = pebs->si;
+		regs->di = pebs->di;
+
+		regs->bp = pebs->bp;
+		regs->sp = pebs->sp;
+
+#ifndef CONFIG_X86_32
+		regs->r8 = pebs->r8;
+		regs->r9 = pebs->r9;
+		regs->r10 = pebs->r10;
+		regs->r11 = pebs->r11;
+		regs->r12 = pebs->r12;
+		regs->r13 = pebs->r13;
+		regs->r14 = pebs->r14;
+		regs->r15 = pebs->r15;
+#endif
+	}
+
+	if (event->attr.precise_ip > 1) {
+		/*
+		 * Haswell and later processors have an 'eventing IP'
+		 * (real IP) which fixes the off-by-1 skid in hardware.
+		 * Use it when precise_ip >= 2 :
+		 */
+		if (x86_pmu.intel_cap.pebs_format >= 2) {
+			set_linear_ip(regs, pebs->real_ip);
+			regs->flags |= PERF_EFLAGS_EXACT;
+		} else {
+			/* Otherwise, use PEBS off-by-1 IP: */
+			set_linear_ip(regs, pebs->ip);
+
+			/*
+			 * With precise_ip >= 2, try to fix up the off-by-1 IP
+			 * using the LBR. If successful, the fixup function
+			 * corrects regs->ip and calls set_linear_ip() on regs:
+			 */
+			if (intel_pmu_pebs_fixup_ip(regs))
+				regs->flags |= PERF_EFLAGS_EXACT;
+		}
+	} else {
+		/*
+		 * When precise_ip == 1, return the PEBS off-by-1 IP,
+		 * no fixup attempted:
+		 */
+		set_linear_ip(regs, pebs->ip);
+	}
+
+
+	if ((sample_type & (PERF_SAMPLE_ADDR | PERF_SAMPLE_PHYS_ADDR)) &&
+	    x86_pmu.intel_cap.pebs_format >= 1)
+		data->addr = pebs->dla;
+
+	if (x86_pmu.intel_cap.pebs_format >= 2) {
+		/* Only set the TSX weight when no memory weight. */
+		if ((sample_type & PERF_SAMPLE_WEIGHT) && !fll)
+			data->weight = intel_get_tsx_weight(pebs->tsx_tuning);
+
+		if (sample_type & PERF_SAMPLE_TRANSACTION)
+			data->txn = intel_get_tsx_transaction(pebs->tsx_tuning,
+							      pebs->ax);
+	}
+
+	/*
+	 * v3 supplies an accurate time stamp, so we use that
+	 * for the time stamp.
+	 *
+	 * We can only do this for the default trace clock.
+	 */
+	if (x86_pmu.intel_cap.pebs_format >= 3 &&
+		event->attr.use_clockid == 0)
+		data->time = native_sched_clock_from_tsc(pebs->tsc);
+
+	if (has_branch_stack(event))
+		data->br_stack = &cpuc->lbr_stack;
+}
+
+static void adaptive_pebs_save_regs(struct pt_regs *regs,
+				    struct pebs_gprs *gprs)
+{
+	regs->ax = gprs->ax;
+	regs->bx = gprs->bx;
+	regs->cx = gprs->cx;
+	regs->dx = gprs->dx;
+	regs->si = gprs->si;
+	regs->di = gprs->di;
+	regs->bp = gprs->bp;
+	regs->sp = gprs->sp;
+#ifndef CONFIG_X86_32
+	regs->r8 = gprs->r8;
+	regs->r9 = gprs->r9;
+	regs->r10 = gprs->r10;
+	regs->r11 = gprs->r11;
+	regs->r12 = gprs->r12;
+	regs->r13 = gprs->r13;
+	regs->r14 = gprs->r14;
+	regs->r15 = gprs->r15;
+#endif
+}
+
+/*
+ * With adaptive PEBS the layout depends on what fields are configured.
+ */
+
+static void setup_pebs_adaptive_sample_data(struct perf_event *event,
+					    struct pt_regs *iregs, void *__pebs,
+					    struct perf_sample_data *data,
+					    struct pt_regs *regs)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct pebs_basic *basic = __pebs;
+	void *next_record = basic + 1;
+	u64 sample_type;
+	u64 format_size;
+	struct pebs_meminfo *meminfo = NULL;
+	struct pebs_gprs *gprs = NULL;
+	struct x86_perf_regs *perf_regs;
+
+	if (basic == NULL)
+		return;
+
+	perf_regs = container_of(regs, struct x86_perf_regs, regs);
+	perf_regs->xmm_regs = NULL;
+
+	sample_type = event->attr.sample_type;
+	format_size = basic->format_size;
+	perf_sample_data_init(data, 0, event->hw.last_period);
+	data->period = event->hw.last_period;
+
+	if (event->attr.use_clockid == 0)
+		data->time = native_sched_clock_from_tsc(basic->tsc);
+
+	/*
+	 * We must however always use iregs for the unwinder to stay sane; the
+	 * record BP,SP,IP can point into thin air when the record is from a
+	 * previous PMI context or an (I)RET happened between the record and
+	 * PMI.
+	 */
+	if (sample_type & PERF_SAMPLE_CALLCHAIN)
+		data->callchain = perf_callchain(event, iregs);
+
+	*regs = *iregs;
+	/* The ip in basic is EventingIP */
+	set_linear_ip(regs, basic->ip);
+	regs->flags = PERF_EFLAGS_EXACT;
+
+	/*
+	 * The record for MEMINFO is in front of GP
+	 * But PERF_SAMPLE_TRANSACTION needs gprs->ax.
+	 * Save the pointer here but process later.
+	 */
+	if (format_size & PEBS_DATACFG_MEMINFO) {
+		meminfo = next_record;
+		next_record = meminfo + 1;
+	}
+
+	if (format_size & PEBS_DATACFG_GP) {
+		gprs = next_record;
+		next_record = gprs + 1;
+
+		if (event->attr.precise_ip < 2) {
+			set_linear_ip(regs, gprs->ip);
+			regs->flags &= ~PERF_EFLAGS_EXACT;
+		}
+
+		if (sample_type & PERF_SAMPLE_REGS_INTR)
+			adaptive_pebs_save_regs(regs, gprs);
+	}
+
+	if (format_size & PEBS_DATACFG_MEMINFO) {
+		if (sample_type & PERF_SAMPLE_WEIGHT)
+			data->weight = meminfo->latency ?:
+				intel_get_tsx_weight(meminfo->tsx_tuning);
+
+		if (sample_type & PERF_SAMPLE_DATA_SRC)
+			data->data_src.val = get_data_src(event, meminfo->aux);
+
+		if (sample_type & (PERF_SAMPLE_ADDR | PERF_SAMPLE_PHYS_ADDR))
+			data->addr = meminfo->address;
+
+		if (sample_type & PERF_SAMPLE_TRANSACTION)
+			data->txn = intel_get_tsx_transaction(meminfo->tsx_tuning,
+							  gprs ? gprs->ax : 0);
+	}
+
+	if (format_size & PEBS_DATACFG_XMMS) {
+		struct pebs_xmm *xmm = next_record;
+
+		next_record = xmm + 1;
+		perf_regs->xmm_regs = xmm->xmm;
+	}
+
+	if (format_size & PEBS_DATACFG_LBRS) {
+		struct lbr_entry *lbr = next_record;
+		int num_lbr = ((format_size >> PEBS_DATACFG_LBR_SHIFT)
+					& 0xff) + 1;
+		next_record = next_record + num_lbr * sizeof(struct lbr_entry);
+
+		if (has_branch_stack(event)) {
+			intel_pmu_store_pebs_lbrs(lbr);
+			data->br_stack = &cpuc->lbr_stack;
+		}
+	}
+
+	WARN_ONCE(next_record != __pebs + (format_size >> 48),
+			"PEBS record size %llu, expected %llu, config %llx\n",
+			format_size >> 48,
+			(u64)(next_record - __pebs),
+			basic->format_size);
+}
+
+static inline void *
+get_next_pebs_record_by_bit(void *base, void *top, int bit)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	void *at;
+	u64 pebs_status;
+
+	/*
+	 * fmt0 does not have a status bitfield (does not use
+	 * perf_record_nhm format)
+	 */
+	if (x86_pmu.intel_cap.pebs_format < 1)
+		return base;
+
+	if (base == NULL)
+		return NULL;
+
+	for (at = base; at < top; at += cpuc->pebs_record_size) {
+		unsigned long status = get_pebs_status(at);
+
+		if (test_bit(bit, (unsigned long *)&status)) {
+			/* PEBS v3 has accurate status bits */
+			if (x86_pmu.intel_cap.pebs_format >= 3)
+				return at;
+
+			if (status == (1 << bit))
+				return at;
+
+			/* clear non-PEBS bit and re-check */
+			pebs_status = status & cpuc->pebs_enabled;
+			pebs_status &= PEBS_COUNTER_MASK;
+			if (pebs_status == (1 << bit))
+				return at;
+		}
+	}
+	return NULL;
+}
+
+void intel_pmu_auto_reload_read(struct perf_event *event)
+{
+	WARN_ON(!(event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD));
+
+	perf_pmu_disable(event->pmu);
+	intel_pmu_drain_pebs_buffer();
+	perf_pmu_enable(event->pmu);
+}
+
+/*
+ * Special variant of intel_pmu_save_and_restart() for auto-reload.
+ */
+static int
+intel_pmu_save_and_restart_reload(struct perf_event *event, int count)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int shift = 64 - x86_pmu.cntval_bits;
+	u64 period = hwc->sample_period;
+	u64 prev_raw_count, new_raw_count;
+	s64 new, old;
+
+	WARN_ON(!period);
+
+	/*
+	 * drain_pebs() only happens when the PMU is disabled.
+	 */
+	WARN_ON(this_cpu_read(cpu_hw_events.enabled));
+
+	prev_raw_count = local64_read(&hwc->prev_count);
+	rdpmcl(hwc->event_base_rdpmc, new_raw_count);
+	local64_set(&hwc->prev_count, new_raw_count);
+
+	/*
+	 * Since the counter increments a negative counter value and
+	 * overflows on the sign switch, giving the interval:
+	 *
+	 *   [-period, 0]
+	 *
+	 * the difference between two consequtive reads is:
+	 *
+	 *   A) value2 - value1;
+	 *      when no overflows have happened in between,
+	 *
+	 *   B) (0 - value1) + (value2 - (-period));
+	 *      when one overflow happened in between,
+	 *
+	 *   C) (0 - value1) + (n - 1) * (period) + (value2 - (-period));
+	 *      when @n overflows happened in between.
+	 *
+	 * Here A) is the obvious difference, B) is the extension to the
+	 * discrete interval, where the first term is to the top of the
+	 * interval and the second term is from the bottom of the next
+	 * interval and C) the extension to multiple intervals, where the
+	 * middle term is the whole intervals covered.
+	 *
+	 * An equivalent of C, by reduction, is:
+	 *
+	 *   value2 - value1 + n * period
+	 */
+	new = ((s64)(new_raw_count << shift) >> shift);
+	old = ((s64)(prev_raw_count << shift) >> shift);
+	local64_add(new - old + count * period, &event->count);
+
+	local64_set(&hwc->period_left, -new);
+
+	perf_event_update_userpage(event);
+
+	return 0;
+}
+
+static __always_inline void
+__intel_pmu_pebs_event(struct perf_event *event,
+		       struct pt_regs *iregs,
+		       struct perf_sample_data *data,
+		       void *base, void *top,
+		       int bit, int count,
+		       void (*setup_sample)(struct perf_event *,
+					    struct pt_regs *,
+					    void *,
+					    struct perf_sample_data *,
+					    struct pt_regs *))
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	struct x86_perf_regs perf_regs;
+	struct pt_regs *regs = &perf_regs.regs;
+	void *at = get_next_pebs_record_by_bit(base, top, bit);
+	static struct pt_regs dummy_iregs;
+
+	if (hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) {
+		/*
+		 * Now, auto-reload is only enabled in fixed period mode.
+		 * The reload value is always hwc->sample_period.
+		 * May need to change it, if auto-reload is enabled in
+		 * freq mode later.
+		 */
+		intel_pmu_save_and_restart_reload(event, count);
+	} else if (!intel_pmu_save_and_restart(event))
+		return;
+
+	if (!iregs)
+		iregs = &dummy_iregs;
+
+	while (count > 1) {
+		setup_sample(event, iregs, at, data, regs);
+		perf_event_output(event, data, regs);
+		at += cpuc->pebs_record_size;
+		at = get_next_pebs_record_by_bit(at, top, bit);
+		count--;
+	}
+
+	setup_sample(event, iregs, at, data, regs);
+	if (iregs == &dummy_iregs) {
+		/*
+		 * The PEBS records may be drained in the non-overflow context,
+		 * e.g., large PEBS + context switch. Perf should treat the
+		 * last record the same as other PEBS records, and doesn't
+		 * invoke the generic overflow handler.
+		 */
+		perf_event_output(event, data, regs);
+	} else {
+		/*
+		 * All but the last records are processed.
+		 * The last one is left to be able to call the overflow handler.
+		 */
+		if (perf_event_overflow(event, data, regs))
+			x86_pmu_stop(event, 0);
+	}
+}
+
+static void intel_pmu_drain_pebs_core(struct pt_regs *iregs, struct perf_sample_data *data)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct perf_event *event = cpuc->events[0]; /* PMC0 only */
+	struct pebs_record_core *at, *top;
+	int n;
+
+	if (!x86_pmu.pebs_active)
+		return;
+
+	at  = (struct pebs_record_core *)(unsigned long)ds->pebs_buffer_base;
+	top = (struct pebs_record_core *)(unsigned long)ds->pebs_index;
+
+	/*
+	 * Whatever else happens, drain the thing
+	 */
+	ds->pebs_index = ds->pebs_buffer_base;
+
+	if (!test_bit(0, cpuc->active_mask))
+		return;
+
+	WARN_ON_ONCE(!event);
+
+	if (!event->attr.precise_ip)
+		return;
+
+	n = top - at;
+	if (n <= 0) {
+		if (event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD)
+			intel_pmu_save_and_restart_reload(event, 0);
+		return;
+	}
+
+	__intel_pmu_pebs_event(event, iregs, data, at, top, 0, n,
+			       setup_pebs_fixed_sample_data);
+}
+
+static void intel_pmu_pebs_event_update_no_drain(struct cpu_hw_events *cpuc, int size)
+{
+	struct perf_event *event;
+	int bit;
+
+	/*
+	 * The drain_pebs() could be called twice in a short period
+	 * for auto-reload event in pmu::read(). There are no
+	 * overflows have happened in between.
+	 * It needs to call intel_pmu_save_and_restart_reload() to
+	 * update the event->count for this case.
+	 */
+	for_each_set_bit(bit, (unsigned long *)&cpuc->pebs_enabled, size) {
+		event = cpuc->events[bit];
+		if (event->hw.flags & PERF_X86_EVENT_AUTO_RELOAD)
+			intel_pmu_save_and_restart_reload(event, 0);
+	}
+}
+
+static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs, struct perf_sample_data *data)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct perf_event *event;
+	void *base, *at, *top;
+	short counts[INTEL_PMC_IDX_FIXED + MAX_FIXED_PEBS_EVENTS] = {};
+	short error[INTEL_PMC_IDX_FIXED + MAX_FIXED_PEBS_EVENTS] = {};
+	int bit, i, size;
+	u64 mask;
+
+	if (!x86_pmu.pebs_active)
+		return;
+
+	base = (struct pebs_record_nhm *)(unsigned long)ds->pebs_buffer_base;
+	top = (struct pebs_record_nhm *)(unsigned long)ds->pebs_index;
+
+	ds->pebs_index = ds->pebs_buffer_base;
+
+	mask = (1ULL << x86_pmu.max_pebs_events) - 1;
+	size = x86_pmu.max_pebs_events;
+	if (x86_pmu.flags & PMU_FL_PEBS_ALL) {
+		mask |= ((1ULL << x86_pmu.num_counters_fixed) - 1) << INTEL_PMC_IDX_FIXED;
+		size = INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed;
+	}
+
+	if (unlikely(base >= top)) {
+		intel_pmu_pebs_event_update_no_drain(cpuc, size);
+		return;
+	}
+
+	for (at = base; at < top; at += x86_pmu.pebs_record_size) {
+		struct pebs_record_nhm *p = at;
+		u64 pebs_status;
+
+		pebs_status = p->status & cpuc->pebs_enabled;
+		pebs_status &= mask;
+
+		/* PEBS v3 has more accurate status bits */
+		if (x86_pmu.intel_cap.pebs_format >= 3) {
+			for_each_set_bit(bit, (unsigned long *)&pebs_status, size)
+				counts[bit]++;
+
+			continue;
+		}
+
+		/*
+		 * On some CPUs the PEBS status can be zero when PEBS is
+		 * racing with clearing of GLOBAL_STATUS.
+		 *
+		 * Normally we would drop that record, but in the
+		 * case when there is only a single active PEBS event
+		 * we can assume it's for that event.
+		 */
+		if (!pebs_status && cpuc->pebs_enabled &&
+			!(cpuc->pebs_enabled & (cpuc->pebs_enabled-1)))
+			pebs_status = p->status = cpuc->pebs_enabled;
+
+		bit = find_first_bit((unsigned long *)&pebs_status,
+					x86_pmu.max_pebs_events);
+		if (bit >= x86_pmu.max_pebs_events)
+			continue;
+
+		/*
+		 * The PEBS hardware does not deal well with the situation
+		 * when events happen near to each other and multiple bits
+		 * are set. But it should happen rarely.
+		 *
+		 * If these events include one PEBS and multiple non-PEBS
+		 * events, it doesn't impact PEBS record. The record will
+		 * be handled normally. (slow path)
+		 *
+		 * If these events include two or more PEBS events, the
+		 * records for the events can be collapsed into a single
+		 * one, and it's not possible to reconstruct all events
+		 * that caused the PEBS record. It's called collision.
+		 * If collision happened, the record will be dropped.
+		 */
+		if (pebs_status != (1ULL << bit)) {
+			for_each_set_bit(i, (unsigned long *)&pebs_status, size)
+				error[i]++;
+			continue;
+		}
+
+		counts[bit]++;
+	}
+
+	for_each_set_bit(bit, (unsigned long *)&mask, size) {
+		if ((counts[bit] == 0) && (error[bit] == 0))
+			continue;
+
+		event = cpuc->events[bit];
+		if (WARN_ON_ONCE(!event))
+			continue;
+
+		if (WARN_ON_ONCE(!event->attr.precise_ip))
+			continue;
+
+		/* log dropped samples number */
+		if (error[bit]) {
+			perf_log_lost_samples(event, error[bit]);
+
+			if (iregs && perf_event_account_interrupt(event))
+				x86_pmu_stop(event, 0);
+		}
+
+		if (counts[bit]) {
+			__intel_pmu_pebs_event(event, iregs, data, base,
+					       top, bit, counts[bit],
+					       setup_pebs_fixed_sample_data);
+		}
+	}
+}
+
+static void intel_pmu_drain_pebs_icl(struct pt_regs *iregs, struct perf_sample_data *data)
+{
+	short counts[INTEL_PMC_IDX_FIXED + MAX_FIXED_PEBS_EVENTS] = {};
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct debug_store *ds = cpuc->ds;
+	struct perf_event *event;
+	void *base, *at, *top;
+	int bit, size;
+	u64 mask;
+
+	if (!x86_pmu.pebs_active)
+		return;
+
+	base = (struct pebs_basic *)(unsigned long)ds->pebs_buffer_base;
+	top = (struct pebs_basic *)(unsigned long)ds->pebs_index;
+
+	ds->pebs_index = ds->pebs_buffer_base;
+
+	mask = ((1ULL << x86_pmu.max_pebs_events) - 1) |
+	       (((1ULL << x86_pmu.num_counters_fixed) - 1) << INTEL_PMC_IDX_FIXED);
+	size = INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed;
+
+	if (unlikely(base >= top)) {
+		intel_pmu_pebs_event_update_no_drain(cpuc, size);
+		return;
+	}
+
+	for (at = base; at < top; at += cpuc->pebs_record_size) {
+		u64 pebs_status;
+
+		pebs_status = get_pebs_status(at) & cpuc->pebs_enabled;
+		pebs_status &= mask;
+
+		for_each_set_bit(bit, (unsigned long *)&pebs_status, size)
+			counts[bit]++;
+	}
+
+	for_each_set_bit(bit, (unsigned long *)&mask, size) {
+		if (counts[bit] == 0)
+			continue;
+
+		event = cpuc->events[bit];
+		if (WARN_ON_ONCE(!event))
+			continue;
+
+		if (WARN_ON_ONCE(!event->attr.precise_ip))
+			continue;
+
+		__intel_pmu_pebs_event(event, iregs, data, base,
+				       top, bit, counts[bit],
+				       setup_pebs_adaptive_sample_data);
+	}
+}
+
+/*
+ * BTS, PEBS probe and setup
+ */
+
+void __init intel_ds_init(void)
+{
+	/*
+	 * No support for 32bit formats
+	 */
+	if (!boot_cpu_has(X86_FEATURE_DTES64))
+		return;
+
+	x86_pmu.bts  = boot_cpu_has(X86_FEATURE_BTS);
+	x86_pmu.pebs = boot_cpu_has(X86_FEATURE_PEBS);
+	x86_pmu.pebs_buffer_size = PEBS_BUFFER_SIZE;
+	if (x86_pmu.version <= 4)
+		x86_pmu.pebs_no_isolation = 1;
+
+	if (x86_pmu.pebs) {
+		char pebs_type = x86_pmu.intel_cap.pebs_trap ?  '+' : '-';
+		char *pebs_qual = "";
+		int format = x86_pmu.intel_cap.pebs_format;
+
+		if (format < 4)
+			x86_pmu.intel_cap.pebs_baseline = 0;
+
+		switch (format) {
+		case 0:
+			pr_cont("PEBS fmt0%c, ", pebs_type);
+			x86_pmu.pebs_record_size = sizeof(struct pebs_record_core);
+			/*
+			 * Using >PAGE_SIZE buffers makes the WRMSR to
+			 * PERF_GLOBAL_CTRL in intel_pmu_enable_all()
+			 * mysteriously hang on Core2.
+			 *
+			 * As a workaround, we don't do this.
+			 */
+			x86_pmu.pebs_buffer_size = PAGE_SIZE;
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_core;
+			break;
+
+		case 1:
+			pr_cont("PEBS fmt1%c, ", pebs_type);
+			x86_pmu.pebs_record_size = sizeof(struct pebs_record_nhm);
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm;
+			break;
+
+		case 2:
+			pr_cont("PEBS fmt2%c, ", pebs_type);
+			x86_pmu.pebs_record_size = sizeof(struct pebs_record_hsw);
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm;
+			break;
+
+		case 3:
+			pr_cont("PEBS fmt3%c, ", pebs_type);
+			x86_pmu.pebs_record_size =
+						sizeof(struct pebs_record_skl);
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_nhm;
+			x86_pmu.large_pebs_flags |= PERF_SAMPLE_TIME;
+			break;
+
+		case 4:
+			x86_pmu.drain_pebs = intel_pmu_drain_pebs_icl;
+			x86_pmu.pebs_record_size = sizeof(struct pebs_basic);
+			if (x86_pmu.intel_cap.pebs_baseline) {
+				x86_pmu.large_pebs_flags |=
+					PERF_SAMPLE_BRANCH_STACK |
+					PERF_SAMPLE_TIME;
+				x86_pmu.flags |= PMU_FL_PEBS_ALL;
+				pebs_qual = "-baseline";
+				x86_get_pmu(smp_processor_id())->capabilities |= PERF_PMU_CAP_EXTENDED_REGS;
+			} else {
+				/* Only basic record supported */
+				x86_pmu.large_pebs_flags &=
+					~(PERF_SAMPLE_ADDR |
+					  PERF_SAMPLE_TIME |
+					  PERF_SAMPLE_DATA_SRC |
+					  PERF_SAMPLE_TRANSACTION |
+					  PERF_SAMPLE_REGS_USER |
+					  PERF_SAMPLE_REGS_INTR);
+			}
+			pr_cont("PEBS fmt4%c%s, ", pebs_type, pebs_qual);
+
+			if (x86_pmu.intel_cap.pebs_output_pt_available) {
+				pr_cont("PEBS-via-PT, ");
+				x86_get_pmu(smp_processor_id())->capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
+			}
+
+			break;
+
+		default:
+			pr_cont("no PEBS fmt%d%c, ", format, pebs_type);
+			x86_pmu.pebs = 0;
+		}
+	}
+}
+
+void perf_restore_debug_store(void)
+{
+	struct debug_store *ds = __this_cpu_read(cpu_hw_events.ds);
+
+	if (!x86_pmu.bts && !x86_pmu.pebs)
+		return;
+
+	wrmsrl(MSR_IA32_DS_AREA, (unsigned long)ds);
+}
diff --git a/arch/x86/events/intel/knc.c b/arch/x86/events/intel/knc.c
new file mode 100644
index 000000000..618001c20
--- /dev/null
+++ b/arch/x86/events/intel/knc.c
@@ -0,0 +1,322 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Driver for Intel Xeon Phi "Knights Corner" PMU */
+
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include <asm/hardirq.h>
+
+#include "../perf_event.h"
+
+static const u64 knc_perfmon_event_map[] =
+{
+  [PERF_COUNT_HW_CPU_CYCLES]		= 0x002a,
+  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x0016,
+  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x0028,
+  [PERF_COUNT_HW_CACHE_MISSES]		= 0x0029,
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x0012,
+  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x002b,
+};
+
+static const u64 __initconst knc_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) ] = {
+		/* On Xeon Phi event "0" is a valid DATA_READ          */
+		/*   (L1 Data Cache Reads) Instruction.                */
+		/* We code this as ARCH_PERFMON_EVENTSEL_INT as this   */
+		/* bit will always be set in x86_pmu_hw_config().      */
+		[ C(RESULT_ACCESS) ] = ARCH_PERFMON_EVENTSEL_INT,
+						/* DATA_READ           */
+		[ C(RESULT_MISS)   ] = 0x0003,	/* DATA_READ_MISS      */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0001,	/* DATA_WRITE          */
+		[ C(RESULT_MISS)   ] = 0x0004,	/* DATA_WRITE_MISS     */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0011,	/* L1_DATA_PF1         */
+		[ C(RESULT_MISS)   ] = 0x001c,	/* L1_DATA_PF1_MISS    */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x000c,	/* CODE_READ          */
+		[ C(RESULT_MISS)   ] = 0x000e,	/* CODE_CACHE_MISS    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x10cb,	/* L2_READ_MISS */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x10cc,	/* L2_WRITE_HIT */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x10fc,	/* L2_DATA_PF2      */
+		[ C(RESULT_MISS)   ] = 0x10fe,	/* L2_DATA_PF2_MISS */
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = ARCH_PERFMON_EVENTSEL_INT,
+						/* DATA_READ */
+						/* see note on L1 OP_READ */
+		[ C(RESULT_MISS)   ] = 0x0002,	/* DATA_PAGE_WALK */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0001,	/* DATA_WRITE */
+		[ C(RESULT_MISS)   ] = 0x0002,	/* DATA_PAGE_WALK */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x000c,	/* CODE_READ */
+		[ C(RESULT_MISS)   ] = 0x000d,	/* CODE_PAGE_WALK */
+	},
+	[ 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) ] = 0x0012,	/* BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x002b,	/* BRANCHES_MISPREDICTED */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+
+static u64 knc_pmu_event_map(int hw_event)
+{
+	return knc_perfmon_event_map[hw_event];
+}
+
+static struct event_constraint knc_event_constraints[] =
+{
+	INTEL_EVENT_CONSTRAINT(0xc3, 0x1),	/* HWP_L2HIT */
+	INTEL_EVENT_CONSTRAINT(0xc4, 0x1),	/* HWP_L2MISS */
+	INTEL_EVENT_CONSTRAINT(0xc8, 0x1),	/* L2_READ_HIT_E */
+	INTEL_EVENT_CONSTRAINT(0xc9, 0x1),	/* L2_READ_HIT_M */
+	INTEL_EVENT_CONSTRAINT(0xca, 0x1),	/* L2_READ_HIT_S */
+	INTEL_EVENT_CONSTRAINT(0xcb, 0x1),	/* L2_READ_MISS */
+	INTEL_EVENT_CONSTRAINT(0xcc, 0x1),	/* L2_WRITE_HIT */
+	INTEL_EVENT_CONSTRAINT(0xce, 0x1),	/* L2_STRONGLY_ORDERED_STREAMING_VSTORES_MISS */
+	INTEL_EVENT_CONSTRAINT(0xcf, 0x1),	/* L2_WEAKLY_ORDERED_STREAMING_VSTORE_MISS */
+	INTEL_EVENT_CONSTRAINT(0xd7, 0x1),	/* L2_VICTIM_REQ_WITH_DATA */
+	INTEL_EVENT_CONSTRAINT(0xe3, 0x1),	/* SNP_HITM_BUNIT */
+	INTEL_EVENT_CONSTRAINT(0xe6, 0x1),	/* SNP_HIT_L2 */
+	INTEL_EVENT_CONSTRAINT(0xe7, 0x1),	/* SNP_HITM_L2 */
+	INTEL_EVENT_CONSTRAINT(0xf1, 0x1),	/* L2_DATA_READ_MISS_CACHE_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf2, 0x1),	/* L2_DATA_WRITE_MISS_CACHE_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf6, 0x1),	/* L2_DATA_READ_MISS_MEM_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf7, 0x1),	/* L2_DATA_WRITE_MISS_MEM_FILL */
+	INTEL_EVENT_CONSTRAINT(0xfc, 0x1),	/* L2_DATA_PF2 */
+	INTEL_EVENT_CONSTRAINT(0xfd, 0x1),	/* L2_DATA_PF2_DROP */
+	INTEL_EVENT_CONSTRAINT(0xfe, 0x1),	/* L2_DATA_PF2_MISS */
+	INTEL_EVENT_CONSTRAINT(0xff, 0x1),	/* L2_DATA_HIT_INFLIGHT_PF2 */
+	EVENT_CONSTRAINT_END
+};
+
+#define MSR_KNC_IA32_PERF_GLOBAL_STATUS		0x0000002d
+#define MSR_KNC_IA32_PERF_GLOBAL_OVF_CONTROL	0x0000002e
+#define MSR_KNC_IA32_PERF_GLOBAL_CTRL		0x0000002f
+
+#define KNC_ENABLE_COUNTER0			0x00000001
+#define KNC_ENABLE_COUNTER1			0x00000002
+
+static void knc_pmu_disable_all(void)
+{
+	u64 val;
+
+	rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+	val &= ~(KNC_ENABLE_COUNTER0|KNC_ENABLE_COUNTER1);
+	wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+}
+
+static void knc_pmu_enable_all(int added)
+{
+	u64 val;
+
+	rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+	val |= (KNC_ENABLE_COUNTER0|KNC_ENABLE_COUNTER1);
+	wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+}
+
+static inline void
+knc_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 val;
+
+	val = hwc->config;
+	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 hw_perf_event *hwc = &event->hw;
+	u64 val;
+
+	val = hwc->config;
+	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 = this_cpu_ptr(&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:
+	/* Only restore PMU state when it's active. See x86_pmu_disable(). */
+	if (cpuc->enabled)
+		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"	);
+PMU_FORMAT_ATTR(inv,	"config:23"	);
+PMU_FORMAT_ATTR(cmask,	"config:24-31"	);
+
+static struct attribute *intel_knc_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+static const struct x86_pmu knc_pmu __initconst = {
+	.name			= "knc",
+	.handle_irq		= knc_pmu_handle_irq,
+	.disable_all		= knc_pmu_disable_all,
+	.enable_all		= knc_pmu_enable_all,
+	.enable			= knc_pmu_enable_event,
+	.disable		= knc_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_KNC_EVNTSEL0,
+	.perfctr		= MSR_KNC_PERFCTR0,
+	.event_map		= knc_pmu_event_map,
+	.max_events             = ARRAY_SIZE(knc_perfmon_event_map),
+	.apic			= 1,
+	.max_period		= (1ULL << 39) - 1,
+	.version		= 0,
+	.num_counters		= 2,
+	.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,
+};
+
+__init int knc_pmu_init(void)
+{
+	x86_pmu = knc_pmu;
+
+	memcpy(hw_cache_event_ids, knc_hw_cache_event_ids, 
+		sizeof(hw_cache_event_ids));
+
+	return 0;
+}
diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c
new file mode 100644
index 000000000..4b6c39c5f
--- /dev/null
+++ b/arch/x86/events/intel/lbr.c
@@ -0,0 +1,1875 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include <asm/perf_event.h>
+#include <asm/msr.h>
+#include <asm/insn.h>
+
+#include "../perf_event.h"
+
+static const enum {
+	LBR_EIP_FLAGS		= 1,
+	LBR_TSX			= 2,
+} lbr_desc[LBR_FORMAT_MAX_KNOWN + 1] = {
+	[LBR_FORMAT_EIP_FLAGS]  = LBR_EIP_FLAGS,
+	[LBR_FORMAT_EIP_FLAGS2] = LBR_EIP_FLAGS | LBR_TSX,
+};
+
+/*
+ * Intel LBR_SELECT bits
+ * Intel Vol3a, April 2011, Section 16.7 Table 16-10
+ *
+ * Hardware branch filter (not available on all CPUs)
+ */
+#define LBR_KERNEL_BIT		0 /* do not capture at ring0 */
+#define LBR_USER_BIT		1 /* do not capture at ring > 0 */
+#define LBR_JCC_BIT		2 /* do not capture conditional branches */
+#define LBR_REL_CALL_BIT	3 /* do not capture relative calls */
+#define LBR_IND_CALL_BIT	4 /* do not capture indirect calls */
+#define LBR_RETURN_BIT		5 /* do not capture near returns */
+#define LBR_IND_JMP_BIT		6 /* do not capture indirect jumps */
+#define LBR_REL_JMP_BIT		7 /* do not capture relative jumps */
+#define LBR_FAR_BIT		8 /* do not capture far branches */
+#define LBR_CALL_STACK_BIT	9 /* enable call stack */
+
+/*
+ * Following bit only exists in Linux; we mask it out before writing it to
+ * the actual MSR. But it helps the constraint perf code to understand
+ * that this is a separate configuration.
+ */
+#define LBR_NO_INFO_BIT	       63 /* don't read LBR_INFO. */
+
+#define LBR_KERNEL	(1 << LBR_KERNEL_BIT)
+#define LBR_USER	(1 << LBR_USER_BIT)
+#define LBR_JCC		(1 << LBR_JCC_BIT)
+#define LBR_REL_CALL	(1 << LBR_REL_CALL_BIT)
+#define LBR_IND_CALL	(1 << LBR_IND_CALL_BIT)
+#define LBR_RETURN	(1 << LBR_RETURN_BIT)
+#define LBR_REL_JMP	(1 << LBR_REL_JMP_BIT)
+#define LBR_IND_JMP	(1 << LBR_IND_JMP_BIT)
+#define LBR_FAR		(1 << LBR_FAR_BIT)
+#define LBR_CALL_STACK	(1 << LBR_CALL_STACK_BIT)
+#define LBR_NO_INFO	(1ULL << LBR_NO_INFO_BIT)
+
+#define LBR_PLM (LBR_KERNEL | LBR_USER)
+
+#define LBR_SEL_MASK	0x3ff	/* valid bits in LBR_SELECT */
+#define LBR_NOT_SUPP	-1	/* LBR filter not supported */
+#define LBR_IGN		0	/* ignored */
+
+#define LBR_ANY		 \
+	(LBR_JCC	|\
+	 LBR_REL_CALL	|\
+	 LBR_IND_CALL	|\
+	 LBR_RETURN	|\
+	 LBR_REL_JMP	|\
+	 LBR_IND_JMP	|\
+	 LBR_FAR)
+
+#define LBR_FROM_FLAG_MISPRED	BIT_ULL(63)
+#define LBR_FROM_FLAG_IN_TX	BIT_ULL(62)
+#define LBR_FROM_FLAG_ABORT	BIT_ULL(61)
+
+#define LBR_FROM_SIGNEXT_2MSB	(BIT_ULL(60) | BIT_ULL(59))
+
+/*
+ * x86control flow change classification
+ * x86control flow changes include branches, interrupts, traps, faults
+ */
+enum {
+	X86_BR_NONE		= 0,      /* unknown */
+
+	X86_BR_USER		= 1 << 0, /* branch target is user */
+	X86_BR_KERNEL		= 1 << 1, /* branch target is kernel */
+
+	X86_BR_CALL		= 1 << 2, /* call */
+	X86_BR_RET		= 1 << 3, /* return */
+	X86_BR_SYSCALL		= 1 << 4, /* syscall */
+	X86_BR_SYSRET		= 1 << 5, /* syscall return */
+	X86_BR_INT		= 1 << 6, /* sw interrupt */
+	X86_BR_IRET		= 1 << 7, /* return from interrupt */
+	X86_BR_JCC		= 1 << 8, /* conditional */
+	X86_BR_JMP		= 1 << 9, /* jump */
+	X86_BR_IRQ		= 1 << 10,/* hw interrupt or trap or fault */
+	X86_BR_IND_CALL		= 1 << 11,/* indirect calls */
+	X86_BR_ABORT		= 1 << 12,/* transaction abort */
+	X86_BR_IN_TX		= 1 << 13,/* in transaction */
+	X86_BR_NO_TX		= 1 << 14,/* not in transaction */
+	X86_BR_ZERO_CALL	= 1 << 15,/* zero length call */
+	X86_BR_CALL_STACK	= 1 << 16,/* call stack */
+	X86_BR_IND_JMP		= 1 << 17,/* indirect jump */
+
+	X86_BR_TYPE_SAVE	= 1 << 18,/* indicate to save branch type */
+
+};
+
+#define X86_BR_PLM (X86_BR_USER | X86_BR_KERNEL)
+#define X86_BR_ANYTX (X86_BR_NO_TX | X86_BR_IN_TX)
+
+#define X86_BR_ANY       \
+	(X86_BR_CALL    |\
+	 X86_BR_RET     |\
+	 X86_BR_SYSCALL |\
+	 X86_BR_SYSRET  |\
+	 X86_BR_INT     |\
+	 X86_BR_IRET    |\
+	 X86_BR_JCC     |\
+	 X86_BR_JMP	 |\
+	 X86_BR_IRQ	 |\
+	 X86_BR_ABORT	 |\
+	 X86_BR_IND_CALL |\
+	 X86_BR_IND_JMP  |\
+	 X86_BR_ZERO_CALL)
+
+#define X86_BR_ALL (X86_BR_PLM | X86_BR_ANY)
+
+#define X86_BR_ANY_CALL		 \
+	(X86_BR_CALL		|\
+	 X86_BR_IND_CALL	|\
+	 X86_BR_ZERO_CALL	|\
+	 X86_BR_SYSCALL		|\
+	 X86_BR_IRQ		|\
+	 X86_BR_INT)
+
+/*
+ * Intel LBR_CTL bits
+ *
+ * Hardware branch filter for Arch LBR
+ */
+#define ARCH_LBR_KERNEL_BIT		1  /* capture at ring0 */
+#define ARCH_LBR_USER_BIT		2  /* capture at ring > 0 */
+#define ARCH_LBR_CALL_STACK_BIT		3  /* enable call stack */
+#define ARCH_LBR_JCC_BIT		16 /* capture conditional branches */
+#define ARCH_LBR_REL_JMP_BIT		17 /* capture relative jumps */
+#define ARCH_LBR_IND_JMP_BIT		18 /* capture indirect jumps */
+#define ARCH_LBR_REL_CALL_BIT		19 /* capture relative calls */
+#define ARCH_LBR_IND_CALL_BIT		20 /* capture indirect calls */
+#define ARCH_LBR_RETURN_BIT		21 /* capture near returns */
+#define ARCH_LBR_OTHER_BRANCH_BIT	22 /* capture other branches */
+
+#define ARCH_LBR_KERNEL			(1ULL << ARCH_LBR_KERNEL_BIT)
+#define ARCH_LBR_USER			(1ULL << ARCH_LBR_USER_BIT)
+#define ARCH_LBR_CALL_STACK		(1ULL << ARCH_LBR_CALL_STACK_BIT)
+#define ARCH_LBR_JCC			(1ULL << ARCH_LBR_JCC_BIT)
+#define ARCH_LBR_REL_JMP		(1ULL << ARCH_LBR_REL_JMP_BIT)
+#define ARCH_LBR_IND_JMP		(1ULL << ARCH_LBR_IND_JMP_BIT)
+#define ARCH_LBR_REL_CALL		(1ULL << ARCH_LBR_REL_CALL_BIT)
+#define ARCH_LBR_IND_CALL		(1ULL << ARCH_LBR_IND_CALL_BIT)
+#define ARCH_LBR_RETURN			(1ULL << ARCH_LBR_RETURN_BIT)
+#define ARCH_LBR_OTHER_BRANCH		(1ULL << ARCH_LBR_OTHER_BRANCH_BIT)
+
+#define ARCH_LBR_ANY			 \
+	(ARCH_LBR_JCC			|\
+	 ARCH_LBR_REL_JMP		|\
+	 ARCH_LBR_IND_JMP		|\
+	 ARCH_LBR_REL_CALL		|\
+	 ARCH_LBR_IND_CALL		|\
+	 ARCH_LBR_RETURN		|\
+	 ARCH_LBR_OTHER_BRANCH)
+
+#define ARCH_LBR_CTL_MASK			0x7f000e
+
+static void intel_pmu_lbr_filter(struct cpu_hw_events *cpuc);
+
+static __always_inline bool is_lbr_call_stack_bit_set(u64 config)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return !!(config & ARCH_LBR_CALL_STACK);
+
+	return !!(config & LBR_CALL_STACK);
+}
+
+/*
+ * We only support LBR implementations that have FREEZE_LBRS_ON_PMI
+ * otherwise it becomes near impossible to get a reliable stack.
+ */
+
+static void __intel_pmu_lbr_enable(bool pmi)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	u64 debugctl, lbr_select = 0, orig_debugctl;
+
+	/*
+	 * No need to unfreeze manually, as v4 can do that as part
+	 * of the GLOBAL_STATUS ack.
+	 */
+	if (pmi && x86_pmu.version >= 4)
+		return;
+
+	/*
+	 * No need to reprogram LBR_SELECT in a PMI, as it
+	 * did not change.
+	 */
+	if (cpuc->lbr_sel)
+		lbr_select = cpuc->lbr_sel->config & x86_pmu.lbr_sel_mask;
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR) && !pmi && cpuc->lbr_sel)
+		wrmsrl(MSR_LBR_SELECT, lbr_select);
+
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+	orig_debugctl = debugctl;
+
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR))
+		debugctl |= DEBUGCTLMSR_LBR;
+	/*
+	 * LBR callstack does not work well with FREEZE_LBRS_ON_PMI.
+	 * If FREEZE_LBRS_ON_PMI is set, PMI near call/return instructions
+	 * may cause superfluous increase/decrease of LBR_TOS.
+	 */
+	if (is_lbr_call_stack_bit_set(lbr_select))
+		debugctl &= ~DEBUGCTLMSR_FREEZE_LBRS_ON_PMI;
+	else
+		debugctl |= DEBUGCTLMSR_FREEZE_LBRS_ON_PMI;
+
+	if (orig_debugctl != debugctl)
+		wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR))
+		wrmsrl(MSR_ARCH_LBR_CTL, lbr_select | ARCH_LBR_CTL_LBREN);
+}
+
+static void __intel_pmu_lbr_disable(void)
+{
+	u64 debugctl;
+
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR)) {
+		wrmsrl(MSR_ARCH_LBR_CTL, 0);
+		return;
+	}
+
+	rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+	debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI);
+	wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+}
+
+void intel_pmu_lbr_reset_32(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++)
+		wrmsrl(x86_pmu.lbr_from + i, 0);
+}
+
+void intel_pmu_lbr_reset_64(void)
+{
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		wrmsrl(x86_pmu.lbr_from + i, 0);
+		wrmsrl(x86_pmu.lbr_to   + i, 0);
+		if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO)
+			wrmsrl(x86_pmu.lbr_info + i, 0);
+	}
+}
+
+static void intel_pmu_arch_lbr_reset(void)
+{
+	/* Write to ARCH_LBR_DEPTH MSR, all LBR entries are reset to 0 */
+	wrmsrl(MSR_ARCH_LBR_DEPTH, x86_pmu.lbr_nr);
+}
+
+void intel_pmu_lbr_reset(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	x86_pmu.lbr_reset();
+
+	cpuc->last_task_ctx = NULL;
+	cpuc->last_log_id = 0;
+}
+
+/*
+ * TOS = most recently recorded branch
+ */
+static inline u64 intel_pmu_lbr_tos(void)
+{
+	u64 tos;
+
+	rdmsrl(x86_pmu.lbr_tos, tos);
+	return tos;
+}
+
+enum {
+	LBR_NONE,
+	LBR_VALID,
+};
+
+/*
+ * For formats with LBR_TSX flags (e.g. LBR_FORMAT_EIP_FLAGS2), bits 61:62 in
+ * MSR_LAST_BRANCH_FROM_x are the TSX flags when TSX is supported, but when
+ * TSX is not supported they have no consistent behavior:
+ *
+ *   - For wrmsr(), bits 61:62 are considered part of the sign extension.
+ *   - For HW updates (branch captures) bits 61:62 are always OFF and are not
+ *     part of the sign extension.
+ *
+ * Therefore, if:
+ *
+ *   1) LBR has TSX format
+ *   2) CPU has no TSX support enabled
+ *
+ * ... then any value passed to wrmsr() must be sign extended to 63 bits and any
+ * value from rdmsr() must be converted to have a 61 bits sign extension,
+ * ignoring the TSX flags.
+ */
+static inline bool lbr_from_signext_quirk_needed(void)
+{
+	int lbr_format = x86_pmu.intel_cap.lbr_format;
+	bool tsx_support = boot_cpu_has(X86_FEATURE_HLE) ||
+			   boot_cpu_has(X86_FEATURE_RTM);
+
+	return !tsx_support && (lbr_desc[lbr_format] & LBR_TSX);
+}
+
+static DEFINE_STATIC_KEY_FALSE(lbr_from_quirk_key);
+
+/* If quirk is enabled, ensure sign extension is 63 bits: */
+inline u64 lbr_from_signext_quirk_wr(u64 val)
+{
+	if (static_branch_unlikely(&lbr_from_quirk_key)) {
+		/*
+		 * Sign extend into bits 61:62 while preserving bit 63.
+		 *
+		 * Quirk is enabled when TSX is disabled. Therefore TSX bits
+		 * in val are always OFF and must be changed to be sign
+		 * extension bits. Since bits 59:60 are guaranteed to be
+		 * part of the sign extension bits, we can just copy them
+		 * to 61:62.
+		 */
+		val |= (LBR_FROM_SIGNEXT_2MSB & val) << 2;
+	}
+	return val;
+}
+
+/*
+ * If quirk is needed, ensure sign extension is 61 bits:
+ */
+static u64 lbr_from_signext_quirk_rd(u64 val)
+{
+	if (static_branch_unlikely(&lbr_from_quirk_key)) {
+		/*
+		 * Quirk is on when TSX is not enabled. Therefore TSX
+		 * flags must be read as OFF.
+		 */
+		val &= ~(LBR_FROM_FLAG_IN_TX | LBR_FROM_FLAG_ABORT);
+	}
+	return val;
+}
+
+static __always_inline void wrlbr_from(unsigned int idx, u64 val)
+{
+	val = lbr_from_signext_quirk_wr(val);
+	wrmsrl(x86_pmu.lbr_from + idx, val);
+}
+
+static __always_inline void wrlbr_to(unsigned int idx, u64 val)
+{
+	wrmsrl(x86_pmu.lbr_to + idx, val);
+}
+
+static __always_inline void wrlbr_info(unsigned int idx, u64 val)
+{
+	wrmsrl(x86_pmu.lbr_info + idx, val);
+}
+
+static __always_inline u64 rdlbr_from(unsigned int idx, struct lbr_entry *lbr)
+{
+	u64 val;
+
+	if (lbr)
+		return lbr->from;
+
+	rdmsrl(x86_pmu.lbr_from + idx, val);
+
+	return lbr_from_signext_quirk_rd(val);
+}
+
+static __always_inline u64 rdlbr_to(unsigned int idx, struct lbr_entry *lbr)
+{
+	u64 val;
+
+	if (lbr)
+		return lbr->to;
+
+	rdmsrl(x86_pmu.lbr_to + idx, val);
+
+	return val;
+}
+
+static __always_inline u64 rdlbr_info(unsigned int idx, struct lbr_entry *lbr)
+{
+	u64 val;
+
+	if (lbr)
+		return lbr->info;
+
+	rdmsrl(x86_pmu.lbr_info + idx, val);
+
+	return val;
+}
+
+static inline void
+wrlbr_all(struct lbr_entry *lbr, unsigned int idx, bool need_info)
+{
+	wrlbr_from(idx, lbr->from);
+	wrlbr_to(idx, lbr->to);
+	if (need_info)
+		wrlbr_info(idx, lbr->info);
+}
+
+static inline bool
+rdlbr_all(struct lbr_entry *lbr, unsigned int idx, bool need_info)
+{
+	u64 from = rdlbr_from(idx, NULL);
+
+	/* Don't read invalid entry */
+	if (!from)
+		return false;
+
+	lbr->from = from;
+	lbr->to = rdlbr_to(idx, NULL);
+	if (need_info)
+		lbr->info = rdlbr_info(idx, NULL);
+
+	return true;
+}
+
+void intel_pmu_lbr_restore(void *ctx)
+{
+	bool need_info = x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO;
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct x86_perf_task_context *task_ctx = ctx;
+	int i;
+	unsigned lbr_idx, mask;
+	u64 tos = task_ctx->tos;
+
+	mask = x86_pmu.lbr_nr - 1;
+	for (i = 0; i < task_ctx->valid_lbrs; i++) {
+		lbr_idx = (tos - i) & mask;
+		wrlbr_all(&task_ctx->lbr[i], lbr_idx, need_info);
+	}
+
+	for (; i < x86_pmu.lbr_nr; i++) {
+		lbr_idx = (tos - i) & mask;
+		wrlbr_from(lbr_idx, 0);
+		wrlbr_to(lbr_idx, 0);
+		if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO)
+			wrlbr_info(lbr_idx, 0);
+	}
+
+	wrmsrl(x86_pmu.lbr_tos, tos);
+
+	if (cpuc->lbr_select)
+		wrmsrl(MSR_LBR_SELECT, task_ctx->lbr_sel);
+}
+
+static void intel_pmu_arch_lbr_restore(void *ctx)
+{
+	struct x86_perf_task_context_arch_lbr *task_ctx = ctx;
+	struct lbr_entry *entries = task_ctx->entries;
+	int i;
+
+	/* Fast reset the LBRs before restore if the call stack is not full. */
+	if (!entries[x86_pmu.lbr_nr - 1].from)
+		intel_pmu_arch_lbr_reset();
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		if (!entries[i].from)
+			break;
+		wrlbr_all(&entries[i], i, true);
+	}
+}
+
+/*
+ * Restore the Architecture LBR state from the xsave area in the perf
+ * context data for the task via the XRSTORS instruction.
+ */
+static void intel_pmu_arch_lbr_xrstors(void *ctx)
+{
+	struct x86_perf_task_context_arch_lbr_xsave *task_ctx = ctx;
+
+	copy_kernel_to_dynamic_supervisor(&task_ctx->xsave, XFEATURE_MASK_LBR);
+}
+
+static __always_inline bool lbr_is_reset_in_cstate(void *ctx)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return x86_pmu.lbr_deep_c_reset && !rdlbr_from(0, NULL);
+
+	return !rdlbr_from(((struct x86_perf_task_context *)ctx)->tos, NULL);
+}
+
+static void __intel_pmu_lbr_restore(void *ctx)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (task_context_opt(ctx)->lbr_callstack_users == 0 ||
+	    task_context_opt(ctx)->lbr_stack_state == LBR_NONE) {
+		intel_pmu_lbr_reset();
+		return;
+	}
+
+	/*
+	 * Does not restore the LBR registers, if
+	 * - No one else touched them, and
+	 * - Was not cleared in Cstate
+	 */
+	if ((ctx == cpuc->last_task_ctx) &&
+	    (task_context_opt(ctx)->log_id == cpuc->last_log_id) &&
+	    !lbr_is_reset_in_cstate(ctx)) {
+		task_context_opt(ctx)->lbr_stack_state = LBR_NONE;
+		return;
+	}
+
+	x86_pmu.lbr_restore(ctx);
+
+	task_context_opt(ctx)->lbr_stack_state = LBR_NONE;
+}
+
+void intel_pmu_lbr_save(void *ctx)
+{
+	bool need_info = x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO;
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	struct x86_perf_task_context *task_ctx = ctx;
+	unsigned lbr_idx, mask;
+	u64 tos;
+	int i;
+
+	mask = x86_pmu.lbr_nr - 1;
+	tos = intel_pmu_lbr_tos();
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		lbr_idx = (tos - i) & mask;
+		if (!rdlbr_all(&task_ctx->lbr[i], lbr_idx, need_info))
+			break;
+	}
+	task_ctx->valid_lbrs = i;
+	task_ctx->tos = tos;
+
+	if (cpuc->lbr_select)
+		rdmsrl(MSR_LBR_SELECT, task_ctx->lbr_sel);
+}
+
+static void intel_pmu_arch_lbr_save(void *ctx)
+{
+	struct x86_perf_task_context_arch_lbr *task_ctx = ctx;
+	struct lbr_entry *entries = task_ctx->entries;
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		if (!rdlbr_all(&entries[i], i, true))
+			break;
+	}
+
+	/* LBR call stack is not full. Reset is required in restore. */
+	if (i < x86_pmu.lbr_nr)
+		entries[x86_pmu.lbr_nr - 1].from = 0;
+}
+
+/*
+ * Save the Architecture LBR state to the xsave area in the perf
+ * context data for the task via the XSAVES instruction.
+ */
+static void intel_pmu_arch_lbr_xsaves(void *ctx)
+{
+	struct x86_perf_task_context_arch_lbr_xsave *task_ctx = ctx;
+
+	copy_dynamic_supervisor_to_kernel(&task_ctx->xsave, XFEATURE_MASK_LBR);
+}
+
+static void __intel_pmu_lbr_save(void *ctx)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (task_context_opt(ctx)->lbr_callstack_users == 0) {
+		task_context_opt(ctx)->lbr_stack_state = LBR_NONE;
+		return;
+	}
+
+	x86_pmu.lbr_save(ctx);
+
+	task_context_opt(ctx)->lbr_stack_state = LBR_VALID;
+
+	cpuc->last_task_ctx = ctx;
+	cpuc->last_log_id = ++task_context_opt(ctx)->log_id;
+}
+
+void intel_pmu_lbr_swap_task_ctx(struct perf_event_context *prev,
+				 struct perf_event_context *next)
+{
+	void *prev_ctx_data, *next_ctx_data;
+
+	swap(prev->task_ctx_data, next->task_ctx_data);
+
+	/*
+	 * Architecture specific synchronization makes sense in
+	 * case both prev->task_ctx_data and next->task_ctx_data
+	 * pointers are allocated.
+	 */
+
+	prev_ctx_data = next->task_ctx_data;
+	next_ctx_data = prev->task_ctx_data;
+
+	if (!prev_ctx_data || !next_ctx_data)
+		return;
+
+	swap(task_context_opt(prev_ctx_data)->lbr_callstack_users,
+	     task_context_opt(next_ctx_data)->lbr_callstack_users);
+}
+
+void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	void *task_ctx;
+
+	if (!cpuc->lbr_users)
+		return;
+
+	/*
+	 * If LBR callstack feature is enabled and the stack was saved when
+	 * the task was scheduled out, restore the stack. Otherwise flush
+	 * the LBR stack.
+	 */
+	task_ctx = ctx ? ctx->task_ctx_data : NULL;
+	if (task_ctx) {
+		if (sched_in)
+			__intel_pmu_lbr_restore(task_ctx);
+		else
+			__intel_pmu_lbr_save(task_ctx);
+		return;
+	}
+
+	/*
+	 * Since a context switch can flip the address space and LBR entries
+	 * are not tagged with an identifier, we need to wipe the LBR, even for
+	 * per-cpu events. You simply cannot resolve the branches from the old
+	 * address space.
+	 */
+	if (sched_in)
+		intel_pmu_lbr_reset();
+}
+
+static inline bool branch_user_callstack(unsigned br_sel)
+{
+	return (br_sel & X86_BR_USER) && (br_sel & X86_BR_CALL_STACK);
+}
+
+void intel_pmu_lbr_add(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	if (event->hw.flags & PERF_X86_EVENT_LBR_SELECT)
+		cpuc->lbr_select = 1;
+
+	cpuc->br_sel = event->hw.branch_reg.reg;
+
+	if (branch_user_callstack(cpuc->br_sel) && event->ctx->task_ctx_data)
+		task_context_opt(event->ctx->task_ctx_data)->lbr_callstack_users++;
+
+	/*
+	 * Request pmu::sched_task() callback, which will fire inside the
+	 * regular perf event scheduling, so that call will:
+	 *
+	 *  - restore or wipe; when LBR-callstack,
+	 *  - wipe; otherwise,
+	 *
+	 * when this is from __perf_event_task_sched_in().
+	 *
+	 * However, if this is from perf_install_in_context(), no such callback
+	 * will follow and we'll need to reset the LBR here if this is the
+	 * first LBR event.
+	 *
+	 * The problem is, we cannot tell these cases apart... but we can
+	 * exclude the biggest chunk of cases by looking at
+	 * event->total_time_running. An event that has accrued runtime cannot
+	 * be 'new'. Conversely, a new event can get installed through the
+	 * context switch path for the first time.
+	 */
+	if (x86_pmu.intel_cap.pebs_baseline && event->attr.precise_ip > 0)
+		cpuc->lbr_pebs_users++;
+	perf_sched_cb_inc(event->ctx->pmu);
+	if (!cpuc->lbr_users++ && !event->total_time_running)
+		intel_pmu_lbr_reset();
+}
+
+void release_lbr_buffers(void)
+{
+	struct kmem_cache *kmem_cache;
+	struct cpu_hw_events *cpuc;
+	int cpu;
+
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return;
+
+	for_each_possible_cpu(cpu) {
+		cpuc = per_cpu_ptr(&cpu_hw_events, cpu);
+		kmem_cache = x86_get_pmu(cpu)->task_ctx_cache;
+		if (kmem_cache && cpuc->lbr_xsave) {
+			kmem_cache_free(kmem_cache, cpuc->lbr_xsave);
+			cpuc->lbr_xsave = NULL;
+		}
+	}
+}
+
+void reserve_lbr_buffers(void)
+{
+	struct kmem_cache *kmem_cache;
+	struct cpu_hw_events *cpuc;
+	int cpu;
+
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return;
+
+	for_each_possible_cpu(cpu) {
+		cpuc = per_cpu_ptr(&cpu_hw_events, cpu);
+		kmem_cache = x86_get_pmu(cpu)->task_ctx_cache;
+		if (!kmem_cache || cpuc->lbr_xsave)
+			continue;
+
+		cpuc->lbr_xsave = kmem_cache_alloc_node(kmem_cache,
+							GFP_KERNEL | __GFP_ZERO,
+							cpu_to_node(cpu));
+	}
+}
+
+void intel_pmu_lbr_del(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (!x86_pmu.lbr_nr)
+		return;
+
+	if (branch_user_callstack(cpuc->br_sel) &&
+	    event->ctx->task_ctx_data)
+		task_context_opt(event->ctx->task_ctx_data)->lbr_callstack_users--;
+
+	if (event->hw.flags & PERF_X86_EVENT_LBR_SELECT)
+		cpuc->lbr_select = 0;
+
+	if (x86_pmu.intel_cap.pebs_baseline && event->attr.precise_ip > 0)
+		cpuc->lbr_pebs_users--;
+	cpuc->lbr_users--;
+	WARN_ON_ONCE(cpuc->lbr_users < 0);
+	WARN_ON_ONCE(cpuc->lbr_pebs_users < 0);
+	perf_sched_cb_dec(event->ctx->pmu);
+}
+
+static inline bool vlbr_exclude_host(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	return test_bit(INTEL_PMC_IDX_FIXED_VLBR,
+		(unsigned long *)&cpuc->intel_ctrl_guest_mask);
+}
+
+void intel_pmu_lbr_enable_all(bool pmi)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (cpuc->lbr_users && !vlbr_exclude_host())
+		__intel_pmu_lbr_enable(pmi);
+}
+
+void intel_pmu_lbr_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	if (cpuc->lbr_users && !vlbr_exclude_host())
+		__intel_pmu_lbr_disable();
+}
+
+void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc)
+{
+	unsigned long mask = x86_pmu.lbr_nr - 1;
+	u64 tos = intel_pmu_lbr_tos();
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		unsigned long lbr_idx = (tos - i) & mask;
+		union {
+			struct {
+				u32 from;
+				u32 to;
+			};
+			u64     lbr;
+		} msr_lastbranch;
+
+		rdmsrl(x86_pmu.lbr_from + lbr_idx, msr_lastbranch.lbr);
+
+		cpuc->lbr_entries[i].from	= msr_lastbranch.from;
+		cpuc->lbr_entries[i].to		= msr_lastbranch.to;
+		cpuc->lbr_entries[i].mispred	= 0;
+		cpuc->lbr_entries[i].predicted	= 0;
+		cpuc->lbr_entries[i].in_tx	= 0;
+		cpuc->lbr_entries[i].abort	= 0;
+		cpuc->lbr_entries[i].cycles	= 0;
+		cpuc->lbr_entries[i].type	= 0;
+		cpuc->lbr_entries[i].reserved	= 0;
+	}
+	cpuc->lbr_stack.nr = i;
+	cpuc->lbr_stack.hw_idx = tos;
+}
+
+/*
+ * Due to lack of segmentation in Linux the effective address (offset)
+ * is the same as the linear address, allowing us to merge the LIP and EIP
+ * LBR formats.
+ */
+void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc)
+{
+	bool need_info = false, call_stack = false;
+	unsigned long mask = x86_pmu.lbr_nr - 1;
+	int lbr_format = x86_pmu.intel_cap.lbr_format;
+	u64 tos = intel_pmu_lbr_tos();
+	int i;
+	int out = 0;
+	int num = x86_pmu.lbr_nr;
+
+	if (cpuc->lbr_sel) {
+		need_info = !(cpuc->lbr_sel->config & LBR_NO_INFO);
+		if (cpuc->lbr_sel->config & LBR_CALL_STACK)
+			call_stack = true;
+	}
+
+	for (i = 0; i < num; i++) {
+		unsigned long lbr_idx = (tos - i) & mask;
+		u64 from, to, mis = 0, pred = 0, in_tx = 0, abort = 0;
+		int skip = 0;
+		u16 cycles = 0;
+		int lbr_flags = lbr_desc[lbr_format];
+
+		from = rdlbr_from(lbr_idx, NULL);
+		to   = rdlbr_to(lbr_idx, NULL);
+
+		/*
+		 * Read LBR call stack entries
+		 * until invalid entry (0s) is detected.
+		 */
+		if (call_stack && !from)
+			break;
+
+		if (lbr_format == LBR_FORMAT_INFO && need_info) {
+			u64 info;
+
+			info = rdlbr_info(lbr_idx, NULL);
+			mis = !!(info & LBR_INFO_MISPRED);
+			pred = !mis;
+			in_tx = !!(info & LBR_INFO_IN_TX);
+			abort = !!(info & LBR_INFO_ABORT);
+			cycles = (info & LBR_INFO_CYCLES);
+		}
+
+		if (lbr_format == LBR_FORMAT_TIME) {
+			mis = !!(from & LBR_FROM_FLAG_MISPRED);
+			pred = !mis;
+			skip = 1;
+			cycles = ((to >> 48) & LBR_INFO_CYCLES);
+
+			to = (u64)((((s64)to) << 16) >> 16);
+		}
+
+		if (lbr_flags & LBR_EIP_FLAGS) {
+			mis = !!(from & LBR_FROM_FLAG_MISPRED);
+			pred = !mis;
+			skip = 1;
+		}
+		if (lbr_flags & LBR_TSX) {
+			in_tx = !!(from & LBR_FROM_FLAG_IN_TX);
+			abort = !!(from & LBR_FROM_FLAG_ABORT);
+			skip = 3;
+		}
+		from = (u64)((((s64)from) << skip) >> skip);
+
+		/*
+		 * Some CPUs report duplicated abort records,
+		 * with the second entry not having an abort bit set.
+		 * Skip them here. This loop runs backwards,
+		 * so we need to undo the previous record.
+		 * If the abort just happened outside the window
+		 * the extra entry cannot be removed.
+		 */
+		if (abort && x86_pmu.lbr_double_abort && out > 0)
+			out--;
+
+		cpuc->lbr_entries[out].from	 = from;
+		cpuc->lbr_entries[out].to	 = to;
+		cpuc->lbr_entries[out].mispred	 = mis;
+		cpuc->lbr_entries[out].predicted = pred;
+		cpuc->lbr_entries[out].in_tx	 = in_tx;
+		cpuc->lbr_entries[out].abort	 = abort;
+		cpuc->lbr_entries[out].cycles	 = cycles;
+		cpuc->lbr_entries[out].type	 = 0;
+		cpuc->lbr_entries[out].reserved	 = 0;
+		out++;
+	}
+	cpuc->lbr_stack.nr = out;
+	cpuc->lbr_stack.hw_idx = tos;
+}
+
+static __always_inline int get_lbr_br_type(u64 info)
+{
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR) || !x86_pmu.lbr_br_type)
+		return 0;
+
+	return (info & LBR_INFO_BR_TYPE) >> LBR_INFO_BR_TYPE_OFFSET;
+}
+
+static __always_inline bool get_lbr_mispred(u64 info)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR) && !x86_pmu.lbr_mispred)
+		return 0;
+
+	return !!(info & LBR_INFO_MISPRED);
+}
+
+static __always_inline bool get_lbr_predicted(u64 info)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR) && !x86_pmu.lbr_mispred)
+		return 0;
+
+	return !(info & LBR_INFO_MISPRED);
+}
+
+static __always_inline u16 get_lbr_cycles(u64 info)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR) &&
+	    !(x86_pmu.lbr_timed_lbr && info & LBR_INFO_CYC_CNT_VALID))
+		return 0;
+
+	return info & LBR_INFO_CYCLES;
+}
+
+static void intel_pmu_store_lbr(struct cpu_hw_events *cpuc,
+				struct lbr_entry *entries)
+{
+	struct perf_branch_entry *e;
+	struct lbr_entry *lbr;
+	u64 from, to, info;
+	int i;
+
+	for (i = 0; i < x86_pmu.lbr_nr; i++) {
+		lbr = entries ? &entries[i] : NULL;
+		e = &cpuc->lbr_entries[i];
+
+		from = rdlbr_from(i, lbr);
+		/*
+		 * Read LBR entries until invalid entry (0s) is detected.
+		 */
+		if (!from)
+			break;
+
+		to = rdlbr_to(i, lbr);
+		info = rdlbr_info(i, lbr);
+
+		e->from		= from;
+		e->to		= to;
+		e->mispred	= get_lbr_mispred(info);
+		e->predicted	= get_lbr_predicted(info);
+		e->in_tx	= !!(info & LBR_INFO_IN_TX);
+		e->abort	= !!(info & LBR_INFO_ABORT);
+		e->cycles	= get_lbr_cycles(info);
+		e->type		= get_lbr_br_type(info);
+		e->reserved	= 0;
+	}
+
+	cpuc->lbr_stack.nr = i;
+}
+
+static void intel_pmu_arch_lbr_read(struct cpu_hw_events *cpuc)
+{
+	intel_pmu_store_lbr(cpuc, NULL);
+}
+
+static void intel_pmu_arch_lbr_read_xsave(struct cpu_hw_events *cpuc)
+{
+	struct x86_perf_task_context_arch_lbr_xsave *xsave = cpuc->lbr_xsave;
+
+	if (!xsave) {
+		intel_pmu_store_lbr(cpuc, NULL);
+		return;
+	}
+	copy_dynamic_supervisor_to_kernel(&xsave->xsave, XFEATURE_MASK_LBR);
+
+	intel_pmu_store_lbr(cpuc, xsave->lbr.entries);
+}
+
+void intel_pmu_lbr_read(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/*
+	 * Don't read when all LBRs users are using adaptive PEBS.
+	 *
+	 * This could be smarter and actually check the event,
+	 * but this simple approach seems to work for now.
+	 */
+	if (!cpuc->lbr_users || vlbr_exclude_host() ||
+	    cpuc->lbr_users == cpuc->lbr_pebs_users)
+		return;
+
+	x86_pmu.lbr_read(cpuc);
+
+	intel_pmu_lbr_filter(cpuc);
+}
+
+/*
+ * SW filter is used:
+ * - in case there is no HW filter
+ * - in case the HW filter has errata or limitations
+ */
+static int intel_pmu_setup_sw_lbr_filter(struct perf_event *event)
+{
+	u64 br_type = event->attr.branch_sample_type;
+	int mask = 0;
+
+	if (br_type & PERF_SAMPLE_BRANCH_USER)
+		mask |= X86_BR_USER;
+
+	if (br_type & PERF_SAMPLE_BRANCH_KERNEL)
+		mask |= X86_BR_KERNEL;
+
+	/* we ignore BRANCH_HV here */
+
+	if (br_type & PERF_SAMPLE_BRANCH_ANY)
+		mask |= X86_BR_ANY;
+
+	if (br_type & PERF_SAMPLE_BRANCH_ANY_CALL)
+		mask |= X86_BR_ANY_CALL;
+
+	if (br_type & PERF_SAMPLE_BRANCH_ANY_RETURN)
+		mask |= X86_BR_RET | X86_BR_IRET | X86_BR_SYSRET;
+
+	if (br_type & PERF_SAMPLE_BRANCH_IND_CALL)
+		mask |= X86_BR_IND_CALL;
+
+	if (br_type & PERF_SAMPLE_BRANCH_ABORT_TX)
+		mask |= X86_BR_ABORT;
+
+	if (br_type & PERF_SAMPLE_BRANCH_IN_TX)
+		mask |= X86_BR_IN_TX;
+
+	if (br_type & PERF_SAMPLE_BRANCH_NO_TX)
+		mask |= X86_BR_NO_TX;
+
+	if (br_type & PERF_SAMPLE_BRANCH_COND)
+		mask |= X86_BR_JCC;
+
+	if (br_type & PERF_SAMPLE_BRANCH_CALL_STACK) {
+		if (!x86_pmu_has_lbr_callstack())
+			return -EOPNOTSUPP;
+		if (mask & ~(X86_BR_USER | X86_BR_KERNEL))
+			return -EINVAL;
+		mask |= X86_BR_CALL | X86_BR_IND_CALL | X86_BR_RET |
+			X86_BR_CALL_STACK;
+	}
+
+	if (br_type & PERF_SAMPLE_BRANCH_IND_JUMP)
+		mask |= X86_BR_IND_JMP;
+
+	if (br_type & PERF_SAMPLE_BRANCH_CALL)
+		mask |= X86_BR_CALL | X86_BR_ZERO_CALL;
+
+	if (br_type & PERF_SAMPLE_BRANCH_TYPE_SAVE)
+		mask |= X86_BR_TYPE_SAVE;
+
+	/*
+	 * stash actual user request into reg, it may
+	 * be used by fixup code for some CPU
+	 */
+	event->hw.branch_reg.reg = mask;
+	return 0;
+}
+
+/*
+ * setup the HW LBR filter
+ * Used only when available, may not be enough to disambiguate
+ * all branches, may need the help of the SW filter
+ */
+static int intel_pmu_setup_hw_lbr_filter(struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg;
+	u64 br_type = event->attr.branch_sample_type;
+	u64 mask = 0, v;
+	int i;
+
+	for (i = 0; i < PERF_SAMPLE_BRANCH_MAX_SHIFT; i++) {
+		if (!(br_type & (1ULL << i)))
+			continue;
+
+		v = x86_pmu.lbr_sel_map[i];
+		if (v == LBR_NOT_SUPP)
+			return -EOPNOTSUPP;
+
+		if (v != LBR_IGN)
+			mask |= v;
+	}
+
+	reg = &event->hw.branch_reg;
+	reg->idx = EXTRA_REG_LBR;
+
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR)) {
+		reg->config = mask;
+
+		/*
+		 * The Arch LBR HW can retrieve the common branch types
+		 * from the LBR_INFO. It doesn't require the high overhead
+		 * SW disassemble.
+		 * Enable the branch type by default for the Arch LBR.
+		 */
+		reg->reg |= X86_BR_TYPE_SAVE;
+		return 0;
+	}
+
+	/*
+	 * The first 9 bits (LBR_SEL_MASK) in LBR_SELECT operate
+	 * in suppress mode. So LBR_SELECT should be set to
+	 * (~mask & LBR_SEL_MASK) | (mask & ~LBR_SEL_MASK)
+	 * But the 10th bit LBR_CALL_STACK does not operate
+	 * in suppress mode.
+	 */
+	reg->config = mask ^ (x86_pmu.lbr_sel_mask & ~LBR_CALL_STACK);
+
+	if ((br_type & PERF_SAMPLE_BRANCH_NO_CYCLES) &&
+	    (br_type & PERF_SAMPLE_BRANCH_NO_FLAGS) &&
+	    (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_INFO))
+		reg->config |= LBR_NO_INFO;
+
+	return 0;
+}
+
+int intel_pmu_setup_lbr_filter(struct perf_event *event)
+{
+	int ret = 0;
+
+	/*
+	 * no LBR on this PMU
+	 */
+	if (!x86_pmu.lbr_nr)
+		return -EOPNOTSUPP;
+
+	/*
+	 * setup SW LBR filter
+	 */
+	ret = intel_pmu_setup_sw_lbr_filter(event);
+	if (ret)
+		return ret;
+
+	/*
+	 * setup HW LBR filter, if any
+	 */
+	if (x86_pmu.lbr_sel_map)
+		ret = intel_pmu_setup_hw_lbr_filter(event);
+
+	return ret;
+}
+
+/*
+ * return the type of control flow change at address "from"
+ * instruction is not necessarily a branch (in case of interrupt).
+ *
+ * The branch type returned also includes the priv level of the
+ * target of the control flow change (X86_BR_USER, X86_BR_KERNEL).
+ *
+ * If a branch type is unknown OR the instruction cannot be
+ * decoded (e.g., text page not present), then X86_BR_NONE is
+ * returned.
+ */
+static int branch_type(unsigned long from, unsigned long to, int abort)
+{
+	struct insn insn;
+	void *addr;
+	int bytes_read, bytes_left;
+	int ret = X86_BR_NONE;
+	int ext, to_plm, from_plm;
+	u8 buf[MAX_INSN_SIZE];
+	int is64 = 0;
+
+	to_plm = kernel_ip(to) ? X86_BR_KERNEL : X86_BR_USER;
+	from_plm = kernel_ip(from) ? X86_BR_KERNEL : X86_BR_USER;
+
+	/*
+	 * maybe zero if lbr did not fill up after a reset by the time
+	 * we get a PMU interrupt
+	 */
+	if (from == 0 || to == 0)
+		return X86_BR_NONE;
+
+	if (abort)
+		return X86_BR_ABORT | to_plm;
+
+	if (from_plm == X86_BR_USER) {
+		/*
+		 * can happen if measuring at the user level only
+		 * and we interrupt in a kernel thread, e.g., idle.
+		 */
+		if (!current->mm)
+			return X86_BR_NONE;
+
+		/* may fail if text not present */
+		bytes_left = copy_from_user_nmi(buf, (void __user *)from,
+						MAX_INSN_SIZE);
+		bytes_read = MAX_INSN_SIZE - bytes_left;
+		if (!bytes_read)
+			return X86_BR_NONE;
+
+		addr = buf;
+	} else {
+		/*
+		 * The LBR logs any address in the IP, even if the IP just
+		 * faulted. This means userspace can control the from address.
+		 * Ensure we don't blindy read any address by validating it is
+		 * a known text address.
+		 */
+		if (kernel_text_address(from)) {
+			addr = (void *)from;
+			/*
+			 * Assume we can get the maximum possible size
+			 * when grabbing kernel data.  This is not
+			 * _strictly_ true since we could possibly be
+			 * executing up next to a memory hole, but
+			 * it is very unlikely to be a problem.
+			 */
+			bytes_read = MAX_INSN_SIZE;
+		} else {
+			return X86_BR_NONE;
+		}
+	}
+
+	/*
+	 * decoder needs to know the ABI especially
+	 * on 64-bit systems running 32-bit apps
+	 */
+#ifdef CONFIG_X86_64
+	is64 = kernel_ip((unsigned long)addr) || !test_thread_flag(TIF_IA32);
+#endif
+	insn_init(&insn, addr, bytes_read, is64);
+	insn_get_opcode(&insn);
+	if (!insn.opcode.got)
+		return X86_BR_ABORT;
+
+	switch (insn.opcode.bytes[0]) {
+	case 0xf:
+		switch (insn.opcode.bytes[1]) {
+		case 0x05: /* syscall */
+		case 0x34: /* sysenter */
+			ret = X86_BR_SYSCALL;
+			break;
+		case 0x07: /* sysret */
+		case 0x35: /* sysexit */
+			ret = X86_BR_SYSRET;
+			break;
+		case 0x80 ... 0x8f: /* conditional */
+			ret = X86_BR_JCC;
+			break;
+		default:
+			ret = X86_BR_NONE;
+		}
+		break;
+	case 0x70 ... 0x7f: /* conditional */
+		ret = X86_BR_JCC;
+		break;
+	case 0xc2: /* near ret */
+	case 0xc3: /* near ret */
+	case 0xca: /* far ret */
+	case 0xcb: /* far ret */
+		ret = X86_BR_RET;
+		break;
+	case 0xcf: /* iret */
+		ret = X86_BR_IRET;
+		break;
+	case 0xcc ... 0xce: /* int */
+		ret = X86_BR_INT;
+		break;
+	case 0xe8: /* call near rel */
+		insn_get_immediate(&insn);
+		if (insn.immediate1.value == 0) {
+			/* zero length call */
+			ret = X86_BR_ZERO_CALL;
+			break;
+		}
+		fallthrough;
+	case 0x9a: /* call far absolute */
+		ret = X86_BR_CALL;
+		break;
+	case 0xe0 ... 0xe3: /* loop jmp */
+		ret = X86_BR_JCC;
+		break;
+	case 0xe9 ... 0xeb: /* jmp */
+		ret = X86_BR_JMP;
+		break;
+	case 0xff: /* call near absolute, call far absolute ind */
+		insn_get_modrm(&insn);
+		ext = (insn.modrm.bytes[0] >> 3) & 0x7;
+		switch (ext) {
+		case 2: /* near ind call */
+		case 3: /* far ind call */
+			ret = X86_BR_IND_CALL;
+			break;
+		case 4:
+		case 5:
+			ret = X86_BR_IND_JMP;
+			break;
+		}
+		break;
+	default:
+		ret = X86_BR_NONE;
+	}
+	/*
+	 * interrupts, traps, faults (and thus ring transition) may
+	 * occur on any instructions. Thus, to classify them correctly,
+	 * we need to first look at the from and to priv levels. If they
+	 * are different and to is in the kernel, then it indicates
+	 * a ring transition. If the from instruction is not a ring
+	 * transition instr (syscall, systenter, int), then it means
+	 * it was a irq, trap or fault.
+	 *
+	 * we have no way of detecting kernel to kernel faults.
+	 */
+	if (from_plm == X86_BR_USER && to_plm == X86_BR_KERNEL
+	    && ret != X86_BR_SYSCALL && ret != X86_BR_INT)
+		ret = X86_BR_IRQ;
+
+	/*
+	 * branch priv level determined by target as
+	 * is done by HW when LBR_SELECT is implemented
+	 */
+	if (ret != X86_BR_NONE)
+		ret |= to_plm;
+
+	return ret;
+}
+
+#define X86_BR_TYPE_MAP_MAX	16
+
+static int branch_map[X86_BR_TYPE_MAP_MAX] = {
+	PERF_BR_CALL,		/* X86_BR_CALL */
+	PERF_BR_RET,		/* X86_BR_RET */
+	PERF_BR_SYSCALL,	/* X86_BR_SYSCALL */
+	PERF_BR_SYSRET,		/* X86_BR_SYSRET */
+	PERF_BR_UNKNOWN,	/* X86_BR_INT */
+	PERF_BR_UNKNOWN,	/* X86_BR_IRET */
+	PERF_BR_COND,		/* X86_BR_JCC */
+	PERF_BR_UNCOND,		/* X86_BR_JMP */
+	PERF_BR_UNKNOWN,	/* X86_BR_IRQ */
+	PERF_BR_IND_CALL,	/* X86_BR_IND_CALL */
+	PERF_BR_UNKNOWN,	/* X86_BR_ABORT */
+	PERF_BR_UNKNOWN,	/* X86_BR_IN_TX */
+	PERF_BR_UNKNOWN,	/* X86_BR_NO_TX */
+	PERF_BR_CALL,		/* X86_BR_ZERO_CALL */
+	PERF_BR_UNKNOWN,	/* X86_BR_CALL_STACK */
+	PERF_BR_IND,		/* X86_BR_IND_JMP */
+};
+
+static int
+common_branch_type(int type)
+{
+	int i;
+
+	type >>= 2; /* skip X86_BR_USER and X86_BR_KERNEL */
+
+	if (type) {
+		i = __ffs(type);
+		if (i < X86_BR_TYPE_MAP_MAX)
+			return branch_map[i];
+	}
+
+	return PERF_BR_UNKNOWN;
+}
+
+enum {
+	ARCH_LBR_BR_TYPE_JCC			= 0,
+	ARCH_LBR_BR_TYPE_NEAR_IND_JMP		= 1,
+	ARCH_LBR_BR_TYPE_NEAR_REL_JMP		= 2,
+	ARCH_LBR_BR_TYPE_NEAR_IND_CALL		= 3,
+	ARCH_LBR_BR_TYPE_NEAR_REL_CALL		= 4,
+	ARCH_LBR_BR_TYPE_NEAR_RET		= 5,
+	ARCH_LBR_BR_TYPE_KNOWN_MAX		= ARCH_LBR_BR_TYPE_NEAR_RET,
+
+	ARCH_LBR_BR_TYPE_MAP_MAX		= 16,
+};
+
+static const int arch_lbr_br_type_map[ARCH_LBR_BR_TYPE_MAP_MAX] = {
+	[ARCH_LBR_BR_TYPE_JCC]			= X86_BR_JCC,
+	[ARCH_LBR_BR_TYPE_NEAR_IND_JMP]		= X86_BR_IND_JMP,
+	[ARCH_LBR_BR_TYPE_NEAR_REL_JMP]		= X86_BR_JMP,
+	[ARCH_LBR_BR_TYPE_NEAR_IND_CALL]	= X86_BR_IND_CALL,
+	[ARCH_LBR_BR_TYPE_NEAR_REL_CALL]	= X86_BR_CALL,
+	[ARCH_LBR_BR_TYPE_NEAR_RET]		= X86_BR_RET,
+};
+
+/*
+ * implement actual branch filter based on user demand.
+ * Hardware may not exactly satisfy that request, thus
+ * we need to inspect opcodes. Mismatched branches are
+ * discarded. Therefore, the number of branches returned
+ * in PERF_SAMPLE_BRANCH_STACK sample may vary.
+ */
+static void
+intel_pmu_lbr_filter(struct cpu_hw_events *cpuc)
+{
+	u64 from, to;
+	int br_sel = cpuc->br_sel;
+	int i, j, type, to_plm;
+	bool compress = false;
+
+	/* if sampling all branches, then nothing to filter */
+	if (((br_sel & X86_BR_ALL) == X86_BR_ALL) &&
+	    ((br_sel & X86_BR_TYPE_SAVE) != X86_BR_TYPE_SAVE))
+		return;
+
+	for (i = 0; i < cpuc->lbr_stack.nr; i++) {
+
+		from = cpuc->lbr_entries[i].from;
+		to = cpuc->lbr_entries[i].to;
+		type = cpuc->lbr_entries[i].type;
+
+		/*
+		 * Parse the branch type recorded in LBR_x_INFO MSR.
+		 * Doesn't support OTHER_BRANCH decoding for now.
+		 * OTHER_BRANCH branch type still rely on software decoding.
+		 */
+		if (static_cpu_has(X86_FEATURE_ARCH_LBR) &&
+		    type <= ARCH_LBR_BR_TYPE_KNOWN_MAX) {
+			to_plm = kernel_ip(to) ? X86_BR_KERNEL : X86_BR_USER;
+			type = arch_lbr_br_type_map[type] | to_plm;
+		} else
+			type = branch_type(from, to, cpuc->lbr_entries[i].abort);
+		if (type != X86_BR_NONE && (br_sel & X86_BR_ANYTX)) {
+			if (cpuc->lbr_entries[i].in_tx)
+				type |= X86_BR_IN_TX;
+			else
+				type |= X86_BR_NO_TX;
+		}
+
+		/* if type does not correspond, then discard */
+		if (type == X86_BR_NONE || (br_sel & type) != type) {
+			cpuc->lbr_entries[i].from = 0;
+			compress = true;
+		}
+
+		if ((br_sel & X86_BR_TYPE_SAVE) == X86_BR_TYPE_SAVE)
+			cpuc->lbr_entries[i].type = common_branch_type(type);
+	}
+
+	if (!compress)
+		return;
+
+	/* remove all entries with from=0 */
+	for (i = 0; i < cpuc->lbr_stack.nr; ) {
+		if (!cpuc->lbr_entries[i].from) {
+			j = i;
+			while (++j < cpuc->lbr_stack.nr)
+				cpuc->lbr_entries[j-1] = cpuc->lbr_entries[j];
+			cpuc->lbr_stack.nr--;
+			if (!cpuc->lbr_entries[i].from)
+				continue;
+		}
+		i++;
+	}
+}
+
+void intel_pmu_store_pebs_lbrs(struct lbr_entry *lbr)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	/* Cannot get TOS for large PEBS and Arch LBR */
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR) ||
+	    (cpuc->n_pebs == cpuc->n_large_pebs))
+		cpuc->lbr_stack.hw_idx = -1ULL;
+	else
+		cpuc->lbr_stack.hw_idx = intel_pmu_lbr_tos();
+
+	intel_pmu_store_lbr(cpuc, lbr);
+	intel_pmu_lbr_filter(cpuc);
+}
+
+/*
+ * Map interface branch filters onto LBR filters
+ */
+static const int nhm_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = {
+	[PERF_SAMPLE_BRANCH_ANY_SHIFT]		= LBR_ANY,
+	[PERF_SAMPLE_BRANCH_USER_SHIFT]		= LBR_USER,
+	[PERF_SAMPLE_BRANCH_KERNEL_SHIFT]	= LBR_KERNEL,
+	[PERF_SAMPLE_BRANCH_HV_SHIFT]		= LBR_IGN,
+	[PERF_SAMPLE_BRANCH_ANY_RETURN_SHIFT]	= LBR_RETURN | LBR_REL_JMP
+						| LBR_IND_JMP | LBR_FAR,
+	/*
+	 * NHM/WSM erratum: must include REL_JMP+IND_JMP to get CALL branches
+	 */
+	[PERF_SAMPLE_BRANCH_ANY_CALL_SHIFT] =
+	 LBR_REL_CALL | LBR_IND_CALL | LBR_REL_JMP | LBR_IND_JMP | LBR_FAR,
+	/*
+	 * NHM/WSM erratum: must include IND_JMP to capture IND_CALL
+	 */
+	[PERF_SAMPLE_BRANCH_IND_CALL_SHIFT] = LBR_IND_CALL | LBR_IND_JMP,
+	[PERF_SAMPLE_BRANCH_COND_SHIFT]     = LBR_JCC,
+	[PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_IND_JMP,
+};
+
+static const int snb_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = {
+	[PERF_SAMPLE_BRANCH_ANY_SHIFT]		= LBR_ANY,
+	[PERF_SAMPLE_BRANCH_USER_SHIFT]		= LBR_USER,
+	[PERF_SAMPLE_BRANCH_KERNEL_SHIFT]	= LBR_KERNEL,
+	[PERF_SAMPLE_BRANCH_HV_SHIFT]		= LBR_IGN,
+	[PERF_SAMPLE_BRANCH_ANY_RETURN_SHIFT]	= LBR_RETURN | LBR_FAR,
+	[PERF_SAMPLE_BRANCH_ANY_CALL_SHIFT]	= LBR_REL_CALL | LBR_IND_CALL
+						| LBR_FAR,
+	[PERF_SAMPLE_BRANCH_IND_CALL_SHIFT]	= LBR_IND_CALL,
+	[PERF_SAMPLE_BRANCH_COND_SHIFT]		= LBR_JCC,
+	[PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT]	= LBR_IND_JMP,
+	[PERF_SAMPLE_BRANCH_CALL_SHIFT]		= LBR_REL_CALL,
+};
+
+static const int hsw_lbr_sel_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = {
+	[PERF_SAMPLE_BRANCH_ANY_SHIFT]		= LBR_ANY,
+	[PERF_SAMPLE_BRANCH_USER_SHIFT]		= LBR_USER,
+	[PERF_SAMPLE_BRANCH_KERNEL_SHIFT]	= LBR_KERNEL,
+	[PERF_SAMPLE_BRANCH_HV_SHIFT]		= LBR_IGN,
+	[PERF_SAMPLE_BRANCH_ANY_RETURN_SHIFT]	= LBR_RETURN | LBR_FAR,
+	[PERF_SAMPLE_BRANCH_ANY_CALL_SHIFT]	= LBR_REL_CALL | LBR_IND_CALL
+						| LBR_FAR,
+	[PERF_SAMPLE_BRANCH_IND_CALL_SHIFT]	= LBR_IND_CALL,
+	[PERF_SAMPLE_BRANCH_COND_SHIFT]		= LBR_JCC,
+	[PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT]	= LBR_REL_CALL | LBR_IND_CALL
+						| LBR_RETURN | LBR_CALL_STACK,
+	[PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT]	= LBR_IND_JMP,
+	[PERF_SAMPLE_BRANCH_CALL_SHIFT]		= LBR_REL_CALL,
+};
+
+static int arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_MAX_SHIFT] = {
+	[PERF_SAMPLE_BRANCH_ANY_SHIFT]		= ARCH_LBR_ANY,
+	[PERF_SAMPLE_BRANCH_USER_SHIFT]		= ARCH_LBR_USER,
+	[PERF_SAMPLE_BRANCH_KERNEL_SHIFT]	= ARCH_LBR_KERNEL,
+	[PERF_SAMPLE_BRANCH_HV_SHIFT]		= LBR_IGN,
+	[PERF_SAMPLE_BRANCH_ANY_RETURN_SHIFT]	= ARCH_LBR_RETURN |
+						  ARCH_LBR_OTHER_BRANCH,
+	[PERF_SAMPLE_BRANCH_ANY_CALL_SHIFT]     = ARCH_LBR_REL_CALL |
+						  ARCH_LBR_IND_CALL |
+						  ARCH_LBR_OTHER_BRANCH,
+	[PERF_SAMPLE_BRANCH_IND_CALL_SHIFT]     = ARCH_LBR_IND_CALL,
+	[PERF_SAMPLE_BRANCH_COND_SHIFT]         = ARCH_LBR_JCC,
+	[PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT]   = ARCH_LBR_REL_CALL |
+						  ARCH_LBR_IND_CALL |
+						  ARCH_LBR_RETURN |
+						  ARCH_LBR_CALL_STACK,
+	[PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT]	= ARCH_LBR_IND_JMP,
+	[PERF_SAMPLE_BRANCH_CALL_SHIFT]		= ARCH_LBR_REL_CALL,
+};
+
+/* core */
+void __init intel_pmu_lbr_init_core(void)
+{
+	x86_pmu.lbr_nr     = 4;
+	x86_pmu.lbr_tos    = MSR_LBR_TOS;
+	x86_pmu.lbr_from   = MSR_LBR_CORE_FROM;
+	x86_pmu.lbr_to     = MSR_LBR_CORE_TO;
+
+	/*
+	 * SW branch filter usage:
+	 * - compensate for lack of HW filter
+	 */
+}
+
+/* nehalem/westmere */
+void __init intel_pmu_lbr_init_nhm(void)
+{
+	x86_pmu.lbr_nr     = 16;
+	x86_pmu.lbr_tos    = MSR_LBR_TOS;
+	x86_pmu.lbr_from   = MSR_LBR_NHM_FROM;
+	x86_pmu.lbr_to     = MSR_LBR_NHM_TO;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = nhm_lbr_sel_map;
+
+	/*
+	 * SW branch filter usage:
+	 * - workaround LBR_SEL errata (see above)
+	 * - support syscall, sysret capture.
+	 *   That requires LBR_FAR but that means far
+	 *   jmp need to be filtered out
+	 */
+}
+
+/* sandy bridge */
+void __init intel_pmu_lbr_init_snb(void)
+{
+	x86_pmu.lbr_nr	 = 16;
+	x86_pmu.lbr_tos	 = MSR_LBR_TOS;
+	x86_pmu.lbr_from = MSR_LBR_NHM_FROM;
+	x86_pmu.lbr_to   = MSR_LBR_NHM_TO;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = snb_lbr_sel_map;
+
+	/*
+	 * SW branch filter usage:
+	 * - support syscall, sysret capture.
+	 *   That requires LBR_FAR but that means far
+	 *   jmp need to be filtered out
+	 */
+}
+
+static inline struct kmem_cache *
+create_lbr_kmem_cache(size_t size, size_t align)
+{
+	return kmem_cache_create("x86_lbr", size, align, 0, NULL);
+}
+
+/* haswell */
+void intel_pmu_lbr_init_hsw(void)
+{
+	size_t size = sizeof(struct x86_perf_task_context);
+
+	x86_pmu.lbr_nr	 = 16;
+	x86_pmu.lbr_tos	 = MSR_LBR_TOS;
+	x86_pmu.lbr_from = MSR_LBR_NHM_FROM;
+	x86_pmu.lbr_to   = MSR_LBR_NHM_TO;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = hsw_lbr_sel_map;
+
+	x86_get_pmu(smp_processor_id())->task_ctx_cache = create_lbr_kmem_cache(size, 0);
+
+	if (lbr_from_signext_quirk_needed())
+		static_branch_enable(&lbr_from_quirk_key);
+}
+
+/* skylake */
+__init void intel_pmu_lbr_init_skl(void)
+{
+	size_t size = sizeof(struct x86_perf_task_context);
+
+	x86_pmu.lbr_nr	 = 32;
+	x86_pmu.lbr_tos	 = MSR_LBR_TOS;
+	x86_pmu.lbr_from = MSR_LBR_NHM_FROM;
+	x86_pmu.lbr_to   = MSR_LBR_NHM_TO;
+	x86_pmu.lbr_info = MSR_LBR_INFO_0;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = hsw_lbr_sel_map;
+
+	x86_get_pmu(smp_processor_id())->task_ctx_cache = create_lbr_kmem_cache(size, 0);
+
+	/*
+	 * SW branch filter usage:
+	 * - support syscall, sysret capture.
+	 *   That requires LBR_FAR but that means far
+	 *   jmp need to be filtered out
+	 */
+}
+
+/* atom */
+void __init intel_pmu_lbr_init_atom(void)
+{
+	/*
+	 * only models starting at stepping 10 seems
+	 * to have an operational LBR which can freeze
+	 * on PMU interrupt
+	 */
+	if (boot_cpu_data.x86_model == 28
+	    && boot_cpu_data.x86_stepping < 10) {
+		pr_cont("LBR disabled due to erratum");
+		return;
+	}
+
+	x86_pmu.lbr_nr	   = 8;
+	x86_pmu.lbr_tos    = MSR_LBR_TOS;
+	x86_pmu.lbr_from   = MSR_LBR_CORE_FROM;
+	x86_pmu.lbr_to     = MSR_LBR_CORE_TO;
+
+	/*
+	 * SW branch filter usage:
+	 * - compensate for lack of HW filter
+	 */
+}
+
+/* slm */
+void __init intel_pmu_lbr_init_slm(void)
+{
+	x86_pmu.lbr_nr	   = 8;
+	x86_pmu.lbr_tos    = MSR_LBR_TOS;
+	x86_pmu.lbr_from   = MSR_LBR_CORE_FROM;
+	x86_pmu.lbr_to     = MSR_LBR_CORE_TO;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = nhm_lbr_sel_map;
+
+	/*
+	 * SW branch filter usage:
+	 * - compensate for lack of HW filter
+	 */
+	pr_cont("8-deep LBR, ");
+}
+
+/* Knights Landing */
+void intel_pmu_lbr_init_knl(void)
+{
+	x86_pmu.lbr_nr	   = 8;
+	x86_pmu.lbr_tos    = MSR_LBR_TOS;
+	x86_pmu.lbr_from   = MSR_LBR_NHM_FROM;
+	x86_pmu.lbr_to     = MSR_LBR_NHM_TO;
+
+	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
+	x86_pmu.lbr_sel_map  = snb_lbr_sel_map;
+
+	/* Knights Landing does have MISPREDICT bit */
+	if (x86_pmu.intel_cap.lbr_format == LBR_FORMAT_LIP)
+		x86_pmu.intel_cap.lbr_format = LBR_FORMAT_EIP_FLAGS;
+}
+
+/*
+ * LBR state size is variable based on the max number of registers.
+ * This calculates the expected state size, which should match
+ * what the hardware enumerates for the size of XFEATURE_LBR.
+ */
+static inline unsigned int get_lbr_state_size(void)
+{
+	return sizeof(struct arch_lbr_state) +
+	       x86_pmu.lbr_nr * sizeof(struct lbr_entry);
+}
+
+static bool is_arch_lbr_xsave_available(void)
+{
+	if (!boot_cpu_has(X86_FEATURE_XSAVES))
+		return false;
+
+	/*
+	 * Check the LBR state with the corresponding software structure.
+	 * Disable LBR XSAVES support if the size doesn't match.
+	 */
+	if (xfeature_size(XFEATURE_LBR) == 0)
+		return false;
+
+	if (WARN_ON(xfeature_size(XFEATURE_LBR) != get_lbr_state_size()))
+		return false;
+
+	return true;
+}
+
+void __init intel_pmu_arch_lbr_init(void)
+{
+	struct pmu *pmu = x86_get_pmu(smp_processor_id());
+	union cpuid28_eax eax;
+	union cpuid28_ebx ebx;
+	union cpuid28_ecx ecx;
+	unsigned int unused_edx;
+	bool arch_lbr_xsave;
+	size_t size;
+	u64 lbr_nr;
+
+	/* Arch LBR Capabilities */
+	cpuid(28, &eax.full, &ebx.full, &ecx.full, &unused_edx);
+
+	lbr_nr = fls(eax.split.lbr_depth_mask) * 8;
+	if (!lbr_nr)
+		goto clear_arch_lbr;
+
+	/* Apply the max depth of Arch LBR */
+	if (wrmsrl_safe(MSR_ARCH_LBR_DEPTH, lbr_nr))
+		goto clear_arch_lbr;
+
+	x86_pmu.lbr_depth_mask = eax.split.lbr_depth_mask;
+	x86_pmu.lbr_deep_c_reset = eax.split.lbr_deep_c_reset;
+	x86_pmu.lbr_lip = eax.split.lbr_lip;
+	x86_pmu.lbr_cpl = ebx.split.lbr_cpl;
+	x86_pmu.lbr_filter = ebx.split.lbr_filter;
+	x86_pmu.lbr_call_stack = ebx.split.lbr_call_stack;
+	x86_pmu.lbr_mispred = ecx.split.lbr_mispred;
+	x86_pmu.lbr_timed_lbr = ecx.split.lbr_timed_lbr;
+	x86_pmu.lbr_br_type = ecx.split.lbr_br_type;
+	x86_pmu.lbr_nr = lbr_nr;
+
+
+	arch_lbr_xsave = is_arch_lbr_xsave_available();
+	if (arch_lbr_xsave) {
+		size = sizeof(struct x86_perf_task_context_arch_lbr_xsave) +
+		       get_lbr_state_size();
+		pmu->task_ctx_cache = create_lbr_kmem_cache(size,
+							    XSAVE_ALIGNMENT);
+	}
+
+	if (!pmu->task_ctx_cache) {
+		arch_lbr_xsave = false;
+
+		size = sizeof(struct x86_perf_task_context_arch_lbr) +
+		       lbr_nr * sizeof(struct lbr_entry);
+		pmu->task_ctx_cache = create_lbr_kmem_cache(size, 0);
+	}
+
+	x86_pmu.lbr_from = MSR_ARCH_LBR_FROM_0;
+	x86_pmu.lbr_to = MSR_ARCH_LBR_TO_0;
+	x86_pmu.lbr_info = MSR_ARCH_LBR_INFO_0;
+
+	/* LBR callstack requires both CPL and Branch Filtering support */
+	if (!x86_pmu.lbr_cpl ||
+	    !x86_pmu.lbr_filter ||
+	    !x86_pmu.lbr_call_stack)
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT] = LBR_NOT_SUPP;
+
+	if (!x86_pmu.lbr_cpl) {
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_USER_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_KERNEL_SHIFT] = LBR_NOT_SUPP;
+	} else if (!x86_pmu.lbr_filter) {
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_ANY_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_ANY_RETURN_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_ANY_CALL_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_IND_CALL_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_COND_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_IND_JUMP_SHIFT] = LBR_NOT_SUPP;
+		arch_lbr_ctl_map[PERF_SAMPLE_BRANCH_CALL_SHIFT] = LBR_NOT_SUPP;
+	}
+
+	x86_pmu.lbr_ctl_mask = ARCH_LBR_CTL_MASK;
+	x86_pmu.lbr_ctl_map  = arch_lbr_ctl_map;
+
+	if (!x86_pmu.lbr_cpl && !x86_pmu.lbr_filter)
+		x86_pmu.lbr_ctl_map = NULL;
+
+	x86_pmu.lbr_reset = intel_pmu_arch_lbr_reset;
+	if (arch_lbr_xsave) {
+		x86_pmu.lbr_save = intel_pmu_arch_lbr_xsaves;
+		x86_pmu.lbr_restore = intel_pmu_arch_lbr_xrstors;
+		x86_pmu.lbr_read = intel_pmu_arch_lbr_read_xsave;
+		pr_cont("XSAVE ");
+	} else {
+		x86_pmu.lbr_save = intel_pmu_arch_lbr_save;
+		x86_pmu.lbr_restore = intel_pmu_arch_lbr_restore;
+		x86_pmu.lbr_read = intel_pmu_arch_lbr_read;
+	}
+
+	pr_cont("Architectural LBR, ");
+
+	return;
+
+clear_arch_lbr:
+	setup_clear_cpu_cap(X86_FEATURE_ARCH_LBR);
+}
+
+/**
+ * x86_perf_get_lbr - get the LBR records information
+ *
+ * @lbr: the caller's memory to store the LBR records information
+ *
+ * Returns: 0 indicates the LBR info has been successfully obtained
+ */
+int x86_perf_get_lbr(struct x86_pmu_lbr *lbr)
+{
+	int lbr_fmt = x86_pmu.intel_cap.lbr_format;
+
+	lbr->nr = x86_pmu.lbr_nr;
+	lbr->from = x86_pmu.lbr_from;
+	lbr->to = x86_pmu.lbr_to;
+	lbr->info = (lbr_fmt == LBR_FORMAT_INFO) ? x86_pmu.lbr_info : 0;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(x86_perf_get_lbr);
+
+struct event_constraint vlbr_constraint =
+	__EVENT_CONSTRAINT(INTEL_FIXED_VLBR_EVENT, (1ULL << INTEL_PMC_IDX_FIXED_VLBR),
+			  FIXED_EVENT_FLAGS, 1, 0, PERF_X86_EVENT_LBR_SELECT);
diff --git a/arch/x86/events/intel/p4.c b/arch/x86/events/intel/p4.c
new file mode 100644
index 000000000..a4cc66005
--- /dev/null
+++ b/arch/x86/events/intel/p4.c
@@ -0,0 +1,1377 @@
+/*
+ * Netburst Performance Events (P4, old Xeon)
+ *
+ *  Copyright (C) 2010 Parallels, Inc., Cyrill Gorcunov <gorcunov@openvz.org>
+ *  Copyright (C) 2010 Intel Corporation, Lin Ming <ming.m.lin@intel.com>
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+
+#include <asm/perf_event_p4.h>
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "../perf_event.h"
+
+#define P4_CNTR_LIMIT 3
+/*
+ * array indices: 0,1 - HT threads, used with HT enabled cpu
+ */
+struct p4_event_bind {
+	unsigned int opcode;			/* Event code and ESCR selector */
+	unsigned int escr_msr[2];		/* ESCR MSR for this event */
+	unsigned int escr_emask;		/* valid ESCR EventMask bits */
+	unsigned int shared;			/* event is shared across threads */
+	char cntr[2][P4_CNTR_LIMIT];		/* counter index (offset), -1 on abscence */
+};
+
+struct p4_pebs_bind {
+	unsigned int metric_pebs;
+	unsigned int metric_vert;
+};
+
+/* it sets P4_PEBS_ENABLE_UOP_TAG as well */
+#define P4_GEN_PEBS_BIND(name, pebs, vert)			\
+	[P4_PEBS_METRIC__##name] = {				\
+		.metric_pebs = pebs | P4_PEBS_ENABLE_UOP_TAG,	\
+		.metric_vert = vert,				\
+	}
+
+/*
+ * note we have P4_PEBS_ENABLE_UOP_TAG always set here
+ *
+ * it's needed for mapping P4_PEBS_CONFIG_METRIC_MASK bits of
+ * event configuration to find out which values are to be
+ * written into MSR_IA32_PEBS_ENABLE and MSR_P4_PEBS_MATRIX_VERT
+ * resgisters
+ */
+static struct p4_pebs_bind p4_pebs_bind_map[] = {
+	P4_GEN_PEBS_BIND(1stl_cache_load_miss_retired,	0x0000001, 0x0000001),
+	P4_GEN_PEBS_BIND(2ndl_cache_load_miss_retired,	0x0000002, 0x0000001),
+	P4_GEN_PEBS_BIND(dtlb_load_miss_retired,	0x0000004, 0x0000001),
+	P4_GEN_PEBS_BIND(dtlb_store_miss_retired,	0x0000004, 0x0000002),
+	P4_GEN_PEBS_BIND(dtlb_all_miss_retired,		0x0000004, 0x0000003),
+	P4_GEN_PEBS_BIND(tagged_mispred_branch,		0x0018000, 0x0000010),
+	P4_GEN_PEBS_BIND(mob_load_replay_retired,	0x0000200, 0x0000001),
+	P4_GEN_PEBS_BIND(split_load_retired,		0x0000400, 0x0000001),
+	P4_GEN_PEBS_BIND(split_store_retired,		0x0000400, 0x0000002),
+};
+
+/*
+ * Note that we don't use CCCR1 here, there is an
+ * exception for P4_BSQ_ALLOCATION but we just have
+ * no workaround
+ *
+ * consider this binding as resources which particular
+ * event may borrow, it doesn't contain EventMask,
+ * Tags and friends -- they are left to a caller
+ */
+static struct p4_event_bind p4_event_bind_map[] = {
+	[P4_EVENT_TC_DELIVER_MODE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_DELIVER_MODE),
+		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, ID),
+		.shared		= 1,
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_BPU_FETCH_REQUEST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST),
+		.escr_msr	= { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BPU_FETCH_REQUEST, TCMISS),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_ITLB_REFERENCE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_ITLB_REFERENCE),
+		.escr_msr	= { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, MISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT_UK),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_MEMORY_CANCEL] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_CANCEL),
+		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, 64K_CONF),
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_MEMORY_COMPLETE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_COMPLETE),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, LSC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, SSC),
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_LOAD_PORT_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD),
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_STORE_PORT_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY),
+		.escr_msr	= { MSR_P4_SAAT_ESCR0 ,  MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST),
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_MOB_LOAD_REPLAY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY),
+		.escr_msr	= { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STA)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STD)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_PAGE_WALK_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE),
+		.escr_msr	= { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, DTMISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, ITMISS),
+		.shared		= 1,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BSQ_CACHE_REFERENCE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE),
+		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_IOQ_ALLOCATION] = {
+		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ALLOCATION),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_READ)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OWN)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, PREFETCH),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_IOQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
+		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES),
+		.escr_msr	= { MSR_P4_FSB_ESCR1,  MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH),
+		.cntr		= { {2, -1, -1}, {3, -1, -1} },
+	},
+	[P4_EVENT_FSB_DATA_ACTIVITY] = {
+		.opcode		= P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER),
+		.shared		= 1,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BSQ_ALLOCATION] = {		/* shared ESCR, broken CCCR1 */
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ALLOCATION),
+		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2),
+		.cntr		= { {0, -1, -1}, {1, -1, -1} },
+	},
+	[P4_EVENT_BSQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
+		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES),
+		.escr_msr	= { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2),
+		.cntr		= { {2, -1, -1}, {3, -1, -1} },
+	},
+	[P4_EVENT_SSE_INPUT_ASSIST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SSE_INPUT_ASSIST, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_PACKED_SP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PACKED_SP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_SP_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_PACKED_DP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_PACKED_DP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_DP_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_SCALAR_SP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_SP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_SP_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_SCALAR_DP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_DP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_DP_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_64BIT_MMX_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_64BIT_MMX_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_64BIT_MMX_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_128BIT_MMX_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_128BIT_MMX_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_128BIT_MMX_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_X87_FP_UOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_X87_FP_UOP),
+		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_FP_UOP, ALL),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_TC_MISC] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_MISC),
+		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MISC, FLUSH),
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_GLOBAL_POWER_EVENTS] = {
+		.opcode		= P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING),
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_TC_MS_XFER] = {
+		.opcode		= P4_OPCODE(P4_EVENT_TC_MS_XFER),
+		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MS_XFER, CISC),
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_UOP_QUEUE_WRITES] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES),
+		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM),
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE),
+		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT),
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RETIRED_BRANCH_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE),
+		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT),
+		.cntr		= { {4, 5, -1}, {6, 7, -1} },
+	},
+	[P4_EVENT_RESOURCE_STALL] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RESOURCE_STALL),
+		.escr_msr	= { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RESOURCE_STALL, SBFULL),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_WC_BUFFER] = {
+		.opcode		= P4_OPCODE(P4_EVENT_WC_BUFFER),
+		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_EVICTS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS),
+		.shared		= 1,
+		.cntr		= { {8, 9, -1}, {10, 11, -1} },
+	},
+	[P4_EVENT_B2B_CYCLES] = {
+		.opcode		= P4_OPCODE(P4_EVENT_B2B_CYCLES),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_BNR] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BNR),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_SNOOP] = {
+		.opcode		= P4_OPCODE(P4_EVENT_SNOOP),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_RESPONSE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_RESPONSE),
+		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
+		.cntr		= { {0, -1, -1}, {2, -1, -1} },
+	},
+	[P4_EVENT_FRONT_END_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_FRONT_END_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, BOGUS),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_EXECUTION_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_EXECUTION_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_REPLAY_EVENT] = {
+		.opcode		= P4_OPCODE(P4_EVENT_REPLAY_EVENT),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, BOGUS),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_INSTR_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_INSTR_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSTAG),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_UOPS_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOPS_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, BOGUS),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_UOP_TYPE] = {
+		.opcode		= P4_OPCODE(P4_EVENT_UOP_TYPE),
+		.escr_msr	= { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGLOADS)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGSTORES),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_BRANCH_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_BRANCH_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNM)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTM),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_MISPRED_BRANCH_RETIRED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_X87_ASSIST] = {
+		.opcode		= P4_OPCODE(P4_EVENT_X87_ASSIST),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, PREA),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_MACHINE_CLEAR] = {
+		.opcode		= P4_OPCODE(P4_EVENT_MACHINE_CLEAR),
+		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, CLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, MOCLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, SMCLEAR),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+	[P4_EVENT_INSTR_COMPLETED] = {
+		.opcode		= P4_OPCODE(P4_EVENT_INSTR_COMPLETED),
+		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, BOGUS),
+		.cntr		= { {12, 13, 16}, {14, 15, 17} },
+	},
+};
+
+#define P4_GEN_CACHE_EVENT(event, bit, metric)				  \
+	p4_config_pack_escr(P4_ESCR_EVENT(event)			| \
+			    P4_ESCR_EMASK_BIT(event, bit))		| \
+	p4_config_pack_cccr(metric					| \
+			    P4_CCCR_ESEL(P4_OPCODE_ESEL(P4_OPCODE(event))))
+
+static __initconst const u64 p4_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) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_PEBS_METRIC__1stl_cache_load_miss_retired),
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_PEBS_METRIC__2ndl_cache_load_miss_retired),
+	},
+},
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_PEBS_METRIC__dtlb_load_miss_retired),
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_REPLAY_EVENT, NBOGUS,
+						P4_PEBS_METRIC__dtlb_store_miss_retired),
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, HIT,
+						P4_PEBS_METRIC__none),
+		[ C(RESULT_MISS)   ] = P4_GEN_CACHE_EVENT(P4_EVENT_ITLB_REFERENCE, MISS,
+						P4_PEBS_METRIC__none),
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(NODE) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+/*
+ * Because of Netburst being quite restricted in how many
+ * identical events may run simultaneously, we introduce event aliases,
+ * ie the different events which have the same functionality but
+ * utilize non-intersected resources (ESCR/CCCR/counter registers).
+ *
+ * This allow us to relax restrictions a bit and run two or more
+ * identical events together.
+ *
+ * Never set any custom internal bits such as P4_CONFIG_HT,
+ * P4_CONFIG_ALIASABLE or bits for P4_PEBS_METRIC, they are
+ * either up to date automatically or not applicable at all.
+ */
+static struct p4_event_alias {
+	u64 original;
+	u64 alternative;
+} p4_event_aliases[] = {
+	{
+		/*
+		 * Non-halted cycles can be substituted with non-sleeping cycles (see
+		 * Intel SDM Vol3b for details). We need this alias to be able
+		 * to run nmi-watchdog and 'perf top' (or any other user space tool
+		 * which is interested in running PERF_COUNT_HW_CPU_CYCLES)
+		 * simultaneously.
+		 */
+	.original	=
+		p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_GLOBAL_POWER_EVENTS)		|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING)),
+	.alternative	=
+		p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_EXECUTION_EVENT)		|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0)|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1)|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2)|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3)|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0)	|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1)	|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2)	|
+				    P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3))|
+		p4_config_pack_cccr(P4_CCCR_THRESHOLD(15) | P4_CCCR_COMPLEMENT		|
+				    P4_CCCR_COMPARE),
+	},
+};
+
+static u64 p4_get_alias_event(u64 config)
+{
+	u64 config_match;
+	int i;
+
+	/*
+	 * Only event with special mark is allowed,
+	 * we're to be sure it didn't come as malformed
+	 * RAW event.
+	 */
+	if (!(config & P4_CONFIG_ALIASABLE))
+		return 0;
+
+	config_match = config & P4_CONFIG_EVENT_ALIAS_MASK;
+
+	for (i = 0; i < ARRAY_SIZE(p4_event_aliases); i++) {
+		if (config_match == p4_event_aliases[i].original) {
+			config_match = p4_event_aliases[i].alternative;
+			break;
+		} else if (config_match == p4_event_aliases[i].alternative) {
+			config_match = p4_event_aliases[i].original;
+			break;
+		}
+	}
+
+	if (i >= ARRAY_SIZE(p4_event_aliases))
+		return 0;
+
+	return config_match | (config & P4_CONFIG_EVENT_ALIAS_IMMUTABLE_BITS);
+}
+
+static u64 p4_general_events[PERF_COUNT_HW_MAX] = {
+  /* non-halted CPU clocks */
+  [PERF_COUNT_HW_CPU_CYCLES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_GLOBAL_POWER_EVENTS)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING))	|
+		P4_CONFIG_ALIASABLE,
+
+  /*
+   * retired instructions
+   * in a sake of simplicity we don't use the FSB tagging
+   */
+  [PERF_COUNT_HW_INSTRUCTIONS] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_INSTR_RETIRED)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)),
+
+  /* cache hits */
+  [PERF_COUNT_HW_CACHE_REFERENCES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)),
+
+  /* cache misses */
+  [PERF_COUNT_HW_CACHE_MISSES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_BSQ_CACHE_REFERENCE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS)),
+
+  /* branch instructions retired */
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_RETIRED_BRANCH_TYPE)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT)),
+
+  /* mispredicted branches retired */
+  [PERF_COUNT_HW_BRANCH_MISSES]	=
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_MISPRED_BRANCH_RETIRED)	|
+		P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS)),
+
+  /* bus ready clocks (cpu is driving #DRDY_DRV\#DRDY_OWN):  */
+  [PERF_COUNT_HW_BUS_CYCLES] =
+	p4_config_pack_escr(P4_ESCR_EVENT(P4_EVENT_FSB_DATA_ACTIVITY)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV)		|
+		P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN))	|
+	p4_config_pack_cccr(P4_CCCR_EDGE | P4_CCCR_COMPARE),
+};
+
+static struct p4_event_bind *p4_config_get_bind(u64 config)
+{
+	unsigned int evnt = p4_config_unpack_event(config);
+	struct p4_event_bind *bind = NULL;
+
+	if (evnt < ARRAY_SIZE(p4_event_bind_map))
+		bind = &p4_event_bind_map[evnt];
+
+	return bind;
+}
+
+static u64 p4_pmu_event_map(int hw_event)
+{
+	struct p4_event_bind *bind;
+	unsigned int esel;
+	u64 config;
+
+	config = p4_general_events[hw_event];
+	bind = p4_config_get_bind(config);
+	esel = P4_OPCODE_ESEL(bind->opcode);
+	config |= p4_config_pack_cccr(P4_CCCR_ESEL(esel));
+
+	return config;
+}
+
+/* check cpu model specifics */
+static bool p4_event_match_cpu_model(unsigned int event_idx)
+{
+	/* INSTR_COMPLETED event only exist for model 3, 4, 6 (Prescott) */
+	if (event_idx == P4_EVENT_INSTR_COMPLETED) {
+		if (boot_cpu_data.x86_model != 3 &&
+			boot_cpu_data.x86_model != 4 &&
+			boot_cpu_data.x86_model != 6)
+			return false;
+	}
+
+	/*
+	 * For info
+	 * - IQ_ESCR0, IQ_ESCR1 only for models 1 and 2
+	 */
+
+	return true;
+}
+
+static int p4_validate_raw_event(struct perf_event *event)
+{
+	unsigned int v, emask;
+
+	/* User data may have out-of-bound event index */
+	v = p4_config_unpack_event(event->attr.config);
+	if (v >= ARRAY_SIZE(p4_event_bind_map))
+		return -EINVAL;
+
+	/* It may be unsupported: */
+	if (!p4_event_match_cpu_model(v))
+		return -EINVAL;
+
+	/*
+	 * NOTE: P4_CCCR_THREAD_ANY has not the same meaning as
+	 * in Architectural Performance Monitoring, it means not
+	 * on _which_ logical cpu to count but rather _when_, ie it
+	 * depends on logical cpu state -- count event if one cpu active,
+	 * none, both or any, so we just allow user to pass any value
+	 * desired.
+	 *
+	 * In turn we always set Tx_OS/Tx_USR bits bound to logical
+	 * cpu without their propagation to another cpu
+	 */
+
+	/*
+	 * if an event is shared across the logical threads
+	 * the user needs special permissions to be able to use it
+	 */
+	if (p4_ht_active() && p4_event_bind_map[v].shared) {
+		v = perf_allow_cpu(&event->attr);
+		if (v)
+			return v;
+	}
+
+	/* ESCR EventMask bits may be invalid */
+	emask = p4_config_unpack_escr(event->attr.config) & P4_ESCR_EVENTMASK_MASK;
+	if (emask & ~p4_event_bind_map[v].escr_emask)
+		return -EINVAL;
+
+	/*
+	 * it may have some invalid PEBS bits
+	 */
+	if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE))
+		return -EINVAL;
+
+	v = p4_config_unpack_metric(event->attr.config);
+	if (v >= ARRAY_SIZE(p4_pebs_bind_map))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int p4_hw_config(struct perf_event *event)
+{
+	int cpu = get_cpu();
+	int rc = 0;
+	u32 escr, cccr;
+
+	/*
+	 * the reason we use cpu that early is that: if we get scheduled
+	 * first time on the same cpu -- we will not need swap thread
+	 * specific flags in config (and will save some cpu cycles)
+	 */
+
+	cccr = p4_default_cccr_conf(cpu);
+	escr = p4_default_escr_conf(cpu, event->attr.exclude_kernel,
+					 event->attr.exclude_user);
+	event->hw.config = p4_config_pack_escr(escr) |
+			   p4_config_pack_cccr(cccr);
+
+	if (p4_ht_active() && p4_ht_thread(cpu))
+		event->hw.config = p4_set_ht_bit(event->hw.config);
+
+	if (event->attr.type == PERF_TYPE_RAW) {
+		struct p4_event_bind *bind;
+		unsigned int esel;
+		/*
+		 * Clear bits we reserve to be managed by kernel itself
+		 * and never allowed from a user space
+		 */
+		event->attr.config &= P4_CONFIG_MASK;
+
+		rc = p4_validate_raw_event(event);
+		if (rc)
+			goto out;
+
+		/*
+		 * Note that for RAW events we allow user to use P4_CCCR_RESERVED
+		 * bits since we keep additional info here (for cache events and etc)
+		 */
+		event->hw.config |= event->attr.config;
+		bind = p4_config_get_bind(event->attr.config);
+		if (!bind) {
+			rc = -EINVAL;
+			goto out;
+		}
+		esel = P4_OPCODE_ESEL(bind->opcode);
+		event->hw.config |= p4_config_pack_cccr(P4_CCCR_ESEL(esel));
+	}
+
+	rc = x86_setup_perfctr(event);
+out:
+	put_cpu();
+	return rc;
+}
+
+static inline int p4_pmu_clear_cccr_ovf(struct hw_perf_event *hwc)
+{
+	u64 v;
+
+	/* an official way for overflow indication */
+	rdmsrl(hwc->config_base, v);
+	if (v & P4_CCCR_OVF) {
+		wrmsrl(hwc->config_base, v & ~P4_CCCR_OVF);
+		return 1;
+	}
+
+	/*
+	 * In some circumstances the overflow might issue an NMI but did
+	 * not set P4_CCCR_OVF bit. Because a counter holds a negative value
+	 * we simply check for high bit being set, if it's cleared it means
+	 * the counter has reached zero value and continued counting before
+	 * real NMI signal was received:
+	 */
+	rdmsrl(hwc->event_base, v);
+	if (!(v & ARCH_P4_UNFLAGGED_BIT))
+		return 1;
+
+	return 0;
+}
+
+static void p4_pmu_disable_pebs(void)
+{
+	/*
+	 * FIXME
+	 *
+	 * It's still allowed that two threads setup same cache
+	 * events so we can't simply clear metrics until we knew
+	 * no one is depending on us, so we need kind of counter
+	 * for "ReplayEvent" users.
+	 *
+	 * What is more complex -- RAW events, if user (for some
+	 * reason) will pass some cache event metric with improper
+	 * event opcode -- it's fine from hardware point of view
+	 * but completely nonsense from "meaning" of such action.
+	 *
+	 * So at moment let leave metrics turned on forever -- it's
+	 * ok for now but need to be revisited!
+	 *
+	 * (void)wrmsrl_safe(MSR_IA32_PEBS_ENABLE, 0);
+	 * (void)wrmsrl_safe(MSR_P4_PEBS_MATRIX_VERT, 0);
+	 */
+}
+
+static inline void p4_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/*
+	 * If event gets disabled while counter is in overflowed
+	 * state we need to clear P4_CCCR_OVF, otherwise interrupt get
+	 * asserted again and again
+	 */
+	(void)wrmsrl_safe(hwc->config_base,
+		p4_config_unpack_cccr(hwc->config) & ~P4_CCCR_ENABLE & ~P4_CCCR_OVF & ~P4_CCCR_RESERVED);
+}
+
+static void p4_pmu_disable_all(void)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct perf_event *event = cpuc->events[idx];
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		p4_pmu_disable_event(event);
+	}
+
+	p4_pmu_disable_pebs();
+}
+
+/* configuration must be valid */
+static void p4_pmu_enable_pebs(u64 config)
+{
+	struct p4_pebs_bind *bind;
+	unsigned int idx;
+
+	BUILD_BUG_ON(P4_PEBS_METRIC__max > P4_PEBS_CONFIG_METRIC_MASK);
+
+	idx = p4_config_unpack_metric(config);
+	if (idx == P4_PEBS_METRIC__none)
+		return;
+
+	bind = &p4_pebs_bind_map[idx];
+
+	(void)wrmsrl_safe(MSR_IA32_PEBS_ENABLE,	(u64)bind->metric_pebs);
+	(void)wrmsrl_safe(MSR_P4_PEBS_MATRIX_VERT,	(u64)bind->metric_vert);
+}
+
+static void p4_pmu_enable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	int thread = p4_ht_config_thread(hwc->config);
+	u64 escr_conf = p4_config_unpack_escr(p4_clear_ht_bit(hwc->config));
+	unsigned int idx = p4_config_unpack_event(hwc->config);
+	struct p4_event_bind *bind;
+	u64 escr_addr, cccr;
+
+	bind = &p4_event_bind_map[idx];
+	escr_addr = bind->escr_msr[thread];
+
+	/*
+	 * - we dont support cascaded counters yet
+	 * - and counter 1 is broken (erratum)
+	 */
+	WARN_ON_ONCE(p4_is_event_cascaded(hwc->config));
+	WARN_ON_ONCE(hwc->idx == 1);
+
+	/* we need a real Event value */
+	escr_conf &= ~P4_ESCR_EVENT_MASK;
+	escr_conf |= P4_ESCR_EVENT(P4_OPCODE_EVNT(bind->opcode));
+
+	cccr = p4_config_unpack_cccr(hwc->config);
+
+	/*
+	 * it could be Cache event so we need to write metrics
+	 * into additional MSRs
+	 */
+	p4_pmu_enable_pebs(hwc->config);
+
+	(void)wrmsrl_safe(escr_addr, escr_conf);
+	(void)wrmsrl_safe(hwc->config_base,
+				(cccr & ~P4_CCCR_RESERVED) | P4_CCCR_ENABLE);
+}
+
+static void p4_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct perf_event *event = cpuc->events[idx];
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+		p4_pmu_enable_event(event);
+	}
+}
+
+static int p4_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc;
+	struct perf_event *event;
+	struct hw_perf_event *hwc;
+	int idx, handled = 0;
+	u64 val;
+
+	cpuc = this_cpu_ptr(&cpu_hw_events);
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		int overflow;
+
+		if (!test_bit(idx, cpuc->active_mask)) {
+			/* catch in-flight IRQs */
+			if (__test_and_clear_bit(idx, cpuc->running))
+				handled++;
+			continue;
+		}
+
+		event = cpuc->events[idx];
+		hwc = &event->hw;
+
+		WARN_ON_ONCE(hwc->idx != idx);
+
+		/* it might be unflagged overflow */
+		overflow = p4_pmu_clear_cccr_ovf(hwc);
+
+		val = x86_perf_event_update(event);
+		if (!overflow && (val & (1ULL << (x86_pmu.cntval_bits - 1))))
+			continue;
+
+		handled += overflow;
+
+		/* event overflow for sure */
+		perf_sample_data_init(&data, 0, hwc->last_period);
+
+		if (!x86_perf_event_set_period(event))
+			continue;
+
+
+		if (perf_event_overflow(event, &data, regs))
+			x86_pmu_stop(event, 0);
+	}
+
+	if (handled)
+		inc_irq_stat(apic_perf_irqs);
+
+	/*
+	 * When dealing with the unmasking of the LVTPC on P4 perf hw, it has
+	 * been observed that the OVF bit flag has to be cleared first _before_
+	 * the LVTPC can be unmasked.
+	 *
+	 * The reason is the NMI line will continue to be asserted while the OVF
+	 * bit is set.  This causes a second NMI to generate if the LVTPC is
+	 * unmasked before the OVF bit is cleared, leading to unknown NMI
+	 * messages.
+	 */
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+
+	return handled;
+}
+
+/*
+ * swap thread specific fields according to a thread
+ * we are going to run on
+ */
+static void p4_pmu_swap_config_ts(struct hw_perf_event *hwc, int cpu)
+{
+	u32 escr, cccr;
+
+	/*
+	 * we either lucky and continue on same cpu or no HT support
+	 */
+	if (!p4_should_swap_ts(hwc->config, cpu))
+		return;
+
+	/*
+	 * the event is migrated from an another logical
+	 * cpu, so we need to swap thread specific flags
+	 */
+
+	escr = p4_config_unpack_escr(hwc->config);
+	cccr = p4_config_unpack_cccr(hwc->config);
+
+	if (p4_ht_thread(cpu)) {
+		cccr &= ~P4_CCCR_OVF_PMI_T0;
+		cccr |= P4_CCCR_OVF_PMI_T1;
+		if (escr & P4_ESCR_T0_OS) {
+			escr &= ~P4_ESCR_T0_OS;
+			escr |= P4_ESCR_T1_OS;
+		}
+		if (escr & P4_ESCR_T0_USR) {
+			escr &= ~P4_ESCR_T0_USR;
+			escr |= P4_ESCR_T1_USR;
+		}
+		hwc->config  = p4_config_pack_escr(escr);
+		hwc->config |= p4_config_pack_cccr(cccr);
+		hwc->config |= P4_CONFIG_HT;
+	} else {
+		cccr &= ~P4_CCCR_OVF_PMI_T1;
+		cccr |= P4_CCCR_OVF_PMI_T0;
+		if (escr & P4_ESCR_T1_OS) {
+			escr &= ~P4_ESCR_T1_OS;
+			escr |= P4_ESCR_T0_OS;
+		}
+		if (escr & P4_ESCR_T1_USR) {
+			escr &= ~P4_ESCR_T1_USR;
+			escr |= P4_ESCR_T0_USR;
+		}
+		hwc->config  = p4_config_pack_escr(escr);
+		hwc->config |= p4_config_pack_cccr(cccr);
+		hwc->config &= ~P4_CONFIG_HT;
+	}
+}
+
+/*
+ * ESCR address hashing is tricky, ESCRs are not sequential
+ * in memory but all starts from MSR_P4_BSU_ESCR0 (0x03a0) and
+ * the metric between any ESCRs is laid in range [0xa0,0xe1]
+ *
+ * so we make ~70% filled hashtable
+ */
+
+#define P4_ESCR_MSR_BASE		0x000003a0
+#define P4_ESCR_MSR_MAX			0x000003e1
+#define P4_ESCR_MSR_TABLE_SIZE		(P4_ESCR_MSR_MAX - P4_ESCR_MSR_BASE + 1)
+#define P4_ESCR_MSR_IDX(msr)		(msr - P4_ESCR_MSR_BASE)
+#define P4_ESCR_MSR_TABLE_ENTRY(msr)	[P4_ESCR_MSR_IDX(msr)] = msr
+
+static const unsigned int p4_escr_table[P4_ESCR_MSR_TABLE_SIZE] = {
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_ALF_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_ALF_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_BPU_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_BPU_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_BSU_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_BSU_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR2),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR3),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR4),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_CRU_ESCR5),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_DAC_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_DAC_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FIRM_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FIRM_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FLAME_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FLAME_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FSB_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_FSB_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IQ_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IQ_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IS_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IS_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_ITLB_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_ITLB_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IX_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_IX_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_MOB_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_MOB_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_MS_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_MS_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_PMH_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_PMH_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_RAT_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_RAT_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_SAAT_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_SAAT_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_SSU_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_SSU_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_TBPU_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_TBPU_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_TC_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_TC_ESCR1),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_U2L_ESCR0),
+	P4_ESCR_MSR_TABLE_ENTRY(MSR_P4_U2L_ESCR1),
+};
+
+static int p4_get_escr_idx(unsigned int addr)
+{
+	unsigned int idx = P4_ESCR_MSR_IDX(addr);
+
+	if (unlikely(idx >= P4_ESCR_MSR_TABLE_SIZE	||
+			!p4_escr_table[idx]		||
+			p4_escr_table[idx] != addr)) {
+		WARN_ONCE(1, "P4 PMU: Wrong address passed: %x\n", addr);
+		return -1;
+	}
+
+	return idx;
+}
+
+static int p4_next_cntr(int thread, unsigned long *used_mask,
+			struct p4_event_bind *bind)
+{
+	int i, j;
+
+	for (i = 0; i < P4_CNTR_LIMIT; i++) {
+		j = bind->cntr[thread][i];
+		if (j != -1 && !test_bit(j, used_mask))
+			return j;
+	}
+
+	return -1;
+}
+
+static int p4_pmu_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+{
+	unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long escr_mask[BITS_TO_LONGS(P4_ESCR_MSR_TABLE_SIZE)];
+	int cpu = smp_processor_id();
+	struct hw_perf_event *hwc;
+	struct p4_event_bind *bind;
+	unsigned int i, thread, num;
+	int cntr_idx, escr_idx;
+	u64 config_alias;
+	int pass;
+
+	bitmap_zero(used_mask, X86_PMC_IDX_MAX);
+	bitmap_zero(escr_mask, P4_ESCR_MSR_TABLE_SIZE);
+
+	for (i = 0, num = n; i < n; i++, num--) {
+
+		hwc = &cpuc->event_list[i]->hw;
+		thread = p4_ht_thread(cpu);
+		pass = 0;
+
+again:
+		/*
+		 * It's possible to hit a circular lock
+		 * between original and alternative events
+		 * if both are scheduled already.
+		 */
+		if (pass > 2)
+			goto done;
+
+		bind = p4_config_get_bind(hwc->config);
+		escr_idx = p4_get_escr_idx(bind->escr_msr[thread]);
+		if (unlikely(escr_idx == -1))
+			goto done;
+
+		if (hwc->idx != -1 && !p4_should_swap_ts(hwc->config, cpu)) {
+			cntr_idx = hwc->idx;
+			if (assign)
+				assign[i] = hwc->idx;
+			goto reserve;
+		}
+
+		cntr_idx = p4_next_cntr(thread, used_mask, bind);
+		if (cntr_idx == -1 || test_bit(escr_idx, escr_mask)) {
+			/*
+			 * Check whether an event alias is still available.
+			 */
+			config_alias = p4_get_alias_event(hwc->config);
+			if (!config_alias)
+				goto done;
+			hwc->config = config_alias;
+			pass++;
+			goto again;
+		}
+		/*
+		 * Perf does test runs to see if a whole group can be assigned
+		 * together successfully.  There can be multiple rounds of this.
+		 * Unfortunately, p4_pmu_swap_config_ts touches the hwc->config
+		 * bits, such that the next round of group assignments will
+		 * cause the above p4_should_swap_ts to pass instead of fail.
+		 * This leads to counters exclusive to thread0 being used by
+		 * thread1.
+		 *
+		 * Solve this with a cheap hack, reset the idx back to -1 to
+		 * force a new lookup (p4_next_cntr) to get the right counter
+		 * for the right thread.
+		 *
+		 * This probably doesn't comply with the general spirit of how
+		 * perf wants to work, but P4 is special. :-(
+		 */
+		if (p4_should_swap_ts(hwc->config, cpu))
+			hwc->idx = -1;
+		p4_pmu_swap_config_ts(hwc, cpu);
+		if (assign)
+			assign[i] = cntr_idx;
+reserve:
+		set_bit(cntr_idx, used_mask);
+		set_bit(escr_idx, escr_mask);
+	}
+
+done:
+	return num ? -EINVAL : 0;
+}
+
+PMU_FORMAT_ATTR(cccr, "config:0-31" );
+PMU_FORMAT_ATTR(escr, "config:32-62");
+PMU_FORMAT_ATTR(ht,   "config:63"   );
+
+static struct attribute *intel_p4_formats_attr[] = {
+	&format_attr_cccr.attr,
+	&format_attr_escr.attr,
+	&format_attr_ht.attr,
+	NULL,
+};
+
+static __initconst const struct x86_pmu p4_pmu = {
+	.name			= "Netburst P4/Xeon",
+	.handle_irq		= p4_pmu_handle_irq,
+	.disable_all		= p4_pmu_disable_all,
+	.enable_all		= p4_pmu_enable_all,
+	.enable			= p4_pmu_enable_event,
+	.disable		= p4_pmu_disable_event,
+	.eventsel		= MSR_P4_BPU_CCCR0,
+	.perfctr		= MSR_P4_BPU_PERFCTR0,
+	.event_map		= p4_pmu_event_map,
+	.max_events		= ARRAY_SIZE(p4_general_events),
+	.get_event_constraints	= x86_get_event_constraints,
+	/*
+	 * IF HT disabled we may need to use all
+	 * ARCH_P4_MAX_CCCR counters simulaneously
+	 * though leave it restricted at moment assuming
+	 * HT is on
+	 */
+	.num_counters		= ARCH_P4_MAX_CCCR,
+	.apic			= 1,
+	.cntval_bits		= ARCH_P4_CNTRVAL_BITS,
+	.cntval_mask		= ARCH_P4_CNTRVAL_MASK,
+	.max_period		= (1ULL << (ARCH_P4_CNTRVAL_BITS - 1)) - 1,
+	.hw_config		= p4_hw_config,
+	.schedule_events	= p4_pmu_schedule_events,
+	/*
+	 * This handles erratum N15 in intel doc 249199-029,
+	 * the counter may not be updated correctly on write
+	 * so we need a second write operation to do the trick
+	 * (the official workaround didn't work)
+	 *
+	 * the former idea is taken from OProfile code
+	 */
+	.perfctr_second_write	= 1,
+
+	.format_attrs		= intel_p4_formats_attr,
+};
+
+__init int p4_pmu_init(void)
+{
+	unsigned int low, high;
+	int i, reg;
+
+	/* If we get stripped -- indexing fails */
+	BUILD_BUG_ON(ARCH_P4_MAX_CCCR > INTEL_PMC_MAX_GENERIC);
+
+	rdmsr(MSR_IA32_MISC_ENABLE, low, high);
+	if (!(low & (1 << 7))) {
+		pr_cont("unsupported Netburst CPU model %d ",
+			boot_cpu_data.x86_model);
+		return -ENODEV;
+	}
+
+	memcpy(hw_cache_event_ids, p4_hw_cache_event_ids,
+		sizeof(hw_cache_event_ids));
+
+	pr_cont("Netburst events, ");
+
+	x86_pmu = p4_pmu;
+
+	/*
+	 * Even though the counters are configured to interrupt a particular
+	 * logical processor when an overflow happens, testing has shown that
+	 * on kdump kernels (which uses a single cpu), thread1's counter
+	 * continues to run and will report an NMI on thread0.  Due to the
+	 * overflow bug, this leads to a stream of unknown NMIs.
+	 *
+	 * Solve this by zero'ing out the registers to mimic a reset.
+	 */
+	for (i = 0; i < x86_pmu.num_counters; i++) {
+		reg = x86_pmu_config_addr(i);
+		wrmsrl_safe(reg, 0ULL);
+	}
+
+	return 0;
+}
diff --git a/arch/x86/events/intel/p6.c b/arch/x86/events/intel/p6.c
new file mode 100644
index 000000000..408879b0c
--- /dev/null
+++ b/arch/x86/events/intel/p6.c
@@ -0,0 +1,280 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include "../perf_event.h"
+
+/*
+ * Not sure about some of these
+ */
+static const u64 p6_perfmon_event_map[] =
+{
+  [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 const u64 __initconst 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)
+{
+	return p6_perfmon_event_map[hw_event];
+}
+
+/*
+ * Event setting that is specified not to count anything.
+ * We use this to effectively disable a counter.
+ *
+ * L2_RQSTS with 0 MESI unit mask.
+ */
+#define P6_NOP_EVENT			0x0000002EULL
+
+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, 0x2),	/* FP_ASSIST */
+	INTEL_EVENT_CONSTRAINT(0x12, 0x2),	/* MUL */
+	INTEL_EVENT_CONSTRAINT(0x13, 0x2),	/* DIV */
+	INTEL_EVENT_CONSTRAINT(0x14, 0x1),	/* CYCLES_DIV_BUSY */
+	EVENT_CONSTRAINT_END
+};
+
+static void p6_pmu_disable_all(void)
+{
+	u64 val;
+
+	/* p6 only has one enable register */
+	rdmsrl(MSR_P6_EVNTSEL0, val);
+	val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+	wrmsrl(MSR_P6_EVNTSEL0, val);
+}
+
+static void p6_pmu_enable_all(int added)
+{
+	unsigned long val;
+
+	/* p6 only has one enable register */
+	rdmsrl(MSR_P6_EVNTSEL0, val);
+	val |= ARCH_PERFMON_EVENTSEL_ENABLE;
+	wrmsrl(MSR_P6_EVNTSEL0, val);
+}
+
+static inline void
+p6_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 val = P6_NOP_EVENT;
+
+	(void)wrmsrl_safe(hwc->config_base, val);
+}
+
+static void p6_pmu_enable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 val;
+
+	val = hwc->config;
+
+	/*
+	 * 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);
+}
+
+PMU_FORMAT_ATTR(event,	"config:0-7"	);
+PMU_FORMAT_ATTR(umask,	"config:8-15"	);
+PMU_FORMAT_ATTR(edge,	"config:18"	);
+PMU_FORMAT_ATTR(pc,	"config:19"	);
+PMU_FORMAT_ATTR(inv,	"config:23"	);
+PMU_FORMAT_ATTR(cmask,	"config:24-31"	);
+
+static struct attribute *intel_p6_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_pc.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+static __initconst const struct x86_pmu p6_pmu = {
+	.name			= "p6",
+	.handle_irq		= x86_pmu_handle_irq,
+	.disable_all		= p6_pmu_disable_all,
+	.enable_all		= p6_pmu_enable_all,
+	.enable			= p6_pmu_enable_event,
+	.disable		= p6_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_P6_EVNTSEL0,
+	.perfctr		= MSR_P6_PERFCTR0,
+	.event_map		= p6_pmu_event_map,
+	.max_events		= ARRAY_SIZE(p6_perfmon_event_map),
+	.apic			= 1,
+	.max_period		= (1ULL << 31) - 1,
+	.version		= 0,
+	.num_counters		= 2,
+	/*
+	 * Events have 40 bits implemented. However they are designed such
+	 * that bits [32-39] are sign extensions of bit 31. As such the
+	 * effective width of a event for P6-like PMU is 32 bits only.
+	 *
+	 * See IA-32 Intel Architecture Software developer manual Vol 3B
+	 */
+	.cntval_bits		= 32,
+	.cntval_mask		= (1ULL << 32) - 1,
+	.get_event_constraints	= x86_get_event_constraints,
+	.event_constraints	= p6_event_constraints,
+
+	.format_attrs		= intel_p6_formats_attr,
+	.events_sysfs_show	= intel_event_sysfs_show,
+
+};
+
+static __init void p6_pmu_rdpmc_quirk(void)
+{
+	if (boot_cpu_data.x86_stepping < 9) {
+		/*
+		 * PPro erratum 26; fixed in stepping 9 and above.
+		 */
+		pr_warn("Userspace RDPMC support disabled due to a CPU erratum\n");
+		x86_pmu.attr_rdpmc_broken = 1;
+		x86_pmu.attr_rdpmc = 0;
+	}
+}
+
+__init int p6_pmu_init(void)
+{
+	x86_pmu = p6_pmu;
+
+	switch (boot_cpu_data.x86_model) {
+	case  1: /* Pentium Pro */
+		x86_add_quirk(p6_pmu_rdpmc_quirk);
+		break;
+
+	case  3: /* Pentium II - Klamath */
+	case  5: /* Pentium II - Deschutes */
+	case  6: /* Pentium II - Mendocino */
+		break;
+
+	case  7: /* Pentium III - Katmai */
+	case  8: /* Pentium III - Coppermine */
+	case 10: /* Pentium III Xeon */
+	case 11: /* Pentium III - Tualatin */
+		break;
+
+	case  9: /* Pentium M - Banias */
+	case 13: /* Pentium M - Dothan */
+		break;
+
+	default:
+		pr_cont("unsupported p6 CPU model %d ", boot_cpu_data.x86_model);
+		return -ENODEV;
+	}
+
+	memcpy(hw_cache_event_ids, p6_hw_cache_event_ids,
+		sizeof(hw_cache_event_ids));
+
+	return 0;
+}
diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c
new file mode 100644
index 000000000..5667b8b99
--- /dev/null
+++ b/arch/x86/events/intel/pt.c
@@ -0,0 +1,1798 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Intel(R) Processor Trace PMU driver for perf
+ * Copyright (c) 2013-2014, Intel Corporation.
+ *
+ * Intel PT is specified in the Intel Architecture Instruction Set Extensions
+ * Programming Reference:
+ * http://software.intel.com/en-us/intel-isa-extensions
+ */
+
+#undef DEBUG
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/types.h>
+#include <linux/bits.h>
+#include <linux/limits.h>
+#include <linux/slab.h>
+#include <linux/device.h>
+
+#include <asm/perf_event.h>
+#include <asm/insn.h>
+#include <asm/io.h>
+#include <asm/intel_pt.h>
+#include <asm/intel-family.h>
+
+#include "../perf_event.h"
+#include "pt.h"
+
+static DEFINE_PER_CPU(struct pt, pt_ctx);
+
+static struct pt_pmu pt_pmu;
+
+/*
+ * Capabilities of Intel PT hardware, such as number of address bits or
+ * supported output schemes, are cached and exported to userspace as "caps"
+ * attribute group of pt pmu device
+ * (/sys/bus/event_source/devices/intel_pt/caps/) so that userspace can store
+ * relevant bits together with intel_pt traces.
+ *
+ * These are necessary for both trace decoding (payloads_lip, contains address
+ * width encoded in IP-related packets), and event configuration (bitmasks with
+ * permitted values for certain bit fields).
+ */
+#define PT_CAP(_n, _l, _r, _m)						\
+	[PT_CAP_ ## _n] = { .name = __stringify(_n), .leaf = _l,	\
+			    .reg = _r, .mask = _m }
+
+static struct pt_cap_desc {
+	const char	*name;
+	u32		leaf;
+	u8		reg;
+	u32		mask;
+} pt_caps[] = {
+	PT_CAP(max_subleaf,		0, CPUID_EAX, 0xffffffff),
+	PT_CAP(cr3_filtering,		0, CPUID_EBX, BIT(0)),
+	PT_CAP(psb_cyc,			0, CPUID_EBX, BIT(1)),
+	PT_CAP(ip_filtering,		0, CPUID_EBX, BIT(2)),
+	PT_CAP(mtc,			0, CPUID_EBX, BIT(3)),
+	PT_CAP(ptwrite,			0, CPUID_EBX, BIT(4)),
+	PT_CAP(power_event_trace,	0, CPUID_EBX, BIT(5)),
+	PT_CAP(topa_output,		0, CPUID_ECX, BIT(0)),
+	PT_CAP(topa_multiple_entries,	0, CPUID_ECX, BIT(1)),
+	PT_CAP(single_range_output,	0, CPUID_ECX, BIT(2)),
+	PT_CAP(output_subsys,		0, CPUID_ECX, BIT(3)),
+	PT_CAP(payloads_lip,		0, CPUID_ECX, BIT(31)),
+	PT_CAP(num_address_ranges,	1, CPUID_EAX, 0x7),
+	PT_CAP(mtc_periods,		1, CPUID_EAX, 0xffff0000),
+	PT_CAP(cycle_thresholds,	1, CPUID_EBX, 0xffff),
+	PT_CAP(psb_periods,		1, CPUID_EBX, 0xffff0000),
+};
+
+u32 intel_pt_validate_cap(u32 *caps, enum pt_capabilities capability)
+{
+	struct pt_cap_desc *cd = &pt_caps[capability];
+	u32 c = caps[cd->leaf * PT_CPUID_REGS_NUM + cd->reg];
+	unsigned int shift = __ffs(cd->mask);
+
+	return (c & cd->mask) >> shift;
+}
+EXPORT_SYMBOL_GPL(intel_pt_validate_cap);
+
+u32 intel_pt_validate_hw_cap(enum pt_capabilities cap)
+{
+	return intel_pt_validate_cap(pt_pmu.caps, cap);
+}
+EXPORT_SYMBOL_GPL(intel_pt_validate_hw_cap);
+
+static ssize_t pt_cap_show(struct device *cdev,
+			   struct device_attribute *attr,
+			   char *buf)
+{
+	struct dev_ext_attribute *ea =
+		container_of(attr, struct dev_ext_attribute, attr);
+	enum pt_capabilities cap = (long)ea->var;
+
+	return snprintf(buf, PAGE_SIZE, "%x\n", intel_pt_validate_hw_cap(cap));
+}
+
+static struct attribute_group pt_cap_group __ro_after_init = {
+	.name	= "caps",
+};
+
+PMU_FORMAT_ATTR(pt,		"config:0"	);
+PMU_FORMAT_ATTR(cyc,		"config:1"	);
+PMU_FORMAT_ATTR(pwr_evt,	"config:4"	);
+PMU_FORMAT_ATTR(fup_on_ptw,	"config:5"	);
+PMU_FORMAT_ATTR(mtc,		"config:9"	);
+PMU_FORMAT_ATTR(tsc,		"config:10"	);
+PMU_FORMAT_ATTR(noretcomp,	"config:11"	);
+PMU_FORMAT_ATTR(ptw,		"config:12"	);
+PMU_FORMAT_ATTR(branch,		"config:13"	);
+PMU_FORMAT_ATTR(mtc_period,	"config:14-17"	);
+PMU_FORMAT_ATTR(cyc_thresh,	"config:19-22"	);
+PMU_FORMAT_ATTR(psb_period,	"config:24-27"	);
+
+static struct attribute *pt_formats_attr[] = {
+	&format_attr_pt.attr,
+	&format_attr_cyc.attr,
+	&format_attr_pwr_evt.attr,
+	&format_attr_fup_on_ptw.attr,
+	&format_attr_mtc.attr,
+	&format_attr_tsc.attr,
+	&format_attr_noretcomp.attr,
+	&format_attr_ptw.attr,
+	&format_attr_branch.attr,
+	&format_attr_mtc_period.attr,
+	&format_attr_cyc_thresh.attr,
+	&format_attr_psb_period.attr,
+	NULL,
+};
+
+static struct attribute_group pt_format_group = {
+	.name	= "format",
+	.attrs	= pt_formats_attr,
+};
+
+static ssize_t
+pt_timing_attr_show(struct device *dev, struct device_attribute *attr,
+		    char *page)
+{
+	struct perf_pmu_events_attr *pmu_attr =
+		container_of(attr, struct perf_pmu_events_attr, attr);
+
+	switch (pmu_attr->id) {
+	case 0:
+		return sprintf(page, "%lu\n", pt_pmu.max_nonturbo_ratio);
+	case 1:
+		return sprintf(page, "%u:%u\n",
+			       pt_pmu.tsc_art_num,
+			       pt_pmu.tsc_art_den);
+	default:
+		break;
+	}
+
+	return -EINVAL;
+}
+
+PMU_EVENT_ATTR(max_nonturbo_ratio, timing_attr_max_nonturbo_ratio, 0,
+	       pt_timing_attr_show);
+PMU_EVENT_ATTR(tsc_art_ratio, timing_attr_tsc_art_ratio, 1,
+	       pt_timing_attr_show);
+
+static struct attribute *pt_timing_attr[] = {
+	&timing_attr_max_nonturbo_ratio.attr.attr,
+	&timing_attr_tsc_art_ratio.attr.attr,
+	NULL,
+};
+
+static struct attribute_group pt_timing_group = {
+	.attrs	= pt_timing_attr,
+};
+
+static const struct attribute_group *pt_attr_groups[] = {
+	&pt_cap_group,
+	&pt_format_group,
+	&pt_timing_group,
+	NULL,
+};
+
+static int __init pt_pmu_hw_init(void)
+{
+	struct dev_ext_attribute *de_attrs;
+	struct attribute **attrs;
+	size_t size;
+	u64 reg;
+	int ret;
+	long i;
+
+	rdmsrl(MSR_PLATFORM_INFO, reg);
+	pt_pmu.max_nonturbo_ratio = (reg & 0xff00) >> 8;
+
+	/*
+	 * if available, read in TSC to core crystal clock ratio,
+	 * otherwise, zero for numerator stands for "not enumerated"
+	 * as per SDM
+	 */
+	if (boot_cpu_data.cpuid_level >= CPUID_TSC_LEAF) {
+		u32 eax, ebx, ecx, edx;
+
+		cpuid(CPUID_TSC_LEAF, &eax, &ebx, &ecx, &edx);
+
+		pt_pmu.tsc_art_num = ebx;
+		pt_pmu.tsc_art_den = eax;
+	}
+
+	/* model-specific quirks */
+	switch (boot_cpu_data.x86_model) {
+	case INTEL_FAM6_BROADWELL:
+	case INTEL_FAM6_BROADWELL_D:
+	case INTEL_FAM6_BROADWELL_G:
+	case INTEL_FAM6_BROADWELL_X:
+		/* not setting BRANCH_EN will #GP, erratum BDM106 */
+		pt_pmu.branch_en_always_on = true;
+		break;
+	default:
+		break;
+	}
+
+	if (boot_cpu_has(X86_FEATURE_VMX)) {
+		/*
+		 * Intel SDM, 36.5 "Tracing post-VMXON" says that
+		 * "IA32_VMX_MISC[bit 14]" being 1 means PT can trace
+		 * post-VMXON.
+		 */
+		rdmsrl(MSR_IA32_VMX_MISC, reg);
+		if (reg & BIT(14))
+			pt_pmu.vmx = true;
+	}
+
+	for (i = 0; i < PT_CPUID_LEAVES; i++) {
+		cpuid_count(20, i,
+			    &pt_pmu.caps[CPUID_EAX + i*PT_CPUID_REGS_NUM],
+			    &pt_pmu.caps[CPUID_EBX + i*PT_CPUID_REGS_NUM],
+			    &pt_pmu.caps[CPUID_ECX + i*PT_CPUID_REGS_NUM],
+			    &pt_pmu.caps[CPUID_EDX + i*PT_CPUID_REGS_NUM]);
+	}
+
+	ret = -ENOMEM;
+	size = sizeof(struct attribute *) * (ARRAY_SIZE(pt_caps)+1);
+	attrs = kzalloc(size, GFP_KERNEL);
+	if (!attrs)
+		goto fail;
+
+	size = sizeof(struct dev_ext_attribute) * (ARRAY_SIZE(pt_caps)+1);
+	de_attrs = kzalloc(size, GFP_KERNEL);
+	if (!de_attrs)
+		goto fail;
+
+	for (i = 0; i < ARRAY_SIZE(pt_caps); i++) {
+		struct dev_ext_attribute *de_attr = de_attrs + i;
+
+		de_attr->attr.attr.name = pt_caps[i].name;
+
+		sysfs_attr_init(&de_attr->attr.attr);
+
+		de_attr->attr.attr.mode		= S_IRUGO;
+		de_attr->attr.show		= pt_cap_show;
+		de_attr->var			= (void *)i;
+
+		attrs[i] = &de_attr->attr.attr;
+	}
+
+	pt_cap_group.attrs = attrs;
+
+	return 0;
+
+fail:
+	kfree(attrs);
+
+	return ret;
+}
+
+#define RTIT_CTL_CYC_PSB (RTIT_CTL_CYCLEACC	| \
+			  RTIT_CTL_CYC_THRESH	| \
+			  RTIT_CTL_PSB_FREQ)
+
+#define RTIT_CTL_MTC	(RTIT_CTL_MTC_EN	| \
+			 RTIT_CTL_MTC_RANGE)
+
+#define RTIT_CTL_PTW	(RTIT_CTL_PTW_EN	| \
+			 RTIT_CTL_FUP_ON_PTW)
+
+/*
+ * Bit 0 (TraceEn) in the attr.config is meaningless as the
+ * corresponding bit in the RTIT_CTL can only be controlled
+ * by the driver; therefore, repurpose it to mean: pass
+ * through the bit that was previously assumed to be always
+ * on for PT, thereby allowing the user to *not* set it if
+ * they so wish. See also pt_event_valid() and pt_config().
+ */
+#define RTIT_CTL_PASSTHROUGH RTIT_CTL_TRACEEN
+
+#define PT_CONFIG_MASK (RTIT_CTL_TRACEEN	| \
+			RTIT_CTL_TSC_EN		| \
+			RTIT_CTL_DISRETC	| \
+			RTIT_CTL_BRANCH_EN	| \
+			RTIT_CTL_CYC_PSB	| \
+			RTIT_CTL_MTC		| \
+			RTIT_CTL_PWR_EVT_EN	| \
+			RTIT_CTL_FUP_ON_PTW	| \
+			RTIT_CTL_PTW_EN)
+
+static bool pt_event_valid(struct perf_event *event)
+{
+	u64 config = event->attr.config;
+	u64 allowed, requested;
+
+	if ((config & PT_CONFIG_MASK) != config)
+		return false;
+
+	if (config & RTIT_CTL_CYC_PSB) {
+		if (!intel_pt_validate_hw_cap(PT_CAP_psb_cyc))
+			return false;
+
+		allowed = intel_pt_validate_hw_cap(PT_CAP_psb_periods);
+		requested = (config & RTIT_CTL_PSB_FREQ) >>
+			RTIT_CTL_PSB_FREQ_OFFSET;
+		if (requested && (!(allowed & BIT(requested))))
+			return false;
+
+		allowed = intel_pt_validate_hw_cap(PT_CAP_cycle_thresholds);
+		requested = (config & RTIT_CTL_CYC_THRESH) >>
+			RTIT_CTL_CYC_THRESH_OFFSET;
+		if (requested && (!(allowed & BIT(requested))))
+			return false;
+	}
+
+	if (config & RTIT_CTL_MTC) {
+		/*
+		 * In the unlikely case that CPUID lists valid mtc periods,
+		 * but not the mtc capability, drop out here.
+		 *
+		 * Spec says that setting mtc period bits while mtc bit in
+		 * CPUID is 0 will #GP, so better safe than sorry.
+		 */
+		if (!intel_pt_validate_hw_cap(PT_CAP_mtc))
+			return false;
+
+		allowed = intel_pt_validate_hw_cap(PT_CAP_mtc_periods);
+		if (!allowed)
+			return false;
+
+		requested = (config & RTIT_CTL_MTC_RANGE) >>
+			RTIT_CTL_MTC_RANGE_OFFSET;
+
+		if (!(allowed & BIT(requested)))
+			return false;
+	}
+
+	if (config & RTIT_CTL_PWR_EVT_EN &&
+	    !intel_pt_validate_hw_cap(PT_CAP_power_event_trace))
+		return false;
+
+	if (config & RTIT_CTL_PTW) {
+		if (!intel_pt_validate_hw_cap(PT_CAP_ptwrite))
+			return false;
+
+		/* FUPonPTW without PTW doesn't make sense */
+		if ((config & RTIT_CTL_FUP_ON_PTW) &&
+		    !(config & RTIT_CTL_PTW_EN))
+			return false;
+	}
+
+	/*
+	 * Setting bit 0 (TraceEn in RTIT_CTL MSR) in the attr.config
+	 * clears the assomption that BranchEn must always be enabled,
+	 * as was the case with the first implementation of PT.
+	 * If this bit is not set, the legacy behavior is preserved
+	 * for compatibility with the older userspace.
+	 *
+	 * Re-using bit 0 for this purpose is fine because it is never
+	 * directly set by the user; previous attempts at setting it in
+	 * the attr.config resulted in -EINVAL.
+	 */
+	if (config & RTIT_CTL_PASSTHROUGH) {
+		/*
+		 * Disallow not setting BRANCH_EN where BRANCH_EN is
+		 * always required.
+		 */
+		if (pt_pmu.branch_en_always_on &&
+		    !(config & RTIT_CTL_BRANCH_EN))
+			return false;
+	} else {
+		/*
+		 * Disallow BRANCH_EN without the PASSTHROUGH.
+		 */
+		if (config & RTIT_CTL_BRANCH_EN)
+			return false;
+	}
+
+	return true;
+}
+
+/*
+ * PT configuration helpers
+ * These all are cpu affine and operate on a local PT
+ */
+
+static void pt_config_start(struct perf_event *event)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	u64 ctl = event->hw.config;
+
+	ctl |= RTIT_CTL_TRACEEN;
+	if (READ_ONCE(pt->vmx_on))
+		perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL);
+	else
+		wrmsrl(MSR_IA32_RTIT_CTL, ctl);
+
+	WRITE_ONCE(event->hw.config, ctl);
+}
+
+/* Address ranges and their corresponding msr configuration registers */
+static const struct pt_address_range {
+	unsigned long	msr_a;
+	unsigned long	msr_b;
+	unsigned int	reg_off;
+} pt_address_ranges[] = {
+	{
+		.msr_a	 = MSR_IA32_RTIT_ADDR0_A,
+		.msr_b	 = MSR_IA32_RTIT_ADDR0_B,
+		.reg_off = RTIT_CTL_ADDR0_OFFSET,
+	},
+	{
+		.msr_a	 = MSR_IA32_RTIT_ADDR1_A,
+		.msr_b	 = MSR_IA32_RTIT_ADDR1_B,
+		.reg_off = RTIT_CTL_ADDR1_OFFSET,
+	},
+	{
+		.msr_a	 = MSR_IA32_RTIT_ADDR2_A,
+		.msr_b	 = MSR_IA32_RTIT_ADDR2_B,
+		.reg_off = RTIT_CTL_ADDR2_OFFSET,
+	},
+	{
+		.msr_a	 = MSR_IA32_RTIT_ADDR3_A,
+		.msr_b	 = MSR_IA32_RTIT_ADDR3_B,
+		.reg_off = RTIT_CTL_ADDR3_OFFSET,
+	}
+};
+
+static u64 pt_config_filters(struct perf_event *event)
+{
+	struct pt_filters *filters = event->hw.addr_filters;
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	unsigned int range = 0;
+	u64 rtit_ctl = 0;
+
+	if (!filters)
+		return 0;
+
+	perf_event_addr_filters_sync(event);
+
+	for (range = 0; range < filters->nr_filters; range++) {
+		struct pt_filter *filter = &filters->filter[range];
+
+		/*
+		 * Note, if the range has zero start/end addresses due
+		 * to its dynamic object not being loaded yet, we just
+		 * go ahead and program zeroed range, which will simply
+		 * produce no data. Note^2: if executable code at 0x0
+		 * is a concern, we can set up an "invalid" configuration
+		 * such as msr_b < msr_a.
+		 */
+
+		/* avoid redundant msr writes */
+		if (pt->filters.filter[range].msr_a != filter->msr_a) {
+			wrmsrl(pt_address_ranges[range].msr_a, filter->msr_a);
+			pt->filters.filter[range].msr_a = filter->msr_a;
+		}
+
+		if (pt->filters.filter[range].msr_b != filter->msr_b) {
+			wrmsrl(pt_address_ranges[range].msr_b, filter->msr_b);
+			pt->filters.filter[range].msr_b = filter->msr_b;
+		}
+
+		rtit_ctl |= (u64)filter->config << pt_address_ranges[range].reg_off;
+	}
+
+	return rtit_ctl;
+}
+
+static void pt_config(struct perf_event *event)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	u64 reg;
+
+	/* First round: clear STATUS, in particular the PSB byte counter. */
+	if (!event->hw.config) {
+		perf_event_itrace_started(event);
+		wrmsrl(MSR_IA32_RTIT_STATUS, 0);
+	}
+
+	reg = pt_config_filters(event);
+	reg |= RTIT_CTL_TRACEEN;
+	if (!buf->single)
+		reg |= RTIT_CTL_TOPA;
+
+	/*
+	 * Previously, we had BRANCH_EN on by default, but now that PT has
+	 * grown features outside of branch tracing, it is useful to allow
+	 * the user to disable it. Setting bit 0 in the event's attr.config
+	 * allows BRANCH_EN to pass through instead of being always on. See
+	 * also the comment in pt_event_valid().
+	 */
+	if (event->attr.config & BIT(0)) {
+		reg |= event->attr.config & RTIT_CTL_BRANCH_EN;
+	} else {
+		reg |= RTIT_CTL_BRANCH_EN;
+	}
+
+	if (!event->attr.exclude_kernel)
+		reg |= RTIT_CTL_OS;
+	if (!event->attr.exclude_user)
+		reg |= RTIT_CTL_USR;
+
+	reg |= (event->attr.config & PT_CONFIG_MASK);
+
+	event->hw.config = reg;
+	pt_config_start(event);
+}
+
+static void pt_config_stop(struct perf_event *event)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	u64 ctl = READ_ONCE(event->hw.config);
+
+	/* may be already stopped by a PMI */
+	if (!(ctl & RTIT_CTL_TRACEEN))
+		return;
+
+	ctl &= ~RTIT_CTL_TRACEEN;
+	if (!READ_ONCE(pt->vmx_on))
+		wrmsrl(MSR_IA32_RTIT_CTL, ctl);
+
+	WRITE_ONCE(event->hw.config, ctl);
+
+	/*
+	 * A wrmsr that disables trace generation serializes other PT
+	 * registers and causes all data packets to be written to memory,
+	 * but a fence is required for the data to become globally visible.
+	 *
+	 * The below WMB, separating data store and aux_head store matches
+	 * the consumer's RMB that separates aux_head load and data load.
+	 */
+	wmb();
+}
+
+/**
+ * struct topa - ToPA metadata
+ * @list:	linkage to struct pt_buffer's list of tables
+ * @offset:	offset of the first entry in this table in the buffer
+ * @size:	total size of all entries in this table
+ * @last:	index of the last initialized entry in this table
+ * @z_count:	how many times the first entry repeats
+ */
+struct topa {
+	struct list_head	list;
+	u64			offset;
+	size_t			size;
+	int			last;
+	unsigned int		z_count;
+};
+
+/*
+ * Keep ToPA table-related metadata on the same page as the actual table,
+ * taking up a few words from the top
+ */
+
+#define TENTS_PER_PAGE	\
+	((PAGE_SIZE - sizeof(struct topa)) / sizeof(struct topa_entry))
+
+/**
+ * struct topa_page - page-sized ToPA table with metadata at the top
+ * @table:	actual ToPA table entries, as understood by PT hardware
+ * @topa:	metadata
+ */
+struct topa_page {
+	struct topa_entry	table[TENTS_PER_PAGE];
+	struct topa		topa;
+};
+
+static inline struct topa_page *topa_to_page(struct topa *topa)
+{
+	return container_of(topa, struct topa_page, topa);
+}
+
+static inline struct topa_page *topa_entry_to_page(struct topa_entry *te)
+{
+	return (struct topa_page *)((unsigned long)te & PAGE_MASK);
+}
+
+static inline phys_addr_t topa_pfn(struct topa *topa)
+{
+	return PFN_DOWN(virt_to_phys(topa_to_page(topa)));
+}
+
+/* make -1 stand for the last table entry */
+#define TOPA_ENTRY(t, i)				\
+	((i) == -1					\
+		? &topa_to_page(t)->table[(t)->last]	\
+		: &topa_to_page(t)->table[(i)])
+#define TOPA_ENTRY_SIZE(t, i) (sizes(TOPA_ENTRY((t), (i))->size))
+#define TOPA_ENTRY_PAGES(t, i) (1 << TOPA_ENTRY((t), (i))->size)
+
+static void pt_config_buffer(struct pt_buffer *buf)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	u64 reg, mask;
+	void *base;
+
+	if (buf->single) {
+		base = buf->data_pages[0];
+		mask = (buf->nr_pages * PAGE_SIZE - 1) >> 7;
+	} else {
+		base = topa_to_page(buf->cur)->table;
+		mask = (u64)buf->cur_idx;
+	}
+
+	reg = virt_to_phys(base);
+	if (pt->output_base != reg) {
+		pt->output_base = reg;
+		wrmsrl(MSR_IA32_RTIT_OUTPUT_BASE, reg);
+	}
+
+	reg = 0x7f | (mask << 7) | ((u64)buf->output_off << 32);
+	if (pt->output_mask != reg) {
+		pt->output_mask = reg;
+		wrmsrl(MSR_IA32_RTIT_OUTPUT_MASK, reg);
+	}
+}
+
+/**
+ * topa_alloc() - allocate page-sized ToPA table
+ * @cpu:	CPU on which to allocate.
+ * @gfp:	Allocation flags.
+ *
+ * Return:	On success, return the pointer to ToPA table page.
+ */
+static struct topa *topa_alloc(int cpu, gfp_t gfp)
+{
+	int node = cpu_to_node(cpu);
+	struct topa_page *tp;
+	struct page *p;
+
+	p = alloc_pages_node(node, gfp | __GFP_ZERO, 0);
+	if (!p)
+		return NULL;
+
+	tp = page_address(p);
+	tp->topa.last = 0;
+
+	/*
+	 * In case of singe-entry ToPA, always put the self-referencing END
+	 * link as the 2nd entry in the table
+	 */
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) {
+		TOPA_ENTRY(&tp->topa, 1)->base = page_to_phys(p) >> TOPA_SHIFT;
+		TOPA_ENTRY(&tp->topa, 1)->end = 1;
+	}
+
+	return &tp->topa;
+}
+
+/**
+ * topa_free() - free a page-sized ToPA table
+ * @topa:	Table to deallocate.
+ */
+static void topa_free(struct topa *topa)
+{
+	free_page((unsigned long)topa);
+}
+
+/**
+ * topa_insert_table() - insert a ToPA table into a buffer
+ * @buf:	 PT buffer that's being extended.
+ * @topa:	 New topa table to be inserted.
+ *
+ * If it's the first table in this buffer, set up buffer's pointers
+ * accordingly; otherwise, add a END=1 link entry to @topa to the current
+ * "last" table and adjust the last table pointer to @topa.
+ */
+static void topa_insert_table(struct pt_buffer *buf, struct topa *topa)
+{
+	struct topa *last = buf->last;
+
+	list_add_tail(&topa->list, &buf->tables);
+
+	if (!buf->first) {
+		buf->first = buf->last = buf->cur = topa;
+		return;
+	}
+
+	topa->offset = last->offset + last->size;
+	buf->last = topa;
+
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries))
+		return;
+
+	BUG_ON(last->last != TENTS_PER_PAGE - 1);
+
+	TOPA_ENTRY(last, -1)->base = topa_pfn(topa);
+	TOPA_ENTRY(last, -1)->end = 1;
+}
+
+/**
+ * topa_table_full() - check if a ToPA table is filled up
+ * @topa:	ToPA table.
+ */
+static bool topa_table_full(struct topa *topa)
+{
+	/* single-entry ToPA is a special case */
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries))
+		return !!topa->last;
+
+	return topa->last == TENTS_PER_PAGE - 1;
+}
+
+/**
+ * topa_insert_pages() - create a list of ToPA tables
+ * @buf:	PT buffer being initialized.
+ * @gfp:	Allocation flags.
+ *
+ * This initializes a list of ToPA tables with entries from
+ * the data_pages provided by rb_alloc_aux().
+ *
+ * Return:	0 on success or error code.
+ */
+static int topa_insert_pages(struct pt_buffer *buf, int cpu, gfp_t gfp)
+{
+	struct topa *topa = buf->last;
+	int order = 0;
+	struct page *p;
+
+	p = virt_to_page(buf->data_pages[buf->nr_pages]);
+	if (PagePrivate(p))
+		order = page_private(p);
+
+	if (topa_table_full(topa)) {
+		topa = topa_alloc(cpu, gfp);
+		if (!topa)
+			return -ENOMEM;
+
+		topa_insert_table(buf, topa);
+	}
+
+	if (topa->z_count == topa->last - 1) {
+		if (order == TOPA_ENTRY(topa, topa->last - 1)->size)
+			topa->z_count++;
+	}
+
+	TOPA_ENTRY(topa, -1)->base = page_to_phys(p) >> TOPA_SHIFT;
+	TOPA_ENTRY(topa, -1)->size = order;
+	if (!buf->snapshot &&
+	    !intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) {
+		TOPA_ENTRY(topa, -1)->intr = 1;
+		TOPA_ENTRY(topa, -1)->stop = 1;
+	}
+
+	topa->last++;
+	topa->size += sizes(order);
+
+	buf->nr_pages += 1ul << order;
+
+	return 0;
+}
+
+/**
+ * pt_topa_dump() - print ToPA tables and their entries
+ * @buf:	PT buffer.
+ */
+static void pt_topa_dump(struct pt_buffer *buf)
+{
+	struct topa *topa;
+
+	list_for_each_entry(topa, &buf->tables, list) {
+		struct topa_page *tp = topa_to_page(topa);
+		int i;
+
+		pr_debug("# table @%p, off %llx size %zx\n", tp->table,
+			 topa->offset, topa->size);
+		for (i = 0; i < TENTS_PER_PAGE; i++) {
+			pr_debug("# entry @%p (%lx sz %u %c%c%c) raw=%16llx\n",
+				 &tp->table[i],
+				 (unsigned long)tp->table[i].base << TOPA_SHIFT,
+				 sizes(tp->table[i].size),
+				 tp->table[i].end ?  'E' : ' ',
+				 tp->table[i].intr ? 'I' : ' ',
+				 tp->table[i].stop ? 'S' : ' ',
+				 *(u64 *)&tp->table[i]);
+			if ((intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) &&
+			     tp->table[i].stop) ||
+			    tp->table[i].end)
+				break;
+			if (!i && topa->z_count)
+				i += topa->z_count;
+		}
+	}
+}
+
+/**
+ * pt_buffer_advance() - advance to the next output region
+ * @buf:	PT buffer.
+ *
+ * Advance the current pointers in the buffer to the next ToPA entry.
+ */
+static void pt_buffer_advance(struct pt_buffer *buf)
+{
+	buf->output_off = 0;
+	buf->cur_idx++;
+
+	if (buf->cur_idx == buf->cur->last) {
+		if (buf->cur == buf->last)
+			buf->cur = buf->first;
+		else
+			buf->cur = list_entry(buf->cur->list.next, struct topa,
+					      list);
+		buf->cur_idx = 0;
+	}
+}
+
+/**
+ * pt_update_head() - calculate current offsets and sizes
+ * @pt:		Per-cpu pt context.
+ *
+ * Update buffer's current write pointer position and data size.
+ */
+static void pt_update_head(struct pt *pt)
+{
+	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	u64 topa_idx, base, old;
+
+	if (buf->single) {
+		local_set(&buf->data_size, buf->output_off);
+		return;
+	}
+
+	/* offset of the first region in this table from the beginning of buf */
+	base = buf->cur->offset + buf->output_off;
+
+	/* offset of the current output region within this table */
+	for (topa_idx = 0; topa_idx < buf->cur_idx; topa_idx++)
+		base += TOPA_ENTRY_SIZE(buf->cur, topa_idx);
+
+	if (buf->snapshot) {
+		local_set(&buf->data_size, base);
+	} else {
+		old = (local64_xchg(&buf->head, base) &
+		       ((buf->nr_pages << PAGE_SHIFT) - 1));
+		if (base < old)
+			base += buf->nr_pages << PAGE_SHIFT;
+
+		local_add(base - old, &buf->data_size);
+	}
+}
+
+/**
+ * pt_buffer_region() - obtain current output region's address
+ * @buf:	PT buffer.
+ */
+static void *pt_buffer_region(struct pt_buffer *buf)
+{
+	return phys_to_virt(TOPA_ENTRY(buf->cur, buf->cur_idx)->base << TOPA_SHIFT);
+}
+
+/**
+ * pt_buffer_region_size() - obtain current output region's size
+ * @buf:	PT buffer.
+ */
+static size_t pt_buffer_region_size(struct pt_buffer *buf)
+{
+	return TOPA_ENTRY_SIZE(buf->cur, buf->cur_idx);
+}
+
+/**
+ * pt_handle_status() - take care of possible status conditions
+ * @pt:		Per-cpu pt context.
+ */
+static void pt_handle_status(struct pt *pt)
+{
+	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	int advance = 0;
+	u64 status;
+
+	rdmsrl(MSR_IA32_RTIT_STATUS, status);
+
+	if (status & RTIT_STATUS_ERROR) {
+		pr_err_ratelimited("ToPA ERROR encountered, trying to recover\n");
+		pt_topa_dump(buf);
+		status &= ~RTIT_STATUS_ERROR;
+	}
+
+	if (status & RTIT_STATUS_STOPPED) {
+		status &= ~RTIT_STATUS_STOPPED;
+
+		/*
+		 * On systems that only do single-entry ToPA, hitting STOP
+		 * means we are already losing data; need to let the decoder
+		 * know.
+		 */
+		if (!buf->single &&
+		    (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) ||
+		     buf->output_off == pt_buffer_region_size(buf))) {
+			perf_aux_output_flag(&pt->handle,
+			                     PERF_AUX_FLAG_TRUNCATED);
+			advance++;
+		}
+	}
+
+	/*
+	 * Also on single-entry ToPA implementations, interrupt will come
+	 * before the output reaches its output region's boundary.
+	 */
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) &&
+	    !buf->snapshot &&
+	    pt_buffer_region_size(buf) - buf->output_off <= TOPA_PMI_MARGIN) {
+		void *head = pt_buffer_region(buf);
+
+		/* everything within this margin needs to be zeroed out */
+		memset(head + buf->output_off, 0,
+		       pt_buffer_region_size(buf) -
+		       buf->output_off);
+		advance++;
+	}
+
+	if (advance)
+		pt_buffer_advance(buf);
+
+	wrmsrl(MSR_IA32_RTIT_STATUS, status);
+}
+
+/**
+ * pt_read_offset() - translate registers into buffer pointers
+ * @buf:	PT buffer.
+ *
+ * Set buffer's output pointers from MSR values.
+ */
+static void pt_read_offset(struct pt_buffer *buf)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct topa_page *tp;
+
+	if (!buf->single) {
+		rdmsrl(MSR_IA32_RTIT_OUTPUT_BASE, pt->output_base);
+		tp = phys_to_virt(pt->output_base);
+		buf->cur = &tp->topa;
+	}
+
+	rdmsrl(MSR_IA32_RTIT_OUTPUT_MASK, pt->output_mask);
+	/* offset within current output region */
+	buf->output_off = pt->output_mask >> 32;
+	/* index of current output region within this table */
+	if (!buf->single)
+		buf->cur_idx = (pt->output_mask & 0xffffff80) >> 7;
+}
+
+static struct topa_entry *
+pt_topa_entry_for_page(struct pt_buffer *buf, unsigned int pg)
+{
+	struct topa_page *tp;
+	struct topa *topa;
+	unsigned int idx, cur_pg = 0, z_pg = 0, start_idx = 0;
+
+	/*
+	 * Indicates a bug in the caller.
+	 */
+	if (WARN_ON_ONCE(pg >= buf->nr_pages))
+		return NULL;
+
+	/*
+	 * First, find the ToPA table where @pg fits. With high
+	 * order allocations, there shouldn't be many of these.
+	 */
+	list_for_each_entry(topa, &buf->tables, list) {
+		if (topa->offset + topa->size > pg << PAGE_SHIFT)
+			goto found;
+	}
+
+	/*
+	 * Hitting this means we have a problem in the ToPA
+	 * allocation code.
+	 */
+	WARN_ON_ONCE(1);
+
+	return NULL;
+
+found:
+	/*
+	 * Indicates a problem in the ToPA allocation code.
+	 */
+	if (WARN_ON_ONCE(topa->last == -1))
+		return NULL;
+
+	tp = topa_to_page(topa);
+	cur_pg = PFN_DOWN(topa->offset);
+	if (topa->z_count) {
+		z_pg = TOPA_ENTRY_PAGES(topa, 0) * (topa->z_count + 1);
+		start_idx = topa->z_count + 1;
+	}
+
+	/*
+	 * Multiple entries at the beginning of the table have the same size,
+	 * ideally all of them; if @pg falls there, the search is done.
+	 */
+	if (pg >= cur_pg && pg < cur_pg + z_pg) {
+		idx = (pg - cur_pg) / TOPA_ENTRY_PAGES(topa, 0);
+		return &tp->table[idx];
+	}
+
+	/*
+	 * Otherwise, slow path: iterate through the remaining entries.
+	 */
+	for (idx = start_idx, cur_pg += z_pg; idx < topa->last; idx++) {
+		if (cur_pg + TOPA_ENTRY_PAGES(topa, idx) > pg)
+			return &tp->table[idx];
+
+		cur_pg += TOPA_ENTRY_PAGES(topa, idx);
+	}
+
+	/*
+	 * Means we couldn't find a ToPA entry in the table that does match.
+	 */
+	WARN_ON_ONCE(1);
+
+	return NULL;
+}
+
+static struct topa_entry *
+pt_topa_prev_entry(struct pt_buffer *buf, struct topa_entry *te)
+{
+	unsigned long table = (unsigned long)te & ~(PAGE_SIZE - 1);
+	struct topa_page *tp;
+	struct topa *topa;
+
+	tp = (struct topa_page *)table;
+	if (tp->table != te)
+		return --te;
+
+	topa = &tp->topa;
+	if (topa == buf->first)
+		topa = buf->last;
+	else
+		topa = list_prev_entry(topa, list);
+
+	tp = topa_to_page(topa);
+
+	return &tp->table[topa->last - 1];
+}
+
+/**
+ * pt_buffer_reset_markers() - place interrupt and stop bits in the buffer
+ * @buf:	PT buffer.
+ * @handle:	Current output handle.
+ *
+ * Place INT and STOP marks to prevent overwriting old data that the consumer
+ * hasn't yet collected and waking up the consumer after a certain fraction of
+ * the buffer has filled up. Only needed and sensible for non-snapshot counters.
+ *
+ * This obviously relies on buf::head to figure out buffer markers, so it has
+ * to be called after pt_buffer_reset_offsets() and before the hardware tracing
+ * is enabled.
+ */
+static int pt_buffer_reset_markers(struct pt_buffer *buf,
+				   struct perf_output_handle *handle)
+
+{
+	unsigned long head = local64_read(&buf->head);
+	unsigned long idx, npages, wakeup;
+
+	if (buf->single)
+		return 0;
+
+	/* can't stop in the middle of an output region */
+	if (buf->output_off + handle->size + 1 < pt_buffer_region_size(buf)) {
+		perf_aux_output_flag(handle, PERF_AUX_FLAG_TRUNCATED);
+		return -EINVAL;
+	}
+
+
+	/* single entry ToPA is handled by marking all regions STOP=1 INT=1 */
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries))
+		return 0;
+
+	/* clear STOP and INT from current entry */
+	if (buf->stop_te) {
+		buf->stop_te->stop = 0;
+		buf->stop_te->intr = 0;
+	}
+
+	if (buf->intr_te)
+		buf->intr_te->intr = 0;
+
+	/* how many pages till the STOP marker */
+	npages = handle->size >> PAGE_SHIFT;
+
+	/* if it's on a page boundary, fill up one more page */
+	if (!offset_in_page(head + handle->size + 1))
+		npages++;
+
+	idx = (head >> PAGE_SHIFT) + npages;
+	idx &= buf->nr_pages - 1;
+
+	if (idx != buf->stop_pos) {
+		buf->stop_pos = idx;
+		buf->stop_te = pt_topa_entry_for_page(buf, idx);
+		buf->stop_te = pt_topa_prev_entry(buf, buf->stop_te);
+	}
+
+	wakeup = handle->wakeup >> PAGE_SHIFT;
+
+	/* in the worst case, wake up the consumer one page before hard stop */
+	idx = (head >> PAGE_SHIFT) + npages - 1;
+	if (idx > wakeup)
+		idx = wakeup;
+
+	idx &= buf->nr_pages - 1;
+	if (idx != buf->intr_pos) {
+		buf->intr_pos = idx;
+		buf->intr_te = pt_topa_entry_for_page(buf, idx);
+		buf->intr_te = pt_topa_prev_entry(buf, buf->intr_te);
+	}
+
+	buf->stop_te->stop = 1;
+	buf->stop_te->intr = 1;
+	buf->intr_te->intr = 1;
+
+	return 0;
+}
+
+/**
+ * pt_buffer_reset_offsets() - adjust buffer's write pointers from aux_head
+ * @buf:	PT buffer.
+ * @head:	Write pointer (aux_head) from AUX buffer.
+ *
+ * Find the ToPA table and entry corresponding to given @head and set buffer's
+ * "current" pointers accordingly. This is done after we have obtained the
+ * current aux_head position from a successful call to perf_aux_output_begin()
+ * to make sure the hardware is writing to the right place.
+ *
+ * This function modifies buf::{cur,cur_idx,output_off} that will be programmed
+ * into PT msrs when the tracing is enabled and buf::head and buf::data_size,
+ * which are used to determine INT and STOP markers' locations by a subsequent
+ * call to pt_buffer_reset_markers().
+ */
+static void pt_buffer_reset_offsets(struct pt_buffer *buf, unsigned long head)
+{
+	struct topa_page *cur_tp;
+	struct topa_entry *te;
+	int pg;
+
+	if (buf->snapshot)
+		head &= (buf->nr_pages << PAGE_SHIFT) - 1;
+
+	if (!buf->single) {
+		pg = (head >> PAGE_SHIFT) & (buf->nr_pages - 1);
+		te = pt_topa_entry_for_page(buf, pg);
+
+		cur_tp = topa_entry_to_page(te);
+		buf->cur = &cur_tp->topa;
+		buf->cur_idx = te - TOPA_ENTRY(buf->cur, 0);
+		buf->output_off = head & (pt_buffer_region_size(buf) - 1);
+	} else {
+		buf->output_off = head;
+	}
+
+	local64_set(&buf->head, head);
+	local_set(&buf->data_size, 0);
+}
+
+/**
+ * pt_buffer_fini_topa() - deallocate ToPA structure of a buffer
+ * @buf:	PT buffer.
+ */
+static void pt_buffer_fini_topa(struct pt_buffer *buf)
+{
+	struct topa *topa, *iter;
+
+	if (buf->single)
+		return;
+
+	list_for_each_entry_safe(topa, iter, &buf->tables, list) {
+		/*
+		 * right now, this is in free_aux() path only, so
+		 * no need to unlink this table from the list
+		 */
+		topa_free(topa);
+	}
+}
+
+/**
+ * pt_buffer_init_topa() - initialize ToPA table for pt buffer
+ * @buf:	PT buffer.
+ * @size:	Total size of all regions within this ToPA.
+ * @gfp:	Allocation flags.
+ */
+static int pt_buffer_init_topa(struct pt_buffer *buf, int cpu,
+			       unsigned long nr_pages, gfp_t gfp)
+{
+	struct topa *topa;
+	int err;
+
+	topa = topa_alloc(cpu, gfp);
+	if (!topa)
+		return -ENOMEM;
+
+	topa_insert_table(buf, topa);
+
+	while (buf->nr_pages < nr_pages) {
+		err = topa_insert_pages(buf, cpu, gfp);
+		if (err) {
+			pt_buffer_fini_topa(buf);
+			return -ENOMEM;
+		}
+	}
+
+	/* link last table to the first one, unless we're double buffering */
+	if (intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) {
+		TOPA_ENTRY(buf->last, -1)->base = topa_pfn(buf->first);
+		TOPA_ENTRY(buf->last, -1)->end = 1;
+	}
+
+	pt_topa_dump(buf);
+	return 0;
+}
+
+static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages)
+{
+	struct page *p = virt_to_page(buf->data_pages[0]);
+	int ret = -ENOTSUPP, order = 0;
+
+	/*
+	 * We can use single range output mode
+	 * + in snapshot mode, where we don't need interrupts;
+	 * + if the hardware supports it;
+	 * + if the entire buffer is one contiguous allocation.
+	 */
+	if (!buf->snapshot)
+		goto out;
+
+	if (!intel_pt_validate_hw_cap(PT_CAP_single_range_output))
+		goto out;
+
+	if (PagePrivate(p))
+		order = page_private(p);
+
+	if (1 << order != nr_pages)
+		goto out;
+
+	/*
+	 * Some processors cannot always support single range for more than
+	 * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might
+	 * also be affected, so for now rather than trying to keep track of
+	 * which ones, just disable it for all.
+	 */
+	if (nr_pages > 1)
+		goto out;
+
+	buf->single = true;
+	buf->nr_pages = nr_pages;
+	ret = 0;
+out:
+	return ret;
+}
+
+/**
+ * pt_buffer_setup_aux() - set up topa tables for a PT buffer
+ * @cpu:	Cpu on which to allocate, -1 means current.
+ * @pages:	Array of pointers to buffer pages passed from perf core.
+ * @nr_pages:	Number of pages in the buffer.
+ * @snapshot:	If this is a snapshot/overwrite counter.
+ *
+ * This is a pmu::setup_aux callback that sets up ToPA tables and all the
+ * bookkeeping for an AUX buffer.
+ *
+ * Return:	Our private PT buffer structure.
+ */
+static void *
+pt_buffer_setup_aux(struct perf_event *event, void **pages,
+		    int nr_pages, bool snapshot)
+{
+	struct pt_buffer *buf;
+	int node, ret, cpu = event->cpu;
+
+	if (!nr_pages)
+		return NULL;
+
+	/*
+	 * Only support AUX sampling in snapshot mode, where we don't
+	 * generate NMIs.
+	 */
+	if (event->attr.aux_sample_size && !snapshot)
+		return NULL;
+
+	if (cpu == -1)
+		cpu = raw_smp_processor_id();
+	node = cpu_to_node(cpu);
+
+	buf = kzalloc_node(sizeof(struct pt_buffer), GFP_KERNEL, node);
+	if (!buf)
+		return NULL;
+
+	buf->snapshot = snapshot;
+	buf->data_pages = pages;
+	buf->stop_pos = -1;
+	buf->intr_pos = -1;
+
+	INIT_LIST_HEAD(&buf->tables);
+
+	ret = pt_buffer_try_single(buf, nr_pages);
+	if (!ret)
+		return buf;
+
+	ret = pt_buffer_init_topa(buf, cpu, nr_pages, GFP_KERNEL);
+	if (ret) {
+		kfree(buf);
+		return NULL;
+	}
+
+	return buf;
+}
+
+/**
+ * pt_buffer_free_aux() - perf AUX deallocation path callback
+ * @data:	PT buffer.
+ */
+static void pt_buffer_free_aux(void *data)
+{
+	struct pt_buffer *buf = data;
+
+	pt_buffer_fini_topa(buf);
+	kfree(buf);
+}
+
+static int pt_addr_filters_init(struct perf_event *event)
+{
+	struct pt_filters *filters;
+	int node = event->cpu == -1 ? -1 : cpu_to_node(event->cpu);
+
+	if (!intel_pt_validate_hw_cap(PT_CAP_num_address_ranges))
+		return 0;
+
+	filters = kzalloc_node(sizeof(struct pt_filters), GFP_KERNEL, node);
+	if (!filters)
+		return -ENOMEM;
+
+	if (event->parent)
+		memcpy(filters, event->parent->hw.addr_filters,
+		       sizeof(*filters));
+
+	event->hw.addr_filters = filters;
+
+	return 0;
+}
+
+static void pt_addr_filters_fini(struct perf_event *event)
+{
+	kfree(event->hw.addr_filters);
+	event->hw.addr_filters = NULL;
+}
+
+#ifdef CONFIG_X86_64
+/* Clamp to a canonical address greater-than-or-equal-to the address given */
+static u64 clamp_to_ge_canonical_addr(u64 vaddr, u8 vaddr_bits)
+{
+	return __is_canonical_address(vaddr, vaddr_bits) ?
+	       vaddr :
+	       -BIT_ULL(vaddr_bits - 1);
+}
+
+/* Clamp to a canonical address less-than-or-equal-to the address given */
+static u64 clamp_to_le_canonical_addr(u64 vaddr, u8 vaddr_bits)
+{
+	return __is_canonical_address(vaddr, vaddr_bits) ?
+	       vaddr :
+	       BIT_ULL(vaddr_bits - 1) - 1;
+}
+#else
+#define clamp_to_ge_canonical_addr(x, y) (x)
+#define clamp_to_le_canonical_addr(x, y) (x)
+#endif
+
+static int pt_event_addr_filters_validate(struct list_head *filters)
+{
+	struct perf_addr_filter *filter;
+	int range = 0;
+
+	list_for_each_entry(filter, filters, entry) {
+		/*
+		 * PT doesn't support single address triggers and
+		 * 'start' filters.
+		 */
+		if (!filter->size ||
+		    filter->action == PERF_ADDR_FILTER_ACTION_START)
+			return -EOPNOTSUPP;
+
+		if (++range > intel_pt_validate_hw_cap(PT_CAP_num_address_ranges))
+			return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static void pt_event_addr_filters_sync(struct perf_event *event)
+{
+	struct perf_addr_filters_head *head = perf_event_addr_filters(event);
+	unsigned long msr_a, msr_b;
+	struct perf_addr_filter_range *fr = event->addr_filter_ranges;
+	struct pt_filters *filters = event->hw.addr_filters;
+	struct perf_addr_filter *filter;
+	int range = 0;
+
+	if (!filters)
+		return;
+
+	list_for_each_entry(filter, &head->list, entry) {
+		if (filter->path.dentry && !fr[range].start) {
+			msr_a = msr_b = 0;
+		} else {
+			unsigned long n = fr[range].size - 1;
+			unsigned long a = fr[range].start;
+			unsigned long b;
+
+			if (a > ULONG_MAX - n)
+				b = ULONG_MAX;
+			else
+				b = a + n;
+			/*
+			 * Apply the offset. 64-bit addresses written to the
+			 * MSRs must be canonical, but the range can encompass
+			 * non-canonical addresses. Since software cannot
+			 * execute at non-canonical addresses, adjusting to
+			 * canonical addresses does not affect the result of the
+			 * address filter.
+			 */
+			msr_a = clamp_to_ge_canonical_addr(a, boot_cpu_data.x86_virt_bits);
+			msr_b = clamp_to_le_canonical_addr(b, boot_cpu_data.x86_virt_bits);
+			if (msr_b < msr_a)
+				msr_a = msr_b = 0;
+		}
+
+		filters->filter[range].msr_a  = msr_a;
+		filters->filter[range].msr_b  = msr_b;
+		if (filter->action == PERF_ADDR_FILTER_ACTION_FILTER)
+			filters->filter[range].config = 1;
+		else
+			filters->filter[range].config = 2;
+		range++;
+	}
+
+	filters->nr_filters = range;
+}
+
+/**
+ * intel_pt_interrupt() - PT PMI handler
+ */
+void intel_pt_interrupt(void)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct pt_buffer *buf;
+	struct perf_event *event = pt->handle.event;
+
+	/*
+	 * There may be a dangling PT bit in the interrupt status register
+	 * after PT has been disabled by pt_event_stop(). Make sure we don't
+	 * do anything (particularly, re-enable) for this event here.
+	 */
+	if (!READ_ONCE(pt->handle_nmi))
+		return;
+
+	if (!event)
+		return;
+
+	pt_config_stop(event);
+
+	buf = perf_get_aux(&pt->handle);
+	if (!buf)
+		return;
+
+	pt_read_offset(buf);
+
+	pt_handle_status(pt);
+
+	pt_update_head(pt);
+
+	perf_aux_output_end(&pt->handle, local_xchg(&buf->data_size, 0));
+
+	if (!event->hw.state) {
+		int ret;
+
+		buf = perf_aux_output_begin(&pt->handle, event);
+		if (!buf) {
+			event->hw.state = PERF_HES_STOPPED;
+			return;
+		}
+
+		pt_buffer_reset_offsets(buf, pt->handle.head);
+		/* snapshot counters don't use PMI, so it's safe */
+		ret = pt_buffer_reset_markers(buf, &pt->handle);
+		if (ret) {
+			perf_aux_output_end(&pt->handle, 0);
+			return;
+		}
+
+		pt_config_buffer(buf);
+		pt_config_start(event);
+	}
+}
+
+void intel_pt_handle_vmx(int on)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct perf_event *event;
+	unsigned long flags;
+
+	/* PT plays nice with VMX, do nothing */
+	if (pt_pmu.vmx)
+		return;
+
+	/*
+	 * VMXON will clear RTIT_CTL.TraceEn; we need to make
+	 * sure to not try to set it while VMX is on. Disable
+	 * interrupts to avoid racing with pmu callbacks;
+	 * concurrent PMI should be handled fine.
+	 */
+	local_irq_save(flags);
+	WRITE_ONCE(pt->vmx_on, on);
+
+	/*
+	 * If an AUX transaction is in progress, it will contain
+	 * gap(s), so flag it PARTIAL to inform the user.
+	 */
+	event = pt->handle.event;
+	if (event)
+		perf_aux_output_flag(&pt->handle,
+		                     PERF_AUX_FLAG_PARTIAL);
+
+	/* Turn PTs back on */
+	if (!on && event)
+		wrmsrl(MSR_IA32_RTIT_CTL, event->hw.config);
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(intel_pt_handle_vmx);
+
+/*
+ * PMU callbacks
+ */
+
+static void pt_event_start(struct perf_event *event, int mode)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct pt_buffer *buf;
+
+	buf = perf_aux_output_begin(&pt->handle, event);
+	if (!buf)
+		goto fail_stop;
+
+	pt_buffer_reset_offsets(buf, pt->handle.head);
+	if (!buf->snapshot) {
+		if (pt_buffer_reset_markers(buf, &pt->handle))
+			goto fail_end_stop;
+	}
+
+	WRITE_ONCE(pt->handle_nmi, 1);
+	hwc->state = 0;
+
+	pt_config_buffer(buf);
+	pt_config(event);
+
+	return;
+
+fail_end_stop:
+	perf_aux_output_end(&pt->handle, 0);
+fail_stop:
+	hwc->state = PERF_HES_STOPPED;
+}
+
+static void pt_event_stop(struct perf_event *event, int mode)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+
+	/*
+	 * Protect against the PMI racing with disabling wrmsr,
+	 * see comment in intel_pt_interrupt().
+	 */
+	WRITE_ONCE(pt->handle_nmi, 0);
+
+	pt_config_stop(event);
+
+	if (event->hw.state == PERF_HES_STOPPED)
+		return;
+
+	event->hw.state = PERF_HES_STOPPED;
+
+	if (mode & PERF_EF_UPDATE) {
+		struct pt_buffer *buf = perf_get_aux(&pt->handle);
+
+		if (!buf)
+			return;
+
+		if (WARN_ON_ONCE(pt->handle.event != event))
+			return;
+
+		pt_read_offset(buf);
+
+		pt_handle_status(pt);
+
+		pt_update_head(pt);
+
+		if (buf->snapshot)
+			pt->handle.head =
+				local_xchg(&buf->data_size,
+					   buf->nr_pages << PAGE_SHIFT);
+		perf_aux_output_end(&pt->handle, local_xchg(&buf->data_size, 0));
+	}
+}
+
+static long pt_event_snapshot_aux(struct perf_event *event,
+				  struct perf_output_handle *handle,
+				  unsigned long size)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct pt_buffer *buf = perf_get_aux(&pt->handle);
+	unsigned long from = 0, to;
+	long ret;
+
+	if (WARN_ON_ONCE(!buf))
+		return 0;
+
+	/*
+	 * Sampling is only allowed on snapshot events;
+	 * see pt_buffer_setup_aux().
+	 */
+	if (WARN_ON_ONCE(!buf->snapshot))
+		return 0;
+
+	/*
+	 * Here, handle_nmi tells us if the tracing is on
+	 */
+	if (READ_ONCE(pt->handle_nmi))
+		pt_config_stop(event);
+
+	pt_read_offset(buf);
+	pt_update_head(pt);
+
+	to = local_read(&buf->data_size);
+	if (to < size)
+		from = buf->nr_pages << PAGE_SHIFT;
+	from += to - size;
+
+	ret = perf_output_copy_aux(&pt->handle, handle, from, to);
+
+	/*
+	 * If the tracing was on when we turned up, restart it.
+	 * Compiler barrier not needed as we couldn't have been
+	 * preempted by anything that touches pt->handle_nmi.
+	 */
+	if (pt->handle_nmi)
+		pt_config_start(event);
+
+	return ret;
+}
+
+static void pt_event_del(struct perf_event *event, int mode)
+{
+	pt_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int pt_event_add(struct perf_event *event, int mode)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+	struct hw_perf_event *hwc = &event->hw;
+	int ret = -EBUSY;
+
+	if (pt->handle.event)
+		goto fail;
+
+	if (mode & PERF_EF_START) {
+		pt_event_start(event, 0);
+		ret = -EINVAL;
+		if (hwc->state == PERF_HES_STOPPED)
+			goto fail;
+	} else {
+		hwc->state = PERF_HES_STOPPED;
+	}
+
+	ret = 0;
+fail:
+
+	return ret;
+}
+
+static void pt_event_read(struct perf_event *event)
+{
+}
+
+static void pt_event_destroy(struct perf_event *event)
+{
+	pt_addr_filters_fini(event);
+	x86_del_exclusive(x86_lbr_exclusive_pt);
+}
+
+static int pt_event_init(struct perf_event *event)
+{
+	if (event->attr.type != pt_pmu.pmu.type)
+		return -ENOENT;
+
+	if (!pt_event_valid(event))
+		return -EINVAL;
+
+	if (x86_add_exclusive(x86_lbr_exclusive_pt))
+		return -EBUSY;
+
+	if (pt_addr_filters_init(event)) {
+		x86_del_exclusive(x86_lbr_exclusive_pt);
+		return -ENOMEM;
+	}
+
+	event->destroy = pt_event_destroy;
+
+	return 0;
+}
+
+void cpu_emergency_stop_pt(void)
+{
+	struct pt *pt = this_cpu_ptr(&pt_ctx);
+
+	if (pt->handle.event)
+		pt_event_stop(pt->handle.event, PERF_EF_UPDATE);
+}
+
+int is_intel_pt_event(struct perf_event *event)
+{
+	return event->pmu == &pt_pmu.pmu;
+}
+
+static __init int pt_init(void)
+{
+	int ret, cpu, prior_warn = 0;
+
+	BUILD_BUG_ON(sizeof(struct topa) > PAGE_SIZE);
+
+	if (!boot_cpu_has(X86_FEATURE_INTEL_PT))
+		return -ENODEV;
+
+	get_online_cpus();
+	for_each_online_cpu(cpu) {
+		u64 ctl;
+
+		ret = rdmsrl_safe_on_cpu(cpu, MSR_IA32_RTIT_CTL, &ctl);
+		if (!ret && (ctl & RTIT_CTL_TRACEEN))
+			prior_warn++;
+	}
+	put_online_cpus();
+
+	if (prior_warn) {
+		x86_add_exclusive(x86_lbr_exclusive_pt);
+		pr_warn("PT is enabled at boot time, doing nothing\n");
+
+		return -EBUSY;
+	}
+
+	ret = pt_pmu_hw_init();
+	if (ret)
+		return ret;
+
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_output)) {
+		pr_warn("ToPA output is not supported on this CPU\n");
+		return -ENODEV;
+	}
+
+	if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries))
+		pt_pmu.pmu.capabilities = PERF_PMU_CAP_AUX_NO_SG;
+
+	pt_pmu.pmu.capabilities	|= PERF_PMU_CAP_EXCLUSIVE | PERF_PMU_CAP_ITRACE;
+	pt_pmu.pmu.attr_groups		 = pt_attr_groups;
+	pt_pmu.pmu.task_ctx_nr		 = perf_sw_context;
+	pt_pmu.pmu.event_init		 = pt_event_init;
+	pt_pmu.pmu.add			 = pt_event_add;
+	pt_pmu.pmu.del			 = pt_event_del;
+	pt_pmu.pmu.start		 = pt_event_start;
+	pt_pmu.pmu.stop			 = pt_event_stop;
+	pt_pmu.pmu.snapshot_aux		 = pt_event_snapshot_aux;
+	pt_pmu.pmu.read			 = pt_event_read;
+	pt_pmu.pmu.setup_aux		 = pt_buffer_setup_aux;
+	pt_pmu.pmu.free_aux		 = pt_buffer_free_aux;
+	pt_pmu.pmu.addr_filters_sync     = pt_event_addr_filters_sync;
+	pt_pmu.pmu.addr_filters_validate = pt_event_addr_filters_validate;
+	pt_pmu.pmu.nr_addr_filters       =
+		intel_pt_validate_hw_cap(PT_CAP_num_address_ranges);
+
+	ret = perf_pmu_register(&pt_pmu.pmu, "intel_pt", -1);
+
+	return ret;
+}
+arch_initcall(pt_init);
diff --git a/arch/x86/events/intel/pt.h b/arch/x86/events/intel/pt.h
new file mode 100644
index 000000000..96906a62a
--- /dev/null
+++ b/arch/x86/events/intel/pt.h
@@ -0,0 +1,132 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Intel(R) Processor Trace PMU driver for perf
+ * Copyright (c) 2013-2014, Intel Corporation.
+ *
+ * Intel PT is specified in the Intel Architecture Instruction Set Extensions
+ * Programming Reference:
+ * http://software.intel.com/en-us/intel-isa-extensions
+ */
+
+#ifndef __INTEL_PT_H__
+#define __INTEL_PT_H__
+
+/*
+ * Single-entry ToPA: when this close to region boundary, switch
+ * buffers to avoid losing data.
+ */
+#define TOPA_PMI_MARGIN 512
+
+#define TOPA_SHIFT 12
+
+static inline unsigned int sizes(unsigned int tsz)
+{
+	return 1 << (tsz + TOPA_SHIFT);
+};
+
+struct topa_entry {
+	u64	end	: 1;
+	u64	rsvd0	: 1;
+	u64	intr	: 1;
+	u64	rsvd1	: 1;
+	u64	stop	: 1;
+	u64	rsvd2	: 1;
+	u64	size	: 4;
+	u64	rsvd3	: 2;
+	u64	base	: 36;
+	u64	rsvd4	: 16;
+};
+
+/* TSC to Core Crystal Clock Ratio */
+#define CPUID_TSC_LEAF		0x15
+
+struct pt_pmu {
+	struct pmu		pmu;
+	u32			caps[PT_CPUID_REGS_NUM * PT_CPUID_LEAVES];
+	bool			vmx;
+	bool			branch_en_always_on;
+	unsigned long		max_nonturbo_ratio;
+	unsigned int		tsc_art_num;
+	unsigned int		tsc_art_den;
+};
+
+/**
+ * struct pt_buffer - buffer configuration; one buffer per task_struct or
+ *		cpu, depending on perf event configuration
+ * @tables:	list of ToPA tables in this buffer
+ * @first:	shorthand for first topa table
+ * @last:	shorthand for last topa table
+ * @cur:	current topa table
+ * @nr_pages:	buffer size in pages
+ * @cur_idx:	current output region's index within @cur table
+ * @output_off:	offset within the current output region
+ * @data_size:	running total of the amount of data in this buffer
+ * @lost:	if data was lost/truncated
+ * @head:	logical write offset inside the buffer
+ * @snapshot:	if this is for a snapshot/overwrite counter
+ * @single:	use Single Range Output instead of ToPA
+ * @stop_pos:	STOP topa entry index
+ * @intr_pos:	INT topa entry index
+ * @stop_te:	STOP topa entry pointer
+ * @intr_te:	INT topa entry pointer
+ * @data_pages:	array of pages from perf
+ * @topa_index:	table of topa entries indexed by page offset
+ */
+struct pt_buffer {
+	struct list_head	tables;
+	struct topa		*first, *last, *cur;
+	unsigned int		cur_idx;
+	size_t			output_off;
+	unsigned long		nr_pages;
+	local_t			data_size;
+	local64_t		head;
+	bool			snapshot;
+	bool			single;
+	long			stop_pos, intr_pos;
+	struct topa_entry	*stop_te, *intr_te;
+	void			**data_pages;
+};
+
+#define PT_FILTERS_NUM	4
+
+/**
+ * struct pt_filter - IP range filter configuration
+ * @msr_a:	range start, goes to RTIT_ADDRn_A
+ * @msr_b:	range end, goes to RTIT_ADDRn_B
+ * @config:	4-bit field in RTIT_CTL
+ */
+struct pt_filter {
+	unsigned long	msr_a;
+	unsigned long	msr_b;
+	unsigned long	config;
+};
+
+/**
+ * struct pt_filters - IP range filtering context
+ * @filter:	filters defined for this context
+ * @nr_filters:	number of defined filters in the @filter array
+ */
+struct pt_filters {
+	struct pt_filter	filter[PT_FILTERS_NUM];
+	unsigned int		nr_filters;
+};
+
+/**
+ * struct pt - per-cpu pt context
+ * @handle:		perf output handle
+ * @filters:		last configured filters
+ * @handle_nmi:		do handle PT PMI on this cpu, there's an active event
+ * @vmx_on:		1 if VMX is ON on this cpu
+ * @output_base:	cached RTIT_OUTPUT_BASE MSR value
+ * @output_mask:	cached RTIT_OUTPUT_MASK MSR value
+ */
+struct pt {
+	struct perf_output_handle handle;
+	struct pt_filters	filters;
+	int			handle_nmi;
+	int			vmx_on;
+	u64			output_base;
+	u64			output_mask;
+};
+
+#endif /* __INTEL_PT_H__ */
diff --git a/arch/x86/events/intel/uncore.c b/arch/x86/events/intel/uncore.c
new file mode 100644
index 000000000..80d52cbe2
--- /dev/null
+++ b/arch/x86/events/intel/uncore.c
@@ -0,0 +1,1751 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/module.h>
+
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include "uncore.h"
+
+static struct intel_uncore_type *empty_uncore[] = { NULL, };
+struct intel_uncore_type **uncore_msr_uncores = empty_uncore;
+struct intel_uncore_type **uncore_pci_uncores = empty_uncore;
+struct intel_uncore_type **uncore_mmio_uncores = empty_uncore;
+
+static bool pcidrv_registered;
+struct pci_driver *uncore_pci_driver;
+/* The PCI driver for the device which the uncore doesn't own. */
+struct pci_driver *uncore_pci_sub_driver;
+/* pci bus to socket mapping */
+DEFINE_RAW_SPINLOCK(pci2phy_map_lock);
+struct list_head pci2phy_map_head = LIST_HEAD_INIT(pci2phy_map_head);
+struct pci_extra_dev *uncore_extra_pci_dev;
+int __uncore_max_dies;
+
+/* mask of cpus that collect uncore events */
+static cpumask_t uncore_cpu_mask;
+
+/* constraint for the fixed counter */
+static struct event_constraint uncore_constraint_fixed =
+	EVENT_CONSTRAINT(~0ULL, 1 << UNCORE_PMC_IDX_FIXED, ~0ULL);
+struct event_constraint uncore_constraint_empty =
+	EVENT_CONSTRAINT(0, 0, 0);
+
+MODULE_LICENSE("GPL");
+
+int uncore_pcibus_to_physid(struct pci_bus *bus)
+{
+	struct pci2phy_map *map;
+	int phys_id = -1;
+
+	raw_spin_lock(&pci2phy_map_lock);
+	list_for_each_entry(map, &pci2phy_map_head, list) {
+		if (map->segment == pci_domain_nr(bus)) {
+			phys_id = map->pbus_to_physid[bus->number];
+			break;
+		}
+	}
+	raw_spin_unlock(&pci2phy_map_lock);
+
+	return phys_id;
+}
+
+static void uncore_free_pcibus_map(void)
+{
+	struct pci2phy_map *map, *tmp;
+
+	list_for_each_entry_safe(map, tmp, &pci2phy_map_head, list) {
+		list_del(&map->list);
+		kfree(map);
+	}
+}
+
+struct pci2phy_map *__find_pci2phy_map(int segment)
+{
+	struct pci2phy_map *map, *alloc = NULL;
+	int i;
+
+	lockdep_assert_held(&pci2phy_map_lock);
+
+lookup:
+	list_for_each_entry(map, &pci2phy_map_head, list) {
+		if (map->segment == segment)
+			goto end;
+	}
+
+	if (!alloc) {
+		raw_spin_unlock(&pci2phy_map_lock);
+		alloc = kmalloc(sizeof(struct pci2phy_map), GFP_KERNEL);
+		raw_spin_lock(&pci2phy_map_lock);
+
+		if (!alloc)
+			return NULL;
+
+		goto lookup;
+	}
+
+	map = alloc;
+	alloc = NULL;
+	map->segment = segment;
+	for (i = 0; i < 256; i++)
+		map->pbus_to_physid[i] = -1;
+	list_add_tail(&map->list, &pci2phy_map_head);
+
+end:
+	kfree(alloc);
+	return map;
+}
+
+ssize_t uncore_event_show(struct device *dev,
+			  struct device_attribute *attr, char *buf)
+{
+	struct uncore_event_desc *event =
+		container_of(attr, struct uncore_event_desc, attr);
+	return sprintf(buf, "%s", event->config);
+}
+
+struct intel_uncore_box *uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu)
+{
+	unsigned int dieid = topology_logical_die_id(cpu);
+
+	/*
+	 * The unsigned check also catches the '-1' return value for non
+	 * existent mappings in the topology map.
+	 */
+	return dieid < uncore_max_dies() ? pmu->boxes[dieid] : NULL;
+}
+
+u64 uncore_msr_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	u64 count;
+
+	rdmsrl(event->hw.event_base, count);
+
+	return count;
+}
+
+void uncore_mmio_exit_box(struct intel_uncore_box *box)
+{
+	if (box->io_addr)
+		iounmap(box->io_addr);
+}
+
+u64 uncore_mmio_read_counter(struct intel_uncore_box *box,
+			     struct perf_event *event)
+{
+	if (!box->io_addr)
+		return 0;
+
+	if (!uncore_mmio_is_valid_offset(box, event->hw.event_base))
+		return 0;
+
+	return readq(box->io_addr + event->hw.event_base);
+}
+
+/*
+ * generic get constraint function for shared match/mask registers.
+ */
+struct event_constraint *
+uncore_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct intel_uncore_extra_reg *er;
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct hw_perf_event_extra *reg2 = &event->hw.branch_reg;
+	unsigned long flags;
+	bool ok = false;
+
+	/*
+	 * reg->alloc can be set due to existing state, so for fake box we
+	 * need to ignore this, otherwise we might fail to allocate proper
+	 * fake state for this extra reg constraint.
+	 */
+	if (reg1->idx == EXTRA_REG_NONE ||
+	    (!uncore_box_is_fake(box) && reg1->alloc))
+		return NULL;
+
+	er = &box->shared_regs[reg1->idx];
+	raw_spin_lock_irqsave(&er->lock, flags);
+	if (!atomic_read(&er->ref) ||
+	    (er->config1 == reg1->config && er->config2 == reg2->config)) {
+		atomic_inc(&er->ref);
+		er->config1 = reg1->config;
+		er->config2 = reg2->config;
+		ok = true;
+	}
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+
+	if (ok) {
+		if (!uncore_box_is_fake(box))
+			reg1->alloc = 1;
+		return NULL;
+	}
+
+	return &uncore_constraint_empty;
+}
+
+void uncore_put_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct intel_uncore_extra_reg *er;
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+
+	/*
+	 * Only put constraint if extra reg was actually allocated. Also
+	 * takes care of event which do not use an extra shared reg.
+	 *
+	 * Also, if this is a fake box we shouldn't touch any event state
+	 * (reg->alloc) and we don't care about leaving inconsistent box
+	 * state either since it will be thrown out.
+	 */
+	if (uncore_box_is_fake(box) || !reg1->alloc)
+		return;
+
+	er = &box->shared_regs[reg1->idx];
+	atomic_dec(&er->ref);
+	reg1->alloc = 0;
+}
+
+u64 uncore_shared_reg_config(struct intel_uncore_box *box, int idx)
+{
+	struct intel_uncore_extra_reg *er;
+	unsigned long flags;
+	u64 config;
+
+	er = &box->shared_regs[idx];
+
+	raw_spin_lock_irqsave(&er->lock, flags);
+	config = er->config;
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+
+	return config;
+}
+
+static void uncore_assign_hw_event(struct intel_uncore_box *box,
+				   struct perf_event *event, int idx)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	hwc->idx = idx;
+	hwc->last_tag = ++box->tags[idx];
+
+	if (uncore_pmc_fixed(hwc->idx)) {
+		hwc->event_base = uncore_fixed_ctr(box);
+		hwc->config_base = uncore_fixed_ctl(box);
+		return;
+	}
+
+	hwc->config_base = uncore_event_ctl(box, hwc->idx);
+	hwc->event_base  = uncore_perf_ctr(box, hwc->idx);
+}
+
+void uncore_perf_event_update(struct intel_uncore_box *box, struct perf_event *event)
+{
+	u64 prev_count, new_count, delta;
+	int shift;
+
+	if (uncore_pmc_freerunning(event->hw.idx))
+		shift = 64 - uncore_freerunning_bits(box, event);
+	else if (uncore_pmc_fixed(event->hw.idx))
+		shift = 64 - uncore_fixed_ctr_bits(box);
+	else
+		shift = 64 - uncore_perf_ctr_bits(box);
+
+	/* the hrtimer might modify the previous event value */
+again:
+	prev_count = local64_read(&event->hw.prev_count);
+	new_count = uncore_read_counter(box, event);
+	if (local64_xchg(&event->hw.prev_count, new_count) != prev_count)
+		goto again;
+
+	delta = (new_count << shift) - (prev_count << shift);
+	delta >>= shift;
+
+	local64_add(delta, &event->count);
+}
+
+/*
+ * The overflow interrupt is unavailable for SandyBridge-EP, is broken
+ * for SandyBridge. So we use hrtimer to periodically poll the counter
+ * to avoid overflow.
+ */
+static enum hrtimer_restart uncore_pmu_hrtimer(struct hrtimer *hrtimer)
+{
+	struct intel_uncore_box *box;
+	struct perf_event *event;
+	unsigned long flags;
+	int bit;
+
+	box = container_of(hrtimer, struct intel_uncore_box, hrtimer);
+	if (!box->n_active || box->cpu != smp_processor_id())
+		return HRTIMER_NORESTART;
+	/*
+	 * disable local interrupt to prevent uncore_pmu_event_start/stop
+	 * to interrupt the update process
+	 */
+	local_irq_save(flags);
+
+	/*
+	 * handle boxes with an active event list as opposed to active
+	 * counters
+	 */
+	list_for_each_entry(event, &box->active_list, active_entry) {
+		uncore_perf_event_update(box, event);
+	}
+
+	for_each_set_bit(bit, box->active_mask, UNCORE_PMC_IDX_MAX)
+		uncore_perf_event_update(box, box->events[bit]);
+
+	local_irq_restore(flags);
+
+	hrtimer_forward_now(hrtimer, ns_to_ktime(box->hrtimer_duration));
+	return HRTIMER_RESTART;
+}
+
+void uncore_pmu_start_hrtimer(struct intel_uncore_box *box)
+{
+	hrtimer_start(&box->hrtimer, ns_to_ktime(box->hrtimer_duration),
+		      HRTIMER_MODE_REL_PINNED);
+}
+
+void uncore_pmu_cancel_hrtimer(struct intel_uncore_box *box)
+{
+	hrtimer_cancel(&box->hrtimer);
+}
+
+static void uncore_pmu_init_hrtimer(struct intel_uncore_box *box)
+{
+	hrtimer_init(&box->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	box->hrtimer.function = uncore_pmu_hrtimer;
+}
+
+static struct intel_uncore_box *uncore_alloc_box(struct intel_uncore_type *type,
+						 int node)
+{
+	int i, size, numshared = type->num_shared_regs ;
+	struct intel_uncore_box *box;
+
+	size = sizeof(*box) + numshared * sizeof(struct intel_uncore_extra_reg);
+
+	box = kzalloc_node(size, GFP_KERNEL, node);
+	if (!box)
+		return NULL;
+
+	for (i = 0; i < numshared; i++)
+		raw_spin_lock_init(&box->shared_regs[i].lock);
+
+	uncore_pmu_init_hrtimer(box);
+	box->cpu = -1;
+	box->pci_phys_id = -1;
+	box->dieid = -1;
+
+	/* set default hrtimer timeout */
+	box->hrtimer_duration = UNCORE_PMU_HRTIMER_INTERVAL;
+
+	INIT_LIST_HEAD(&box->active_list);
+
+	return box;
+}
+
+/*
+ * Using uncore_pmu_event_init pmu event_init callback
+ * as a detection point for uncore events.
+ */
+static int uncore_pmu_event_init(struct perf_event *event);
+
+static bool is_box_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return &box->pmu->pmu == event->pmu;
+}
+
+static int
+uncore_collect_events(struct intel_uncore_box *box, struct perf_event *leader,
+		      bool dogrp)
+{
+	struct perf_event *event;
+	int n, max_count;
+
+	max_count = box->pmu->type->num_counters;
+	if (box->pmu->type->fixed_ctl)
+		max_count++;
+
+	if (box->n_events >= max_count)
+		return -EINVAL;
+
+	n = box->n_events;
+
+	if (is_box_event(box, leader)) {
+		box->event_list[n] = leader;
+		n++;
+	}
+
+	if (!dogrp)
+		return n;
+
+	for_each_sibling_event(event, leader) {
+		if (!is_box_event(box, event) ||
+		    event->state <= PERF_EVENT_STATE_OFF)
+			continue;
+
+		if (n >= max_count)
+			return -EINVAL;
+
+		box->event_list[n] = event;
+		n++;
+	}
+	return n;
+}
+
+static struct event_constraint *
+uncore_get_event_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct intel_uncore_type *type = box->pmu->type;
+	struct event_constraint *c;
+
+	if (type->ops->get_constraint) {
+		c = type->ops->get_constraint(box, event);
+		if (c)
+			return c;
+	}
+
+	if (event->attr.config == UNCORE_FIXED_EVENT)
+		return &uncore_constraint_fixed;
+
+	if (type->constraints) {
+		for_each_event_constraint(c, type->constraints) {
+			if ((event->hw.config & c->cmask) == c->code)
+				return c;
+		}
+	}
+
+	return &type->unconstrainted;
+}
+
+static void uncore_put_event_constraint(struct intel_uncore_box *box,
+					struct perf_event *event)
+{
+	if (box->pmu->type->ops->put_constraint)
+		box->pmu->type->ops->put_constraint(box, event);
+}
+
+static int uncore_assign_events(struct intel_uncore_box *box, int assign[], int n)
+{
+	unsigned long used_mask[BITS_TO_LONGS(UNCORE_PMC_IDX_MAX)];
+	struct event_constraint *c;
+	int i, wmin, wmax, ret = 0;
+	struct hw_perf_event *hwc;
+
+	bitmap_zero(used_mask, UNCORE_PMC_IDX_MAX);
+
+	for (i = 0, wmin = UNCORE_PMC_IDX_MAX, wmax = 0; i < n; i++) {
+		c = uncore_get_event_constraint(box, box->event_list[i]);
+		box->event_constraint[i] = c;
+		wmin = min(wmin, c->weight);
+		wmax = max(wmax, c->weight);
+	}
+
+	/* fastpath, try to reuse previous register */
+	for (i = 0; i < n; i++) {
+		hwc = &box->event_list[i]->hw;
+		c = box->event_constraint[i];
+
+		/* never assigned */
+		if (hwc->idx == -1)
+			break;
+
+		/* constraint still honored */
+		if (!test_bit(hwc->idx, c->idxmsk))
+			break;
+
+		/* not already used */
+		if (test_bit(hwc->idx, used_mask))
+			break;
+
+		__set_bit(hwc->idx, used_mask);
+		if (assign)
+			assign[i] = hwc->idx;
+	}
+	/* slow path */
+	if (i != n)
+		ret = perf_assign_events(box->event_constraint, n,
+					 wmin, wmax, n, assign);
+
+	if (!assign || ret) {
+		for (i = 0; i < n; i++)
+			uncore_put_event_constraint(box, box->event_list[i]);
+	}
+	return ret ? -EINVAL : 0;
+}
+
+void uncore_pmu_event_start(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	int idx = event->hw.idx;
+
+	if (WARN_ON_ONCE(idx == -1 || idx >= UNCORE_PMC_IDX_MAX))
+		return;
+
+	/*
+	 * Free running counter is read-only and always active.
+	 * Use the current counter value as start point.
+	 * There is no overflow interrupt for free running counter.
+	 * Use hrtimer to periodically poll the counter to avoid overflow.
+	 */
+	if (uncore_pmc_freerunning(event->hw.idx)) {
+		list_add_tail(&event->active_entry, &box->active_list);
+		local64_set(&event->hw.prev_count,
+			    uncore_read_counter(box, event));
+		if (box->n_active++ == 0)
+			uncore_pmu_start_hrtimer(box);
+		return;
+	}
+
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	event->hw.state = 0;
+	box->events[idx] = event;
+	box->n_active++;
+	__set_bit(idx, box->active_mask);
+
+	local64_set(&event->hw.prev_count, uncore_read_counter(box, event));
+	uncore_enable_event(box, event);
+
+	if (box->n_active == 1)
+		uncore_pmu_start_hrtimer(box);
+}
+
+void uncore_pmu_event_stop(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	struct hw_perf_event *hwc = &event->hw;
+
+	/* Cannot disable free running counter which is read-only */
+	if (uncore_pmc_freerunning(hwc->idx)) {
+		list_del(&event->active_entry);
+		if (--box->n_active == 0)
+			uncore_pmu_cancel_hrtimer(box);
+		uncore_perf_event_update(box, event);
+		return;
+	}
+
+	if (__test_and_clear_bit(hwc->idx, box->active_mask)) {
+		uncore_disable_event(box, event);
+		box->n_active--;
+		box->events[hwc->idx] = NULL;
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+
+		if (box->n_active == 0)
+			uncore_pmu_cancel_hrtimer(box);
+	}
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		uncore_perf_event_update(box, event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+}
+
+int uncore_pmu_event_add(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	struct hw_perf_event *hwc = &event->hw;
+	int assign[UNCORE_PMC_IDX_MAX];
+	int i, n, ret;
+
+	if (!box)
+		return -ENODEV;
+
+	/*
+	 * The free funning counter is assigned in event_init().
+	 * The free running counter event and free running counter
+	 * are 1:1 mapped. It doesn't need to be tracked in event_list.
+	 */
+	if (uncore_pmc_freerunning(hwc->idx)) {
+		if (flags & PERF_EF_START)
+			uncore_pmu_event_start(event, 0);
+		return 0;
+	}
+
+	ret = n = uncore_collect_events(box, event, false);
+	if (ret < 0)
+		return ret;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
+
+	ret = uncore_assign_events(box, assign, n);
+	if (ret)
+		return ret;
+
+	/* save events moving to new counters */
+	for (i = 0; i < box->n_events; i++) {
+		event = box->event_list[i];
+		hwc = &event->hw;
+
+		if (hwc->idx == assign[i] &&
+			hwc->last_tag == box->tags[assign[i]])
+			continue;
+		/*
+		 * Ensure we don't accidentally enable a stopped
+		 * counter simply because we rescheduled.
+		 */
+		if (hwc->state & PERF_HES_STOPPED)
+			hwc->state |= PERF_HES_ARCH;
+
+		uncore_pmu_event_stop(event, PERF_EF_UPDATE);
+	}
+
+	/* reprogram moved events into new counters */
+	for (i = 0; i < n; i++) {
+		event = box->event_list[i];
+		hwc = &event->hw;
+
+		if (hwc->idx != assign[i] ||
+			hwc->last_tag != box->tags[assign[i]])
+			uncore_assign_hw_event(box, event, assign[i]);
+		else if (i < box->n_events)
+			continue;
+
+		if (hwc->state & PERF_HES_ARCH)
+			continue;
+
+		uncore_pmu_event_start(event, 0);
+	}
+	box->n_events = n;
+
+	return 0;
+}
+
+void uncore_pmu_event_del(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	int i;
+
+	uncore_pmu_event_stop(event, PERF_EF_UPDATE);
+
+	/*
+	 * The event for free running counter is not tracked by event_list.
+	 * It doesn't need to force event->hw.idx = -1 to reassign the counter.
+	 * Because the event and the free running counter are 1:1 mapped.
+	 */
+	if (uncore_pmc_freerunning(event->hw.idx))
+		return;
+
+	for (i = 0; i < box->n_events; i++) {
+		if (event == box->event_list[i]) {
+			uncore_put_event_constraint(box, event);
+
+			for (++i; i < box->n_events; i++)
+				box->event_list[i - 1] = box->event_list[i];
+
+			--box->n_events;
+			break;
+		}
+	}
+
+	event->hw.idx = -1;
+	event->hw.last_tag = ~0ULL;
+}
+
+void uncore_pmu_event_read(struct perf_event *event)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	uncore_perf_event_update(box, event);
+}
+
+/*
+ * validation ensures the group can be loaded onto the
+ * PMU if it was the only group available.
+ */
+static int uncore_validate_group(struct intel_uncore_pmu *pmu,
+				struct perf_event *event)
+{
+	struct perf_event *leader = event->group_leader;
+	struct intel_uncore_box *fake_box;
+	int ret = -EINVAL, n;
+
+	/* The free running counter is always active. */
+	if (uncore_pmc_freerunning(event->hw.idx))
+		return 0;
+
+	fake_box = uncore_alloc_box(pmu->type, NUMA_NO_NODE);
+	if (!fake_box)
+		return -ENOMEM;
+
+	fake_box->pmu = pmu;
+	/*
+	 * the event is not yet connected with its
+	 * siblings therefore we must first collect
+	 * existing siblings, then add the new event
+	 * before we can simulate the scheduling
+	 */
+	n = uncore_collect_events(fake_box, leader, true);
+	if (n < 0)
+		goto out;
+
+	fake_box->n_events = n;
+	n = uncore_collect_events(fake_box, event, false);
+	if (n < 0)
+		goto out;
+
+	fake_box->n_events = n;
+
+	ret = uncore_assign_events(fake_box, NULL, n);
+out:
+	kfree(fake_box);
+	return ret;
+}
+
+static int uncore_pmu_event_init(struct perf_event *event)
+{
+	struct intel_uncore_pmu *pmu;
+	struct intel_uncore_box *box;
+	struct hw_perf_event *hwc = &event->hw;
+	int ret;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	pmu = uncore_event_to_pmu(event);
+	/* no device found for this pmu */
+	if (pmu->func_id < 0)
+		return -ENOENT;
+
+	/* Sampling not supported yet */
+	if (hwc->sample_period)
+		return -EINVAL;
+
+	/*
+	 * Place all uncore events for a particular physical package
+	 * onto a single cpu
+	 */
+	if (event->cpu < 0)
+		return -EINVAL;
+	box = uncore_pmu_to_box(pmu, event->cpu);
+	if (!box || box->cpu < 0)
+		return -EINVAL;
+	event->cpu = box->cpu;
+	event->pmu_private = box;
+
+	event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG;
+
+	event->hw.idx = -1;
+	event->hw.last_tag = ~0ULL;
+	event->hw.extra_reg.idx = EXTRA_REG_NONE;
+	event->hw.branch_reg.idx = EXTRA_REG_NONE;
+
+	if (event->attr.config == UNCORE_FIXED_EVENT) {
+		/* no fixed counter */
+		if (!pmu->type->fixed_ctl)
+			return -EINVAL;
+		/*
+		 * if there is only one fixed counter, only the first pmu
+		 * can access the fixed counter
+		 */
+		if (pmu->type->single_fixed && pmu->pmu_idx > 0)
+			return -EINVAL;
+
+		/* fixed counters have event field hardcoded to zero */
+		hwc->config = 0ULL;
+	} else if (is_freerunning_event(event)) {
+		hwc->config = event->attr.config;
+		if (!check_valid_freerunning_event(box, event))
+			return -EINVAL;
+		event->hw.idx = UNCORE_PMC_IDX_FREERUNNING;
+		/*
+		 * The free running counter event and free running counter
+		 * are always 1:1 mapped.
+		 * The free running counter is always active.
+		 * Assign the free running counter here.
+		 */
+		event->hw.event_base = uncore_freerunning_counter(box, event);
+	} else {
+		hwc->config = event->attr.config &
+			      (pmu->type->event_mask | ((u64)pmu->type->event_mask_ext << 32));
+		if (pmu->type->ops->hw_config) {
+			ret = pmu->type->ops->hw_config(box, event);
+			if (ret)
+				return ret;
+		}
+	}
+
+	if (event->group_leader != event)
+		ret = uncore_validate_group(pmu, event);
+	else
+		ret = 0;
+
+	return ret;
+}
+
+static void uncore_pmu_enable(struct pmu *pmu)
+{
+	struct intel_uncore_pmu *uncore_pmu;
+	struct intel_uncore_box *box;
+
+	uncore_pmu = container_of(pmu, struct intel_uncore_pmu, pmu);
+	if (!uncore_pmu)
+		return;
+
+	box = uncore_pmu_to_box(uncore_pmu, smp_processor_id());
+	if (!box)
+		return;
+
+	if (uncore_pmu->type->ops->enable_box)
+		uncore_pmu->type->ops->enable_box(box);
+}
+
+static void uncore_pmu_disable(struct pmu *pmu)
+{
+	struct intel_uncore_pmu *uncore_pmu;
+	struct intel_uncore_box *box;
+
+	uncore_pmu = container_of(pmu, struct intel_uncore_pmu, pmu);
+	if (!uncore_pmu)
+		return;
+
+	box = uncore_pmu_to_box(uncore_pmu, smp_processor_id());
+	if (!box)
+		return;
+
+	if (uncore_pmu->type->ops->disable_box)
+		uncore_pmu->type->ops->disable_box(box);
+}
+
+static ssize_t uncore_get_attr_cpumask(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	return cpumap_print_to_pagebuf(true, buf, &uncore_cpu_mask);
+}
+
+static DEVICE_ATTR(cpumask, S_IRUGO, uncore_get_attr_cpumask, NULL);
+
+static struct attribute *uncore_pmu_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static const struct attribute_group uncore_pmu_attr_group = {
+	.attrs = uncore_pmu_attrs,
+};
+
+static int uncore_pmu_register(struct intel_uncore_pmu *pmu)
+{
+	int ret;
+
+	if (!pmu->type->pmu) {
+		pmu->pmu = (struct pmu) {
+			.attr_groups	= pmu->type->attr_groups,
+			.task_ctx_nr	= perf_invalid_context,
+			.pmu_enable	= uncore_pmu_enable,
+			.pmu_disable	= uncore_pmu_disable,
+			.event_init	= uncore_pmu_event_init,
+			.add		= uncore_pmu_event_add,
+			.del		= uncore_pmu_event_del,
+			.start		= uncore_pmu_event_start,
+			.stop		= uncore_pmu_event_stop,
+			.read		= uncore_pmu_event_read,
+			.module		= THIS_MODULE,
+			.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+			.attr_update	= pmu->type->attr_update,
+		};
+	} else {
+		pmu->pmu = *pmu->type->pmu;
+		pmu->pmu.attr_groups = pmu->type->attr_groups;
+		pmu->pmu.attr_update = pmu->type->attr_update;
+	}
+
+	if (pmu->type->num_boxes == 1) {
+		if (strlen(pmu->type->name) > 0)
+			sprintf(pmu->name, "uncore_%s", pmu->type->name);
+		else
+			sprintf(pmu->name, "uncore");
+	} else {
+		sprintf(pmu->name, "uncore_%s_%d", pmu->type->name,
+			pmu->pmu_idx);
+	}
+
+	ret = perf_pmu_register(&pmu->pmu, pmu->name, -1);
+	if (!ret)
+		pmu->registered = true;
+	return ret;
+}
+
+static void uncore_pmu_unregister(struct intel_uncore_pmu *pmu)
+{
+	if (!pmu->registered)
+		return;
+	perf_pmu_unregister(&pmu->pmu);
+	pmu->registered = false;
+}
+
+static void uncore_free_boxes(struct intel_uncore_pmu *pmu)
+{
+	int die;
+
+	for (die = 0; die < uncore_max_dies(); die++)
+		kfree(pmu->boxes[die]);
+	kfree(pmu->boxes);
+}
+
+static void uncore_type_exit(struct intel_uncore_type *type)
+{
+	struct intel_uncore_pmu *pmu = type->pmus;
+	int i;
+
+	if (type->cleanup_mapping)
+		type->cleanup_mapping(type);
+
+	if (pmu) {
+		for (i = 0; i < type->num_boxes; i++, pmu++) {
+			uncore_pmu_unregister(pmu);
+			uncore_free_boxes(pmu);
+		}
+		kfree(type->pmus);
+		type->pmus = NULL;
+	}
+	kfree(type->events_group);
+	type->events_group = NULL;
+}
+
+static void uncore_types_exit(struct intel_uncore_type **types)
+{
+	for (; *types; types++)
+		uncore_type_exit(*types);
+}
+
+static int __init uncore_type_init(struct intel_uncore_type *type, bool setid)
+{
+	struct intel_uncore_pmu *pmus;
+	size_t size;
+	int i, j;
+
+	pmus = kcalloc(type->num_boxes, sizeof(*pmus), GFP_KERNEL);
+	if (!pmus)
+		return -ENOMEM;
+
+	size = uncore_max_dies() * sizeof(struct intel_uncore_box *);
+
+	for (i = 0; i < type->num_boxes; i++) {
+		pmus[i].func_id	= setid ? i : -1;
+		pmus[i].pmu_idx	= i;
+		pmus[i].type	= type;
+		pmus[i].boxes	= kzalloc(size, GFP_KERNEL);
+		if (!pmus[i].boxes)
+			goto err;
+	}
+
+	type->pmus = pmus;
+	type->unconstrainted = (struct event_constraint)
+		__EVENT_CONSTRAINT(0, (1ULL << type->num_counters) - 1,
+				0, type->num_counters, 0, 0);
+
+	if (type->event_descs) {
+		struct {
+			struct attribute_group group;
+			struct attribute *attrs[];
+		} *attr_group;
+		for (i = 0; type->event_descs[i].attr.attr.name; i++);
+
+		attr_group = kzalloc(struct_size(attr_group, attrs, i + 1),
+								GFP_KERNEL);
+		if (!attr_group)
+			goto err;
+
+		attr_group->group.name = "events";
+		attr_group->group.attrs = attr_group->attrs;
+
+		for (j = 0; j < i; j++)
+			attr_group->attrs[j] = &type->event_descs[j].attr.attr;
+
+		type->events_group = &attr_group->group;
+	}
+
+	type->pmu_group = &uncore_pmu_attr_group;
+
+	if (type->set_mapping)
+		type->set_mapping(type);
+
+	return 0;
+
+err:
+	for (i = 0; i < type->num_boxes; i++)
+		kfree(pmus[i].boxes);
+	kfree(pmus);
+
+	return -ENOMEM;
+}
+
+static int __init
+uncore_types_init(struct intel_uncore_type **types, bool setid)
+{
+	int ret;
+
+	for (; *types; types++) {
+		ret = uncore_type_init(*types, setid);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+/*
+ * Get the die information of a PCI device.
+ * @pdev: The PCI device.
+ * @phys_id: The physical socket id which the device maps to.
+ * @die: The die id which the device maps to.
+ */
+static int uncore_pci_get_dev_die_info(struct pci_dev *pdev,
+				       int *phys_id, int *die)
+{
+	*phys_id = uncore_pcibus_to_physid(pdev->bus);
+	if (*phys_id < 0)
+		return -ENODEV;
+
+	*die = (topology_max_die_per_package() > 1) ? *phys_id :
+				topology_phys_to_logical_pkg(*phys_id);
+	if (*die < 0)
+		return -EINVAL;
+
+	return 0;
+}
+
+/*
+ * Find the PMU of a PCI device.
+ * @pdev: The PCI device.
+ * @ids: The ID table of the available PCI devices with a PMU.
+ */
+static struct intel_uncore_pmu *
+uncore_pci_find_dev_pmu(struct pci_dev *pdev, const struct pci_device_id *ids)
+{
+	struct intel_uncore_pmu *pmu = NULL;
+	struct intel_uncore_type *type;
+	kernel_ulong_t data;
+	unsigned int devfn;
+
+	while (ids && ids->vendor) {
+		if ((ids->vendor == pdev->vendor) &&
+		    (ids->device == pdev->device)) {
+			data = ids->driver_data;
+			devfn = PCI_DEVFN(UNCORE_PCI_DEV_DEV(data),
+					  UNCORE_PCI_DEV_FUNC(data));
+			if (devfn == pdev->devfn) {
+				type = uncore_pci_uncores[UNCORE_PCI_DEV_TYPE(data)];
+				pmu = &type->pmus[UNCORE_PCI_DEV_IDX(data)];
+				break;
+			}
+		}
+		ids++;
+	}
+	return pmu;
+}
+
+/*
+ * Register the PMU for a PCI device
+ * @pdev: The PCI device.
+ * @type: The corresponding PMU type of the device.
+ * @pmu: The corresponding PMU of the device.
+ * @phys_id: The physical socket id which the device maps to.
+ * @die: The die id which the device maps to.
+ */
+static int uncore_pci_pmu_register(struct pci_dev *pdev,
+				   struct intel_uncore_type *type,
+				   struct intel_uncore_pmu *pmu,
+				   int phys_id, int die)
+{
+	struct intel_uncore_box *box;
+	int ret;
+
+	if (WARN_ON_ONCE(pmu->boxes[die] != NULL))
+		return -EINVAL;
+
+	box = uncore_alloc_box(type, NUMA_NO_NODE);
+	if (!box)
+		return -ENOMEM;
+
+	if (pmu->func_id < 0)
+		pmu->func_id = pdev->devfn;
+	else
+		WARN_ON_ONCE(pmu->func_id != pdev->devfn);
+
+	atomic_inc(&box->refcnt);
+	box->pci_phys_id = phys_id;
+	box->dieid = die;
+	box->pci_dev = pdev;
+	box->pmu = pmu;
+	uncore_box_init(box);
+
+	pmu->boxes[die] = box;
+	if (atomic_inc_return(&pmu->activeboxes) > 1)
+		return 0;
+
+	/* First active box registers the pmu */
+	ret = uncore_pmu_register(pmu);
+	if (ret) {
+		pmu->boxes[die] = NULL;
+		uncore_box_exit(box);
+		kfree(box);
+	}
+	return ret;
+}
+
+/*
+ * add a pci uncore device
+ */
+static int uncore_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+	struct intel_uncore_type *type;
+	struct intel_uncore_pmu *pmu = NULL;
+	int phys_id, die, ret;
+
+	ret = uncore_pci_get_dev_die_info(pdev, &phys_id, &die);
+	if (ret)
+		return ret;
+
+	if (UNCORE_PCI_DEV_TYPE(id->driver_data) == UNCORE_EXTRA_PCI_DEV) {
+		int idx = UNCORE_PCI_DEV_IDX(id->driver_data);
+
+		uncore_extra_pci_dev[die].dev[idx] = pdev;
+		pci_set_drvdata(pdev, NULL);
+		return 0;
+	}
+
+	type = uncore_pci_uncores[UNCORE_PCI_DEV_TYPE(id->driver_data)];
+
+	/*
+	 * Some platforms, e.g.  Knights Landing, use a common PCI device ID
+	 * for multiple instances of an uncore PMU device type. We should check
+	 * PCI slot and func to indicate the uncore box.
+	 */
+	if (id->driver_data & ~0xffff) {
+		struct pci_driver *pci_drv = pdev->driver;
+
+		pmu = uncore_pci_find_dev_pmu(pdev, pci_drv->id_table);
+		if (pmu == NULL)
+			return -ENODEV;
+	} else {
+		/*
+		 * for performance monitoring unit with multiple boxes,
+		 * each box has a different function id.
+		 */
+		pmu = &type->pmus[UNCORE_PCI_DEV_IDX(id->driver_data)];
+	}
+
+	ret = uncore_pci_pmu_register(pdev, type, pmu, phys_id, die);
+
+	pci_set_drvdata(pdev, pmu->boxes[die]);
+
+	return ret;
+}
+
+/*
+ * Unregister the PMU of a PCI device
+ * @pmu: The corresponding PMU is unregistered.
+ * @phys_id: The physical socket id which the device maps to.
+ * @die: The die id which the device maps to.
+ */
+static void uncore_pci_pmu_unregister(struct intel_uncore_pmu *pmu,
+				      int phys_id, int die)
+{
+	struct intel_uncore_box *box = pmu->boxes[die];
+
+	if (WARN_ON_ONCE(phys_id != box->pci_phys_id))
+		return;
+
+	pmu->boxes[die] = NULL;
+	if (atomic_dec_return(&pmu->activeboxes) == 0)
+		uncore_pmu_unregister(pmu);
+	uncore_box_exit(box);
+	kfree(box);
+}
+
+static void uncore_pci_remove(struct pci_dev *pdev)
+{
+	struct intel_uncore_box *box;
+	struct intel_uncore_pmu *pmu;
+	int i, phys_id, die;
+
+	if (uncore_pci_get_dev_die_info(pdev, &phys_id, &die))
+		return;
+
+	box = pci_get_drvdata(pdev);
+	if (!box) {
+		for (i = 0; i < UNCORE_EXTRA_PCI_DEV_MAX; i++) {
+			if (uncore_extra_pci_dev[die].dev[i] == pdev) {
+				uncore_extra_pci_dev[die].dev[i] = NULL;
+				break;
+			}
+		}
+		WARN_ON_ONCE(i >= UNCORE_EXTRA_PCI_DEV_MAX);
+		return;
+	}
+
+	pmu = box->pmu;
+
+	pci_set_drvdata(pdev, NULL);
+
+	uncore_pci_pmu_unregister(pmu, phys_id, die);
+}
+
+static int uncore_bus_notify(struct notifier_block *nb,
+			     unsigned long action, void *data)
+{
+	struct device *dev = data;
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct intel_uncore_pmu *pmu;
+	int phys_id, die;
+
+	/* Unregister the PMU when the device is going to be deleted. */
+	if (action != BUS_NOTIFY_DEL_DEVICE)
+		return NOTIFY_DONE;
+
+	pmu = uncore_pci_find_dev_pmu(pdev, uncore_pci_sub_driver->id_table);
+	if (!pmu)
+		return NOTIFY_DONE;
+
+	if (uncore_pci_get_dev_die_info(pdev, &phys_id, &die))
+		return NOTIFY_DONE;
+
+	uncore_pci_pmu_unregister(pmu, phys_id, die);
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block uncore_notifier = {
+	.notifier_call = uncore_bus_notify,
+};
+
+static void uncore_pci_sub_driver_init(void)
+{
+	const struct pci_device_id *ids = uncore_pci_sub_driver->id_table;
+	struct intel_uncore_type *type;
+	struct intel_uncore_pmu *pmu;
+	struct pci_dev *pci_sub_dev;
+	bool notify = false;
+	unsigned int devfn;
+	int phys_id, die;
+
+	while (ids && ids->vendor) {
+		pci_sub_dev = NULL;
+		type = uncore_pci_uncores[UNCORE_PCI_DEV_TYPE(ids->driver_data)];
+		/*
+		 * Search the available device, and register the
+		 * corresponding PMU.
+		 */
+		while ((pci_sub_dev = pci_get_device(PCI_VENDOR_ID_INTEL,
+						     ids->device, pci_sub_dev))) {
+			devfn = PCI_DEVFN(UNCORE_PCI_DEV_DEV(ids->driver_data),
+					  UNCORE_PCI_DEV_FUNC(ids->driver_data));
+			if (devfn != pci_sub_dev->devfn)
+				continue;
+
+			pmu = &type->pmus[UNCORE_PCI_DEV_IDX(ids->driver_data)];
+			if (!pmu)
+				continue;
+
+			if (uncore_pci_get_dev_die_info(pci_sub_dev,
+							&phys_id, &die))
+				continue;
+
+			if (!uncore_pci_pmu_register(pci_sub_dev, type, pmu,
+						     phys_id, die))
+				notify = true;
+		}
+		ids++;
+	}
+
+	if (notify && bus_register_notifier(&pci_bus_type, &uncore_notifier))
+		notify = false;
+
+	if (!notify)
+		uncore_pci_sub_driver = NULL;
+}
+
+static int __init uncore_pci_init(void)
+{
+	size_t size;
+	int ret;
+
+	size = uncore_max_dies() * sizeof(struct pci_extra_dev);
+	uncore_extra_pci_dev = kzalloc(size, GFP_KERNEL);
+	if (!uncore_extra_pci_dev) {
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	ret = uncore_types_init(uncore_pci_uncores, false);
+	if (ret)
+		goto errtype;
+
+	uncore_pci_driver->probe = uncore_pci_probe;
+	uncore_pci_driver->remove = uncore_pci_remove;
+
+	ret = pci_register_driver(uncore_pci_driver);
+	if (ret)
+		goto errtype;
+
+	if (uncore_pci_sub_driver)
+		uncore_pci_sub_driver_init();
+
+	pcidrv_registered = true;
+	return 0;
+
+errtype:
+	uncore_types_exit(uncore_pci_uncores);
+	kfree(uncore_extra_pci_dev);
+	uncore_extra_pci_dev = NULL;
+	uncore_free_pcibus_map();
+err:
+	uncore_pci_uncores = empty_uncore;
+	return ret;
+}
+
+static void uncore_pci_exit(void)
+{
+	if (pcidrv_registered) {
+		pcidrv_registered = false;
+		if (uncore_pci_sub_driver)
+			bus_unregister_notifier(&pci_bus_type, &uncore_notifier);
+		pci_unregister_driver(uncore_pci_driver);
+		uncore_types_exit(uncore_pci_uncores);
+		kfree(uncore_extra_pci_dev);
+		uncore_free_pcibus_map();
+	}
+}
+
+static void uncore_change_type_ctx(struct intel_uncore_type *type, int old_cpu,
+				   int new_cpu)
+{
+	struct intel_uncore_pmu *pmu = type->pmus;
+	struct intel_uncore_box *box;
+	int i, die;
+
+	die = topology_logical_die_id(old_cpu < 0 ? new_cpu : old_cpu);
+	for (i = 0; i < type->num_boxes; i++, pmu++) {
+		box = pmu->boxes[die];
+		if (!box)
+			continue;
+
+		if (old_cpu < 0) {
+			WARN_ON_ONCE(box->cpu != -1);
+			box->cpu = new_cpu;
+			continue;
+		}
+
+		WARN_ON_ONCE(box->cpu != old_cpu);
+		box->cpu = -1;
+		if (new_cpu < 0)
+			continue;
+
+		uncore_pmu_cancel_hrtimer(box);
+		perf_pmu_migrate_context(&pmu->pmu, old_cpu, new_cpu);
+		box->cpu = new_cpu;
+	}
+}
+
+static void uncore_change_context(struct intel_uncore_type **uncores,
+				  int old_cpu, int new_cpu)
+{
+	for (; *uncores; uncores++)
+		uncore_change_type_ctx(*uncores, old_cpu, new_cpu);
+}
+
+static void uncore_box_unref(struct intel_uncore_type **types, int id)
+{
+	struct intel_uncore_type *type;
+	struct intel_uncore_pmu *pmu;
+	struct intel_uncore_box *box;
+	int i;
+
+	for (; *types; types++) {
+		type = *types;
+		pmu = type->pmus;
+		for (i = 0; i < type->num_boxes; i++, pmu++) {
+			box = pmu->boxes[id];
+			if (box && atomic_dec_return(&box->refcnt) == 0)
+				uncore_box_exit(box);
+		}
+	}
+}
+
+static int uncore_event_cpu_offline(unsigned int cpu)
+{
+	int die, target;
+
+	/* Check if exiting cpu is used for collecting uncore events */
+	if (!cpumask_test_and_clear_cpu(cpu, &uncore_cpu_mask))
+		goto unref;
+	/* Find a new cpu to collect uncore events */
+	target = cpumask_any_but(topology_die_cpumask(cpu), cpu);
+
+	/* Migrate uncore events to the new target */
+	if (target < nr_cpu_ids)
+		cpumask_set_cpu(target, &uncore_cpu_mask);
+	else
+		target = -1;
+
+	uncore_change_context(uncore_msr_uncores, cpu, target);
+	uncore_change_context(uncore_mmio_uncores, cpu, target);
+	uncore_change_context(uncore_pci_uncores, cpu, target);
+
+unref:
+	/* Clear the references */
+	die = topology_logical_die_id(cpu);
+	uncore_box_unref(uncore_msr_uncores, die);
+	uncore_box_unref(uncore_mmio_uncores, die);
+	return 0;
+}
+
+static int allocate_boxes(struct intel_uncore_type **types,
+			 unsigned int die, unsigned int cpu)
+{
+	struct intel_uncore_box *box, *tmp;
+	struct intel_uncore_type *type;
+	struct intel_uncore_pmu *pmu;
+	LIST_HEAD(allocated);
+	int i;
+
+	/* Try to allocate all required boxes */
+	for (; *types; types++) {
+		type = *types;
+		pmu = type->pmus;
+		for (i = 0; i < type->num_boxes; i++, pmu++) {
+			if (pmu->boxes[die])
+				continue;
+			box = uncore_alloc_box(type, cpu_to_node(cpu));
+			if (!box)
+				goto cleanup;
+			box->pmu = pmu;
+			box->dieid = die;
+			list_add(&box->active_list, &allocated);
+		}
+	}
+	/* Install them in the pmus */
+	list_for_each_entry_safe(box, tmp, &allocated, active_list) {
+		list_del_init(&box->active_list);
+		box->pmu->boxes[die] = box;
+	}
+	return 0;
+
+cleanup:
+	list_for_each_entry_safe(box, tmp, &allocated, active_list) {
+		list_del_init(&box->active_list);
+		kfree(box);
+	}
+	return -ENOMEM;
+}
+
+static int uncore_box_ref(struct intel_uncore_type **types,
+			  int id, unsigned int cpu)
+{
+	struct intel_uncore_type *type;
+	struct intel_uncore_pmu *pmu;
+	struct intel_uncore_box *box;
+	int i, ret;
+
+	ret = allocate_boxes(types, id, cpu);
+	if (ret)
+		return ret;
+
+	for (; *types; types++) {
+		type = *types;
+		pmu = type->pmus;
+		for (i = 0; i < type->num_boxes; i++, pmu++) {
+			box = pmu->boxes[id];
+			if (box && atomic_inc_return(&box->refcnt) == 1)
+				uncore_box_init(box);
+		}
+	}
+	return 0;
+}
+
+static int uncore_event_cpu_online(unsigned int cpu)
+{
+	int die, target, msr_ret, mmio_ret;
+
+	die = topology_logical_die_id(cpu);
+	msr_ret = uncore_box_ref(uncore_msr_uncores, die, cpu);
+	mmio_ret = uncore_box_ref(uncore_mmio_uncores, die, cpu);
+	if (msr_ret && mmio_ret)
+		return -ENOMEM;
+
+	/*
+	 * Check if there is an online cpu in the package
+	 * which collects uncore events already.
+	 */
+	target = cpumask_any_and(&uncore_cpu_mask, topology_die_cpumask(cpu));
+	if (target < nr_cpu_ids)
+		return 0;
+
+	cpumask_set_cpu(cpu, &uncore_cpu_mask);
+
+	if (!msr_ret)
+		uncore_change_context(uncore_msr_uncores, -1, cpu);
+	if (!mmio_ret)
+		uncore_change_context(uncore_mmio_uncores, -1, cpu);
+	uncore_change_context(uncore_pci_uncores, -1, cpu);
+	return 0;
+}
+
+static int __init type_pmu_register(struct intel_uncore_type *type)
+{
+	int i, ret;
+
+	for (i = 0; i < type->num_boxes; i++) {
+		ret = uncore_pmu_register(&type->pmus[i]);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int __init uncore_msr_pmus_register(void)
+{
+	struct intel_uncore_type **types = uncore_msr_uncores;
+	int ret;
+
+	for (; *types; types++) {
+		ret = type_pmu_register(*types);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+static int __init uncore_cpu_init(void)
+{
+	int ret;
+
+	ret = uncore_types_init(uncore_msr_uncores, true);
+	if (ret)
+		goto err;
+
+	ret = uncore_msr_pmus_register();
+	if (ret)
+		goto err;
+	return 0;
+err:
+	uncore_types_exit(uncore_msr_uncores);
+	uncore_msr_uncores = empty_uncore;
+	return ret;
+}
+
+static int __init uncore_mmio_init(void)
+{
+	struct intel_uncore_type **types = uncore_mmio_uncores;
+	int ret;
+
+	ret = uncore_types_init(types, true);
+	if (ret)
+		goto err;
+
+	for (; *types; types++) {
+		ret = type_pmu_register(*types);
+		if (ret)
+			goto err;
+	}
+	return 0;
+err:
+	uncore_types_exit(uncore_mmio_uncores);
+	uncore_mmio_uncores = empty_uncore;
+	return ret;
+}
+
+struct intel_uncore_init_fun {
+	void	(*cpu_init)(void);
+	int	(*pci_init)(void);
+	void	(*mmio_init)(void);
+};
+
+static const struct intel_uncore_init_fun nhm_uncore_init __initconst = {
+	.cpu_init = nhm_uncore_cpu_init,
+};
+
+static const struct intel_uncore_init_fun snb_uncore_init __initconst = {
+	.cpu_init = snb_uncore_cpu_init,
+	.pci_init = snb_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun ivb_uncore_init __initconst = {
+	.cpu_init = snb_uncore_cpu_init,
+	.pci_init = ivb_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun hsw_uncore_init __initconst = {
+	.cpu_init = snb_uncore_cpu_init,
+	.pci_init = hsw_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun bdw_uncore_init __initconst = {
+	.cpu_init = snb_uncore_cpu_init,
+	.pci_init = bdw_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun snbep_uncore_init __initconst = {
+	.cpu_init = snbep_uncore_cpu_init,
+	.pci_init = snbep_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun nhmex_uncore_init __initconst = {
+	.cpu_init = nhmex_uncore_cpu_init,
+};
+
+static const struct intel_uncore_init_fun ivbep_uncore_init __initconst = {
+	.cpu_init = ivbep_uncore_cpu_init,
+	.pci_init = ivbep_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun hswep_uncore_init __initconst = {
+	.cpu_init = hswep_uncore_cpu_init,
+	.pci_init = hswep_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun bdx_uncore_init __initconst = {
+	.cpu_init = bdx_uncore_cpu_init,
+	.pci_init = bdx_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun knl_uncore_init __initconst = {
+	.cpu_init = knl_uncore_cpu_init,
+	.pci_init = knl_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun skl_uncore_init __initconst = {
+	.cpu_init = skl_uncore_cpu_init,
+	.pci_init = skl_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun skx_uncore_init __initconst = {
+	.cpu_init = skx_uncore_cpu_init,
+	.pci_init = skx_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun icl_uncore_init __initconst = {
+	.cpu_init = icl_uncore_cpu_init,
+	.pci_init = skl_uncore_pci_init,
+};
+
+static const struct intel_uncore_init_fun tgl_uncore_init __initconst = {
+	.cpu_init = tgl_uncore_cpu_init,
+	.mmio_init = tgl_uncore_mmio_init,
+};
+
+static const struct intel_uncore_init_fun tgl_l_uncore_init __initconst = {
+	.cpu_init = tgl_uncore_cpu_init,
+	.mmio_init = tgl_l_uncore_mmio_init,
+};
+
+static const struct intel_uncore_init_fun icx_uncore_init __initconst = {
+	.cpu_init = icx_uncore_cpu_init,
+	.pci_init = icx_uncore_pci_init,
+	.mmio_init = icx_uncore_mmio_init,
+};
+
+static const struct intel_uncore_init_fun snr_uncore_init __initconst = {
+	.cpu_init = snr_uncore_cpu_init,
+	.pci_init = snr_uncore_pci_init,
+	.mmio_init = snr_uncore_mmio_init,
+};
+
+static const struct x86_cpu_id intel_uncore_match[] __initconst = {
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EP,		&nhm_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM,		&nhm_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE,		&nhm_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EP,		&nhm_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE,		&snb_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE,		&ivb_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL,		&hsw_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_L,		&hsw_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_G,		&hsw_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL,		&bdw_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_G,		&bdw_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE_X,	&snbep_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EX,		&nhmex_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EX,		&nhmex_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE_X,		&ivbep_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_X,		&hswep_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_X,		&bdx_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_D,		&bdx_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL,	&knl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM,	&knl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X,		&skx_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE_L,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE,		&skl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L,		&icl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_NNPI,	&icl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE,		&icl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D,		&icx_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X,		&icx_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE_L,		&tgl_l_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(TIGERLAKE,		&tgl_uncore_init),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_D,	&snr_uncore_init),
+	{},
+};
+MODULE_DEVICE_TABLE(x86cpu, intel_uncore_match);
+
+static int __init intel_uncore_init(void)
+{
+	const struct x86_cpu_id *id;
+	struct intel_uncore_init_fun *uncore_init;
+	int pret = 0, cret = 0, mret = 0, ret;
+
+	id = x86_match_cpu(intel_uncore_match);
+	if (!id)
+		return -ENODEV;
+
+	if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
+		return -ENODEV;
+
+	__uncore_max_dies =
+		topology_max_packages() * topology_max_die_per_package();
+
+	uncore_init = (struct intel_uncore_init_fun *)id->driver_data;
+	if (uncore_init->pci_init) {
+		pret = uncore_init->pci_init();
+		if (!pret)
+			pret = uncore_pci_init();
+	}
+
+	if (uncore_init->cpu_init) {
+		uncore_init->cpu_init();
+		cret = uncore_cpu_init();
+	}
+
+	if (uncore_init->mmio_init) {
+		uncore_init->mmio_init();
+		mret = uncore_mmio_init();
+	}
+
+	if (cret && pret && mret)
+		return -ENODEV;
+
+	/* Install hotplug callbacks to setup the targets for each package */
+	ret = cpuhp_setup_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE,
+				"perf/x86/intel/uncore:online",
+				uncore_event_cpu_online,
+				uncore_event_cpu_offline);
+	if (ret)
+		goto err;
+	return 0;
+
+err:
+	uncore_types_exit(uncore_msr_uncores);
+	uncore_types_exit(uncore_mmio_uncores);
+	uncore_pci_exit();
+	return ret;
+}
+module_init(intel_uncore_init);
+
+static void __exit intel_uncore_exit(void)
+{
+	cpuhp_remove_state(CPUHP_AP_PERF_X86_UNCORE_ONLINE);
+	uncore_types_exit(uncore_msr_uncores);
+	uncore_types_exit(uncore_mmio_uncores);
+	uncore_pci_exit();
+}
+module_exit(intel_uncore_exit);
diff --git a/arch/x86/events/intel/uncore.h b/arch/x86/events/intel/uncore.h
new file mode 100644
index 000000000..4e2953a9e
--- /dev/null
+++ b/arch/x86/events/intel/uncore.h
@@ -0,0 +1,598 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/slab.h>
+#include <linux/pci.h>
+#include <asm/apicdef.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+
+#include <linux/perf_event.h>
+#include "../perf_event.h"
+
+#define UNCORE_PMU_NAME_LEN		32
+#define UNCORE_PMU_HRTIMER_INTERVAL	(60LL * NSEC_PER_SEC)
+#define UNCORE_SNB_IMC_HRTIMER_INTERVAL (5ULL * NSEC_PER_SEC)
+
+#define UNCORE_FIXED_EVENT		0xff
+#define UNCORE_PMC_IDX_MAX_GENERIC	8
+#define UNCORE_PMC_IDX_MAX_FIXED	1
+#define UNCORE_PMC_IDX_MAX_FREERUNNING	1
+#define UNCORE_PMC_IDX_FIXED		UNCORE_PMC_IDX_MAX_GENERIC
+#define UNCORE_PMC_IDX_FREERUNNING	(UNCORE_PMC_IDX_FIXED + \
+					UNCORE_PMC_IDX_MAX_FIXED)
+#define UNCORE_PMC_IDX_MAX		(UNCORE_PMC_IDX_FREERUNNING + \
+					UNCORE_PMC_IDX_MAX_FREERUNNING)
+
+#define UNCORE_PCI_DEV_FULL_DATA(dev, func, type, idx)	\
+		((dev << 24) | (func << 16) | (type << 8) | idx)
+#define UNCORE_PCI_DEV_DATA(type, idx)	((type << 8) | idx)
+#define UNCORE_PCI_DEV_DEV(data)	((data >> 24) & 0xff)
+#define UNCORE_PCI_DEV_FUNC(data)	((data >> 16) & 0xff)
+#define UNCORE_PCI_DEV_TYPE(data)	((data >> 8) & 0xff)
+#define UNCORE_PCI_DEV_IDX(data)	(data & 0xff)
+#define UNCORE_EXTRA_PCI_DEV		0xff
+#define UNCORE_EXTRA_PCI_DEV_MAX	4
+
+#define UNCORE_EVENT_CONSTRAINT(c, n) EVENT_CONSTRAINT(c, n, 0xff)
+
+struct pci_extra_dev {
+	struct pci_dev *dev[UNCORE_EXTRA_PCI_DEV_MAX];
+};
+
+struct intel_uncore_ops;
+struct intel_uncore_pmu;
+struct intel_uncore_box;
+struct uncore_event_desc;
+struct freerunning_counters;
+
+struct intel_uncore_type {
+	const char *name;
+	int num_counters;
+	int num_boxes;
+	int perf_ctr_bits;
+	int fixed_ctr_bits;
+	int num_freerunning_types;
+	unsigned perf_ctr;
+	unsigned event_ctl;
+	unsigned event_mask;
+	unsigned event_mask_ext;
+	unsigned fixed_ctr;
+	unsigned fixed_ctl;
+	unsigned box_ctl;
+	union {
+		unsigned msr_offset;
+		unsigned mmio_offset;
+	};
+	unsigned mmio_map_size;
+	unsigned num_shared_regs:8;
+	unsigned single_fixed:1;
+	unsigned pair_ctr_ctl:1;
+	unsigned *msr_offsets;
+	struct event_constraint unconstrainted;
+	struct event_constraint *constraints;
+	struct intel_uncore_pmu *pmus;
+	struct intel_uncore_ops *ops;
+	struct uncore_event_desc *event_descs;
+	struct freerunning_counters *freerunning;
+	const struct attribute_group *attr_groups[4];
+	const struct attribute_group **attr_update;
+	struct pmu *pmu; /* for custom pmu ops */
+	/*
+	 * Uncore PMU would store relevant platform topology configuration here
+	 * to identify which platform component each PMON block of that type is
+	 * supposed to monitor.
+	 */
+	u64 *topology;
+	/*
+	 * Optional callbacks for managing mapping of Uncore units to PMONs
+	 */
+	int (*get_topology)(struct intel_uncore_type *type);
+	int (*set_mapping)(struct intel_uncore_type *type);
+	void (*cleanup_mapping)(struct intel_uncore_type *type);
+};
+
+#define pmu_group attr_groups[0]
+#define format_group attr_groups[1]
+#define events_group attr_groups[2]
+
+struct intel_uncore_ops {
+	void (*init_box)(struct intel_uncore_box *);
+	void (*exit_box)(struct intel_uncore_box *);
+	void (*disable_box)(struct intel_uncore_box *);
+	void (*enable_box)(struct intel_uncore_box *);
+	void (*disable_event)(struct intel_uncore_box *, struct perf_event *);
+	void (*enable_event)(struct intel_uncore_box *, struct perf_event *);
+	u64 (*read_counter)(struct intel_uncore_box *, struct perf_event *);
+	int (*hw_config)(struct intel_uncore_box *, struct perf_event *);
+	struct event_constraint *(*get_constraint)(struct intel_uncore_box *,
+						   struct perf_event *);
+	void (*put_constraint)(struct intel_uncore_box *, struct perf_event *);
+};
+
+struct intel_uncore_pmu {
+	struct pmu			pmu;
+	char				name[UNCORE_PMU_NAME_LEN];
+	int				pmu_idx;
+	int				func_id;
+	bool				registered;
+	atomic_t			activeboxes;
+	struct intel_uncore_type	*type;
+	struct intel_uncore_box		**boxes;
+};
+
+struct intel_uncore_extra_reg {
+	raw_spinlock_t lock;
+	u64 config, config1, config2;
+	atomic_t ref;
+};
+
+struct intel_uncore_box {
+	int pci_phys_id;
+	int dieid;	/* Logical die ID */
+	int n_active;	/* number of active events */
+	int n_events;
+	int cpu;	/* cpu to collect events */
+	unsigned long flags;
+	atomic_t refcnt;
+	struct perf_event *events[UNCORE_PMC_IDX_MAX];
+	struct perf_event *event_list[UNCORE_PMC_IDX_MAX];
+	struct event_constraint *event_constraint[UNCORE_PMC_IDX_MAX];
+	unsigned long active_mask[BITS_TO_LONGS(UNCORE_PMC_IDX_MAX)];
+	u64 tags[UNCORE_PMC_IDX_MAX];
+	struct pci_dev *pci_dev;
+	struct intel_uncore_pmu *pmu;
+	u64 hrtimer_duration; /* hrtimer timeout for this box */
+	struct hrtimer hrtimer;
+	struct list_head list;
+	struct list_head active_list;
+	void __iomem *io_addr;
+	struct intel_uncore_extra_reg shared_regs[];
+};
+
+/* CFL uncore 8th cbox MSRs */
+#define CFL_UNC_CBO_7_PERFEVTSEL0		0xf70
+#define CFL_UNC_CBO_7_PER_CTR0			0xf76
+
+#define UNCORE_BOX_FLAG_INITIATED		0
+/* event config registers are 8-byte apart */
+#define UNCORE_BOX_FLAG_CTL_OFFS8		1
+/* CFL 8th CBOX has different MSR space */
+#define UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS	2
+
+struct uncore_event_desc {
+	struct device_attribute attr;
+	const char *config;
+};
+
+struct freerunning_counters {
+	unsigned int counter_base;
+	unsigned int counter_offset;
+	unsigned int box_offset;
+	unsigned int num_counters;
+	unsigned int bits;
+	unsigned *box_offsets;
+};
+
+struct pci2phy_map {
+	struct list_head list;
+	int segment;
+	int pbus_to_physid[256];
+};
+
+struct pci2phy_map *__find_pci2phy_map(int segment);
+int uncore_pcibus_to_physid(struct pci_bus *bus);
+
+ssize_t uncore_event_show(struct device *dev,
+			  struct device_attribute *attr, char *buf);
+
+static inline struct intel_uncore_pmu *dev_to_uncore_pmu(struct device *dev)
+{
+	return container_of(dev_get_drvdata(dev), struct intel_uncore_pmu, pmu);
+}
+
+#define to_device_attribute(n)	container_of(n, struct device_attribute, attr)
+#define to_dev_ext_attribute(n)	container_of(n, struct dev_ext_attribute, attr)
+#define attr_to_ext_attr(n)	to_dev_ext_attribute(to_device_attribute(n))
+
+extern int __uncore_max_dies;
+#define uncore_max_dies()	(__uncore_max_dies)
+
+#define INTEL_UNCORE_EVENT_DESC(_name, _config)			\
+{								\
+	.attr	= __ATTR(_name, 0444, uncore_event_show, NULL),	\
+	.config	= _config,					\
+}
+
+#define DEFINE_UNCORE_FORMAT_ATTR(_var, _name, _format)			\
+static ssize_t __uncore_##_var##_show(struct device *dev,		\
+				struct device_attribute *attr,		\
+				char *page)				\
+{									\
+	BUILD_BUG_ON(sizeof(_format) >= PAGE_SIZE);			\
+	return sprintf(page, _format "\n");				\
+}									\
+static struct device_attribute format_attr_##_var =			\
+	__ATTR(_name, 0444, __uncore_##_var##_show, NULL)
+
+static inline bool uncore_pmc_fixed(int idx)
+{
+	return idx == UNCORE_PMC_IDX_FIXED;
+}
+
+static inline bool uncore_pmc_freerunning(int idx)
+{
+	return idx == UNCORE_PMC_IDX_FREERUNNING;
+}
+
+static inline bool uncore_mmio_is_valid_offset(struct intel_uncore_box *box,
+					       unsigned long offset)
+{
+	if (offset < box->pmu->type->mmio_map_size)
+		return true;
+
+	pr_warn_once("perf uncore: Invalid offset 0x%lx exceeds mapped area of %s.\n",
+		     offset, box->pmu->type->name);
+
+	return false;
+}
+
+static inline
+unsigned int uncore_mmio_box_ctl(struct intel_uncore_box *box)
+{
+	return box->pmu->type->box_ctl +
+	       box->pmu->type->mmio_offset * box->pmu->pmu_idx;
+}
+
+static inline unsigned uncore_pci_box_ctl(struct intel_uncore_box *box)
+{
+	return box->pmu->type->box_ctl;
+}
+
+static inline unsigned uncore_pci_fixed_ctl(struct intel_uncore_box *box)
+{
+	return box->pmu->type->fixed_ctl;
+}
+
+static inline unsigned uncore_pci_fixed_ctr(struct intel_uncore_box *box)
+{
+	return box->pmu->type->fixed_ctr;
+}
+
+static inline
+unsigned uncore_pci_event_ctl(struct intel_uncore_box *box, int idx)
+{
+	if (test_bit(UNCORE_BOX_FLAG_CTL_OFFS8, &box->flags))
+		return idx * 8 + box->pmu->type->event_ctl;
+
+	return idx * 4 + box->pmu->type->event_ctl;
+}
+
+static inline
+unsigned uncore_pci_perf_ctr(struct intel_uncore_box *box, int idx)
+{
+	return idx * 8 + box->pmu->type->perf_ctr;
+}
+
+static inline unsigned uncore_msr_box_offset(struct intel_uncore_box *box)
+{
+	struct intel_uncore_pmu *pmu = box->pmu;
+	return pmu->type->msr_offsets ?
+		pmu->type->msr_offsets[pmu->pmu_idx] :
+		pmu->type->msr_offset * pmu->pmu_idx;
+}
+
+static inline unsigned uncore_msr_box_ctl(struct intel_uncore_box *box)
+{
+	if (!box->pmu->type->box_ctl)
+		return 0;
+	return box->pmu->type->box_ctl + uncore_msr_box_offset(box);
+}
+
+static inline unsigned uncore_msr_fixed_ctl(struct intel_uncore_box *box)
+{
+	if (!box->pmu->type->fixed_ctl)
+		return 0;
+	return box->pmu->type->fixed_ctl + uncore_msr_box_offset(box);
+}
+
+static inline unsigned uncore_msr_fixed_ctr(struct intel_uncore_box *box)
+{
+	return box->pmu->type->fixed_ctr + uncore_msr_box_offset(box);
+}
+
+
+/*
+ * In the uncore document, there is no event-code assigned to free running
+ * counters. Some events need to be defined to indicate the free running
+ * counters. The events are encoded as event-code + umask-code.
+ *
+ * The event-code for all free running counters is 0xff, which is the same as
+ * the fixed counters.
+ *
+ * The umask-code is used to distinguish a fixed counter and a free running
+ * counter, and different types of free running counters.
+ * - For fixed counters, the umask-code is 0x0X.
+ *   X indicates the index of the fixed counter, which starts from 0.
+ * - For free running counters, the umask-code uses the rest of the space.
+ *   It would bare the format of 0xXY.
+ *   X stands for the type of free running counters, which starts from 1.
+ *   Y stands for the index of free running counters of same type, which
+ *   starts from 0.
+ *
+ * For example, there are three types of IIO free running counters on Skylake
+ * server, IO CLOCKS counters, BANDWIDTH counters and UTILIZATION counters.
+ * The event-code for all the free running counters is 0xff.
+ * 'ioclk' is the first counter of IO CLOCKS. IO CLOCKS is the first type,
+ * which umask-code starts from 0x10.
+ * So 'ioclk' is encoded as event=0xff,umask=0x10
+ * 'bw_in_port2' is the third counter of BANDWIDTH counters. BANDWIDTH is
+ * the second type, which umask-code starts from 0x20.
+ * So 'bw_in_port2' is encoded as event=0xff,umask=0x22
+ */
+static inline unsigned int uncore_freerunning_idx(u64 config)
+{
+	return ((config >> 8) & 0xf);
+}
+
+#define UNCORE_FREERUNNING_UMASK_START		0x10
+
+static inline unsigned int uncore_freerunning_type(u64 config)
+{
+	return ((((config >> 8) - UNCORE_FREERUNNING_UMASK_START) >> 4) & 0xf);
+}
+
+static inline
+unsigned int uncore_freerunning_counter(struct intel_uncore_box *box,
+					struct perf_event *event)
+{
+	unsigned int type = uncore_freerunning_type(event->hw.config);
+	unsigned int idx = uncore_freerunning_idx(event->hw.config);
+	struct intel_uncore_pmu *pmu = box->pmu;
+
+	return pmu->type->freerunning[type].counter_base +
+	       pmu->type->freerunning[type].counter_offset * idx +
+	       (pmu->type->freerunning[type].box_offsets ?
+	        pmu->type->freerunning[type].box_offsets[pmu->pmu_idx] :
+	        pmu->type->freerunning[type].box_offset * pmu->pmu_idx);
+}
+
+static inline
+unsigned uncore_msr_event_ctl(struct intel_uncore_box *box, int idx)
+{
+	if (test_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags)) {
+		return CFL_UNC_CBO_7_PERFEVTSEL0 +
+		       (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx);
+	} else {
+		return box->pmu->type->event_ctl +
+		       (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
+		       uncore_msr_box_offset(box);
+	}
+}
+
+static inline
+unsigned uncore_msr_perf_ctr(struct intel_uncore_box *box, int idx)
+{
+	if (test_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags)) {
+		return CFL_UNC_CBO_7_PER_CTR0 +
+		       (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx);
+	} else {
+		return box->pmu->type->perf_ctr +
+		       (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
+		       uncore_msr_box_offset(box);
+	}
+}
+
+static inline
+unsigned uncore_fixed_ctl(struct intel_uncore_box *box)
+{
+	if (box->pci_dev || box->io_addr)
+		return uncore_pci_fixed_ctl(box);
+	else
+		return uncore_msr_fixed_ctl(box);
+}
+
+static inline
+unsigned uncore_fixed_ctr(struct intel_uncore_box *box)
+{
+	if (box->pci_dev || box->io_addr)
+		return uncore_pci_fixed_ctr(box);
+	else
+		return uncore_msr_fixed_ctr(box);
+}
+
+static inline
+unsigned uncore_event_ctl(struct intel_uncore_box *box, int idx)
+{
+	if (box->pci_dev || box->io_addr)
+		return uncore_pci_event_ctl(box, idx);
+	else
+		return uncore_msr_event_ctl(box, idx);
+}
+
+static inline
+unsigned uncore_perf_ctr(struct intel_uncore_box *box, int idx)
+{
+	if (box->pci_dev || box->io_addr)
+		return uncore_pci_perf_ctr(box, idx);
+	else
+		return uncore_msr_perf_ctr(box, idx);
+}
+
+static inline int uncore_perf_ctr_bits(struct intel_uncore_box *box)
+{
+	return box->pmu->type->perf_ctr_bits;
+}
+
+static inline int uncore_fixed_ctr_bits(struct intel_uncore_box *box)
+{
+	return box->pmu->type->fixed_ctr_bits;
+}
+
+static inline
+unsigned int uncore_freerunning_bits(struct intel_uncore_box *box,
+				     struct perf_event *event)
+{
+	unsigned int type = uncore_freerunning_type(event->hw.config);
+
+	return box->pmu->type->freerunning[type].bits;
+}
+
+static inline int uncore_num_freerunning(struct intel_uncore_box *box,
+					 struct perf_event *event)
+{
+	unsigned int type = uncore_freerunning_type(event->hw.config);
+
+	return box->pmu->type->freerunning[type].num_counters;
+}
+
+static inline int uncore_num_freerunning_types(struct intel_uncore_box *box,
+					       struct perf_event *event)
+{
+	return box->pmu->type->num_freerunning_types;
+}
+
+static inline bool check_valid_freerunning_event(struct intel_uncore_box *box,
+						 struct perf_event *event)
+{
+	unsigned int type = uncore_freerunning_type(event->hw.config);
+	unsigned int idx = uncore_freerunning_idx(event->hw.config);
+
+	return (type < uncore_num_freerunning_types(box, event)) &&
+	       (idx < uncore_num_freerunning(box, event));
+}
+
+static inline int uncore_num_counters(struct intel_uncore_box *box)
+{
+	return box->pmu->type->num_counters;
+}
+
+static inline bool is_freerunning_event(struct perf_event *event)
+{
+	u64 cfg = event->attr.config;
+
+	return ((cfg & UNCORE_FIXED_EVENT) == UNCORE_FIXED_EVENT) &&
+	       (((cfg >> 8) & 0xff) >= UNCORE_FREERUNNING_UMASK_START);
+}
+
+/* Check and reject invalid config */
+static inline int uncore_freerunning_hw_config(struct intel_uncore_box *box,
+					       struct perf_event *event)
+{
+	if (is_freerunning_event(event))
+		return 0;
+
+	return -EINVAL;
+}
+
+static inline void uncore_disable_event(struct intel_uncore_box *box,
+				struct perf_event *event)
+{
+	box->pmu->type->ops->disable_event(box, event);
+}
+
+static inline void uncore_enable_event(struct intel_uncore_box *box,
+				struct perf_event *event)
+{
+	box->pmu->type->ops->enable_event(box, event);
+}
+
+static inline u64 uncore_read_counter(struct intel_uncore_box *box,
+				struct perf_event *event)
+{
+	return box->pmu->type->ops->read_counter(box, event);
+}
+
+static inline void uncore_box_init(struct intel_uncore_box *box)
+{
+	if (!test_and_set_bit(UNCORE_BOX_FLAG_INITIATED, &box->flags)) {
+		if (box->pmu->type->ops->init_box)
+			box->pmu->type->ops->init_box(box);
+	}
+}
+
+static inline void uncore_box_exit(struct intel_uncore_box *box)
+{
+	if (test_and_clear_bit(UNCORE_BOX_FLAG_INITIATED, &box->flags)) {
+		if (box->pmu->type->ops->exit_box)
+			box->pmu->type->ops->exit_box(box);
+	}
+}
+
+static inline bool uncore_box_is_fake(struct intel_uncore_box *box)
+{
+	return (box->dieid < 0);
+}
+
+static inline struct intel_uncore_pmu *uncore_event_to_pmu(struct perf_event *event)
+{
+	return container_of(event->pmu, struct intel_uncore_pmu, pmu);
+}
+
+static inline struct intel_uncore_box *uncore_event_to_box(struct perf_event *event)
+{
+	return event->pmu_private;
+}
+
+struct intel_uncore_box *uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu);
+u64 uncore_msr_read_counter(struct intel_uncore_box *box, struct perf_event *event);
+void uncore_mmio_exit_box(struct intel_uncore_box *box);
+u64 uncore_mmio_read_counter(struct intel_uncore_box *box,
+			     struct perf_event *event);
+void uncore_pmu_start_hrtimer(struct intel_uncore_box *box);
+void uncore_pmu_cancel_hrtimer(struct intel_uncore_box *box);
+void uncore_pmu_event_start(struct perf_event *event, int flags);
+void uncore_pmu_event_stop(struct perf_event *event, int flags);
+int uncore_pmu_event_add(struct perf_event *event, int flags);
+void uncore_pmu_event_del(struct perf_event *event, int flags);
+void uncore_pmu_event_read(struct perf_event *event);
+void uncore_perf_event_update(struct intel_uncore_box *box, struct perf_event *event);
+struct event_constraint *
+uncore_get_constraint(struct intel_uncore_box *box, struct perf_event *event);
+void uncore_put_constraint(struct intel_uncore_box *box, struct perf_event *event);
+u64 uncore_shared_reg_config(struct intel_uncore_box *box, int idx);
+
+extern struct intel_uncore_type **uncore_msr_uncores;
+extern struct intel_uncore_type **uncore_pci_uncores;
+extern struct intel_uncore_type **uncore_mmio_uncores;
+extern struct pci_driver *uncore_pci_driver;
+extern struct pci_driver *uncore_pci_sub_driver;
+extern raw_spinlock_t pci2phy_map_lock;
+extern struct list_head pci2phy_map_head;
+extern struct pci_extra_dev *uncore_extra_pci_dev;
+extern struct event_constraint uncore_constraint_empty;
+
+/* uncore_snb.c */
+int snb_uncore_pci_init(void);
+int ivb_uncore_pci_init(void);
+int hsw_uncore_pci_init(void);
+int bdw_uncore_pci_init(void);
+int skl_uncore_pci_init(void);
+void snb_uncore_cpu_init(void);
+void nhm_uncore_cpu_init(void);
+void skl_uncore_cpu_init(void);
+void icl_uncore_cpu_init(void);
+void tgl_uncore_cpu_init(void);
+void tgl_uncore_mmio_init(void);
+void tgl_l_uncore_mmio_init(void);
+int snb_pci2phy_map_init(int devid);
+
+/* uncore_snbep.c */
+int snbep_uncore_pci_init(void);
+void snbep_uncore_cpu_init(void);
+int ivbep_uncore_pci_init(void);
+void ivbep_uncore_cpu_init(void);
+int hswep_uncore_pci_init(void);
+void hswep_uncore_cpu_init(void);
+int bdx_uncore_pci_init(void);
+void bdx_uncore_cpu_init(void);
+int knl_uncore_pci_init(void);
+void knl_uncore_cpu_init(void);
+int skx_uncore_pci_init(void);
+void skx_uncore_cpu_init(void);
+int snr_uncore_pci_init(void);
+void snr_uncore_cpu_init(void);
+void snr_uncore_mmio_init(void);
+int icx_uncore_pci_init(void);
+void icx_uncore_cpu_init(void);
+void icx_uncore_mmio_init(void);
+
+/* uncore_nhmex.c */
+void nhmex_uncore_cpu_init(void);
diff --git a/arch/x86/events/intel/uncore_nhmex.c b/arch/x86/events/intel/uncore_nhmex.c
new file mode 100644
index 000000000..173e2674b
--- /dev/null
+++ b/arch/x86/events/intel/uncore_nhmex.c
@@ -0,0 +1,1228 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Nehalem-EX/Westmere-EX uncore support */
+#include "uncore.h"
+
+/* NHM-EX event control */
+#define NHMEX_PMON_CTL_EV_SEL_MASK	0x000000ff
+#define NHMEX_PMON_CTL_UMASK_MASK	0x0000ff00
+#define NHMEX_PMON_CTL_EN_BIT0		(1 << 0)
+#define NHMEX_PMON_CTL_EDGE_DET		(1 << 18)
+#define NHMEX_PMON_CTL_PMI_EN		(1 << 20)
+#define NHMEX_PMON_CTL_EN_BIT22		(1 << 22)
+#define NHMEX_PMON_CTL_INVERT		(1 << 23)
+#define NHMEX_PMON_CTL_TRESH_MASK	0xff000000
+#define NHMEX_PMON_RAW_EVENT_MASK	(NHMEX_PMON_CTL_EV_SEL_MASK | \
+					 NHMEX_PMON_CTL_UMASK_MASK | \
+					 NHMEX_PMON_CTL_EDGE_DET | \
+					 NHMEX_PMON_CTL_INVERT | \
+					 NHMEX_PMON_CTL_TRESH_MASK)
+
+/* NHM-EX Ubox */
+#define NHMEX_U_MSR_PMON_GLOBAL_CTL		0xc00
+#define NHMEX_U_MSR_PMON_CTR			0xc11
+#define NHMEX_U_MSR_PMON_EV_SEL			0xc10
+
+#define NHMEX_U_PMON_GLOBAL_EN			(1 << 0)
+#define NHMEX_U_PMON_GLOBAL_PMI_CORE_SEL	0x0000001e
+#define NHMEX_U_PMON_GLOBAL_EN_ALL		(1 << 28)
+#define NHMEX_U_PMON_GLOBAL_RST_ALL		(1 << 29)
+#define NHMEX_U_PMON_GLOBAL_FRZ_ALL		(1 << 31)
+
+#define NHMEX_U_PMON_RAW_EVENT_MASK		\
+		(NHMEX_PMON_CTL_EV_SEL_MASK |	\
+		 NHMEX_PMON_CTL_EDGE_DET)
+
+/* NHM-EX Cbox */
+#define NHMEX_C0_MSR_PMON_GLOBAL_CTL		0xd00
+#define NHMEX_C0_MSR_PMON_CTR0			0xd11
+#define NHMEX_C0_MSR_PMON_EV_SEL0		0xd10
+#define NHMEX_C_MSR_OFFSET			0x20
+
+/* NHM-EX Bbox */
+#define NHMEX_B0_MSR_PMON_GLOBAL_CTL		0xc20
+#define NHMEX_B0_MSR_PMON_CTR0			0xc31
+#define NHMEX_B0_MSR_PMON_CTL0			0xc30
+#define NHMEX_B_MSR_OFFSET			0x40
+#define NHMEX_B0_MSR_MATCH			0xe45
+#define NHMEX_B0_MSR_MASK			0xe46
+#define NHMEX_B1_MSR_MATCH			0xe4d
+#define NHMEX_B1_MSR_MASK			0xe4e
+
+#define NHMEX_B_PMON_CTL_EN			(1 << 0)
+#define NHMEX_B_PMON_CTL_EV_SEL_SHIFT		1
+#define NHMEX_B_PMON_CTL_EV_SEL_MASK		\
+		(0x1f << NHMEX_B_PMON_CTL_EV_SEL_SHIFT)
+#define NHMEX_B_PMON_CTR_SHIFT		6
+#define NHMEX_B_PMON_CTR_MASK		\
+		(0x3 << NHMEX_B_PMON_CTR_SHIFT)
+#define NHMEX_B_PMON_RAW_EVENT_MASK		\
+		(NHMEX_B_PMON_CTL_EV_SEL_MASK | \
+		 NHMEX_B_PMON_CTR_MASK)
+
+/* NHM-EX Sbox */
+#define NHMEX_S0_MSR_PMON_GLOBAL_CTL		0xc40
+#define NHMEX_S0_MSR_PMON_CTR0			0xc51
+#define NHMEX_S0_MSR_PMON_CTL0			0xc50
+#define NHMEX_S_MSR_OFFSET			0x80
+#define NHMEX_S0_MSR_MM_CFG			0xe48
+#define NHMEX_S0_MSR_MATCH			0xe49
+#define NHMEX_S0_MSR_MASK			0xe4a
+#define NHMEX_S1_MSR_MM_CFG			0xe58
+#define NHMEX_S1_MSR_MATCH			0xe59
+#define NHMEX_S1_MSR_MASK			0xe5a
+
+#define NHMEX_S_PMON_MM_CFG_EN			(0x1ULL << 63)
+#define NHMEX_S_EVENT_TO_R_PROG_EV		0
+
+/* NHM-EX Mbox */
+#define NHMEX_M0_MSR_GLOBAL_CTL			0xca0
+#define NHMEX_M0_MSR_PMU_DSP			0xca5
+#define NHMEX_M0_MSR_PMU_ISS			0xca6
+#define NHMEX_M0_MSR_PMU_MAP			0xca7
+#define NHMEX_M0_MSR_PMU_MSC_THR		0xca8
+#define NHMEX_M0_MSR_PMU_PGT			0xca9
+#define NHMEX_M0_MSR_PMU_PLD			0xcaa
+#define NHMEX_M0_MSR_PMU_ZDP_CTL_FVC		0xcab
+#define NHMEX_M0_MSR_PMU_CTL0			0xcb0
+#define NHMEX_M0_MSR_PMU_CNT0			0xcb1
+#define NHMEX_M_MSR_OFFSET			0x40
+#define NHMEX_M0_MSR_PMU_MM_CFG			0xe54
+#define NHMEX_M1_MSR_PMU_MM_CFG			0xe5c
+
+#define NHMEX_M_PMON_MM_CFG_EN			(1ULL << 63)
+#define NHMEX_M_PMON_ADDR_MATCH_MASK		0x3ffffffffULL
+#define NHMEX_M_PMON_ADDR_MASK_MASK		0x7ffffffULL
+#define NHMEX_M_PMON_ADDR_MASK_SHIFT		34
+
+#define NHMEX_M_PMON_CTL_EN			(1 << 0)
+#define NHMEX_M_PMON_CTL_PMI_EN			(1 << 1)
+#define NHMEX_M_PMON_CTL_COUNT_MODE_SHIFT	2
+#define NHMEX_M_PMON_CTL_COUNT_MODE_MASK	\
+	(0x3 << NHMEX_M_PMON_CTL_COUNT_MODE_SHIFT)
+#define NHMEX_M_PMON_CTL_STORAGE_MODE_SHIFT	4
+#define NHMEX_M_PMON_CTL_STORAGE_MODE_MASK	\
+	(0x3 << NHMEX_M_PMON_CTL_STORAGE_MODE_SHIFT)
+#define NHMEX_M_PMON_CTL_WRAP_MODE		(1 << 6)
+#define NHMEX_M_PMON_CTL_FLAG_MODE		(1 << 7)
+#define NHMEX_M_PMON_CTL_INC_SEL_SHIFT		9
+#define NHMEX_M_PMON_CTL_INC_SEL_MASK		\
+	(0x1f << NHMEX_M_PMON_CTL_INC_SEL_SHIFT)
+#define NHMEX_M_PMON_CTL_SET_FLAG_SEL_SHIFT	19
+#define NHMEX_M_PMON_CTL_SET_FLAG_SEL_MASK	\
+	(0x7 << NHMEX_M_PMON_CTL_SET_FLAG_SEL_SHIFT)
+#define NHMEX_M_PMON_RAW_EVENT_MASK			\
+		(NHMEX_M_PMON_CTL_COUNT_MODE_MASK |	\
+		 NHMEX_M_PMON_CTL_STORAGE_MODE_MASK |	\
+		 NHMEX_M_PMON_CTL_WRAP_MODE |		\
+		 NHMEX_M_PMON_CTL_FLAG_MODE |		\
+		 NHMEX_M_PMON_CTL_INC_SEL_MASK |	\
+		 NHMEX_M_PMON_CTL_SET_FLAG_SEL_MASK)
+
+#define NHMEX_M_PMON_ZDP_CTL_FVC_MASK		(((1 << 11) - 1) | (1 << 23))
+#define NHMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(n)	(0x7ULL << (11 + 3 * (n)))
+
+#define WSMEX_M_PMON_ZDP_CTL_FVC_MASK		(((1 << 12) - 1) | (1 << 24))
+#define WSMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(n)	(0x7ULL << (12 + 3 * (n)))
+
+/*
+ * use the 9~13 bits to select event If the 7th bit is not set,
+ * otherwise use the 19~21 bits to select event.
+ */
+#define MBOX_INC_SEL(x) ((x) << NHMEX_M_PMON_CTL_INC_SEL_SHIFT)
+#define MBOX_SET_FLAG_SEL(x) (((x) << NHMEX_M_PMON_CTL_SET_FLAG_SEL_SHIFT) | \
+				NHMEX_M_PMON_CTL_FLAG_MODE)
+#define MBOX_INC_SEL_MASK (NHMEX_M_PMON_CTL_INC_SEL_MASK | \
+			   NHMEX_M_PMON_CTL_FLAG_MODE)
+#define MBOX_SET_FLAG_SEL_MASK (NHMEX_M_PMON_CTL_SET_FLAG_SEL_MASK | \
+				NHMEX_M_PMON_CTL_FLAG_MODE)
+#define MBOX_INC_SEL_EXTAR_REG(c, r) \
+		EVENT_EXTRA_REG(MBOX_INC_SEL(c), NHMEX_M0_MSR_PMU_##r, \
+				MBOX_INC_SEL_MASK, (u64)-1, NHMEX_M_##r)
+#define MBOX_SET_FLAG_SEL_EXTRA_REG(c, r) \
+		EVENT_EXTRA_REG(MBOX_SET_FLAG_SEL(c), NHMEX_M0_MSR_PMU_##r, \
+				MBOX_SET_FLAG_SEL_MASK, \
+				(u64)-1, NHMEX_M_##r)
+
+/* NHM-EX Rbox */
+#define NHMEX_R_MSR_GLOBAL_CTL			0xe00
+#define NHMEX_R_MSR_PMON_CTL0			0xe10
+#define NHMEX_R_MSR_PMON_CNT0			0xe11
+#define NHMEX_R_MSR_OFFSET			0x20
+
+#define NHMEX_R_MSR_PORTN_QLX_CFG(n)		\
+		((n) < 4 ? (0xe0c + (n)) : (0xe2c + (n) - 4))
+#define NHMEX_R_MSR_PORTN_IPERF_CFG0(n)		(0xe04 + (n))
+#define NHMEX_R_MSR_PORTN_IPERF_CFG1(n)		(0xe24 + (n))
+#define NHMEX_R_MSR_PORTN_XBR_OFFSET(n)		\
+		(((n) < 4 ? 0 : 0x10) + (n) * 4)
+#define NHMEX_R_MSR_PORTN_XBR_SET1_MM_CFG(n)	\
+		(0xe60 + NHMEX_R_MSR_PORTN_XBR_OFFSET(n))
+#define NHMEX_R_MSR_PORTN_XBR_SET1_MATCH(n)	\
+		(NHMEX_R_MSR_PORTN_XBR_SET1_MM_CFG(n) + 1)
+#define NHMEX_R_MSR_PORTN_XBR_SET1_MASK(n)	\
+		(NHMEX_R_MSR_PORTN_XBR_SET1_MM_CFG(n) + 2)
+#define NHMEX_R_MSR_PORTN_XBR_SET2_MM_CFG(n)	\
+		(0xe70 + NHMEX_R_MSR_PORTN_XBR_OFFSET(n))
+#define NHMEX_R_MSR_PORTN_XBR_SET2_MATCH(n)	\
+		(NHMEX_R_MSR_PORTN_XBR_SET2_MM_CFG(n) + 1)
+#define NHMEX_R_MSR_PORTN_XBR_SET2_MASK(n)	\
+		(NHMEX_R_MSR_PORTN_XBR_SET2_MM_CFG(n) + 2)
+
+#define NHMEX_R_PMON_CTL_EN			(1 << 0)
+#define NHMEX_R_PMON_CTL_EV_SEL_SHIFT		1
+#define NHMEX_R_PMON_CTL_EV_SEL_MASK		\
+		(0x1f << NHMEX_R_PMON_CTL_EV_SEL_SHIFT)
+#define NHMEX_R_PMON_CTL_PMI_EN			(1 << 6)
+#define NHMEX_R_PMON_RAW_EVENT_MASK		NHMEX_R_PMON_CTL_EV_SEL_MASK
+
+/* NHM-EX Wbox */
+#define NHMEX_W_MSR_GLOBAL_CTL			0xc80
+#define NHMEX_W_MSR_PMON_CNT0			0xc90
+#define NHMEX_W_MSR_PMON_EVT_SEL0		0xc91
+#define NHMEX_W_MSR_PMON_FIXED_CTR		0x394
+#define NHMEX_W_MSR_PMON_FIXED_CTL		0x395
+
+#define NHMEX_W_PMON_GLOBAL_FIXED_EN		(1ULL << 31)
+
+#define __BITS_VALUE(x, i, n)  ((typeof(x))(((x) >> ((i) * (n))) & \
+				((1ULL << (n)) - 1)))
+
+DEFINE_UNCORE_FORMAT_ATTR(event, event, "config:0-7");
+DEFINE_UNCORE_FORMAT_ATTR(event5, event, "config:1-5");
+DEFINE_UNCORE_FORMAT_ATTR(umask, umask, "config:8-15");
+DEFINE_UNCORE_FORMAT_ATTR(edge, edge, "config:18");
+DEFINE_UNCORE_FORMAT_ATTR(inv, inv, "config:23");
+DEFINE_UNCORE_FORMAT_ATTR(thresh8, thresh, "config:24-31");
+DEFINE_UNCORE_FORMAT_ATTR(counter, counter, "config:6-7");
+DEFINE_UNCORE_FORMAT_ATTR(match, match, "config1:0-63");
+DEFINE_UNCORE_FORMAT_ATTR(mask, mask, "config2:0-63");
+
+static void nhmex_uncore_msr_init_box(struct intel_uncore_box *box)
+{
+	wrmsrl(NHMEX_U_MSR_PMON_GLOBAL_CTL, NHMEX_U_PMON_GLOBAL_EN_ALL);
+}
+
+static void nhmex_uncore_msr_exit_box(struct intel_uncore_box *box)
+{
+	wrmsrl(NHMEX_U_MSR_PMON_GLOBAL_CTL, 0);
+}
+
+static void nhmex_uncore_msr_disable_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+	u64 config;
+
+	if (msr) {
+		rdmsrl(msr, config);
+		config &= ~((1ULL << uncore_num_counters(box)) - 1);
+		/* WBox has a fixed counter */
+		if (uncore_msr_fixed_ctl(box))
+			config &= ~NHMEX_W_PMON_GLOBAL_FIXED_EN;
+		wrmsrl(msr, config);
+	}
+}
+
+static void nhmex_uncore_msr_enable_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+	u64 config;
+
+	if (msr) {
+		rdmsrl(msr, config);
+		config |= (1ULL << uncore_num_counters(box)) - 1;
+		/* WBox has a fixed counter */
+		if (uncore_msr_fixed_ctl(box))
+			config |= NHMEX_W_PMON_GLOBAL_FIXED_EN;
+		wrmsrl(msr, config);
+	}
+}
+
+static void nhmex_uncore_msr_disable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	wrmsrl(event->hw.config_base, 0);
+}
+
+static void nhmex_uncore_msr_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->idx == UNCORE_PMC_IDX_FIXED)
+		wrmsrl(hwc->config_base, NHMEX_PMON_CTL_EN_BIT0);
+	else if (box->pmu->type->event_mask & NHMEX_PMON_CTL_EN_BIT0)
+		wrmsrl(hwc->config_base, hwc->config | NHMEX_PMON_CTL_EN_BIT22);
+	else
+		wrmsrl(hwc->config_base, hwc->config | NHMEX_PMON_CTL_EN_BIT0);
+}
+
+#define NHMEX_UNCORE_OPS_COMMON_INIT()				\
+	.init_box	= nhmex_uncore_msr_init_box,		\
+	.exit_box	= nhmex_uncore_msr_exit_box,		\
+	.disable_box	= nhmex_uncore_msr_disable_box,		\
+	.enable_box	= nhmex_uncore_msr_enable_box,		\
+	.disable_event	= nhmex_uncore_msr_disable_event,	\
+	.read_counter	= uncore_msr_read_counter
+
+static struct intel_uncore_ops nhmex_uncore_ops = {
+	NHMEX_UNCORE_OPS_COMMON_INIT(),
+	.enable_event	= nhmex_uncore_msr_enable_event,
+};
+
+static struct attribute *nhmex_uncore_ubox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_edge.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_ubox_format_group = {
+	.name		= "format",
+	.attrs		= nhmex_uncore_ubox_formats_attr,
+};
+
+static struct intel_uncore_type nhmex_uncore_ubox = {
+	.name		= "ubox",
+	.num_counters	= 1,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.event_ctl	= NHMEX_U_MSR_PMON_EV_SEL,
+	.perf_ctr	= NHMEX_U_MSR_PMON_CTR,
+	.event_mask	= NHMEX_U_PMON_RAW_EVENT_MASK,
+	.box_ctl	= NHMEX_U_MSR_PMON_GLOBAL_CTL,
+	.ops		= &nhmex_uncore_ops,
+	.format_group	= &nhmex_uncore_ubox_format_group
+};
+
+static struct attribute *nhmex_uncore_cbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_cbox_format_group = {
+	.name = "format",
+	.attrs = nhmex_uncore_cbox_formats_attr,
+};
+
+/* msr offset for each instance of cbox */
+static unsigned nhmex_cbox_msr_offsets[] = {
+	0x0, 0x80, 0x40, 0xc0, 0x20, 0xa0, 0x60, 0xe0, 0x240, 0x2c0,
+};
+
+static struct intel_uncore_type nhmex_uncore_cbox = {
+	.name			= "cbox",
+	.num_counters		= 6,
+	.num_boxes		= 10,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_C0_MSR_PMON_EV_SEL0,
+	.perf_ctr		= NHMEX_C0_MSR_PMON_CTR0,
+	.event_mask		= NHMEX_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_C0_MSR_PMON_GLOBAL_CTL,
+	.msr_offsets		= nhmex_cbox_msr_offsets,
+	.pair_ctr_ctl		= 1,
+	.ops			= &nhmex_uncore_ops,
+	.format_group		= &nhmex_uncore_cbox_format_group
+};
+
+static struct uncore_event_desc nhmex_uncore_wbox_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks, "event=0xff,umask=0"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_type nhmex_uncore_wbox = {
+	.name			= "wbox",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_W_MSR_PMON_CNT0,
+	.perf_ctr		= NHMEX_W_MSR_PMON_EVT_SEL0,
+	.fixed_ctr		= NHMEX_W_MSR_PMON_FIXED_CTR,
+	.fixed_ctl		= NHMEX_W_MSR_PMON_FIXED_CTL,
+	.event_mask		= NHMEX_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_W_MSR_GLOBAL_CTL,
+	.pair_ctr_ctl		= 1,
+	.event_descs		= nhmex_uncore_wbox_events,
+	.ops			= &nhmex_uncore_ops,
+	.format_group		= &nhmex_uncore_cbox_format_group
+};
+
+static int nhmex_bbox_hw_config(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+	int ctr, ev_sel;
+
+	ctr = (hwc->config & NHMEX_B_PMON_CTR_MASK) >>
+		NHMEX_B_PMON_CTR_SHIFT;
+	ev_sel = (hwc->config & NHMEX_B_PMON_CTL_EV_SEL_MASK) >>
+		  NHMEX_B_PMON_CTL_EV_SEL_SHIFT;
+
+	/* events that do not use the match/mask registers */
+	if ((ctr == 0 && ev_sel > 0x3) || (ctr == 1 && ev_sel > 0x6) ||
+	    (ctr == 2 && ev_sel != 0x4) || ctr == 3)
+		return 0;
+
+	if (box->pmu->pmu_idx == 0)
+		reg1->reg = NHMEX_B0_MSR_MATCH;
+	else
+		reg1->reg = NHMEX_B1_MSR_MATCH;
+	reg1->idx = 0;
+	reg1->config = event->attr.config1;
+	reg2->config = event->attr.config2;
+	return 0;
+}
+
+static void nhmex_bbox_msr_enable_event(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+
+	if (reg1->idx != EXTRA_REG_NONE) {
+		wrmsrl(reg1->reg, reg1->config);
+		wrmsrl(reg1->reg + 1, reg2->config);
+	}
+	wrmsrl(hwc->config_base, NHMEX_PMON_CTL_EN_BIT0 |
+		(hwc->config & NHMEX_B_PMON_CTL_EV_SEL_MASK));
+}
+
+/*
+ * The Bbox has 4 counters, but each counter monitors different events.
+ * Use bits 6-7 in the event config to select counter.
+ */
+static struct event_constraint nhmex_uncore_bbox_constraints[] = {
+	EVENT_CONSTRAINT(0 , 1, 0xc0),
+	EVENT_CONSTRAINT(0x40, 2, 0xc0),
+	EVENT_CONSTRAINT(0x80, 4, 0xc0),
+	EVENT_CONSTRAINT(0xc0, 8, 0xc0),
+	EVENT_CONSTRAINT_END,
+};
+
+static struct attribute *nhmex_uncore_bbox_formats_attr[] = {
+	&format_attr_event5.attr,
+	&format_attr_counter.attr,
+	&format_attr_match.attr,
+	&format_attr_mask.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_bbox_format_group = {
+	.name = "format",
+	.attrs = nhmex_uncore_bbox_formats_attr,
+};
+
+static struct intel_uncore_ops nhmex_uncore_bbox_ops = {
+	NHMEX_UNCORE_OPS_COMMON_INIT(),
+	.enable_event		= nhmex_bbox_msr_enable_event,
+	.hw_config		= nhmex_bbox_hw_config,
+	.get_constraint		= uncore_get_constraint,
+	.put_constraint		= uncore_put_constraint,
+};
+
+static struct intel_uncore_type nhmex_uncore_bbox = {
+	.name			= "bbox",
+	.num_counters		= 4,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_B0_MSR_PMON_CTL0,
+	.perf_ctr		= NHMEX_B0_MSR_PMON_CTR0,
+	.event_mask		= NHMEX_B_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_B0_MSR_PMON_GLOBAL_CTL,
+	.msr_offset		= NHMEX_B_MSR_OFFSET,
+	.pair_ctr_ctl		= 1,
+	.num_shared_regs	= 1,
+	.constraints		= nhmex_uncore_bbox_constraints,
+	.ops			= &nhmex_uncore_bbox_ops,
+	.format_group		= &nhmex_uncore_bbox_format_group
+};
+
+static int nhmex_sbox_hw_config(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+
+	/* only TO_R_PROG_EV event uses the match/mask register */
+	if ((hwc->config & NHMEX_PMON_CTL_EV_SEL_MASK) !=
+	    NHMEX_S_EVENT_TO_R_PROG_EV)
+		return 0;
+
+	if (box->pmu->pmu_idx == 0)
+		reg1->reg = NHMEX_S0_MSR_MM_CFG;
+	else
+		reg1->reg = NHMEX_S1_MSR_MM_CFG;
+	reg1->idx = 0;
+	reg1->config = event->attr.config1;
+	reg2->config = event->attr.config2;
+	return 0;
+}
+
+static void nhmex_sbox_msr_enable_event(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+
+	if (reg1->idx != EXTRA_REG_NONE) {
+		wrmsrl(reg1->reg, 0);
+		wrmsrl(reg1->reg + 1, reg1->config);
+		wrmsrl(reg1->reg + 2, reg2->config);
+		wrmsrl(reg1->reg, NHMEX_S_PMON_MM_CFG_EN);
+	}
+	wrmsrl(hwc->config_base, hwc->config | NHMEX_PMON_CTL_EN_BIT22);
+}
+
+static struct attribute *nhmex_uncore_sbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_match.attr,
+	&format_attr_mask.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_sbox_format_group = {
+	.name			= "format",
+	.attrs			= nhmex_uncore_sbox_formats_attr,
+};
+
+static struct intel_uncore_ops nhmex_uncore_sbox_ops = {
+	NHMEX_UNCORE_OPS_COMMON_INIT(),
+	.enable_event		= nhmex_sbox_msr_enable_event,
+	.hw_config		= nhmex_sbox_hw_config,
+	.get_constraint		= uncore_get_constraint,
+	.put_constraint		= uncore_put_constraint,
+};
+
+static struct intel_uncore_type nhmex_uncore_sbox = {
+	.name			= "sbox",
+	.num_counters		= 4,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_S0_MSR_PMON_CTL0,
+	.perf_ctr		= NHMEX_S0_MSR_PMON_CTR0,
+	.event_mask		= NHMEX_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_S0_MSR_PMON_GLOBAL_CTL,
+	.msr_offset		= NHMEX_S_MSR_OFFSET,
+	.pair_ctr_ctl		= 1,
+	.num_shared_regs	= 1,
+	.ops			= &nhmex_uncore_sbox_ops,
+	.format_group		= &nhmex_uncore_sbox_format_group
+};
+
+enum {
+	EXTRA_REG_NHMEX_M_FILTER,
+	EXTRA_REG_NHMEX_M_DSP,
+	EXTRA_REG_NHMEX_M_ISS,
+	EXTRA_REG_NHMEX_M_MAP,
+	EXTRA_REG_NHMEX_M_MSC_THR,
+	EXTRA_REG_NHMEX_M_PGT,
+	EXTRA_REG_NHMEX_M_PLD,
+	EXTRA_REG_NHMEX_M_ZDP_CTL_FVC,
+};
+
+static struct extra_reg nhmex_uncore_mbox_extra_regs[] = {
+	MBOX_INC_SEL_EXTAR_REG(0x0, DSP),
+	MBOX_INC_SEL_EXTAR_REG(0x4, MSC_THR),
+	MBOX_INC_SEL_EXTAR_REG(0x5, MSC_THR),
+	MBOX_INC_SEL_EXTAR_REG(0x9, ISS),
+	/* event 0xa uses two extra registers */
+	MBOX_INC_SEL_EXTAR_REG(0xa, ISS),
+	MBOX_INC_SEL_EXTAR_REG(0xa, PLD),
+	MBOX_INC_SEL_EXTAR_REG(0xb, PLD),
+	/* events 0xd ~ 0x10 use the same extra register */
+	MBOX_INC_SEL_EXTAR_REG(0xd, ZDP_CTL_FVC),
+	MBOX_INC_SEL_EXTAR_REG(0xe, ZDP_CTL_FVC),
+	MBOX_INC_SEL_EXTAR_REG(0xf, ZDP_CTL_FVC),
+	MBOX_INC_SEL_EXTAR_REG(0x10, ZDP_CTL_FVC),
+	MBOX_INC_SEL_EXTAR_REG(0x16, PGT),
+	MBOX_SET_FLAG_SEL_EXTRA_REG(0x0, DSP),
+	MBOX_SET_FLAG_SEL_EXTRA_REG(0x1, ISS),
+	MBOX_SET_FLAG_SEL_EXTRA_REG(0x5, PGT),
+	MBOX_SET_FLAG_SEL_EXTRA_REG(0x6, MAP),
+	EVENT_EXTRA_END
+};
+
+/* Nehalem-EX or Westmere-EX ? */
+static bool uncore_nhmex;
+
+static bool nhmex_mbox_get_shared_reg(struct intel_uncore_box *box, int idx, u64 config)
+{
+	struct intel_uncore_extra_reg *er;
+	unsigned long flags;
+	bool ret = false;
+	u64 mask;
+
+	if (idx < EXTRA_REG_NHMEX_M_ZDP_CTL_FVC) {
+		er = &box->shared_regs[idx];
+		raw_spin_lock_irqsave(&er->lock, flags);
+		if (!atomic_read(&er->ref) || er->config == config) {
+			atomic_inc(&er->ref);
+			er->config = config;
+			ret = true;
+		}
+		raw_spin_unlock_irqrestore(&er->lock, flags);
+
+		return ret;
+	}
+	/*
+	 * The ZDP_CTL_FVC MSR has 4 fields which are used to control
+	 * events 0xd ~ 0x10. Besides these 4 fields, there are additional
+	 * fields which are shared.
+	 */
+	idx -= EXTRA_REG_NHMEX_M_ZDP_CTL_FVC;
+	if (WARN_ON_ONCE(idx >= 4))
+		return false;
+
+	/* mask of the shared fields */
+	if (uncore_nhmex)
+		mask = NHMEX_M_PMON_ZDP_CTL_FVC_MASK;
+	else
+		mask = WSMEX_M_PMON_ZDP_CTL_FVC_MASK;
+	er = &box->shared_regs[EXTRA_REG_NHMEX_M_ZDP_CTL_FVC];
+
+	raw_spin_lock_irqsave(&er->lock, flags);
+	/* add mask of the non-shared field if it's in use */
+	if (__BITS_VALUE(atomic_read(&er->ref), idx, 8)) {
+		if (uncore_nhmex)
+			mask |= NHMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+		else
+			mask |= WSMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+	}
+
+	if (!atomic_read(&er->ref) || !((er->config ^ config) & mask)) {
+		atomic_add(1 << (idx * 8), &er->ref);
+		if (uncore_nhmex)
+			mask = NHMEX_M_PMON_ZDP_CTL_FVC_MASK |
+				NHMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+		else
+			mask = WSMEX_M_PMON_ZDP_CTL_FVC_MASK |
+				WSMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+		er->config &= ~mask;
+		er->config |= (config & mask);
+		ret = true;
+	}
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+
+	return ret;
+}
+
+static void nhmex_mbox_put_shared_reg(struct intel_uncore_box *box, int idx)
+{
+	struct intel_uncore_extra_reg *er;
+
+	if (idx < EXTRA_REG_NHMEX_M_ZDP_CTL_FVC) {
+		er = &box->shared_regs[idx];
+		atomic_dec(&er->ref);
+		return;
+	}
+
+	idx -= EXTRA_REG_NHMEX_M_ZDP_CTL_FVC;
+	er = &box->shared_regs[EXTRA_REG_NHMEX_M_ZDP_CTL_FVC];
+	atomic_sub(1 << (idx * 8), &er->ref);
+}
+
+static u64 nhmex_mbox_alter_er(struct perf_event *event, int new_idx, bool modify)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct hw_perf_event_extra *reg1 = &hwc->extra_reg;
+	u64 idx, orig_idx = __BITS_VALUE(reg1->idx, 0, 8);
+	u64 config = reg1->config;
+
+	/* get the non-shared control bits and shift them */
+	idx = orig_idx - EXTRA_REG_NHMEX_M_ZDP_CTL_FVC;
+	if (uncore_nhmex)
+		config &= NHMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+	else
+		config &= WSMEX_M_PMON_ZDP_CTL_FVC_EVENT_MASK(idx);
+	if (new_idx > orig_idx) {
+		idx = new_idx - orig_idx;
+		config <<= 3 * idx;
+	} else {
+		idx = orig_idx - new_idx;
+		config >>= 3 * idx;
+	}
+
+	/* add the shared control bits back */
+	if (uncore_nhmex)
+		config |= NHMEX_M_PMON_ZDP_CTL_FVC_MASK & reg1->config;
+	else
+		config |= WSMEX_M_PMON_ZDP_CTL_FVC_MASK & reg1->config;
+	config |= NHMEX_M_PMON_ZDP_CTL_FVC_MASK & reg1->config;
+	if (modify) {
+		/* adjust the main event selector */
+		if (new_idx > orig_idx)
+			hwc->config += idx << NHMEX_M_PMON_CTL_INC_SEL_SHIFT;
+		else
+			hwc->config -= idx << NHMEX_M_PMON_CTL_INC_SEL_SHIFT;
+		reg1->config = config;
+		reg1->idx = ~0xff | new_idx;
+	}
+	return config;
+}
+
+static struct event_constraint *
+nhmex_mbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct hw_perf_event_extra *reg2 = &event->hw.branch_reg;
+	int i, idx[2], alloc = 0;
+	u64 config1 = reg1->config;
+
+	idx[0] = __BITS_VALUE(reg1->idx, 0, 8);
+	idx[1] = __BITS_VALUE(reg1->idx, 1, 8);
+again:
+	for (i = 0; i < 2; i++) {
+		if (!uncore_box_is_fake(box) && (reg1->alloc & (0x1 << i)))
+			idx[i] = 0xff;
+
+		if (idx[i] == 0xff)
+			continue;
+
+		if (!nhmex_mbox_get_shared_reg(box, idx[i],
+				__BITS_VALUE(config1, i, 32)))
+			goto fail;
+		alloc |= (0x1 << i);
+	}
+
+	/* for the match/mask registers */
+	if (reg2->idx != EXTRA_REG_NONE &&
+	    (uncore_box_is_fake(box) || !reg2->alloc) &&
+	    !nhmex_mbox_get_shared_reg(box, reg2->idx, reg2->config))
+		goto fail;
+
+	/*
+	 * If it's a fake box -- as per validate_{group,event}() we
+	 * shouldn't touch event state and we can avoid doing so
+	 * since both will only call get_event_constraints() once
+	 * on each event, this avoids the need for reg->alloc.
+	 */
+	if (!uncore_box_is_fake(box)) {
+		if (idx[0] != 0xff && idx[0] != __BITS_VALUE(reg1->idx, 0, 8))
+			nhmex_mbox_alter_er(event, idx[0], true);
+		reg1->alloc |= alloc;
+		if (reg2->idx != EXTRA_REG_NONE)
+			reg2->alloc = 1;
+	}
+	return NULL;
+fail:
+	if (idx[0] != 0xff && !(alloc & 0x1) &&
+	    idx[0] >= EXTRA_REG_NHMEX_M_ZDP_CTL_FVC) {
+		/*
+		 * events 0xd ~ 0x10 are functional identical, but are
+		 * controlled by different fields in the ZDP_CTL_FVC
+		 * register. If we failed to take one field, try the
+		 * rest 3 choices.
+		 */
+		BUG_ON(__BITS_VALUE(reg1->idx, 1, 8) != 0xff);
+		idx[0] -= EXTRA_REG_NHMEX_M_ZDP_CTL_FVC;
+		idx[0] = (idx[0] + 1) % 4;
+		idx[0] += EXTRA_REG_NHMEX_M_ZDP_CTL_FVC;
+		if (idx[0] != __BITS_VALUE(reg1->idx, 0, 8)) {
+			config1 = nhmex_mbox_alter_er(event, idx[0], false);
+			goto again;
+		}
+	}
+
+	if (alloc & 0x1)
+		nhmex_mbox_put_shared_reg(box, idx[0]);
+	if (alloc & 0x2)
+		nhmex_mbox_put_shared_reg(box, idx[1]);
+	return &uncore_constraint_empty;
+}
+
+static void nhmex_mbox_put_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct hw_perf_event_extra *reg2 = &event->hw.branch_reg;
+
+	if (uncore_box_is_fake(box))
+		return;
+
+	if (reg1->alloc & 0x1)
+		nhmex_mbox_put_shared_reg(box, __BITS_VALUE(reg1->idx, 0, 8));
+	if (reg1->alloc & 0x2)
+		nhmex_mbox_put_shared_reg(box, __BITS_VALUE(reg1->idx, 1, 8));
+	reg1->alloc = 0;
+
+	if (reg2->alloc) {
+		nhmex_mbox_put_shared_reg(box, reg2->idx);
+		reg2->alloc = 0;
+	}
+}
+
+static int nhmex_mbox_extra_reg_idx(struct extra_reg *er)
+{
+	if (er->idx < EXTRA_REG_NHMEX_M_ZDP_CTL_FVC)
+		return er->idx;
+	return er->idx + (er->event >> NHMEX_M_PMON_CTL_INC_SEL_SHIFT) - 0xd;
+}
+
+static int nhmex_mbox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct intel_uncore_type *type = box->pmu->type;
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct hw_perf_event_extra *reg2 = &event->hw.branch_reg;
+	struct extra_reg *er;
+	unsigned msr;
+	int reg_idx = 0;
+	/*
+	 * The mbox events may require 2 extra MSRs at the most. But only
+	 * the lower 32 bits in these MSRs are significant, so we can use
+	 * config1 to pass two MSRs' config.
+	 */
+	for (er = nhmex_uncore_mbox_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		if (event->attr.config1 & ~er->valid_mask)
+			return -EINVAL;
+
+		msr = er->msr + type->msr_offset * box->pmu->pmu_idx;
+		if (WARN_ON_ONCE(msr >= 0xffff || er->idx >= 0xff))
+			return -EINVAL;
+
+		/* always use the 32~63 bits to pass the PLD config */
+		if (er->idx == EXTRA_REG_NHMEX_M_PLD)
+			reg_idx = 1;
+		else if (WARN_ON_ONCE(reg_idx > 0))
+			return -EINVAL;
+
+		reg1->idx &= ~(0xff << (reg_idx * 8));
+		reg1->reg &= ~(0xffff << (reg_idx * 16));
+		reg1->idx |= nhmex_mbox_extra_reg_idx(er) << (reg_idx * 8);
+		reg1->reg |= msr << (reg_idx * 16);
+		reg1->config = event->attr.config1;
+		reg_idx++;
+	}
+	/*
+	 * The mbox only provides ability to perform address matching
+	 * for the PLD events.
+	 */
+	if (reg_idx == 2) {
+		reg2->idx = EXTRA_REG_NHMEX_M_FILTER;
+		if (event->attr.config2 & NHMEX_M_PMON_MM_CFG_EN)
+			reg2->config = event->attr.config2;
+		else
+			reg2->config = ~0ULL;
+		if (box->pmu->pmu_idx == 0)
+			reg2->reg = NHMEX_M0_MSR_PMU_MM_CFG;
+		else
+			reg2->reg = NHMEX_M1_MSR_PMU_MM_CFG;
+	}
+	return 0;
+}
+
+static u64 nhmex_mbox_shared_reg_config(struct intel_uncore_box *box, int idx)
+{
+	struct intel_uncore_extra_reg *er;
+	unsigned long flags;
+	u64 config;
+
+	if (idx < EXTRA_REG_NHMEX_M_ZDP_CTL_FVC)
+		return box->shared_regs[idx].config;
+
+	er = &box->shared_regs[EXTRA_REG_NHMEX_M_ZDP_CTL_FVC];
+	raw_spin_lock_irqsave(&er->lock, flags);
+	config = er->config;
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+	return config;
+}
+
+static void nhmex_mbox_msr_enable_event(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+	int idx;
+
+	idx = __BITS_VALUE(reg1->idx, 0, 8);
+	if (idx != 0xff)
+		wrmsrl(__BITS_VALUE(reg1->reg, 0, 16),
+			nhmex_mbox_shared_reg_config(box, idx));
+	idx = __BITS_VALUE(reg1->idx, 1, 8);
+	if (idx != 0xff)
+		wrmsrl(__BITS_VALUE(reg1->reg, 1, 16),
+			nhmex_mbox_shared_reg_config(box, idx));
+
+	if (reg2->idx != EXTRA_REG_NONE) {
+		wrmsrl(reg2->reg, 0);
+		if (reg2->config != ~0ULL) {
+			wrmsrl(reg2->reg + 1,
+				reg2->config & NHMEX_M_PMON_ADDR_MATCH_MASK);
+			wrmsrl(reg2->reg + 2, NHMEX_M_PMON_ADDR_MASK_MASK &
+				(reg2->config >> NHMEX_M_PMON_ADDR_MASK_SHIFT));
+			wrmsrl(reg2->reg, NHMEX_M_PMON_MM_CFG_EN);
+		}
+	}
+
+	wrmsrl(hwc->config_base, hwc->config | NHMEX_PMON_CTL_EN_BIT0);
+}
+
+DEFINE_UNCORE_FORMAT_ATTR(count_mode,		count_mode,	"config:2-3");
+DEFINE_UNCORE_FORMAT_ATTR(storage_mode,		storage_mode,	"config:4-5");
+DEFINE_UNCORE_FORMAT_ATTR(wrap_mode,		wrap_mode,	"config:6");
+DEFINE_UNCORE_FORMAT_ATTR(flag_mode,		flag_mode,	"config:7");
+DEFINE_UNCORE_FORMAT_ATTR(inc_sel,		inc_sel,	"config:9-13");
+DEFINE_UNCORE_FORMAT_ATTR(set_flag_sel,		set_flag_sel,	"config:19-21");
+DEFINE_UNCORE_FORMAT_ATTR(filter_cfg_en,	filter_cfg_en,	"config2:63");
+DEFINE_UNCORE_FORMAT_ATTR(filter_match,		filter_match,	"config2:0-33");
+DEFINE_UNCORE_FORMAT_ATTR(filter_mask,		filter_mask,	"config2:34-61");
+DEFINE_UNCORE_FORMAT_ATTR(dsp,			dsp,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(thr,			thr,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(fvc,			fvc,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(pgt,			pgt,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(map,			map,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(iss,			iss,		"config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(pld,			pld,		"config1:32-63");
+
+static struct attribute *nhmex_uncore_mbox_formats_attr[] = {
+	&format_attr_count_mode.attr,
+	&format_attr_storage_mode.attr,
+	&format_attr_wrap_mode.attr,
+	&format_attr_flag_mode.attr,
+	&format_attr_inc_sel.attr,
+	&format_attr_set_flag_sel.attr,
+	&format_attr_filter_cfg_en.attr,
+	&format_attr_filter_match.attr,
+	&format_attr_filter_mask.attr,
+	&format_attr_dsp.attr,
+	&format_attr_thr.attr,
+	&format_attr_fvc.attr,
+	&format_attr_pgt.attr,
+	&format_attr_map.attr,
+	&format_attr_iss.attr,
+	&format_attr_pld.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_mbox_format_group = {
+	.name		= "format",
+	.attrs		= nhmex_uncore_mbox_formats_attr,
+};
+
+static struct uncore_event_desc nhmex_uncore_mbox_events[] = {
+	INTEL_UNCORE_EVENT_DESC(bbox_cmds_read, "inc_sel=0xd,fvc=0x2800"),
+	INTEL_UNCORE_EVENT_DESC(bbox_cmds_write, "inc_sel=0xd,fvc=0x2820"),
+	{ /* end: all zeroes */ },
+};
+
+static struct uncore_event_desc wsmex_uncore_mbox_events[] = {
+	INTEL_UNCORE_EVENT_DESC(bbox_cmds_read, "inc_sel=0xd,fvc=0x5000"),
+	INTEL_UNCORE_EVENT_DESC(bbox_cmds_write, "inc_sel=0xd,fvc=0x5040"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_ops nhmex_uncore_mbox_ops = {
+	NHMEX_UNCORE_OPS_COMMON_INIT(),
+	.enable_event	= nhmex_mbox_msr_enable_event,
+	.hw_config	= nhmex_mbox_hw_config,
+	.get_constraint	= nhmex_mbox_get_constraint,
+	.put_constraint	= nhmex_mbox_put_constraint,
+};
+
+static struct intel_uncore_type nhmex_uncore_mbox = {
+	.name			= "mbox",
+	.num_counters		= 6,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_M0_MSR_PMU_CTL0,
+	.perf_ctr		= NHMEX_M0_MSR_PMU_CNT0,
+	.event_mask		= NHMEX_M_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_M0_MSR_GLOBAL_CTL,
+	.msr_offset		= NHMEX_M_MSR_OFFSET,
+	.pair_ctr_ctl		= 1,
+	.num_shared_regs	= 8,
+	.event_descs		= nhmex_uncore_mbox_events,
+	.ops			= &nhmex_uncore_mbox_ops,
+	.format_group		= &nhmex_uncore_mbox_format_group,
+};
+
+static 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;
+
+	/* adjust the main event selector and extra register index */
+	if (reg1->idx % 2) {
+		reg1->idx--;
+		hwc->config -= 1 << NHMEX_R_PMON_CTL_EV_SEL_SHIFT;
+	} else {
+		reg1->idx++;
+		hwc->config += 1 << NHMEX_R_PMON_CTL_EV_SEL_SHIFT;
+	}
+
+	/* adjust extra register config */
+	switch (reg1->idx % 6) {
+	case 2:
+		/* shift the 8~15 bits to the 0~7 bits */
+		reg1->config >>= 8;
+		break;
+	case 3:
+		/* shift the 0~7 bits to the 8~15 bits */
+		reg1->config <<= 8;
+		break;
+	}
+}
+
+/*
+ * Each rbox has 4 event set which monitor PQI port 0~3 or 4~7.
+ * An event set consists of 6 events, the 3rd and 4th events in
+ * an event set use the same extra register. So an event set uses
+ * 5 extra registers.
+ */
+static struct event_constraint *
+nhmex_rbox_get_constraint(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+	struct intel_uncore_extra_reg *er;
+	unsigned long flags;
+	int idx, er_idx;
+	u64 config1;
+	bool ok = false;
+
+	if (!uncore_box_is_fake(box) && reg1->alloc)
+		return NULL;
+
+	idx = reg1->idx % 6;
+	config1 = reg1->config;
+again:
+	er_idx = idx;
+	/* the 3rd and 4th events use the same extra register */
+	if (er_idx > 2)
+		er_idx--;
+	er_idx += (reg1->idx / 6) * 5;
+
+	er = &box->shared_regs[er_idx];
+	raw_spin_lock_irqsave(&er->lock, flags);
+	if (idx < 2) {
+		if (!atomic_read(&er->ref) || er->config == reg1->config) {
+			atomic_inc(&er->ref);
+			er->config = reg1->config;
+			ok = true;
+		}
+	} else if (idx == 2 || idx == 3) {
+		/*
+		 * these two events use different fields in a extra register,
+		 * the 0~7 bits and the 8~15 bits respectively.
+		 */
+		u64 mask = 0xff << ((idx - 2) * 8);
+		if (!__BITS_VALUE(atomic_read(&er->ref), idx - 2, 8) ||
+				!((er->config ^ config1) & mask)) {
+			atomic_add(1 << ((idx - 2) * 8), &er->ref);
+			er->config &= ~mask;
+			er->config |= config1 & mask;
+			ok = true;
+		}
+	} else {
+		if (!atomic_read(&er->ref) ||
+				(er->config == (hwc->config >> 32) &&
+				 er->config1 == reg1->config &&
+				 er->config2 == reg2->config)) {
+			atomic_inc(&er->ref);
+			er->config = (hwc->config >> 32);
+			er->config1 = reg1->config;
+			er->config2 = reg2->config;
+			ok = true;
+		}
+	}
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+
+	if (!ok) {
+		/*
+		 * The Rbox events are always in pairs. The paired
+		 * events are functional identical, but use different
+		 * extra registers. If we failed to take an extra
+		 * register, try the alternative.
+		 */
+		idx ^= 1;
+		if (idx != reg1->idx % 6) {
+			if (idx == 2)
+				config1 >>= 8;
+			else if (idx == 3)
+				config1 <<= 8;
+			goto again;
+		}
+	} else {
+		if (!uncore_box_is_fake(box)) {
+			if (idx != reg1->idx % 6)
+				nhmex_rbox_alter_er(box, event);
+			reg1->alloc = 1;
+		}
+		return NULL;
+	}
+	return &uncore_constraint_empty;
+}
+
+static void nhmex_rbox_put_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct intel_uncore_extra_reg *er;
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	int idx, er_idx;
+
+	if (uncore_box_is_fake(box) || !reg1->alloc)
+		return;
+
+	idx = reg1->idx % 6;
+	er_idx = idx;
+	if (er_idx > 2)
+		er_idx--;
+	er_idx += (reg1->idx / 6) * 5;
+
+	er = &box->shared_regs[er_idx];
+	if (idx == 2 || idx == 3)
+		atomic_sub(1 << ((idx - 2) * 8), &er->ref);
+	else
+		atomic_dec(&er->ref);
+
+	reg1->alloc = 0;
+}
+
+static int nhmex_rbox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct hw_perf_event_extra *reg2 = &event->hw.branch_reg;
+	int idx;
+
+	idx = (event->hw.config & NHMEX_R_PMON_CTL_EV_SEL_MASK) >>
+		NHMEX_R_PMON_CTL_EV_SEL_SHIFT;
+	if (idx >= 0x18)
+		return -EINVAL;
+
+	reg1->idx = idx;
+	reg1->config = event->attr.config1;
+
+	switch (idx % 6) {
+	case 4:
+	case 5:
+		hwc->config |= event->attr.config & (~0ULL << 32);
+		reg2->config = event->attr.config2;
+		break;
+	}
+	return 0;
+}
+
+static void nhmex_rbox_msr_enable_event(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+	int idx, port;
+
+	idx = reg1->idx;
+	port = idx / 6 + box->pmu->pmu_idx * 4;
+
+	switch (idx % 6) {
+	case 0:
+		wrmsrl(NHMEX_R_MSR_PORTN_IPERF_CFG0(port), reg1->config);
+		break;
+	case 1:
+		wrmsrl(NHMEX_R_MSR_PORTN_IPERF_CFG1(port), reg1->config);
+		break;
+	case 2:
+	case 3:
+		wrmsrl(NHMEX_R_MSR_PORTN_QLX_CFG(port),
+			uncore_shared_reg_config(box, 2 + (idx / 6) * 5));
+		break;
+	case 4:
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET1_MM_CFG(port),
+			hwc->config >> 32);
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET1_MATCH(port), reg1->config);
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET1_MASK(port), reg2->config);
+		break;
+	case 5:
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET2_MM_CFG(port),
+			hwc->config >> 32);
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET2_MATCH(port), reg1->config);
+		wrmsrl(NHMEX_R_MSR_PORTN_XBR_SET2_MASK(port), reg2->config);
+		break;
+	}
+
+	wrmsrl(hwc->config_base, NHMEX_PMON_CTL_EN_BIT0 |
+		(hwc->config & NHMEX_R_PMON_CTL_EV_SEL_MASK));
+}
+
+DEFINE_UNCORE_FORMAT_ATTR(xbr_mm_cfg, xbr_mm_cfg, "config:32-63");
+DEFINE_UNCORE_FORMAT_ATTR(xbr_match, xbr_match, "config1:0-63");
+DEFINE_UNCORE_FORMAT_ATTR(xbr_mask, xbr_mask, "config2:0-63");
+DEFINE_UNCORE_FORMAT_ATTR(qlx_cfg, qlx_cfg, "config1:0-15");
+DEFINE_UNCORE_FORMAT_ATTR(iperf_cfg, iperf_cfg, "config1:0-31");
+
+static struct attribute *nhmex_uncore_rbox_formats_attr[] = {
+	&format_attr_event5.attr,
+	&format_attr_xbr_mm_cfg.attr,
+	&format_attr_xbr_match.attr,
+	&format_attr_xbr_mask.attr,
+	&format_attr_qlx_cfg.attr,
+	&format_attr_iperf_cfg.attr,
+	NULL,
+};
+
+static const struct attribute_group nhmex_uncore_rbox_format_group = {
+	.name = "format",
+	.attrs = nhmex_uncore_rbox_formats_attr,
+};
+
+static struct uncore_event_desc nhmex_uncore_rbox_events[] = {
+	INTEL_UNCORE_EVENT_DESC(qpi0_flit_send,		"event=0x0,iperf_cfg=0x80000000"),
+	INTEL_UNCORE_EVENT_DESC(qpi1_filt_send,		"event=0x6,iperf_cfg=0x80000000"),
+	INTEL_UNCORE_EVENT_DESC(qpi0_idle_filt,		"event=0x0,iperf_cfg=0x40000000"),
+	INTEL_UNCORE_EVENT_DESC(qpi1_idle_filt,		"event=0x6,iperf_cfg=0x40000000"),
+	INTEL_UNCORE_EVENT_DESC(qpi0_date_response,	"event=0x0,iperf_cfg=0xc4"),
+	INTEL_UNCORE_EVENT_DESC(qpi1_date_response,	"event=0x6,iperf_cfg=0xc4"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_ops nhmex_uncore_rbox_ops = {
+	NHMEX_UNCORE_OPS_COMMON_INIT(),
+	.enable_event		= nhmex_rbox_msr_enable_event,
+	.hw_config		= nhmex_rbox_hw_config,
+	.get_constraint		= nhmex_rbox_get_constraint,
+	.put_constraint		= nhmex_rbox_put_constraint,
+};
+
+static struct intel_uncore_type nhmex_uncore_rbox = {
+	.name			= "rbox",
+	.num_counters		= 8,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= NHMEX_R_MSR_PMON_CTL0,
+	.perf_ctr		= NHMEX_R_MSR_PMON_CNT0,
+	.event_mask		= NHMEX_R_PMON_RAW_EVENT_MASK,
+	.box_ctl		= NHMEX_R_MSR_GLOBAL_CTL,
+	.msr_offset		= NHMEX_R_MSR_OFFSET,
+	.pair_ctr_ctl		= 1,
+	.num_shared_regs	= 20,
+	.event_descs		= nhmex_uncore_rbox_events,
+	.ops			= &nhmex_uncore_rbox_ops,
+	.format_group		= &nhmex_uncore_rbox_format_group
+};
+
+static struct intel_uncore_type *nhmex_msr_uncores[] = {
+	&nhmex_uncore_ubox,
+	&nhmex_uncore_cbox,
+	&nhmex_uncore_bbox,
+	&nhmex_uncore_sbox,
+	&nhmex_uncore_mbox,
+	&nhmex_uncore_rbox,
+	&nhmex_uncore_wbox,
+	NULL,
+};
+
+void nhmex_uncore_cpu_init(void)
+{
+	if (boot_cpu_data.x86_model == 46)
+		uncore_nhmex = true;
+	else
+		nhmex_uncore_mbox.event_descs = wsmex_uncore_mbox_events;
+	if (nhmex_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		nhmex_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+	uncore_msr_uncores = nhmex_msr_uncores;
+}
+/* end of Nehalem-EX uncore support */
diff --git a/arch/x86/events/intel/uncore_snb.c b/arch/x86/events/intel/uncore_snb.c
new file mode 100644
index 000000000..a4c20e37b
--- /dev/null
+++ b/arch/x86/events/intel/uncore_snb.c
@@ -0,0 +1,1341 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Nehalem/SandBridge/Haswell/Broadwell/Skylake uncore support */
+#include "uncore.h"
+
+/* Uncore IMC PCI IDs */
+#define PCI_DEVICE_ID_INTEL_SNB_IMC		0x0100
+#define PCI_DEVICE_ID_INTEL_IVB_IMC		0x0154
+#define PCI_DEVICE_ID_INTEL_IVB_E3_IMC		0x0150
+#define PCI_DEVICE_ID_INTEL_HSW_IMC		0x0c00
+#define PCI_DEVICE_ID_INTEL_HSW_U_IMC		0x0a04
+#define PCI_DEVICE_ID_INTEL_BDW_IMC		0x1604
+#define PCI_DEVICE_ID_INTEL_SKL_U_IMC		0x1904
+#define PCI_DEVICE_ID_INTEL_SKL_Y_IMC		0x190c
+#define PCI_DEVICE_ID_INTEL_SKL_HD_IMC		0x1900
+#define PCI_DEVICE_ID_INTEL_SKL_HQ_IMC		0x1910
+#define PCI_DEVICE_ID_INTEL_SKL_SD_IMC		0x190f
+#define PCI_DEVICE_ID_INTEL_SKL_SQ_IMC		0x191f
+#define PCI_DEVICE_ID_INTEL_SKL_E3_IMC		0x1918
+#define PCI_DEVICE_ID_INTEL_KBL_Y_IMC		0x590c
+#define PCI_DEVICE_ID_INTEL_KBL_U_IMC		0x5904
+#define PCI_DEVICE_ID_INTEL_KBL_UQ_IMC		0x5914
+#define PCI_DEVICE_ID_INTEL_KBL_SD_IMC		0x590f
+#define PCI_DEVICE_ID_INTEL_KBL_SQ_IMC		0x591f
+#define PCI_DEVICE_ID_INTEL_KBL_HQ_IMC		0x5910
+#define PCI_DEVICE_ID_INTEL_KBL_WQ_IMC		0x5918
+#define PCI_DEVICE_ID_INTEL_CFL_2U_IMC		0x3ecc
+#define PCI_DEVICE_ID_INTEL_CFL_4U_IMC		0x3ed0
+#define PCI_DEVICE_ID_INTEL_CFL_4H_IMC		0x3e10
+#define PCI_DEVICE_ID_INTEL_CFL_6H_IMC		0x3ec4
+#define PCI_DEVICE_ID_INTEL_CFL_2S_D_IMC	0x3e0f
+#define PCI_DEVICE_ID_INTEL_CFL_4S_D_IMC	0x3e1f
+#define PCI_DEVICE_ID_INTEL_CFL_6S_D_IMC	0x3ec2
+#define PCI_DEVICE_ID_INTEL_CFL_8S_D_IMC	0x3e30
+#define PCI_DEVICE_ID_INTEL_CFL_4S_W_IMC	0x3e18
+#define PCI_DEVICE_ID_INTEL_CFL_6S_W_IMC	0x3ec6
+#define PCI_DEVICE_ID_INTEL_CFL_8S_W_IMC	0x3e31
+#define PCI_DEVICE_ID_INTEL_CFL_4S_S_IMC	0x3e33
+#define PCI_DEVICE_ID_INTEL_CFL_6S_S_IMC	0x3eca
+#define PCI_DEVICE_ID_INTEL_CFL_8S_S_IMC	0x3e32
+#define PCI_DEVICE_ID_INTEL_AML_YD_IMC		0x590c
+#define PCI_DEVICE_ID_INTEL_AML_YQ_IMC		0x590d
+#define PCI_DEVICE_ID_INTEL_WHL_UQ_IMC		0x3ed0
+#define PCI_DEVICE_ID_INTEL_WHL_4_UQ_IMC	0x3e34
+#define PCI_DEVICE_ID_INTEL_WHL_UD_IMC		0x3e35
+#define PCI_DEVICE_ID_INTEL_CML_H1_IMC		0x9b44
+#define PCI_DEVICE_ID_INTEL_CML_H2_IMC		0x9b54
+#define PCI_DEVICE_ID_INTEL_CML_H3_IMC		0x9b64
+#define PCI_DEVICE_ID_INTEL_CML_U1_IMC		0x9b51
+#define PCI_DEVICE_ID_INTEL_CML_U2_IMC		0x9b61
+#define PCI_DEVICE_ID_INTEL_CML_U3_IMC		0x9b71
+#define PCI_DEVICE_ID_INTEL_CML_S1_IMC		0x9b33
+#define PCI_DEVICE_ID_INTEL_CML_S2_IMC		0x9b43
+#define PCI_DEVICE_ID_INTEL_CML_S3_IMC		0x9b53
+#define PCI_DEVICE_ID_INTEL_CML_S4_IMC		0x9b63
+#define PCI_DEVICE_ID_INTEL_CML_S5_IMC		0x9b73
+#define PCI_DEVICE_ID_INTEL_ICL_U_IMC		0x8a02
+#define PCI_DEVICE_ID_INTEL_ICL_U2_IMC		0x8a12
+#define PCI_DEVICE_ID_INTEL_TGL_U1_IMC		0x9a02
+#define PCI_DEVICE_ID_INTEL_TGL_U2_IMC		0x9a04
+#define PCI_DEVICE_ID_INTEL_TGL_U3_IMC		0x9a12
+#define PCI_DEVICE_ID_INTEL_TGL_U4_IMC		0x9a14
+#define PCI_DEVICE_ID_INTEL_TGL_H_IMC		0x9a36
+
+
+/* SNB event control */
+#define SNB_UNC_CTL_EV_SEL_MASK			0x000000ff
+#define SNB_UNC_CTL_UMASK_MASK			0x0000ff00
+#define SNB_UNC_CTL_EDGE_DET			(1 << 18)
+#define SNB_UNC_CTL_EN				(1 << 22)
+#define SNB_UNC_CTL_INVERT			(1 << 23)
+#define SNB_UNC_CTL_CMASK_MASK			0x1f000000
+#define NHM_UNC_CTL_CMASK_MASK			0xff000000
+#define NHM_UNC_FIXED_CTR_CTL_EN		(1 << 0)
+
+#define SNB_UNC_RAW_EVENT_MASK			(SNB_UNC_CTL_EV_SEL_MASK | \
+						 SNB_UNC_CTL_UMASK_MASK | \
+						 SNB_UNC_CTL_EDGE_DET | \
+						 SNB_UNC_CTL_INVERT | \
+						 SNB_UNC_CTL_CMASK_MASK)
+
+#define NHM_UNC_RAW_EVENT_MASK			(SNB_UNC_CTL_EV_SEL_MASK | \
+						 SNB_UNC_CTL_UMASK_MASK | \
+						 SNB_UNC_CTL_EDGE_DET | \
+						 SNB_UNC_CTL_INVERT | \
+						 NHM_UNC_CTL_CMASK_MASK)
+
+/* SNB global control register */
+#define SNB_UNC_PERF_GLOBAL_CTL                 0x391
+#define SNB_UNC_FIXED_CTR_CTRL                  0x394
+#define SNB_UNC_FIXED_CTR                       0x395
+
+/* SNB uncore global control */
+#define SNB_UNC_GLOBAL_CTL_CORE_ALL             ((1 << 4) - 1)
+#define SNB_UNC_GLOBAL_CTL_EN                   (1 << 29)
+
+/* SNB Cbo register */
+#define SNB_UNC_CBO_0_PERFEVTSEL0               0x700
+#define SNB_UNC_CBO_0_PER_CTR0                  0x706
+#define SNB_UNC_CBO_MSR_OFFSET                  0x10
+
+/* SNB ARB register */
+#define SNB_UNC_ARB_PER_CTR0			0x3b0
+#define SNB_UNC_ARB_PERFEVTSEL0			0x3b2
+#define SNB_UNC_ARB_MSR_OFFSET			0x10
+
+/* NHM global control register */
+#define NHM_UNC_PERF_GLOBAL_CTL                 0x391
+#define NHM_UNC_FIXED_CTR                       0x394
+#define NHM_UNC_FIXED_CTR_CTRL                  0x395
+
+/* NHM uncore global control */
+#define NHM_UNC_GLOBAL_CTL_EN_PC_ALL            ((1ULL << 8) - 1)
+#define NHM_UNC_GLOBAL_CTL_EN_FC                (1ULL << 32)
+
+/* NHM uncore register */
+#define NHM_UNC_PERFEVTSEL0                     0x3c0
+#define NHM_UNC_UNCORE_PMC0                     0x3b0
+
+/* SKL uncore global control */
+#define SKL_UNC_PERF_GLOBAL_CTL			0xe01
+#define SKL_UNC_GLOBAL_CTL_CORE_ALL		((1 << 5) - 1)
+
+/* ICL Cbo register */
+#define ICL_UNC_CBO_CONFIG			0x396
+#define ICL_UNC_NUM_CBO_MASK			0xf
+#define ICL_UNC_CBO_0_PER_CTR0			0x702
+#define ICL_UNC_CBO_MSR_OFFSET			0x8
+
+/* ICL ARB register */
+#define ICL_UNC_ARB_PER_CTR			0x3b1
+#define ICL_UNC_ARB_PERFEVTSEL			0x3b3
+
+DEFINE_UNCORE_FORMAT_ATTR(event, event, "config:0-7");
+DEFINE_UNCORE_FORMAT_ATTR(umask, umask, "config:8-15");
+DEFINE_UNCORE_FORMAT_ATTR(edge, edge, "config:18");
+DEFINE_UNCORE_FORMAT_ATTR(inv, inv, "config:23");
+DEFINE_UNCORE_FORMAT_ATTR(cmask5, cmask, "config:24-28");
+DEFINE_UNCORE_FORMAT_ATTR(cmask8, cmask, "config:24-31");
+
+/* Sandy Bridge uncore support */
+static void snb_uncore_msr_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->idx < UNCORE_PMC_IDX_FIXED)
+		wrmsrl(hwc->config_base, hwc->config | SNB_UNC_CTL_EN);
+	else
+		wrmsrl(hwc->config_base, SNB_UNC_CTL_EN);
+}
+
+static void snb_uncore_msr_disable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	wrmsrl(event->hw.config_base, 0);
+}
+
+static void snb_uncore_msr_init_box(struct intel_uncore_box *box)
+{
+	if (box->pmu->pmu_idx == 0) {
+		wrmsrl(SNB_UNC_PERF_GLOBAL_CTL,
+			SNB_UNC_GLOBAL_CTL_EN | SNB_UNC_GLOBAL_CTL_CORE_ALL);
+	}
+}
+
+static void snb_uncore_msr_enable_box(struct intel_uncore_box *box)
+{
+	wrmsrl(SNB_UNC_PERF_GLOBAL_CTL,
+		SNB_UNC_GLOBAL_CTL_EN | SNB_UNC_GLOBAL_CTL_CORE_ALL);
+}
+
+static void snb_uncore_msr_exit_box(struct intel_uncore_box *box)
+{
+	if (box->pmu->pmu_idx == 0)
+		wrmsrl(SNB_UNC_PERF_GLOBAL_CTL, 0);
+}
+
+static struct uncore_event_desc snb_uncore_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks, "event=0xff,umask=0x00"),
+	{ /* end: all zeroes */ },
+};
+
+static struct attribute *snb_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask5.attr,
+	NULL,
+};
+
+static const struct attribute_group snb_uncore_format_group = {
+	.name		= "format",
+	.attrs		= snb_uncore_formats_attr,
+};
+
+static struct intel_uncore_ops snb_uncore_msr_ops = {
+	.init_box	= snb_uncore_msr_init_box,
+	.enable_box	= snb_uncore_msr_enable_box,
+	.exit_box	= snb_uncore_msr_exit_box,
+	.disable_event	= snb_uncore_msr_disable_event,
+	.enable_event	= snb_uncore_msr_enable_event,
+	.read_counter	= uncore_msr_read_counter,
+};
+
+static struct event_constraint snb_uncore_arb_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x80, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x83, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type snb_uncore_cbox = {
+	.name		= "cbox",
+	.num_counters   = 2,
+	.num_boxes	= 4,
+	.perf_ctr_bits	= 44,
+	.fixed_ctr_bits	= 48,
+	.perf_ctr	= SNB_UNC_CBO_0_PER_CTR0,
+	.event_ctl	= SNB_UNC_CBO_0_PERFEVTSEL0,
+	.fixed_ctr	= SNB_UNC_FIXED_CTR,
+	.fixed_ctl	= SNB_UNC_FIXED_CTR_CTRL,
+	.single_fixed	= 1,
+	.event_mask	= SNB_UNC_RAW_EVENT_MASK,
+	.msr_offset	= SNB_UNC_CBO_MSR_OFFSET,
+	.ops		= &snb_uncore_msr_ops,
+	.format_group	= &snb_uncore_format_group,
+	.event_descs	= snb_uncore_events,
+};
+
+static struct intel_uncore_type snb_uncore_arb = {
+	.name		= "arb",
+	.num_counters   = 2,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.perf_ctr	= SNB_UNC_ARB_PER_CTR0,
+	.event_ctl	= SNB_UNC_ARB_PERFEVTSEL0,
+	.event_mask	= SNB_UNC_RAW_EVENT_MASK,
+	.msr_offset	= SNB_UNC_ARB_MSR_OFFSET,
+	.constraints	= snb_uncore_arb_constraints,
+	.ops		= &snb_uncore_msr_ops,
+	.format_group	= &snb_uncore_format_group,
+};
+
+static struct intel_uncore_type *snb_msr_uncores[] = {
+	&snb_uncore_cbox,
+	&snb_uncore_arb,
+	NULL,
+};
+
+void snb_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = snb_msr_uncores;
+	if (snb_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		snb_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+}
+
+static void skl_uncore_msr_init_box(struct intel_uncore_box *box)
+{
+	if (box->pmu->pmu_idx == 0) {
+		wrmsrl(SKL_UNC_PERF_GLOBAL_CTL,
+			SNB_UNC_GLOBAL_CTL_EN | SKL_UNC_GLOBAL_CTL_CORE_ALL);
+	}
+
+	/* The 8th CBOX has different MSR space */
+	if (box->pmu->pmu_idx == 7)
+		__set_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags);
+}
+
+static void skl_uncore_msr_enable_box(struct intel_uncore_box *box)
+{
+	wrmsrl(SKL_UNC_PERF_GLOBAL_CTL,
+		SNB_UNC_GLOBAL_CTL_EN | SKL_UNC_GLOBAL_CTL_CORE_ALL);
+}
+
+static void skl_uncore_msr_exit_box(struct intel_uncore_box *box)
+{
+	if (box->pmu->pmu_idx == 0)
+		wrmsrl(SKL_UNC_PERF_GLOBAL_CTL, 0);
+}
+
+static struct intel_uncore_ops skl_uncore_msr_ops = {
+	.init_box	= skl_uncore_msr_init_box,
+	.enable_box	= skl_uncore_msr_enable_box,
+	.exit_box	= skl_uncore_msr_exit_box,
+	.disable_event	= snb_uncore_msr_disable_event,
+	.enable_event	= snb_uncore_msr_enable_event,
+	.read_counter	= uncore_msr_read_counter,
+};
+
+static struct intel_uncore_type skl_uncore_cbox = {
+	.name		= "cbox",
+	.num_counters   = 4,
+	.num_boxes	= 8,
+	.perf_ctr_bits	= 44,
+	.fixed_ctr_bits	= 48,
+	.perf_ctr	= SNB_UNC_CBO_0_PER_CTR0,
+	.event_ctl	= SNB_UNC_CBO_0_PERFEVTSEL0,
+	.fixed_ctr	= SNB_UNC_FIXED_CTR,
+	.fixed_ctl	= SNB_UNC_FIXED_CTR_CTRL,
+	.single_fixed	= 1,
+	.event_mask	= SNB_UNC_RAW_EVENT_MASK,
+	.msr_offset	= SNB_UNC_CBO_MSR_OFFSET,
+	.ops		= &skl_uncore_msr_ops,
+	.format_group	= &snb_uncore_format_group,
+	.event_descs	= snb_uncore_events,
+};
+
+static struct intel_uncore_type *skl_msr_uncores[] = {
+	&skl_uncore_cbox,
+	&snb_uncore_arb,
+	NULL,
+};
+
+void skl_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = skl_msr_uncores;
+	if (skl_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		skl_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+	snb_uncore_arb.ops = &skl_uncore_msr_ops;
+}
+
+static struct intel_uncore_ops icl_uncore_msr_ops = {
+	.disable_event	= snb_uncore_msr_disable_event,
+	.enable_event	= snb_uncore_msr_enable_event,
+	.read_counter	= uncore_msr_read_counter,
+};
+
+static struct intel_uncore_type icl_uncore_cbox = {
+	.name		= "cbox",
+	.num_counters   = 2,
+	.perf_ctr_bits	= 44,
+	.perf_ctr	= ICL_UNC_CBO_0_PER_CTR0,
+	.event_ctl	= SNB_UNC_CBO_0_PERFEVTSEL0,
+	.event_mask	= SNB_UNC_RAW_EVENT_MASK,
+	.msr_offset	= ICL_UNC_CBO_MSR_OFFSET,
+	.ops		= &icl_uncore_msr_ops,
+	.format_group	= &snb_uncore_format_group,
+};
+
+static struct uncore_event_desc icl_uncore_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks, "event=0xff"),
+	{ /* end: all zeroes */ },
+};
+
+static struct attribute *icl_uncore_clock_formats_attr[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+
+static struct attribute_group icl_uncore_clock_format_group = {
+	.name = "format",
+	.attrs = icl_uncore_clock_formats_attr,
+};
+
+static struct intel_uncore_type icl_uncore_clockbox = {
+	.name		= "clock",
+	.num_counters	= 1,
+	.num_boxes	= 1,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNB_UNC_FIXED_CTR,
+	.fixed_ctl	= SNB_UNC_FIXED_CTR_CTRL,
+	.single_fixed	= 1,
+	.event_mask	= SNB_UNC_CTL_EV_SEL_MASK,
+	.format_group	= &icl_uncore_clock_format_group,
+	.ops		= &icl_uncore_msr_ops,
+	.event_descs	= icl_uncore_events,
+};
+
+static struct intel_uncore_type icl_uncore_arb = {
+	.name		= "arb",
+	.num_counters   = 1,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.perf_ctr	= ICL_UNC_ARB_PER_CTR,
+	.event_ctl	= ICL_UNC_ARB_PERFEVTSEL,
+	.event_mask	= SNB_UNC_RAW_EVENT_MASK,
+	.ops		= &icl_uncore_msr_ops,
+	.format_group	= &snb_uncore_format_group,
+};
+
+static struct intel_uncore_type *icl_msr_uncores[] = {
+	&icl_uncore_cbox,
+	&icl_uncore_arb,
+	&icl_uncore_clockbox,
+	NULL,
+};
+
+static int icl_get_cbox_num(void)
+{
+	u64 num_boxes;
+
+	rdmsrl(ICL_UNC_CBO_CONFIG, num_boxes);
+
+	return num_boxes & ICL_UNC_NUM_CBO_MASK;
+}
+
+void icl_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = icl_msr_uncores;
+	icl_uncore_cbox.num_boxes = icl_get_cbox_num();
+}
+
+static struct intel_uncore_type *tgl_msr_uncores[] = {
+	&icl_uncore_cbox,
+	&snb_uncore_arb,
+	&icl_uncore_clockbox,
+	NULL,
+};
+
+void tgl_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = tgl_msr_uncores;
+	icl_uncore_cbox.num_boxes = icl_get_cbox_num();
+	icl_uncore_cbox.ops = &skl_uncore_msr_ops;
+	icl_uncore_clockbox.ops = &skl_uncore_msr_ops;
+	snb_uncore_arb.ops = &skl_uncore_msr_ops;
+}
+
+enum {
+	SNB_PCI_UNCORE_IMC,
+};
+
+static struct uncore_event_desc snb_uncore_imc_events[] = {
+	INTEL_UNCORE_EVENT_DESC(data_reads,  "event=0x01"),
+	INTEL_UNCORE_EVENT_DESC(data_reads.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(data_reads.unit, "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(data_writes, "event=0x02"),
+	INTEL_UNCORE_EVENT_DESC(data_writes.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(data_writes.unit, "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(gt_requests, "event=0x03"),
+	INTEL_UNCORE_EVENT_DESC(gt_requests.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(gt_requests.unit, "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(ia_requests, "event=0x04"),
+	INTEL_UNCORE_EVENT_DESC(ia_requests.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(ia_requests.unit, "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(io_requests, "event=0x05"),
+	INTEL_UNCORE_EVENT_DESC(io_requests.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(io_requests.unit, "MiB"),
+
+	{ /* end: all zeroes */ },
+};
+
+#define SNB_UNCORE_PCI_IMC_EVENT_MASK		0xff
+#define SNB_UNCORE_PCI_IMC_BAR_OFFSET		0x48
+
+/* page size multiple covering all config regs */
+#define SNB_UNCORE_PCI_IMC_MAP_SIZE		0x6000
+
+#define SNB_UNCORE_PCI_IMC_DATA_READS		0x1
+#define SNB_UNCORE_PCI_IMC_DATA_READS_BASE	0x5050
+#define SNB_UNCORE_PCI_IMC_DATA_WRITES		0x2
+#define SNB_UNCORE_PCI_IMC_DATA_WRITES_BASE	0x5054
+#define SNB_UNCORE_PCI_IMC_CTR_BASE		SNB_UNCORE_PCI_IMC_DATA_READS_BASE
+
+/* BW break down- legacy counters */
+#define SNB_UNCORE_PCI_IMC_GT_REQUESTS		0x3
+#define SNB_UNCORE_PCI_IMC_GT_REQUESTS_BASE	0x5040
+#define SNB_UNCORE_PCI_IMC_IA_REQUESTS		0x4
+#define SNB_UNCORE_PCI_IMC_IA_REQUESTS_BASE	0x5044
+#define SNB_UNCORE_PCI_IMC_IO_REQUESTS		0x5
+#define SNB_UNCORE_PCI_IMC_IO_REQUESTS_BASE	0x5048
+
+enum perf_snb_uncore_imc_freerunning_types {
+	SNB_PCI_UNCORE_IMC_DATA_READS		= 0,
+	SNB_PCI_UNCORE_IMC_DATA_WRITES,
+	SNB_PCI_UNCORE_IMC_GT_REQUESTS,
+	SNB_PCI_UNCORE_IMC_IA_REQUESTS,
+	SNB_PCI_UNCORE_IMC_IO_REQUESTS,
+
+	SNB_PCI_UNCORE_IMC_FREERUNNING_TYPE_MAX,
+};
+
+static struct freerunning_counters snb_uncore_imc_freerunning[] = {
+	[SNB_PCI_UNCORE_IMC_DATA_READS]		= { SNB_UNCORE_PCI_IMC_DATA_READS_BASE,
+							0x0, 0x0, 1, 32 },
+	[SNB_PCI_UNCORE_IMC_DATA_WRITES]	= { SNB_UNCORE_PCI_IMC_DATA_WRITES_BASE,
+							0x0, 0x0, 1, 32 },
+	[SNB_PCI_UNCORE_IMC_GT_REQUESTS]	= { SNB_UNCORE_PCI_IMC_GT_REQUESTS_BASE,
+							0x0, 0x0, 1, 32 },
+	[SNB_PCI_UNCORE_IMC_IA_REQUESTS]	= { SNB_UNCORE_PCI_IMC_IA_REQUESTS_BASE,
+							0x0, 0x0, 1, 32 },
+	[SNB_PCI_UNCORE_IMC_IO_REQUESTS]	= { SNB_UNCORE_PCI_IMC_IO_REQUESTS_BASE,
+							0x0, 0x0, 1, 32 },
+};
+
+static struct attribute *snb_uncore_imc_formats_attr[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+
+static const struct attribute_group snb_uncore_imc_format_group = {
+	.name = "format",
+	.attrs = snb_uncore_imc_formats_attr,
+};
+
+static void snb_uncore_imc_init_box(struct intel_uncore_box *box)
+{
+	struct intel_uncore_type *type = box->pmu->type;
+	struct pci_dev *pdev = box->pci_dev;
+	int where = SNB_UNCORE_PCI_IMC_BAR_OFFSET;
+	resource_size_t addr;
+	u32 pci_dword;
+
+	pci_read_config_dword(pdev, where, &pci_dword);
+	addr = pci_dword;
+
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	pci_read_config_dword(pdev, where + 4, &pci_dword);
+	addr |= ((resource_size_t)pci_dword << 32);
+#endif
+
+	addr &= ~(PAGE_SIZE - 1);
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
+	if (!box->io_addr)
+		pr_warn("perf uncore: Failed to ioremap for %s.\n", type->name);
+
+	box->hrtimer_duration = UNCORE_SNB_IMC_HRTIMER_INTERVAL;
+}
+
+static void snb_uncore_imc_enable_box(struct intel_uncore_box *box)
+{}
+
+static void snb_uncore_imc_disable_box(struct intel_uncore_box *box)
+{}
+
+static void snb_uncore_imc_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{}
+
+static void snb_uncore_imc_disable_event(struct intel_uncore_box *box, struct perf_event *event)
+{}
+
+/*
+ * Keep the custom event_init() function compatible with old event
+ * encoding for free running counters.
+ */
+static int snb_uncore_imc_event_init(struct perf_event *event)
+{
+	struct intel_uncore_pmu *pmu;
+	struct intel_uncore_box *box;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 cfg = event->attr.config & SNB_UNCORE_PCI_IMC_EVENT_MASK;
+	int idx, base;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	pmu = uncore_event_to_pmu(event);
+	/* no device found for this pmu */
+	if (pmu->func_id < 0)
+		return -ENOENT;
+
+	/* Sampling not supported yet */
+	if (hwc->sample_period)
+		return -EINVAL;
+
+	/* unsupported modes and filters */
+	if (event->attr.sample_period) /* no sampling */
+		return -EINVAL;
+
+	/*
+	 * Place all uncore events for a particular physical package
+	 * onto a single cpu
+	 */
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	/* check only supported bits are set */
+	if (event->attr.config & ~SNB_UNCORE_PCI_IMC_EVENT_MASK)
+		return -EINVAL;
+
+	box = uncore_pmu_to_box(pmu, event->cpu);
+	if (!box || box->cpu < 0)
+		return -EINVAL;
+
+	event->cpu = box->cpu;
+	event->pmu_private = box;
+
+	event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG;
+
+	event->hw.idx = -1;
+	event->hw.last_tag = ~0ULL;
+	event->hw.extra_reg.idx = EXTRA_REG_NONE;
+	event->hw.branch_reg.idx = EXTRA_REG_NONE;
+	/*
+	 * check event is known (whitelist, determines counter)
+	 */
+	switch (cfg) {
+	case SNB_UNCORE_PCI_IMC_DATA_READS:
+		base = SNB_UNCORE_PCI_IMC_DATA_READS_BASE;
+		idx = UNCORE_PMC_IDX_FREERUNNING;
+		break;
+	case SNB_UNCORE_PCI_IMC_DATA_WRITES:
+		base = SNB_UNCORE_PCI_IMC_DATA_WRITES_BASE;
+		idx = UNCORE_PMC_IDX_FREERUNNING;
+		break;
+	case SNB_UNCORE_PCI_IMC_GT_REQUESTS:
+		base = SNB_UNCORE_PCI_IMC_GT_REQUESTS_BASE;
+		idx = UNCORE_PMC_IDX_FREERUNNING;
+		break;
+	case SNB_UNCORE_PCI_IMC_IA_REQUESTS:
+		base = SNB_UNCORE_PCI_IMC_IA_REQUESTS_BASE;
+		idx = UNCORE_PMC_IDX_FREERUNNING;
+		break;
+	case SNB_UNCORE_PCI_IMC_IO_REQUESTS:
+		base = SNB_UNCORE_PCI_IMC_IO_REQUESTS_BASE;
+		idx = UNCORE_PMC_IDX_FREERUNNING;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* must be done before validate_group */
+	event->hw.event_base = base;
+	event->hw.idx = idx;
+
+	/* Convert to standard encoding format for freerunning counters */
+	event->hw.config = ((cfg - 1) << 8) | 0x10ff;
+
+	/* no group validation needed, we have free running counters */
+
+	return 0;
+}
+
+static int snb_uncore_imc_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return 0;
+}
+
+int snb_pci2phy_map_init(int devid)
+{
+	struct pci_dev *dev = NULL;
+	struct pci2phy_map *map;
+	int bus, segment;
+
+	dev = pci_get_device(PCI_VENDOR_ID_INTEL, devid, dev);
+	if (!dev)
+		return -ENOTTY;
+
+	bus = dev->bus->number;
+	segment = pci_domain_nr(dev->bus);
+
+	raw_spin_lock(&pci2phy_map_lock);
+	map = __find_pci2phy_map(segment);
+	if (!map) {
+		raw_spin_unlock(&pci2phy_map_lock);
+		pci_dev_put(dev);
+		return -ENOMEM;
+	}
+	map->pbus_to_physid[bus] = 0;
+	raw_spin_unlock(&pci2phy_map_lock);
+
+	pci_dev_put(dev);
+
+	return 0;
+}
+
+static u64 snb_uncore_imc_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	/*
+	 * SNB IMC counters are 32-bit and are laid out back to back
+	 * in MMIO space. Therefore we must use a 32-bit accessor function
+	 * using readq() from uncore_mmio_read_counter() causes problems
+	 * because it is reading 64-bit at a time. This is okay for the
+	 * uncore_perf_event_update() function because it drops the upper
+	 * 32-bits but not okay for plain uncore_read_counter() as invoked
+	 * in uncore_pmu_event_start().
+	 */
+	return (u64)readl(box->io_addr + hwc->event_base);
+}
+
+static struct pmu snb_uncore_imc_pmu = {
+	.task_ctx_nr	= perf_invalid_context,
+	.event_init	= snb_uncore_imc_event_init,
+	.add		= uncore_pmu_event_add,
+	.del		= uncore_pmu_event_del,
+	.start		= uncore_pmu_event_start,
+	.stop		= uncore_pmu_event_stop,
+	.read		= uncore_pmu_event_read,
+	.capabilities	= PERF_PMU_CAP_NO_EXCLUDE,
+};
+
+static struct intel_uncore_ops snb_uncore_imc_ops = {
+	.init_box	= snb_uncore_imc_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.enable_box	= snb_uncore_imc_enable_box,
+	.disable_box	= snb_uncore_imc_disable_box,
+	.disable_event	= snb_uncore_imc_disable_event,
+	.enable_event	= snb_uncore_imc_enable_event,
+	.hw_config	= snb_uncore_imc_hw_config,
+	.read_counter	= snb_uncore_imc_read_counter,
+};
+
+static struct intel_uncore_type snb_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 5,
+	.num_boxes	= 1,
+	.num_freerunning_types	= SNB_PCI_UNCORE_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size	= SNB_UNCORE_PCI_IMC_MAP_SIZE,
+	.freerunning	= snb_uncore_imc_freerunning,
+	.event_descs	= snb_uncore_imc_events,
+	.format_group	= &snb_uncore_imc_format_group,
+	.ops		= &snb_uncore_imc_ops,
+	.pmu		= &snb_uncore_imc_pmu,
+};
+
+static struct intel_uncore_type *snb_pci_uncores[] = {
+	[SNB_PCI_UNCORE_IMC]	= &snb_uncore_imc,
+	NULL,
+};
+
+static const struct pci_device_id snb_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SNB_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static const struct pci_device_id ivb_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_E3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static const struct pci_device_id hsw_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HSW_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HSW_U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static const struct pci_device_id bdw_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BDW_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static const struct pci_device_id skl_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_Y_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_HD_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_HQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_SD_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_SQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_E3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_Y_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_UQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_SD_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_SQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_HQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_WQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_2U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4H_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6H_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_2S_D_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_D_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_D_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_D_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_W_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_W_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_W_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_S_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_S_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_S_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_AML_YD_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_AML_YQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WHL_UQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WHL_4_UQ_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WHL_UD_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_H1_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_H2_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_H3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_U1_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_U2_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_U3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_S1_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_S2_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_S3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_S4_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CML_S5_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static const struct pci_device_id icl_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICL_U_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICL_U2_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static struct pci_driver snb_uncore_pci_driver = {
+	.name		= "snb_uncore",
+	.id_table	= snb_uncore_pci_ids,
+};
+
+static struct pci_driver ivb_uncore_pci_driver = {
+	.name		= "ivb_uncore",
+	.id_table	= ivb_uncore_pci_ids,
+};
+
+static struct pci_driver hsw_uncore_pci_driver = {
+	.name		= "hsw_uncore",
+	.id_table	= hsw_uncore_pci_ids,
+};
+
+static struct pci_driver bdw_uncore_pci_driver = {
+	.name		= "bdw_uncore",
+	.id_table	= bdw_uncore_pci_ids,
+};
+
+static struct pci_driver skl_uncore_pci_driver = {
+	.name		= "skl_uncore",
+	.id_table	= skl_uncore_pci_ids,
+};
+
+static struct pci_driver icl_uncore_pci_driver = {
+	.name		= "icl_uncore",
+	.id_table	= icl_uncore_pci_ids,
+};
+
+struct imc_uncore_pci_dev {
+	__u32 pci_id;
+	struct pci_driver *driver;
+};
+#define IMC_DEV(a, d) \
+	{ .pci_id = PCI_DEVICE_ID_INTEL_##a, .driver = (d) }
+
+static const struct imc_uncore_pci_dev desktop_imc_pci_ids[] = {
+	IMC_DEV(SNB_IMC, &snb_uncore_pci_driver),
+	IMC_DEV(IVB_IMC, &ivb_uncore_pci_driver),    /* 3rd Gen Core processor */
+	IMC_DEV(IVB_E3_IMC, &ivb_uncore_pci_driver), /* Xeon E3-1200 v2/3rd Gen Core processor */
+	IMC_DEV(HSW_IMC, &hsw_uncore_pci_driver),    /* 4th Gen Core Processor */
+	IMC_DEV(HSW_U_IMC, &hsw_uncore_pci_driver),  /* 4th Gen Core ULT Mobile Processor */
+	IMC_DEV(BDW_IMC, &bdw_uncore_pci_driver),    /* 5th Gen Core U */
+	IMC_DEV(SKL_Y_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core Y */
+	IMC_DEV(SKL_U_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core U */
+	IMC_DEV(SKL_HD_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core H Dual Core */
+	IMC_DEV(SKL_HQ_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core H Quad Core */
+	IMC_DEV(SKL_SD_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core S Dual Core */
+	IMC_DEV(SKL_SQ_IMC, &skl_uncore_pci_driver),  /* 6th Gen Core S Quad Core */
+	IMC_DEV(SKL_E3_IMC, &skl_uncore_pci_driver),  /* Xeon E3 V5 Gen Core processor */
+	IMC_DEV(KBL_Y_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core Y */
+	IMC_DEV(KBL_U_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core U */
+	IMC_DEV(KBL_UQ_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core U Quad Core */
+	IMC_DEV(KBL_SD_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core S Dual Core */
+	IMC_DEV(KBL_SQ_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core S Quad Core */
+	IMC_DEV(KBL_HQ_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core H Quad Core */
+	IMC_DEV(KBL_WQ_IMC, &skl_uncore_pci_driver),  /* 7th Gen Core S 4 cores Work Station */
+	IMC_DEV(CFL_2U_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core U 2 Cores */
+	IMC_DEV(CFL_4U_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core U 4 Cores */
+	IMC_DEV(CFL_4H_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core H 4 Cores */
+	IMC_DEV(CFL_6H_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core H 6 Cores */
+	IMC_DEV(CFL_2S_D_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 2 Cores Desktop */
+	IMC_DEV(CFL_4S_D_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 4 Cores Desktop */
+	IMC_DEV(CFL_6S_D_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 6 Cores Desktop */
+	IMC_DEV(CFL_8S_D_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 8 Cores Desktop */
+	IMC_DEV(CFL_4S_W_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 4 Cores Work Station */
+	IMC_DEV(CFL_6S_W_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 6 Cores Work Station */
+	IMC_DEV(CFL_8S_W_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 8 Cores Work Station */
+	IMC_DEV(CFL_4S_S_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 4 Cores Server */
+	IMC_DEV(CFL_6S_S_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 6 Cores Server */
+	IMC_DEV(CFL_8S_S_IMC, &skl_uncore_pci_driver),  /* 8th Gen Core S 8 Cores Server */
+	IMC_DEV(AML_YD_IMC, &skl_uncore_pci_driver),	/* 8th Gen Core Y Mobile Dual Core */
+	IMC_DEV(AML_YQ_IMC, &skl_uncore_pci_driver),	/* 8th Gen Core Y Mobile Quad Core */
+	IMC_DEV(WHL_UQ_IMC, &skl_uncore_pci_driver),	/* 8th Gen Core U Mobile Quad Core */
+	IMC_DEV(WHL_4_UQ_IMC, &skl_uncore_pci_driver),	/* 8th Gen Core U Mobile Quad Core */
+	IMC_DEV(WHL_UD_IMC, &skl_uncore_pci_driver),	/* 8th Gen Core U Mobile Dual Core */
+	IMC_DEV(CML_H1_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_H2_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_H3_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_U1_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_U2_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_U3_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_S1_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_S2_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_S3_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_S4_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(CML_S5_IMC, &skl_uncore_pci_driver),
+	IMC_DEV(ICL_U_IMC, &icl_uncore_pci_driver),	/* 10th Gen Core Mobile */
+	IMC_DEV(ICL_U2_IMC, &icl_uncore_pci_driver),	/* 10th Gen Core Mobile */
+	{  /* end marker */ }
+};
+
+
+#define for_each_imc_pci_id(x, t) \
+	for (x = (t); (x)->pci_id; x++)
+
+static struct pci_driver *imc_uncore_find_dev(void)
+{
+	const struct imc_uncore_pci_dev *p;
+	int ret;
+
+	for_each_imc_pci_id(p, desktop_imc_pci_ids) {
+		ret = snb_pci2phy_map_init(p->pci_id);
+		if (ret == 0)
+			return p->driver;
+	}
+	return NULL;
+}
+
+static int imc_uncore_pci_init(void)
+{
+	struct pci_driver *imc_drv = imc_uncore_find_dev();
+
+	if (!imc_drv)
+		return -ENODEV;
+
+	uncore_pci_uncores = snb_pci_uncores;
+	uncore_pci_driver = imc_drv;
+
+	return 0;
+}
+
+int snb_uncore_pci_init(void)
+{
+	return imc_uncore_pci_init();
+}
+
+int ivb_uncore_pci_init(void)
+{
+	return imc_uncore_pci_init();
+}
+int hsw_uncore_pci_init(void)
+{
+	return imc_uncore_pci_init();
+}
+
+int bdw_uncore_pci_init(void)
+{
+	return imc_uncore_pci_init();
+}
+
+int skl_uncore_pci_init(void)
+{
+	return imc_uncore_pci_init();
+}
+
+/* end of Sandy Bridge uncore support */
+
+/* Nehalem uncore support */
+static void nhm_uncore_msr_disable_box(struct intel_uncore_box *box)
+{
+	wrmsrl(NHM_UNC_PERF_GLOBAL_CTL, 0);
+}
+
+static void nhm_uncore_msr_enable_box(struct intel_uncore_box *box)
+{
+	wrmsrl(NHM_UNC_PERF_GLOBAL_CTL, NHM_UNC_GLOBAL_CTL_EN_PC_ALL | NHM_UNC_GLOBAL_CTL_EN_FC);
+}
+
+static void nhm_uncore_msr_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->idx < UNCORE_PMC_IDX_FIXED)
+		wrmsrl(hwc->config_base, hwc->config | SNB_UNC_CTL_EN);
+	else
+		wrmsrl(hwc->config_base, NHM_UNC_FIXED_CTR_CTL_EN);
+}
+
+static struct attribute *nhm_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask8.attr,
+	NULL,
+};
+
+static const struct attribute_group nhm_uncore_format_group = {
+	.name = "format",
+	.attrs = nhm_uncore_formats_attr,
+};
+
+static struct uncore_event_desc nhm_uncore_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks,                "event=0xff,umask=0x00"),
+	INTEL_UNCORE_EVENT_DESC(qmc_writes_full_any,       "event=0x2f,umask=0x0f"),
+	INTEL_UNCORE_EVENT_DESC(qmc_normal_reads_any,      "event=0x2c,umask=0x0f"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_ioh_reads,     "event=0x20,umask=0x01"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_ioh_writes,    "event=0x20,umask=0x02"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_remote_reads,  "event=0x20,umask=0x04"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_remote_writes, "event=0x20,umask=0x08"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_local_reads,   "event=0x20,umask=0x10"),
+	INTEL_UNCORE_EVENT_DESC(qhl_request_local_writes,  "event=0x20,umask=0x20"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_ops nhm_uncore_msr_ops = {
+	.disable_box	= nhm_uncore_msr_disable_box,
+	.enable_box	= nhm_uncore_msr_enable_box,
+	.disable_event	= snb_uncore_msr_disable_event,
+	.enable_event	= nhm_uncore_msr_enable_event,
+	.read_counter	= uncore_msr_read_counter,
+};
+
+static struct intel_uncore_type nhm_uncore = {
+	.name		= "",
+	.num_counters   = 8,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.event_ctl	= NHM_UNC_PERFEVTSEL0,
+	.perf_ctr	= NHM_UNC_UNCORE_PMC0,
+	.fixed_ctr	= NHM_UNC_FIXED_CTR,
+	.fixed_ctl	= NHM_UNC_FIXED_CTR_CTRL,
+	.event_mask	= NHM_UNC_RAW_EVENT_MASK,
+	.event_descs	= nhm_uncore_events,
+	.ops		= &nhm_uncore_msr_ops,
+	.format_group	= &nhm_uncore_format_group,
+};
+
+static struct intel_uncore_type *nhm_msr_uncores[] = {
+	&nhm_uncore,
+	NULL,
+};
+
+void nhm_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = nhm_msr_uncores;
+}
+
+/* end of Nehalem uncore support */
+
+/* Tiger Lake MMIO uncore support */
+
+static const struct pci_device_id tgl_uncore_pci_ids[] = {
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGL_U1_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGL_U2_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGL_U3_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGL_U4_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* IMC */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_TGL_H_IMC),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+	},
+	{ /* end: all zeroes */ }
+};
+
+enum perf_tgl_uncore_imc_freerunning_types {
+	TGL_MMIO_UNCORE_IMC_DATA_TOTAL,
+	TGL_MMIO_UNCORE_IMC_DATA_READ,
+	TGL_MMIO_UNCORE_IMC_DATA_WRITE,
+	TGL_MMIO_UNCORE_IMC_FREERUNNING_TYPE_MAX
+};
+
+static struct freerunning_counters tgl_l_uncore_imc_freerunning[] = {
+	[TGL_MMIO_UNCORE_IMC_DATA_TOTAL]	= { 0x5040, 0x0, 0x0, 1, 64 },
+	[TGL_MMIO_UNCORE_IMC_DATA_READ]		= { 0x5058, 0x0, 0x0, 1, 64 },
+	[TGL_MMIO_UNCORE_IMC_DATA_WRITE]	= { 0x50A0, 0x0, 0x0, 1, 64 },
+};
+
+static struct freerunning_counters tgl_uncore_imc_freerunning[] = {
+	[TGL_MMIO_UNCORE_IMC_DATA_TOTAL]	= { 0xd840, 0x0, 0x0, 1, 64 },
+	[TGL_MMIO_UNCORE_IMC_DATA_READ]		= { 0xd858, 0x0, 0x0, 1, 64 },
+	[TGL_MMIO_UNCORE_IMC_DATA_WRITE]	= { 0xd8A0, 0x0, 0x0, 1, 64 },
+};
+
+static struct uncore_event_desc tgl_uncore_imc_events[] = {
+	INTEL_UNCORE_EVENT_DESC(data_total,         "event=0xff,umask=0x10"),
+	INTEL_UNCORE_EVENT_DESC(data_total.scale,   "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(data_total.unit,    "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(data_read,         "event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(data_read.scale,   "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(data_read.unit,    "MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(data_write,        "event=0xff,umask=0x30"),
+	INTEL_UNCORE_EVENT_DESC(data_write.scale,  "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(data_write.unit,   "MiB"),
+
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_dev *tgl_uncore_get_mc_dev(void)
+{
+	const struct pci_device_id *ids = tgl_uncore_pci_ids;
+	struct pci_dev *mc_dev = NULL;
+
+	while (ids && ids->vendor) {
+		mc_dev = pci_get_device(PCI_VENDOR_ID_INTEL, ids->device, NULL);
+		if (mc_dev)
+			return mc_dev;
+		ids++;
+	}
+
+	return mc_dev;
+}
+
+#define TGL_UNCORE_MMIO_IMC_MEM_OFFSET		0x10000
+#define TGL_UNCORE_PCI_IMC_MAP_SIZE		0xe000
+
+static void tgl_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = tgl_uncore_get_mc_dev();
+	struct intel_uncore_pmu *pmu = box->pmu;
+	struct intel_uncore_type *type = pmu->type;
+	resource_size_t addr;
+	u32 mch_bar;
+
+	if (!pdev) {
+		pr_warn("perf uncore: Cannot find matched IMC device.\n");
+		return;
+	}
+
+	pci_read_config_dword(pdev, SNB_UNCORE_PCI_IMC_BAR_OFFSET, &mch_bar);
+	/* MCHBAR is disabled */
+	if (!(mch_bar & BIT(0))) {
+		pr_warn("perf uncore: MCHBAR is disabled. Failed to map IMC free-running counters.\n");
+		pci_dev_put(pdev);
+		return;
+	}
+	mch_bar &= ~BIT(0);
+	addr = (resource_size_t)(mch_bar + TGL_UNCORE_MMIO_IMC_MEM_OFFSET * pmu->pmu_idx);
+
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+	pci_read_config_dword(pdev, SNB_UNCORE_PCI_IMC_BAR_OFFSET + 4, &mch_bar);
+	addr |= ((resource_size_t)mch_bar << 32);
+#endif
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
+	if (!box->io_addr)
+		pr_warn("perf uncore: Failed to ioremap for %s.\n", type->name);
+
+	pci_dev_put(pdev);
+}
+
+static struct intel_uncore_ops tgl_uncore_imc_freerunning_ops = {
+	.init_box	= tgl_uncore_imc_freerunning_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.read_counter	= uncore_mmio_read_counter,
+	.hw_config	= uncore_freerunning_hw_config,
+};
+
+static struct attribute *tgl_uncore_imc_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	NULL
+};
+
+static const struct attribute_group tgl_uncore_imc_format_group = {
+	.name = "format",
+	.attrs = tgl_uncore_imc_formats_attr,
+};
+
+static struct intel_uncore_type tgl_uncore_imc_free_running = {
+	.name			= "imc_free_running",
+	.num_counters		= 3,
+	.num_boxes		= 2,
+	.num_freerunning_types	= TGL_MMIO_UNCORE_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= TGL_UNCORE_PCI_IMC_MAP_SIZE,
+	.freerunning		= tgl_uncore_imc_freerunning,
+	.ops			= &tgl_uncore_imc_freerunning_ops,
+	.event_descs		= tgl_uncore_imc_events,
+	.format_group		= &tgl_uncore_imc_format_group,
+};
+
+static struct intel_uncore_type *tgl_mmio_uncores[] = {
+	&tgl_uncore_imc_free_running,
+	NULL
+};
+
+void tgl_l_uncore_mmio_init(void)
+{
+	tgl_uncore_imc_free_running.freerunning = tgl_l_uncore_imc_freerunning;
+	uncore_mmio_uncores = tgl_mmio_uncores;
+}
+
+void tgl_uncore_mmio_init(void)
+{
+	uncore_mmio_uncores = tgl_mmio_uncores;
+}
+
+/* end of Tiger Lake MMIO uncore support */
diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c
new file mode 100644
index 000000000..ad084a5a1
--- /dev/null
+++ b/arch/x86/events/intel/uncore_snbep.c
@@ -0,0 +1,5356 @@
+// SPDX-License-Identifier: GPL-2.0
+/* SandyBridge-EP/IvyTown uncore support */
+#include "uncore.h"
+
+/* SNB-EP pci bus to socket mapping */
+#define SNBEP_CPUNODEID			0x40
+#define SNBEP_GIDNIDMAP			0x54
+
+/* SNB-EP Box level control */
+#define SNBEP_PMON_BOX_CTL_RST_CTRL	(1 << 0)
+#define SNBEP_PMON_BOX_CTL_RST_CTRS	(1 << 1)
+#define SNBEP_PMON_BOX_CTL_FRZ		(1 << 8)
+#define SNBEP_PMON_BOX_CTL_FRZ_EN	(1 << 16)
+#define SNBEP_PMON_BOX_CTL_INT		(SNBEP_PMON_BOX_CTL_RST_CTRL | \
+					 SNBEP_PMON_BOX_CTL_RST_CTRS | \
+					 SNBEP_PMON_BOX_CTL_FRZ_EN)
+/* SNB-EP event control */
+#define SNBEP_PMON_CTL_EV_SEL_MASK	0x000000ff
+#define SNBEP_PMON_CTL_UMASK_MASK	0x0000ff00
+#define SNBEP_PMON_CTL_RST		(1 << 17)
+#define SNBEP_PMON_CTL_EDGE_DET		(1 << 18)
+#define SNBEP_PMON_CTL_EV_SEL_EXT	(1 << 21)
+#define SNBEP_PMON_CTL_EN		(1 << 22)
+#define SNBEP_PMON_CTL_INVERT		(1 << 23)
+#define SNBEP_PMON_CTL_TRESH_MASK	0xff000000
+#define SNBEP_PMON_RAW_EVENT_MASK	(SNBEP_PMON_CTL_EV_SEL_MASK | \
+					 SNBEP_PMON_CTL_UMASK_MASK | \
+					 SNBEP_PMON_CTL_EDGE_DET | \
+					 SNBEP_PMON_CTL_INVERT | \
+					 SNBEP_PMON_CTL_TRESH_MASK)
+
+/* SNB-EP Ubox event control */
+#define SNBEP_U_MSR_PMON_CTL_TRESH_MASK		0x1f000000
+#define SNBEP_U_MSR_PMON_RAW_EVENT_MASK		\
+				(SNBEP_PMON_CTL_EV_SEL_MASK | \
+				 SNBEP_PMON_CTL_UMASK_MASK | \
+				 SNBEP_PMON_CTL_EDGE_DET | \
+				 SNBEP_PMON_CTL_INVERT | \
+				 SNBEP_U_MSR_PMON_CTL_TRESH_MASK)
+
+#define SNBEP_CBO_PMON_CTL_TID_EN		(1 << 19)
+#define SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK	(SNBEP_PMON_RAW_EVENT_MASK | \
+						 SNBEP_CBO_PMON_CTL_TID_EN)
+
+/* SNB-EP PCU event control */
+#define SNBEP_PCU_MSR_PMON_CTL_OCC_SEL_MASK	0x0000c000
+#define SNBEP_PCU_MSR_PMON_CTL_TRESH_MASK	0x1f000000
+#define SNBEP_PCU_MSR_PMON_CTL_OCC_INVERT	(1 << 30)
+#define SNBEP_PCU_MSR_PMON_CTL_OCC_EDGE_DET	(1 << 31)
+#define SNBEP_PCU_MSR_PMON_RAW_EVENT_MASK	\
+				(SNBEP_PMON_CTL_EV_SEL_MASK | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_SEL_MASK | \
+				 SNBEP_PMON_CTL_EDGE_DET | \
+				 SNBEP_PMON_CTL_INVERT | \
+				 SNBEP_PCU_MSR_PMON_CTL_TRESH_MASK | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_INVERT | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_EDGE_DET)
+
+#define SNBEP_QPI_PCI_PMON_RAW_EVENT_MASK	\
+				(SNBEP_PMON_RAW_EVENT_MASK | \
+				 SNBEP_PMON_CTL_EV_SEL_EXT)
+
+/* SNB-EP pci control register */
+#define SNBEP_PCI_PMON_BOX_CTL			0xf4
+#define SNBEP_PCI_PMON_CTL0			0xd8
+/* SNB-EP pci counter register */
+#define SNBEP_PCI_PMON_CTR0			0xa0
+
+/* SNB-EP home agent register */
+#define SNBEP_HA_PCI_PMON_BOX_ADDRMATCH0	0x40
+#define SNBEP_HA_PCI_PMON_BOX_ADDRMATCH1	0x44
+#define SNBEP_HA_PCI_PMON_BOX_OPCODEMATCH	0x48
+/* SNB-EP memory controller register */
+#define SNBEP_MC_CHy_PCI_PMON_FIXED_CTL		0xf0
+#define SNBEP_MC_CHy_PCI_PMON_FIXED_CTR		0xd0
+/* SNB-EP QPI register */
+#define SNBEP_Q_Py_PCI_PMON_PKT_MATCH0		0x228
+#define SNBEP_Q_Py_PCI_PMON_PKT_MATCH1		0x22c
+#define SNBEP_Q_Py_PCI_PMON_PKT_MASK0		0x238
+#define SNBEP_Q_Py_PCI_PMON_PKT_MASK1		0x23c
+
+/* SNB-EP Ubox register */
+#define SNBEP_U_MSR_PMON_CTR0			0xc16
+#define SNBEP_U_MSR_PMON_CTL0			0xc10
+
+#define SNBEP_U_MSR_PMON_UCLK_FIXED_CTL		0xc08
+#define SNBEP_U_MSR_PMON_UCLK_FIXED_CTR		0xc09
+
+/* SNB-EP Cbo register */
+#define SNBEP_C0_MSR_PMON_CTR0			0xd16
+#define SNBEP_C0_MSR_PMON_CTL0			0xd10
+#define SNBEP_C0_MSR_PMON_BOX_CTL		0xd04
+#define SNBEP_C0_MSR_PMON_BOX_FILTER		0xd14
+#define SNBEP_CBO_MSR_OFFSET			0x20
+
+#define SNBEP_CB0_MSR_PMON_BOX_FILTER_TID	0x1f
+#define SNBEP_CB0_MSR_PMON_BOX_FILTER_NID	0x3fc00
+#define SNBEP_CB0_MSR_PMON_BOX_FILTER_STATE	0x7c0000
+#define SNBEP_CB0_MSR_PMON_BOX_FILTER_OPC	0xff800000
+
+#define SNBEP_CBO_EVENT_EXTRA_REG(e, m, i) {	\
+	.event = (e),				\
+	.msr = SNBEP_C0_MSR_PMON_BOX_FILTER,	\
+	.config_mask = (m),			\
+	.idx = (i)				\
+}
+
+/* SNB-EP PCU register */
+#define SNBEP_PCU_MSR_PMON_CTR0			0xc36
+#define SNBEP_PCU_MSR_PMON_CTL0			0xc30
+#define SNBEP_PCU_MSR_PMON_BOX_CTL		0xc24
+#define SNBEP_PCU_MSR_PMON_BOX_FILTER		0xc34
+#define SNBEP_PCU_MSR_PMON_BOX_FILTER_MASK	0xffffffff
+#define SNBEP_PCU_MSR_CORE_C3_CTR		0x3fc
+#define SNBEP_PCU_MSR_CORE_C6_CTR		0x3fd
+
+/* IVBEP event control */
+#define IVBEP_PMON_BOX_CTL_INT		(SNBEP_PMON_BOX_CTL_RST_CTRL | \
+					 SNBEP_PMON_BOX_CTL_RST_CTRS)
+#define IVBEP_PMON_RAW_EVENT_MASK		(SNBEP_PMON_CTL_EV_SEL_MASK | \
+					 SNBEP_PMON_CTL_UMASK_MASK | \
+					 SNBEP_PMON_CTL_EDGE_DET | \
+					 SNBEP_PMON_CTL_TRESH_MASK)
+/* IVBEP Ubox */
+#define IVBEP_U_MSR_PMON_GLOBAL_CTL		0xc00
+#define IVBEP_U_PMON_GLOBAL_FRZ_ALL		(1 << 31)
+#define IVBEP_U_PMON_GLOBAL_UNFRZ_ALL		(1 << 29)
+
+#define IVBEP_U_MSR_PMON_RAW_EVENT_MASK	\
+				(SNBEP_PMON_CTL_EV_SEL_MASK | \
+				 SNBEP_PMON_CTL_UMASK_MASK | \
+				 SNBEP_PMON_CTL_EDGE_DET | \
+				 SNBEP_U_MSR_PMON_CTL_TRESH_MASK)
+/* IVBEP Cbo */
+#define IVBEP_CBO_MSR_PMON_RAW_EVENT_MASK		(IVBEP_PMON_RAW_EVENT_MASK | \
+						 SNBEP_CBO_PMON_CTL_TID_EN)
+
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_TID		(0x1fULL << 0)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_LINK	(0xfULL << 5)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_STATE	(0x3fULL << 17)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_NID		(0xffffULL << 32)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_OPC		(0x1ffULL << 52)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_C6		(0x1ULL << 61)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_NC		(0x1ULL << 62)
+#define IVBEP_CB0_MSR_PMON_BOX_FILTER_ISOC	(0x1ULL << 63)
+
+/* IVBEP home agent */
+#define IVBEP_HA_PCI_PMON_CTL_Q_OCC_RST		(1 << 16)
+#define IVBEP_HA_PCI_PMON_RAW_EVENT_MASK		\
+				(IVBEP_PMON_RAW_EVENT_MASK | \
+				 IVBEP_HA_PCI_PMON_CTL_Q_OCC_RST)
+/* IVBEP PCU */
+#define IVBEP_PCU_MSR_PMON_RAW_EVENT_MASK	\
+				(SNBEP_PMON_CTL_EV_SEL_MASK | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_SEL_MASK | \
+				 SNBEP_PMON_CTL_EDGE_DET | \
+				 SNBEP_PCU_MSR_PMON_CTL_TRESH_MASK | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_INVERT | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_EDGE_DET)
+/* IVBEP QPI */
+#define IVBEP_QPI_PCI_PMON_RAW_EVENT_MASK	\
+				(IVBEP_PMON_RAW_EVENT_MASK | \
+				 SNBEP_PMON_CTL_EV_SEL_EXT)
+
+#define __BITS_VALUE(x, i, n)  ((typeof(x))(((x) >> ((i) * (n))) & \
+				((1ULL << (n)) - 1)))
+
+/* Haswell-EP Ubox */
+#define HSWEP_U_MSR_PMON_CTR0			0x709
+#define HSWEP_U_MSR_PMON_CTL0			0x705
+#define HSWEP_U_MSR_PMON_FILTER			0x707
+
+#define HSWEP_U_MSR_PMON_UCLK_FIXED_CTL		0x703
+#define HSWEP_U_MSR_PMON_UCLK_FIXED_CTR		0x704
+
+#define HSWEP_U_MSR_PMON_BOX_FILTER_TID		(0x1 << 0)
+#define HSWEP_U_MSR_PMON_BOX_FILTER_CID		(0x1fULL << 1)
+#define HSWEP_U_MSR_PMON_BOX_FILTER_MASK \
+					(HSWEP_U_MSR_PMON_BOX_FILTER_TID | \
+					 HSWEP_U_MSR_PMON_BOX_FILTER_CID)
+
+/* Haswell-EP CBo */
+#define HSWEP_C0_MSR_PMON_CTR0			0xe08
+#define HSWEP_C0_MSR_PMON_CTL0			0xe01
+#define HSWEP_C0_MSR_PMON_BOX_CTL			0xe00
+#define HSWEP_C0_MSR_PMON_BOX_FILTER0		0xe05
+#define HSWEP_CBO_MSR_OFFSET			0x10
+
+
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_TID		(0x3fULL << 0)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_LINK	(0xfULL << 6)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_STATE	(0x7fULL << 17)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_NID		(0xffffULL << 32)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_OPC		(0x1ffULL << 52)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_C6		(0x1ULL << 61)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_NC		(0x1ULL << 62)
+#define HSWEP_CB0_MSR_PMON_BOX_FILTER_ISOC	(0x1ULL << 63)
+
+
+/* Haswell-EP Sbox */
+#define HSWEP_S0_MSR_PMON_CTR0			0x726
+#define HSWEP_S0_MSR_PMON_CTL0			0x721
+#define HSWEP_S0_MSR_PMON_BOX_CTL			0x720
+#define HSWEP_SBOX_MSR_OFFSET			0xa
+#define HSWEP_S_MSR_PMON_RAW_EVENT_MASK		(SNBEP_PMON_RAW_EVENT_MASK | \
+						 SNBEP_CBO_PMON_CTL_TID_EN)
+
+/* Haswell-EP PCU */
+#define HSWEP_PCU_MSR_PMON_CTR0			0x717
+#define HSWEP_PCU_MSR_PMON_CTL0			0x711
+#define HSWEP_PCU_MSR_PMON_BOX_CTL		0x710
+#define HSWEP_PCU_MSR_PMON_BOX_FILTER		0x715
+
+/* KNL Ubox */
+#define KNL_U_MSR_PMON_RAW_EVENT_MASK \
+					(SNBEP_U_MSR_PMON_RAW_EVENT_MASK | \
+						SNBEP_CBO_PMON_CTL_TID_EN)
+/* KNL CHA */
+#define KNL_CHA_MSR_OFFSET			0xc
+#define KNL_CHA_MSR_PMON_CTL_QOR		(1 << 16)
+#define KNL_CHA_MSR_PMON_RAW_EVENT_MASK \
+					(SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK | \
+					 KNL_CHA_MSR_PMON_CTL_QOR)
+#define KNL_CHA_MSR_PMON_BOX_FILTER_TID		0x1ff
+#define KNL_CHA_MSR_PMON_BOX_FILTER_STATE	(7 << 18)
+#define KNL_CHA_MSR_PMON_BOX_FILTER_OP		(0xfffffe2aULL << 32)
+#define KNL_CHA_MSR_PMON_BOX_FILTER_REMOTE_NODE	(0x1ULL << 32)
+#define KNL_CHA_MSR_PMON_BOX_FILTER_LOCAL_NODE	(0x1ULL << 33)
+#define KNL_CHA_MSR_PMON_BOX_FILTER_NNC		(0x1ULL << 37)
+
+/* KNL EDC/MC UCLK */
+#define KNL_UCLK_MSR_PMON_CTR0_LOW		0x400
+#define KNL_UCLK_MSR_PMON_CTL0			0x420
+#define KNL_UCLK_MSR_PMON_BOX_CTL		0x430
+#define KNL_UCLK_MSR_PMON_UCLK_FIXED_LOW	0x44c
+#define KNL_UCLK_MSR_PMON_UCLK_FIXED_CTL	0x454
+#define KNL_PMON_FIXED_CTL_EN			0x1
+
+/* KNL EDC */
+#define KNL_EDC0_ECLK_MSR_PMON_CTR0_LOW		0xa00
+#define KNL_EDC0_ECLK_MSR_PMON_CTL0		0xa20
+#define KNL_EDC0_ECLK_MSR_PMON_BOX_CTL		0xa30
+#define KNL_EDC0_ECLK_MSR_PMON_ECLK_FIXED_LOW	0xa3c
+#define KNL_EDC0_ECLK_MSR_PMON_ECLK_FIXED_CTL	0xa44
+
+/* KNL MC */
+#define KNL_MC0_CH0_MSR_PMON_CTR0_LOW		0xb00
+#define KNL_MC0_CH0_MSR_PMON_CTL0		0xb20
+#define KNL_MC0_CH0_MSR_PMON_BOX_CTL		0xb30
+#define KNL_MC0_CH0_MSR_PMON_FIXED_LOW		0xb3c
+#define KNL_MC0_CH0_MSR_PMON_FIXED_CTL		0xb44
+
+/* KNL IRP */
+#define KNL_IRP_PCI_PMON_BOX_CTL		0xf0
+#define KNL_IRP_PCI_PMON_RAW_EVENT_MASK		(SNBEP_PMON_RAW_EVENT_MASK | \
+						 KNL_CHA_MSR_PMON_CTL_QOR)
+/* KNL PCU */
+#define KNL_PCU_PMON_CTL_EV_SEL_MASK		0x0000007f
+#define KNL_PCU_PMON_CTL_USE_OCC_CTR		(1 << 7)
+#define KNL_PCU_MSR_PMON_CTL_TRESH_MASK		0x3f000000
+#define KNL_PCU_MSR_PMON_RAW_EVENT_MASK	\
+				(KNL_PCU_PMON_CTL_EV_SEL_MASK | \
+				 KNL_PCU_PMON_CTL_USE_OCC_CTR | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_SEL_MASK | \
+				 SNBEP_PMON_CTL_EDGE_DET | \
+				 SNBEP_CBO_PMON_CTL_TID_EN | \
+				 SNBEP_PMON_CTL_INVERT | \
+				 KNL_PCU_MSR_PMON_CTL_TRESH_MASK | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_INVERT | \
+				 SNBEP_PCU_MSR_PMON_CTL_OCC_EDGE_DET)
+
+/* SKX pci bus to socket mapping */
+#define SKX_CPUNODEID			0xc0
+#define SKX_GIDNIDMAP			0xd4
+
+/*
+ * The CPU_BUS_NUMBER MSR returns the values of the respective CPUBUSNO CSR
+ * that BIOS programmed. MSR has package scope.
+ * |  Bit  |  Default  |  Description
+ * | [63]  |    00h    | VALID - When set, indicates the CPU bus
+ *                       numbers have been initialized. (RO)
+ * |[62:48]|    ---    | Reserved
+ * |[47:40]|    00h    | BUS_NUM_5 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(5). (RO)
+ * |[39:32]|    00h    | BUS_NUM_4 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(4). (RO)
+ * |[31:24]|    00h    | BUS_NUM_3 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(3). (RO)
+ * |[23:16]|    00h    | BUS_NUM_2 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(2). (RO)
+ * |[15:8] |    00h    | BUS_NUM_1 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(1). (RO)
+ * | [7:0] |    00h    | BUS_NUM_0 — Return the bus number BIOS assigned
+ *                       CPUBUSNO(0). (RO)
+ */
+#define SKX_MSR_CPU_BUS_NUMBER		0x300
+#define SKX_MSR_CPU_BUS_VALID_BIT	(1ULL << 63)
+#define BUS_NUM_STRIDE			8
+
+/* SKX CHA */
+#define SKX_CHA_MSR_PMON_BOX_FILTER_TID		(0x1ffULL << 0)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_LINK	(0xfULL << 9)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_STATE	(0x3ffULL << 17)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_REM		(0x1ULL << 32)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_LOC		(0x1ULL << 33)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_ALL_OPC	(0x1ULL << 35)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_NM		(0x1ULL << 36)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_NOT_NM	(0x1ULL << 37)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_OPC0	(0x3ffULL << 41)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_OPC1	(0x3ffULL << 51)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_C6		(0x1ULL << 61)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_NC		(0x1ULL << 62)
+#define SKX_CHA_MSR_PMON_BOX_FILTER_ISOC	(0x1ULL << 63)
+
+/* SKX IIO */
+#define SKX_IIO0_MSR_PMON_CTL0		0xa48
+#define SKX_IIO0_MSR_PMON_CTR0		0xa41
+#define SKX_IIO0_MSR_PMON_BOX_CTL	0xa40
+#define SKX_IIO_MSR_OFFSET		0x20
+
+#define SKX_PMON_CTL_TRESH_MASK		(0xff << 24)
+#define SKX_PMON_CTL_TRESH_MASK_EXT	(0xf)
+#define SKX_PMON_CTL_CH_MASK		(0xff << 4)
+#define SKX_PMON_CTL_FC_MASK		(0x7 << 12)
+#define SKX_IIO_PMON_RAW_EVENT_MASK	(SNBEP_PMON_CTL_EV_SEL_MASK | \
+					 SNBEP_PMON_CTL_UMASK_MASK | \
+					 SNBEP_PMON_CTL_EDGE_DET | \
+					 SNBEP_PMON_CTL_INVERT | \
+					 SKX_PMON_CTL_TRESH_MASK)
+#define SKX_IIO_PMON_RAW_EVENT_MASK_EXT	(SKX_PMON_CTL_TRESH_MASK_EXT | \
+					 SKX_PMON_CTL_CH_MASK | \
+					 SKX_PMON_CTL_FC_MASK)
+
+/* SKX IRP */
+#define SKX_IRP0_MSR_PMON_CTL0		0xa5b
+#define SKX_IRP0_MSR_PMON_CTR0		0xa59
+#define SKX_IRP0_MSR_PMON_BOX_CTL	0xa58
+#define SKX_IRP_MSR_OFFSET		0x20
+
+/* SKX UPI */
+#define SKX_UPI_PCI_PMON_CTL0		0x350
+#define SKX_UPI_PCI_PMON_CTR0		0x318
+#define SKX_UPI_PCI_PMON_BOX_CTL	0x378
+#define SKX_UPI_CTL_UMASK_EXT		0xffefff
+
+/* SKX M2M */
+#define SKX_M2M_PCI_PMON_CTL0		0x228
+#define SKX_M2M_PCI_PMON_CTR0		0x200
+#define SKX_M2M_PCI_PMON_BOX_CTL	0x258
+
+/* SNR Ubox */
+#define SNR_U_MSR_PMON_CTR0			0x1f98
+#define SNR_U_MSR_PMON_CTL0			0x1f91
+#define SNR_U_MSR_PMON_UCLK_FIXED_CTL		0x1f93
+#define SNR_U_MSR_PMON_UCLK_FIXED_CTR		0x1f94
+
+/* SNR CHA */
+#define SNR_CHA_RAW_EVENT_MASK_EXT		0x3ffffff
+#define SNR_CHA_MSR_PMON_CTL0			0x1c01
+#define SNR_CHA_MSR_PMON_CTR0			0x1c08
+#define SNR_CHA_MSR_PMON_BOX_CTL		0x1c00
+#define SNR_C0_MSR_PMON_BOX_FILTER0		0x1c05
+
+
+/* SNR IIO */
+#define SNR_IIO_MSR_PMON_CTL0			0x1e08
+#define SNR_IIO_MSR_PMON_CTR0			0x1e01
+#define SNR_IIO_MSR_PMON_BOX_CTL		0x1e00
+#define SNR_IIO_MSR_OFFSET			0x10
+#define SNR_IIO_PMON_RAW_EVENT_MASK_EXT		0x7ffff
+
+/* SNR IRP */
+#define SNR_IRP0_MSR_PMON_CTL0			0x1ea8
+#define SNR_IRP0_MSR_PMON_CTR0			0x1ea1
+#define SNR_IRP0_MSR_PMON_BOX_CTL		0x1ea0
+#define SNR_IRP_MSR_OFFSET			0x10
+
+/* SNR M2PCIE */
+#define SNR_M2PCIE_MSR_PMON_CTL0		0x1e58
+#define SNR_M2PCIE_MSR_PMON_CTR0		0x1e51
+#define SNR_M2PCIE_MSR_PMON_BOX_CTL		0x1e50
+#define SNR_M2PCIE_MSR_OFFSET			0x10
+
+/* SNR PCU */
+#define SNR_PCU_MSR_PMON_CTL0			0x1ef1
+#define SNR_PCU_MSR_PMON_CTR0			0x1ef8
+#define SNR_PCU_MSR_PMON_BOX_CTL		0x1ef0
+#define SNR_PCU_MSR_PMON_BOX_FILTER		0x1efc
+
+/* SNR M2M */
+#define SNR_M2M_PCI_PMON_CTL0			0x468
+#define SNR_M2M_PCI_PMON_CTR0			0x440
+#define SNR_M2M_PCI_PMON_BOX_CTL		0x438
+#define SNR_M2M_PCI_PMON_UMASK_EXT		0xff
+
+/* SNR PCIE3 */
+#define SNR_PCIE3_PCI_PMON_CTL0			0x508
+#define SNR_PCIE3_PCI_PMON_CTR0			0x4e8
+#define SNR_PCIE3_PCI_PMON_BOX_CTL		0x4e0
+
+/* SNR IMC */
+#define SNR_IMC_MMIO_PMON_FIXED_CTL		0x54
+#define SNR_IMC_MMIO_PMON_FIXED_CTR		0x38
+#define SNR_IMC_MMIO_PMON_CTL0			0x40
+#define SNR_IMC_MMIO_PMON_CTR0			0x8
+#define SNR_IMC_MMIO_PMON_BOX_CTL		0x22800
+#define SNR_IMC_MMIO_OFFSET			0x4000
+#define SNR_IMC_MMIO_SIZE			0x4000
+#define SNR_IMC_MMIO_BASE_OFFSET		0xd0
+#define SNR_IMC_MMIO_BASE_MASK			0x1FFFFFFF
+#define SNR_IMC_MMIO_MEM0_OFFSET		0xd8
+#define SNR_IMC_MMIO_MEM0_MASK			0x7FF
+
+/* ICX CHA */
+#define ICX_C34_MSR_PMON_CTR0			0xb68
+#define ICX_C34_MSR_PMON_CTL0			0xb61
+#define ICX_C34_MSR_PMON_BOX_CTL		0xb60
+#define ICX_C34_MSR_PMON_BOX_FILTER0		0xb65
+
+/* ICX IIO */
+#define ICX_IIO_MSR_PMON_CTL0			0xa58
+#define ICX_IIO_MSR_PMON_CTR0			0xa51
+#define ICX_IIO_MSR_PMON_BOX_CTL		0xa50
+
+/* ICX IRP */
+#define ICX_IRP0_MSR_PMON_CTL0			0xa4d
+#define ICX_IRP0_MSR_PMON_CTR0			0xa4b
+#define ICX_IRP0_MSR_PMON_BOX_CTL		0xa4a
+
+/* ICX M2PCIE */
+#define ICX_M2PCIE_MSR_PMON_CTL0		0xa46
+#define ICX_M2PCIE_MSR_PMON_CTR0		0xa41
+#define ICX_M2PCIE_MSR_PMON_BOX_CTL		0xa40
+
+/* ICX UPI */
+#define ICX_UPI_PCI_PMON_CTL0			0x350
+#define ICX_UPI_PCI_PMON_CTR0			0x320
+#define ICX_UPI_PCI_PMON_BOX_CTL		0x318
+#define ICX_UPI_CTL_UMASK_EXT			0xffffff
+
+/* ICX M3UPI*/
+#define ICX_M3UPI_PCI_PMON_CTL0			0xd8
+#define ICX_M3UPI_PCI_PMON_CTR0			0xa8
+#define ICX_M3UPI_PCI_PMON_BOX_CTL		0xa0
+
+/* ICX IMC */
+#define ICX_NUMBER_IMC_CHN			3
+#define ICX_IMC_MEM_STRIDE			0x4
+
+DEFINE_UNCORE_FORMAT_ATTR(event, event, "config:0-7");
+DEFINE_UNCORE_FORMAT_ATTR(event2, event, "config:0-6");
+DEFINE_UNCORE_FORMAT_ATTR(event_ext, event, "config:0-7,21");
+DEFINE_UNCORE_FORMAT_ATTR(use_occ_ctr, use_occ_ctr, "config:7");
+DEFINE_UNCORE_FORMAT_ATTR(umask, umask, "config:8-15");
+DEFINE_UNCORE_FORMAT_ATTR(umask_ext, umask, "config:8-15,32-43,45-55");
+DEFINE_UNCORE_FORMAT_ATTR(umask_ext2, umask, "config:8-15,32-57");
+DEFINE_UNCORE_FORMAT_ATTR(umask_ext3, umask, "config:8-15,32-39");
+DEFINE_UNCORE_FORMAT_ATTR(umask_ext4, umask, "config:8-15,32-55");
+DEFINE_UNCORE_FORMAT_ATTR(qor, qor, "config:16");
+DEFINE_UNCORE_FORMAT_ATTR(edge, edge, "config:18");
+DEFINE_UNCORE_FORMAT_ATTR(tid_en, tid_en, "config:19");
+DEFINE_UNCORE_FORMAT_ATTR(inv, inv, "config:23");
+DEFINE_UNCORE_FORMAT_ATTR(thresh9, thresh, "config:24-35");
+DEFINE_UNCORE_FORMAT_ATTR(thresh8, thresh, "config:24-31");
+DEFINE_UNCORE_FORMAT_ATTR(thresh6, thresh, "config:24-29");
+DEFINE_UNCORE_FORMAT_ATTR(thresh5, thresh, "config:24-28");
+DEFINE_UNCORE_FORMAT_ATTR(occ_sel, occ_sel, "config:14-15");
+DEFINE_UNCORE_FORMAT_ATTR(occ_invert, occ_invert, "config:30");
+DEFINE_UNCORE_FORMAT_ATTR(occ_edge, occ_edge, "config:14-51");
+DEFINE_UNCORE_FORMAT_ATTR(occ_edge_det, occ_edge_det, "config:31");
+DEFINE_UNCORE_FORMAT_ATTR(ch_mask, ch_mask, "config:36-43");
+DEFINE_UNCORE_FORMAT_ATTR(ch_mask2, ch_mask, "config:36-47");
+DEFINE_UNCORE_FORMAT_ATTR(fc_mask, fc_mask, "config:44-46");
+DEFINE_UNCORE_FORMAT_ATTR(fc_mask2, fc_mask, "config:48-50");
+DEFINE_UNCORE_FORMAT_ATTR(filter_tid, filter_tid, "config1:0-4");
+DEFINE_UNCORE_FORMAT_ATTR(filter_tid2, filter_tid, "config1:0");
+DEFINE_UNCORE_FORMAT_ATTR(filter_tid3, filter_tid, "config1:0-5");
+DEFINE_UNCORE_FORMAT_ATTR(filter_tid4, filter_tid, "config1:0-8");
+DEFINE_UNCORE_FORMAT_ATTR(filter_tid5, filter_tid, "config1:0-9");
+DEFINE_UNCORE_FORMAT_ATTR(filter_cid, filter_cid, "config1:5");
+DEFINE_UNCORE_FORMAT_ATTR(filter_link, filter_link, "config1:5-8");
+DEFINE_UNCORE_FORMAT_ATTR(filter_link2, filter_link, "config1:6-8");
+DEFINE_UNCORE_FORMAT_ATTR(filter_link3, filter_link, "config1:12");
+DEFINE_UNCORE_FORMAT_ATTR(filter_nid, filter_nid, "config1:10-17");
+DEFINE_UNCORE_FORMAT_ATTR(filter_nid2, filter_nid, "config1:32-47");
+DEFINE_UNCORE_FORMAT_ATTR(filter_state, filter_state, "config1:18-22");
+DEFINE_UNCORE_FORMAT_ATTR(filter_state2, filter_state, "config1:17-22");
+DEFINE_UNCORE_FORMAT_ATTR(filter_state3, filter_state, "config1:17-23");
+DEFINE_UNCORE_FORMAT_ATTR(filter_state4, filter_state, "config1:18-20");
+DEFINE_UNCORE_FORMAT_ATTR(filter_state5, filter_state, "config1:17-26");
+DEFINE_UNCORE_FORMAT_ATTR(filter_rem, filter_rem, "config1:32");
+DEFINE_UNCORE_FORMAT_ATTR(filter_loc, filter_loc, "config1:33");
+DEFINE_UNCORE_FORMAT_ATTR(filter_nm, filter_nm, "config1:36");
+DEFINE_UNCORE_FORMAT_ATTR(filter_not_nm, filter_not_nm, "config1:37");
+DEFINE_UNCORE_FORMAT_ATTR(filter_local, filter_local, "config1:33");
+DEFINE_UNCORE_FORMAT_ATTR(filter_all_op, filter_all_op, "config1:35");
+DEFINE_UNCORE_FORMAT_ATTR(filter_nnm, filter_nnm, "config1:37");
+DEFINE_UNCORE_FORMAT_ATTR(filter_opc, filter_opc, "config1:23-31");
+DEFINE_UNCORE_FORMAT_ATTR(filter_opc2, filter_opc, "config1:52-60");
+DEFINE_UNCORE_FORMAT_ATTR(filter_opc3, filter_opc, "config1:41-60");
+DEFINE_UNCORE_FORMAT_ATTR(filter_opc_0, filter_opc0, "config1:41-50");
+DEFINE_UNCORE_FORMAT_ATTR(filter_opc_1, filter_opc1, "config1:51-60");
+DEFINE_UNCORE_FORMAT_ATTR(filter_nc, filter_nc, "config1:62");
+DEFINE_UNCORE_FORMAT_ATTR(filter_c6, filter_c6, "config1:61");
+DEFINE_UNCORE_FORMAT_ATTR(filter_isoc, filter_isoc, "config1:63");
+DEFINE_UNCORE_FORMAT_ATTR(filter_band0, filter_band0, "config1:0-7");
+DEFINE_UNCORE_FORMAT_ATTR(filter_band1, filter_band1, "config1:8-15");
+DEFINE_UNCORE_FORMAT_ATTR(filter_band2, filter_band2, "config1:16-23");
+DEFINE_UNCORE_FORMAT_ATTR(filter_band3, filter_band3, "config1:24-31");
+DEFINE_UNCORE_FORMAT_ATTR(match_rds, match_rds, "config1:48-51");
+DEFINE_UNCORE_FORMAT_ATTR(match_rnid30, match_rnid30, "config1:32-35");
+DEFINE_UNCORE_FORMAT_ATTR(match_rnid4, match_rnid4, "config1:31");
+DEFINE_UNCORE_FORMAT_ATTR(match_dnid, match_dnid, "config1:13-17");
+DEFINE_UNCORE_FORMAT_ATTR(match_mc, match_mc, "config1:9-12");
+DEFINE_UNCORE_FORMAT_ATTR(match_opc, match_opc, "config1:5-8");
+DEFINE_UNCORE_FORMAT_ATTR(match_vnw, match_vnw, "config1:3-4");
+DEFINE_UNCORE_FORMAT_ATTR(match0, match0, "config1:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(match1, match1, "config1:32-63");
+DEFINE_UNCORE_FORMAT_ATTR(mask_rds, mask_rds, "config2:48-51");
+DEFINE_UNCORE_FORMAT_ATTR(mask_rnid30, mask_rnid30, "config2:32-35");
+DEFINE_UNCORE_FORMAT_ATTR(mask_rnid4, mask_rnid4, "config2:31");
+DEFINE_UNCORE_FORMAT_ATTR(mask_dnid, mask_dnid, "config2:13-17");
+DEFINE_UNCORE_FORMAT_ATTR(mask_mc, mask_mc, "config2:9-12");
+DEFINE_UNCORE_FORMAT_ATTR(mask_opc, mask_opc, "config2:5-8");
+DEFINE_UNCORE_FORMAT_ATTR(mask_vnw, mask_vnw, "config2:3-4");
+DEFINE_UNCORE_FORMAT_ATTR(mask0, mask0, "config2:0-31");
+DEFINE_UNCORE_FORMAT_ATTR(mask1, mask1, "config2:32-63");
+
+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 = 0;
+
+	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 = 0;
+
+	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)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	pci_write_config_dword(pdev, hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static void snbep_uncore_pci_disable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	pci_write_config_dword(pdev, hwc->config_base, hwc->config);
+}
+
+static u64 snbep_uncore_pci_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+	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);
+
+	return count;
+}
+
+static void snbep_uncore_pci_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	int box_ctl = uncore_pci_box_ctl(box);
+
+	pci_write_config_dword(pdev, box_ctl, SNBEP_PMON_BOX_CTL_INT);
+}
+
+static void snbep_uncore_msr_disable_box(struct intel_uncore_box *box)
+{
+	u64 config;
+	unsigned msr;
+
+	msr = uncore_msr_box_ctl(box);
+	if (msr) {
+		rdmsrl(msr, config);
+		config |= SNBEP_PMON_BOX_CTL_FRZ;
+		wrmsrl(msr, config);
+	}
+}
+
+static void snbep_uncore_msr_enable_box(struct intel_uncore_box *box)
+{
+	u64 config;
+	unsigned msr;
+
+	msr = uncore_msr_box_ctl(box);
+	if (msr) {
+		rdmsrl(msr, config);
+		config &= ~SNBEP_PMON_BOX_CTL_FRZ;
+		wrmsrl(msr, config);
+	}
+}
+
+static void snbep_uncore_msr_enable_event(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;
+
+	if (reg1->idx != EXTRA_REG_NONE)
+		wrmsrl(reg1->reg, uncore_shared_reg_config(box, 0));
+
+	wrmsrl(hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static void snbep_uncore_msr_disable_event(struct intel_uncore_box *box,
+					struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	wrmsrl(hwc->config_base, hwc->config);
+}
+
+static void snbep_uncore_msr_init_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+
+	if (msr)
+		wrmsrl(msr, SNBEP_PMON_BOX_CTL_INT);
+}
+
+static struct attribute *snbep_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static struct attribute *snbep_uncore_ubox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh5.attr,
+	NULL,
+};
+
+static struct attribute *snbep_uncore_cbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid.attr,
+	&format_attr_filter_nid.attr,
+	&format_attr_filter_state.attr,
+	&format_attr_filter_opc.attr,
+	NULL,
+};
+
+static struct attribute *snbep_uncore_pcu_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_occ_sel.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh5.attr,
+	&format_attr_occ_invert.attr,
+	&format_attr_occ_edge.attr,
+	&format_attr_filter_band0.attr,
+	&format_attr_filter_band1.attr,
+	&format_attr_filter_band2.attr,
+	&format_attr_filter_band3.attr,
+	NULL,
+};
+
+static struct attribute *snbep_uncore_qpi_formats_attr[] = {
+	&format_attr_event_ext.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_match_rds.attr,
+	&format_attr_match_rnid30.attr,
+	&format_attr_match_rnid4.attr,
+	&format_attr_match_dnid.attr,
+	&format_attr_match_mc.attr,
+	&format_attr_match_opc.attr,
+	&format_attr_match_vnw.attr,
+	&format_attr_match0.attr,
+	&format_attr_match1.attr,
+	&format_attr_mask_rds.attr,
+	&format_attr_mask_rnid30.attr,
+	&format_attr_mask_rnid4.attr,
+	&format_attr_mask_dnid.attr,
+	&format_attr_mask_mc.attr,
+	&format_attr_mask_opc.attr,
+	&format_attr_mask_vnw.attr,
+	&format_attr_mask0.attr,
+	&format_attr_mask1.attr,
+	NULL,
+};
+
+static struct uncore_event_desc snbep_uncore_imc_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks,      "event=0xff,umask=0x00"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read,  "event=0x04,umask=0x03"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.unit, "MiB"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write, "event=0x04,umask=0x0c"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.unit, "MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct uncore_event_desc snbep_uncore_qpi_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks,       "event=0x14"),
+	INTEL_UNCORE_EVENT_DESC(txl_flits_active, "event=0x00,umask=0x06"),
+	INTEL_UNCORE_EVENT_DESC(drs_data,         "event=0x102,umask=0x08"),
+	INTEL_UNCORE_EVENT_DESC(ncb_data,         "event=0x103,umask=0x04"),
+	{ /* end: all zeroes */ },
+};
+
+static const struct attribute_group snbep_uncore_format_group = {
+	.name = "format",
+	.attrs = snbep_uncore_formats_attr,
+};
+
+static const struct attribute_group snbep_uncore_ubox_format_group = {
+	.name = "format",
+	.attrs = snbep_uncore_ubox_formats_attr,
+};
+
+static const struct attribute_group snbep_uncore_cbox_format_group = {
+	.name = "format",
+	.attrs = snbep_uncore_cbox_formats_attr,
+};
+
+static const struct attribute_group snbep_uncore_pcu_format_group = {
+	.name = "format",
+	.attrs = snbep_uncore_pcu_formats_attr,
+};
+
+static const struct attribute_group snbep_uncore_qpi_format_group = {
+	.name = "format",
+	.attrs = snbep_uncore_qpi_formats_attr,
+};
+
+#define __SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
+	.disable_box	= snbep_uncore_msr_disable_box,		\
+	.enable_box	= snbep_uncore_msr_enable_box,		\
+	.disable_event	= snbep_uncore_msr_disable_event,	\
+	.enable_event	= snbep_uncore_msr_enable_event,	\
+	.read_counter	= uncore_msr_read_counter
+
+#define SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),			\
+	.init_box	= snbep_uncore_msr_init_box		\
+
+static struct intel_uncore_ops snbep_uncore_msr_ops = {
+	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+};
+
+#define SNBEP_UNCORE_PCI_OPS_COMMON_INIT()			\
+	.init_box	= snbep_uncore_pci_init_box,		\
+	.disable_box	= snbep_uncore_pci_disable_box,		\
+	.enable_box	= snbep_uncore_pci_enable_box,		\
+	.disable_event	= snbep_uncore_pci_disable_event,	\
+	.read_counter	= snbep_uncore_pci_read_counter
+
+static struct intel_uncore_ops snbep_uncore_pci_ops = {
+	SNBEP_UNCORE_PCI_OPS_COMMON_INIT(),
+	.enable_event	= snbep_uncore_pci_enable_event,	\
+};
+
+static struct event_constraint snbep_uncore_cbox_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x01, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x02, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x04, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x05, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x07, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x09, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x12, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x1b, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0x1c, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0x1d, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0x1e, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0x1f, 0xe),
+	UNCORE_EVENT_CONSTRAINT(0x21, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x31, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x32, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x35, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x37, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x38, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x39, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x3b, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint snbep_uncore_r2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x12, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x24, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x32, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint snbep_uncore_r3qpi_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x12, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x20, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x21, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x22, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x24, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x28, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x29, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2a, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2b, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2c, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2e, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2f, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x30, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x31, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x32, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x37, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x38, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x39, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type snbep_uncore_ubox = {
+	.name		= "ubox",
+	.num_counters   = 2,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.fixed_ctr_bits	= 48,
+	.perf_ctr	= SNBEP_U_MSR_PMON_CTR0,
+	.event_ctl	= SNBEP_U_MSR_PMON_CTL0,
+	.event_mask	= SNBEP_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr	= SNBEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl	= SNBEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.ops		= &snbep_uncore_msr_ops,
+	.format_group	= &snbep_uncore_ubox_format_group,
+};
+
+static struct extra_reg snbep_uncore_cbox_extra_regs[] = {
+	SNBEP_CBO_EVENT_EXTRA_REG(SNBEP_CBO_PMON_CTL_TID_EN,
+				  SNBEP_CBO_PMON_CTL_TID_EN, 0x1),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0334, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4334, 0xffff, 0x6),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0534, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4534, 0xffff, 0x6),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0934, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4934, 0xffff, 0x6),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4134, 0xffff, 0x6),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0135, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0335, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4135, 0xffff, 0xa),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4335, 0xffff, 0xa),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4435, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4835, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a35, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5035, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0136, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0336, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4136, 0xffff, 0xa),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4336, 0xffff, 0xa),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4436, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4836, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a36, 0xffff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4037, 0x40ff, 0x2),
+	EVENT_EXTRA_END
+};
+
+static void snbep_cbox_put_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct intel_uncore_extra_reg *er = &box->shared_regs[0];
+	int i;
+
+	if (uncore_box_is_fake(box))
+		return;
+
+	for (i = 0; i < 5; i++) {
+		if (reg1->alloc & (0x1 << i))
+			atomic_sub(1 << (i * 6), &er->ref);
+	}
+	reg1->alloc = 0;
+}
+
+static struct event_constraint *
+__snbep_cbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event,
+			    u64 (*cbox_filter_mask)(int fields))
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct intel_uncore_extra_reg *er = &box->shared_regs[0];
+	int i, alloc = 0;
+	unsigned long flags;
+	u64 mask;
+
+	if (reg1->idx == EXTRA_REG_NONE)
+		return NULL;
+
+	raw_spin_lock_irqsave(&er->lock, flags);
+	for (i = 0; i < 5; i++) {
+		if (!(reg1->idx & (0x1 << i)))
+			continue;
+		if (!uncore_box_is_fake(box) && (reg1->alloc & (0x1 << i)))
+			continue;
+
+		mask = cbox_filter_mask(0x1 << i);
+		if (!__BITS_VALUE(atomic_read(&er->ref), i, 6) ||
+		    !((reg1->config ^ er->config) & mask)) {
+			atomic_add(1 << (i * 6), &er->ref);
+			er->config &= ~mask;
+			er->config |= reg1->config & mask;
+			alloc |= (0x1 << i);
+		} else {
+			break;
+		}
+	}
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+	if (i < 5)
+		goto fail;
+
+	if (!uncore_box_is_fake(box))
+		reg1->alloc |= alloc;
+
+	return NULL;
+fail:
+	for (; i >= 0; i--) {
+		if (alloc & (0x1 << i))
+			atomic_sub(1 << (i * 6), &er->ref);
+	}
+	return &uncore_constraint_empty;
+}
+
+static u64 snbep_cbox_filter_mask(int fields)
+{
+	u64 mask = 0;
+
+	if (fields & 0x1)
+		mask |= SNBEP_CB0_MSR_PMON_BOX_FILTER_TID;
+	if (fields & 0x2)
+		mask |= SNBEP_CB0_MSR_PMON_BOX_FILTER_NID;
+	if (fields & 0x4)
+		mask |= SNBEP_CB0_MSR_PMON_BOX_FILTER_STATE;
+	if (fields & 0x8)
+		mask |= SNBEP_CB0_MSR_PMON_BOX_FILTER_OPC;
+
+	return mask;
+}
+
+static struct event_constraint *
+snbep_cbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return __snbep_cbox_get_constraint(box, event, snbep_cbox_filter_mask);
+}
+
+static int snbep_cbox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct extra_reg *er;
+	int idx = 0;
+
+	for (er = snbep_uncore_cbox_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		idx |= er->idx;
+	}
+
+	if (idx) {
+		reg1->reg = SNBEP_C0_MSR_PMON_BOX_FILTER +
+			SNBEP_CBO_MSR_OFFSET * box->pmu->pmu_idx;
+		reg1->config = event->attr.config1 & snbep_cbox_filter_mask(idx);
+		reg1->idx = idx;
+	}
+	return 0;
+}
+
+static struct intel_uncore_ops snbep_uncore_cbox_ops = {
+	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= snbep_cbox_hw_config,
+	.get_constraint		= snbep_cbox_get_constraint,
+	.put_constraint		= snbep_cbox_put_constraint,
+};
+
+static struct intel_uncore_type snbep_uncore_cbox = {
+	.name			= "cbox",
+	.num_counters		= 4,
+	.num_boxes		= 8,
+	.perf_ctr_bits		= 44,
+	.event_ctl		= SNBEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= SNBEP_C0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= SNBEP_CBO_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= snbep_uncore_cbox_constraints,
+	.ops			= &snbep_uncore_cbox_ops,
+	.format_group		= &snbep_uncore_cbox_format_group,
+};
+
+static u64 snbep_pcu_alter_er(struct perf_event *event, int new_idx, bool modify)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct hw_perf_event_extra *reg1 = &hwc->extra_reg;
+	u64 config = reg1->config;
+
+	if (new_idx > reg1->idx)
+		config <<= 8 * (new_idx - reg1->idx);
+	else
+		config >>= 8 * (reg1->idx - new_idx);
+
+	if (modify) {
+		hwc->config += new_idx - reg1->idx;
+		reg1->config = config;
+		reg1->idx = new_idx;
+	}
+	return config;
+}
+
+static struct event_constraint *
+snbep_pcu_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct intel_uncore_extra_reg *er = &box->shared_regs[0];
+	unsigned long flags;
+	int idx = reg1->idx;
+	u64 mask, config1 = reg1->config;
+	bool ok = false;
+
+	if (reg1->idx == EXTRA_REG_NONE ||
+	    (!uncore_box_is_fake(box) && reg1->alloc))
+		return NULL;
+again:
+	mask = 0xffULL << (idx * 8);
+	raw_spin_lock_irqsave(&er->lock, flags);
+	if (!__BITS_VALUE(atomic_read(&er->ref), idx, 8) ||
+	    !((config1 ^ er->config) & mask)) {
+		atomic_add(1 << (idx * 8), &er->ref);
+		er->config &= ~mask;
+		er->config |= config1 & mask;
+		ok = true;
+	}
+	raw_spin_unlock_irqrestore(&er->lock, flags);
+
+	if (!ok) {
+		idx = (idx + 1) % 4;
+		if (idx != reg1->idx) {
+			config1 = snbep_pcu_alter_er(event, idx, false);
+			goto again;
+		}
+		return &uncore_constraint_empty;
+	}
+
+	if (!uncore_box_is_fake(box)) {
+		if (idx != reg1->idx)
+			snbep_pcu_alter_er(event, idx, true);
+		reg1->alloc = 1;
+	}
+	return NULL;
+}
+
+static void snbep_pcu_put_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct intel_uncore_extra_reg *er = &box->shared_regs[0];
+
+	if (uncore_box_is_fake(box) || !reg1->alloc)
+		return;
+
+	atomic_sub(1 << (reg1->idx * 8), &er->ref);
+	reg1->alloc = 0;
+}
+
+static int snbep_pcu_hw_config(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 ev_sel = hwc->config & SNBEP_PMON_CTL_EV_SEL_MASK;
+
+	if (ev_sel >= 0xb && ev_sel <= 0xe) {
+		reg1->reg = SNBEP_PCU_MSR_PMON_BOX_FILTER;
+		reg1->idx = ev_sel - 0xb;
+		reg1->config = event->attr.config1 & (0xff << (reg1->idx * 8));
+	}
+	return 0;
+}
+
+static struct intel_uncore_ops snbep_uncore_pcu_ops = {
+	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= snbep_pcu_hw_config,
+	.get_constraint		= snbep_pcu_get_constraint,
+	.put_constraint		= snbep_pcu_put_constraint,
+};
+
+static struct intel_uncore_type snbep_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCU_MSR_PMON_CTR0,
+	.event_ctl		= SNBEP_PCU_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PCU_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCU_MSR_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &snbep_uncore_pcu_ops,
+	.format_group		= &snbep_uncore_pcu_format_group,
+};
+
+static struct intel_uncore_type *snbep_msr_uncores[] = {
+	&snbep_uncore_ubox,
+	&snbep_uncore_cbox,
+	&snbep_uncore_pcu,
+	NULL,
+};
+
+void snbep_uncore_cpu_init(void)
+{
+	if (snbep_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		snbep_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+	uncore_msr_uncores = snbep_msr_uncores;
+}
+
+enum {
+	SNBEP_PCI_QPI_PORT0_FILTER,
+	SNBEP_PCI_QPI_PORT1_FILTER,
+	BDX_PCI_QPI_PORT2_FILTER,
+};
+
+static int snbep_qpi_hw_config(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;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+
+	if ((hwc->config & SNBEP_PMON_CTL_EV_SEL_MASK) == 0x38) {
+		reg1->idx = 0;
+		reg1->reg = SNBEP_Q_Py_PCI_PMON_PKT_MATCH0;
+		reg1->config = event->attr.config1;
+		reg2->reg = SNBEP_Q_Py_PCI_PMON_PKT_MASK0;
+		reg2->config = event->attr.config2;
+	}
+	return 0;
+}
+
+static void snbep_qpi_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+	struct hw_perf_event_extra *reg1 = &hwc->extra_reg;
+	struct hw_perf_event_extra *reg2 = &hwc->branch_reg;
+
+	if (reg1->idx != EXTRA_REG_NONE) {
+		int idx = box->pmu->pmu_idx + SNBEP_PCI_QPI_PORT0_FILTER;
+		int die = box->dieid;
+		struct pci_dev *filter_pdev = uncore_extra_pci_dev[die].dev[idx];
+
+		if (filter_pdev) {
+			pci_write_config_dword(filter_pdev, reg1->reg,
+						(u32)reg1->config);
+			pci_write_config_dword(filter_pdev, reg1->reg + 4,
+						(u32)(reg1->config >> 32));
+			pci_write_config_dword(filter_pdev, reg2->reg,
+						(u32)reg2->config);
+			pci_write_config_dword(filter_pdev, reg2->reg + 4,
+						(u32)(reg2->config >> 32));
+		}
+	}
+
+	pci_write_config_dword(pdev, hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops snbep_uncore_qpi_ops = {
+	SNBEP_UNCORE_PCI_OPS_COMMON_INIT(),
+	.enable_event		= snbep_qpi_enable_event,
+	.hw_config		= snbep_qpi_hw_config,
+	.get_constraint		= uncore_get_constraint,
+	.put_constraint		= uncore_put_constraint,
+};
+
+#define SNBEP_UNCORE_PCI_COMMON_INIT()				\
+	.perf_ctr	= SNBEP_PCI_PMON_CTR0,			\
+	.event_ctl	= SNBEP_PCI_PMON_CTL0,			\
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,		\
+	.box_ctl	= SNBEP_PCI_PMON_BOX_CTL,		\
+	.ops		= &snbep_uncore_pci_ops,		\
+	.format_group	= &snbep_uncore_format_group
+
+static struct intel_uncore_type snbep_uncore_ha = {
+	.name		= "ha",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type snbep_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 4,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTR,
+	.fixed_ctl	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTL,
+	.event_descs	= snbep_uncore_imc_events,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type snbep_uncore_qpi = {
+	.name			= "qpi",
+	.num_counters		= 4,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCI_PMON_CTR0,
+	.event_ctl		= SNBEP_PCI_PMON_CTL0,
+	.event_mask		= SNBEP_QPI_PCI_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &snbep_uncore_qpi_ops,
+	.event_descs		= snbep_uncore_qpi_events,
+	.format_group		= &snbep_uncore_qpi_format_group,
+};
+
+
+static struct intel_uncore_type snbep_uncore_r2pcie = {
+	.name		= "r2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.constraints	= snbep_uncore_r2pcie_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type snbep_uncore_r3qpi = {
+	.name		= "r3qpi",
+	.num_counters   = 3,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 44,
+	.constraints	= snbep_uncore_r3qpi_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+enum {
+	SNBEP_PCI_UNCORE_HA,
+	SNBEP_PCI_UNCORE_IMC,
+	SNBEP_PCI_UNCORE_QPI,
+	SNBEP_PCI_UNCORE_R2PCIE,
+	SNBEP_PCI_UNCORE_R3QPI,
+};
+
+static struct intel_uncore_type *snbep_pci_uncores[] = {
+	[SNBEP_PCI_UNCORE_HA]		= &snbep_uncore_ha,
+	[SNBEP_PCI_UNCORE_IMC]		= &snbep_uncore_imc,
+	[SNBEP_PCI_UNCORE_QPI]		= &snbep_uncore_qpi,
+	[SNBEP_PCI_UNCORE_R2PCIE]	= &snbep_uncore_r2pcie,
+	[SNBEP_PCI_UNCORE_R3QPI]	= &snbep_uncore_r3qpi,
+	NULL,
+};
+
+static const struct pci_device_id snbep_uncore_pci_ids[] = {
+	{ /* Home Agent */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_HA),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_HA, 0),
+	},
+	{ /* MC Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_IMC0),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_IMC, 0),
+	},
+	{ /* MC Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_IMC1),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_IMC, 1),
+	},
+	{ /* MC Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_IMC2),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_IMC, 2),
+	},
+	{ /* MC Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_IMC3),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_IMC, 3),
+	},
+	{ /* QPI Port 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_QPI0),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_QPI, 0),
+	},
+	{ /* QPI Port 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_QPI1),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_QPI, 1),
+	},
+	{ /* R2PCIe */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_R2PCIE),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_R2PCIE, 0),
+	},
+	{ /* R3QPI Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_R3QPI0),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_R3QPI, 0),
+	},
+	{ /* R3QPI Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_R3QPI1),
+		.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_R3QPI, 1),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3c86),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT0_FILTER),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3c96),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT1_FILTER),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver snbep_uncore_pci_driver = {
+	.name		= "snbep_uncore",
+	.id_table	= snbep_uncore_pci_ids,
+};
+
+#define NODE_ID_MASK	0x7
+
+/*
+ * build pci bus to socket mapping
+ */
+static int snbep_pci2phy_map_init(int devid, int nodeid_loc, int idmap_loc, bool reverse)
+{
+	struct pci_dev *ubox_dev = NULL;
+	int i, bus, nodeid, segment;
+	struct pci2phy_map *map;
+	int err = 0;
+	u32 config = 0;
+
+	while (1) {
+		/* find the UBOX device */
+		ubox_dev = pci_get_device(PCI_VENDOR_ID_INTEL, devid, ubox_dev);
+		if (!ubox_dev)
+			break;
+		bus = ubox_dev->bus->number;
+		/* get the Node ID of the local register */
+		err = pci_read_config_dword(ubox_dev, nodeid_loc, &config);
+		if (err)
+			break;
+		nodeid = config & NODE_ID_MASK;
+		/* get the Node ID mapping */
+		err = pci_read_config_dword(ubox_dev, idmap_loc, &config);
+		if (err)
+			break;
+
+		segment = pci_domain_nr(ubox_dev->bus);
+		raw_spin_lock(&pci2phy_map_lock);
+		map = __find_pci2phy_map(segment);
+		if (!map) {
+			raw_spin_unlock(&pci2phy_map_lock);
+			err = -ENOMEM;
+			break;
+		}
+
+		/*
+		 * every three bits in the Node ID mapping register maps
+		 * to a particular node.
+		 */
+		for (i = 0; i < 8; i++) {
+			if (nodeid == ((config >> (3 * i)) & 0x7)) {
+				map->pbus_to_physid[bus] = i;
+				break;
+			}
+		}
+		raw_spin_unlock(&pci2phy_map_lock);
+	}
+
+	if (!err) {
+		/*
+		 * For PCI bus with no UBOX device, find the next bus
+		 * that has UBOX device and use its mapping.
+		 */
+		raw_spin_lock(&pci2phy_map_lock);
+		list_for_each_entry(map, &pci2phy_map_head, list) {
+			i = -1;
+			if (reverse) {
+				for (bus = 255; bus >= 0; bus--) {
+					if (map->pbus_to_physid[bus] >= 0)
+						i = map->pbus_to_physid[bus];
+					else
+						map->pbus_to_physid[bus] = i;
+				}
+			} else {
+				for (bus = 0; bus <= 255; bus++) {
+					if (map->pbus_to_physid[bus] >= 0)
+						i = map->pbus_to_physid[bus];
+					else
+						map->pbus_to_physid[bus] = i;
+				}
+			}
+		}
+		raw_spin_unlock(&pci2phy_map_lock);
+	}
+
+	pci_dev_put(ubox_dev);
+
+	return err ? pcibios_err_to_errno(err) : 0;
+}
+
+int snbep_uncore_pci_init(void)
+{
+	int ret = snbep_pci2phy_map_init(0x3ce0, SNBEP_CPUNODEID, SNBEP_GIDNIDMAP, true);
+	if (ret)
+		return ret;
+	uncore_pci_uncores = snbep_pci_uncores;
+	uncore_pci_driver = &snbep_uncore_pci_driver;
+	return 0;
+}
+/* end of Sandy Bridge-EP uncore support */
+
+/* IvyTown uncore support */
+static void ivbep_uncore_msr_init_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+	if (msr)
+		wrmsrl(msr, IVBEP_PMON_BOX_CTL_INT);
+}
+
+static void ivbep_uncore_pci_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+
+	pci_write_config_dword(pdev, SNBEP_PCI_PMON_BOX_CTL, IVBEP_PMON_BOX_CTL_INT);
+}
+
+#define IVBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
+	.init_box	= ivbep_uncore_msr_init_box,		\
+	.disable_box	= snbep_uncore_msr_disable_box,		\
+	.enable_box	= snbep_uncore_msr_enable_box,		\
+	.disable_event	= snbep_uncore_msr_disable_event,	\
+	.enable_event	= snbep_uncore_msr_enable_event,	\
+	.read_counter	= uncore_msr_read_counter
+
+static struct intel_uncore_ops ivbep_uncore_msr_ops = {
+	IVBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+};
+
+static struct intel_uncore_ops ivbep_uncore_pci_ops = {
+	.init_box	= ivbep_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snbep_uncore_pci_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+};
+
+#define IVBEP_UNCORE_PCI_COMMON_INIT()				\
+	.perf_ctr	= SNBEP_PCI_PMON_CTR0,			\
+	.event_ctl	= SNBEP_PCI_PMON_CTL0,			\
+	.event_mask	= IVBEP_PMON_RAW_EVENT_MASK,		\
+	.box_ctl	= SNBEP_PCI_PMON_BOX_CTL,		\
+	.ops		= &ivbep_uncore_pci_ops,			\
+	.format_group	= &ivbep_uncore_format_group
+
+static struct attribute *ivbep_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static struct attribute *ivbep_uncore_ubox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh5.attr,
+	NULL,
+};
+
+static struct attribute *ivbep_uncore_cbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid.attr,
+	&format_attr_filter_link.attr,
+	&format_attr_filter_state2.attr,
+	&format_attr_filter_nid2.attr,
+	&format_attr_filter_opc2.attr,
+	&format_attr_filter_nc.attr,
+	&format_attr_filter_c6.attr,
+	&format_attr_filter_isoc.attr,
+	NULL,
+};
+
+static struct attribute *ivbep_uncore_pcu_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_occ_sel.attr,
+	&format_attr_edge.attr,
+	&format_attr_thresh5.attr,
+	&format_attr_occ_invert.attr,
+	&format_attr_occ_edge.attr,
+	&format_attr_filter_band0.attr,
+	&format_attr_filter_band1.attr,
+	&format_attr_filter_band2.attr,
+	&format_attr_filter_band3.attr,
+	NULL,
+};
+
+static struct attribute *ivbep_uncore_qpi_formats_attr[] = {
+	&format_attr_event_ext.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_match_rds.attr,
+	&format_attr_match_rnid30.attr,
+	&format_attr_match_rnid4.attr,
+	&format_attr_match_dnid.attr,
+	&format_attr_match_mc.attr,
+	&format_attr_match_opc.attr,
+	&format_attr_match_vnw.attr,
+	&format_attr_match0.attr,
+	&format_attr_match1.attr,
+	&format_attr_mask_rds.attr,
+	&format_attr_mask_rnid30.attr,
+	&format_attr_mask_rnid4.attr,
+	&format_attr_mask_dnid.attr,
+	&format_attr_mask_mc.attr,
+	&format_attr_mask_opc.attr,
+	&format_attr_mask_vnw.attr,
+	&format_attr_mask0.attr,
+	&format_attr_mask1.attr,
+	NULL,
+};
+
+static const struct attribute_group ivbep_uncore_format_group = {
+	.name = "format",
+	.attrs = ivbep_uncore_formats_attr,
+};
+
+static const struct attribute_group ivbep_uncore_ubox_format_group = {
+	.name = "format",
+	.attrs = ivbep_uncore_ubox_formats_attr,
+};
+
+static const struct attribute_group ivbep_uncore_cbox_format_group = {
+	.name = "format",
+	.attrs = ivbep_uncore_cbox_formats_attr,
+};
+
+static const struct attribute_group ivbep_uncore_pcu_format_group = {
+	.name = "format",
+	.attrs = ivbep_uncore_pcu_formats_attr,
+};
+
+static const struct attribute_group ivbep_uncore_qpi_format_group = {
+	.name = "format",
+	.attrs = ivbep_uncore_qpi_formats_attr,
+};
+
+static struct intel_uncore_type ivbep_uncore_ubox = {
+	.name		= "ubox",
+	.num_counters   = 2,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.fixed_ctr_bits	= 48,
+	.perf_ctr	= SNBEP_U_MSR_PMON_CTR0,
+	.event_ctl	= SNBEP_U_MSR_PMON_CTL0,
+	.event_mask	= IVBEP_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr	= SNBEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl	= SNBEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.ops		= &ivbep_uncore_msr_ops,
+	.format_group	= &ivbep_uncore_ubox_format_group,
+};
+
+static struct extra_reg ivbep_uncore_cbox_extra_regs[] = {
+	SNBEP_CBO_EVENT_EXTRA_REG(SNBEP_CBO_PMON_CTL_TID_EN,
+				  SNBEP_CBO_PMON_CTL_TID_EN, 0x1),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x1031, 0x10ff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x1134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4134, 0xffff, 0xc),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5134, 0xffff, 0xc),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0334, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4334, 0xffff, 0xc),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0534, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4534, 0xffff, 0xc),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0934, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4934, 0xffff, 0xc),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0135, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2135, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4135, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4335, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4435, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4835, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a35, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5035, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8135, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4136, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4336, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4436, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4836, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a36, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5036, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4037, 0x40ff, 0x8),
+	EVENT_EXTRA_END
+};
+
+static u64 ivbep_cbox_filter_mask(int fields)
+{
+	u64 mask = 0;
+
+	if (fields & 0x1)
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_TID;
+	if (fields & 0x2)
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_LINK;
+	if (fields & 0x4)
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_STATE;
+	if (fields & 0x8)
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_NID;
+	if (fields & 0x10) {
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_OPC;
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_NC;
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_C6;
+		mask |= IVBEP_CB0_MSR_PMON_BOX_FILTER_ISOC;
+	}
+
+	return mask;
+}
+
+static struct event_constraint *
+ivbep_cbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return __snbep_cbox_get_constraint(box, event, ivbep_cbox_filter_mask);
+}
+
+static int ivbep_cbox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct extra_reg *er;
+	int idx = 0;
+
+	for (er = ivbep_uncore_cbox_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		idx |= er->idx;
+	}
+
+	if (idx) {
+		reg1->reg = SNBEP_C0_MSR_PMON_BOX_FILTER +
+			SNBEP_CBO_MSR_OFFSET * box->pmu->pmu_idx;
+		reg1->config = event->attr.config1 & ivbep_cbox_filter_mask(idx);
+		reg1->idx = idx;
+	}
+	return 0;
+}
+
+static void ivbep_cbox_enable_event(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;
+
+	if (reg1->idx != EXTRA_REG_NONE) {
+		u64 filter = uncore_shared_reg_config(box, 0);
+		wrmsrl(reg1->reg, filter & 0xffffffff);
+		wrmsrl(reg1->reg + 6, filter >> 32);
+	}
+
+	wrmsrl(hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops ivbep_uncore_cbox_ops = {
+	.init_box		= ivbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= ivbep_cbox_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= ivbep_cbox_hw_config,
+	.get_constraint		= ivbep_cbox_get_constraint,
+	.put_constraint		= snbep_cbox_put_constraint,
+};
+
+static struct intel_uncore_type ivbep_uncore_cbox = {
+	.name			= "cbox",
+	.num_counters		= 4,
+	.num_boxes		= 15,
+	.perf_ctr_bits		= 44,
+	.event_ctl		= SNBEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= SNBEP_C0_MSR_PMON_CTR0,
+	.event_mask		= IVBEP_CBO_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= SNBEP_CBO_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= snbep_uncore_cbox_constraints,
+	.ops			= &ivbep_uncore_cbox_ops,
+	.format_group		= &ivbep_uncore_cbox_format_group,
+};
+
+static struct intel_uncore_ops ivbep_uncore_pcu_ops = {
+	IVBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= snbep_pcu_hw_config,
+	.get_constraint		= snbep_pcu_get_constraint,
+	.put_constraint		= snbep_pcu_put_constraint,
+};
+
+static struct intel_uncore_type ivbep_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCU_MSR_PMON_CTR0,
+	.event_ctl		= SNBEP_PCU_MSR_PMON_CTL0,
+	.event_mask		= IVBEP_PCU_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCU_MSR_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &ivbep_uncore_pcu_ops,
+	.format_group		= &ivbep_uncore_pcu_format_group,
+};
+
+static struct intel_uncore_type *ivbep_msr_uncores[] = {
+	&ivbep_uncore_ubox,
+	&ivbep_uncore_cbox,
+	&ivbep_uncore_pcu,
+	NULL,
+};
+
+void ivbep_uncore_cpu_init(void)
+{
+	if (ivbep_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		ivbep_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+	uncore_msr_uncores = ivbep_msr_uncores;
+}
+
+static struct intel_uncore_type ivbep_uncore_ha = {
+	.name		= "ha",
+	.num_counters   = 4,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 48,
+	IVBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type ivbep_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 8,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTR,
+	.fixed_ctl	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTL,
+	.event_descs	= snbep_uncore_imc_events,
+	IVBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+/* registers in IRP boxes are not properly aligned */
+static unsigned ivbep_uncore_irp_ctls[] = {0xd8, 0xdc, 0xe0, 0xe4};
+static unsigned ivbep_uncore_irp_ctrs[] = {0xa0, 0xb0, 0xb8, 0xc0};
+
+static void ivbep_uncore_irp_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	pci_write_config_dword(pdev, ivbep_uncore_irp_ctls[hwc->idx],
+			       hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static void ivbep_uncore_irp_disable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	pci_write_config_dword(pdev, ivbep_uncore_irp_ctls[hwc->idx], hwc->config);
+}
+
+static u64 ivbep_uncore_irp_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 count = 0;
+
+	pci_read_config_dword(pdev, ivbep_uncore_irp_ctrs[hwc->idx], (u32 *)&count);
+	pci_read_config_dword(pdev, ivbep_uncore_irp_ctrs[hwc->idx] + 4, (u32 *)&count + 1);
+
+	return count;
+}
+
+static struct intel_uncore_ops ivbep_uncore_irp_ops = {
+	.init_box	= ivbep_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= ivbep_uncore_irp_disable_event,
+	.enable_event	= ivbep_uncore_irp_enable_event,
+	.read_counter	= ivbep_uncore_irp_read_counter,
+};
+
+static struct intel_uncore_type ivbep_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.event_mask		= IVBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.ops			= &ivbep_uncore_irp_ops,
+	.format_group		= &ivbep_uncore_format_group,
+};
+
+static struct intel_uncore_ops ivbep_uncore_qpi_ops = {
+	.init_box	= ivbep_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snbep_qpi_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+	.hw_config	= snbep_qpi_hw_config,
+	.get_constraint	= uncore_get_constraint,
+	.put_constraint	= uncore_put_constraint,
+};
+
+static struct intel_uncore_type ivbep_uncore_qpi = {
+	.name			= "qpi",
+	.num_counters		= 4,
+	.num_boxes		= 3,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCI_PMON_CTR0,
+	.event_ctl		= SNBEP_PCI_PMON_CTL0,
+	.event_mask		= IVBEP_QPI_PCI_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &ivbep_uncore_qpi_ops,
+	.format_group		= &ivbep_uncore_qpi_format_group,
+};
+
+static struct intel_uncore_type ivbep_uncore_r2pcie = {
+	.name		= "r2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 44,
+	.constraints	= snbep_uncore_r2pcie_constraints,
+	IVBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type ivbep_uncore_r3qpi = {
+	.name		= "r3qpi",
+	.num_counters   = 3,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 44,
+	.constraints	= snbep_uncore_r3qpi_constraints,
+	IVBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+enum {
+	IVBEP_PCI_UNCORE_HA,
+	IVBEP_PCI_UNCORE_IMC,
+	IVBEP_PCI_UNCORE_IRP,
+	IVBEP_PCI_UNCORE_QPI,
+	IVBEP_PCI_UNCORE_R2PCIE,
+	IVBEP_PCI_UNCORE_R3QPI,
+};
+
+static struct intel_uncore_type *ivbep_pci_uncores[] = {
+	[IVBEP_PCI_UNCORE_HA]	= &ivbep_uncore_ha,
+	[IVBEP_PCI_UNCORE_IMC]	= &ivbep_uncore_imc,
+	[IVBEP_PCI_UNCORE_IRP]	= &ivbep_uncore_irp,
+	[IVBEP_PCI_UNCORE_QPI]	= &ivbep_uncore_qpi,
+	[IVBEP_PCI_UNCORE_R2PCIE]	= &ivbep_uncore_r2pcie,
+	[IVBEP_PCI_UNCORE_R3QPI]	= &ivbep_uncore_r3qpi,
+	NULL,
+};
+
+static const struct pci_device_id ivbep_uncore_pci_ids[] = {
+	{ /* Home Agent 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe30),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_HA, 0),
+	},
+	{ /* Home Agent 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe38),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_HA, 1),
+	},
+	{ /* MC0 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xeb4),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 0),
+	},
+	{ /* MC0 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xeb5),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 1),
+	},
+	{ /* MC0 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xeb0),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 2),
+	},
+	{ /* MC0 Channel 4 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xeb1),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 3),
+	},
+	{ /* MC1 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xef4),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 4),
+	},
+	{ /* MC1 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xef5),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 5),
+	},
+	{ /* MC1 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xef0),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 6),
+	},
+	{ /* MC1 Channel 4 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xef1),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IMC, 7),
+	},
+	{ /* IRP */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe39),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_IRP, 0),
+	},
+	{ /* QPI0 Port 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe32),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_QPI, 0),
+	},
+	{ /* QPI0 Port 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe33),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_QPI, 1),
+	},
+	{ /* QPI1 Port 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe3a),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_QPI, 2),
+	},
+	{ /* R2PCIe */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe34),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_R2PCIE, 0),
+	},
+	{ /* R3QPI0 Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe36),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_R3QPI, 0),
+	},
+	{ /* R3QPI0 Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe37),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_R3QPI, 1),
+	},
+	{ /* R3QPI1 Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe3e),
+		.driver_data = UNCORE_PCI_DEV_DATA(IVBEP_PCI_UNCORE_R3QPI, 2),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe86),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT0_FILTER),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe96),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT1_FILTER),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver ivbep_uncore_pci_driver = {
+	.name		= "ivbep_uncore",
+	.id_table	= ivbep_uncore_pci_ids,
+};
+
+int ivbep_uncore_pci_init(void)
+{
+	int ret = snbep_pci2phy_map_init(0x0e1e, SNBEP_CPUNODEID, SNBEP_GIDNIDMAP, true);
+	if (ret)
+		return ret;
+	uncore_pci_uncores = ivbep_pci_uncores;
+	uncore_pci_driver = &ivbep_uncore_pci_driver;
+	return 0;
+}
+/* end of IvyTown uncore support */
+
+/* KNL uncore support */
+static struct attribute *knl_uncore_ubox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh5.attr,
+	NULL,
+};
+
+static const struct attribute_group knl_uncore_ubox_format_group = {
+	.name = "format",
+	.attrs = knl_uncore_ubox_formats_attr,
+};
+
+static struct intel_uncore_type knl_uncore_ubox = {
+	.name			= "ubox",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_U_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_U_MSR_PMON_CTL0,
+	.event_mask		= KNL_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.ops			= &snbep_uncore_msr_ops,
+	.format_group		= &knl_uncore_ubox_format_group,
+};
+
+static struct attribute *knl_uncore_cha_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_qor.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid4.attr,
+	&format_attr_filter_link3.attr,
+	&format_attr_filter_state4.attr,
+	&format_attr_filter_local.attr,
+	&format_attr_filter_all_op.attr,
+	&format_attr_filter_nnm.attr,
+	&format_attr_filter_opc3.attr,
+	&format_attr_filter_nc.attr,
+	&format_attr_filter_isoc.attr,
+	NULL,
+};
+
+static const struct attribute_group knl_uncore_cha_format_group = {
+	.name = "format",
+	.attrs = knl_uncore_cha_formats_attr,
+};
+
+static struct event_constraint knl_uncore_cha_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x1f, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg knl_uncore_cha_extra_regs[] = {
+	SNBEP_CBO_EVENT_EXTRA_REG(SNBEP_CBO_PMON_CTL_TID_EN,
+				  SNBEP_CBO_PMON_CTL_TID_EN, 0x1),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x3d, 0xff, 0x2),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x35, 0xff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x36, 0xff, 0x4),
+	EVENT_EXTRA_END
+};
+
+static u64 knl_cha_filter_mask(int fields)
+{
+	u64 mask = 0;
+
+	if (fields & 0x1)
+		mask |= KNL_CHA_MSR_PMON_BOX_FILTER_TID;
+	if (fields & 0x2)
+		mask |= KNL_CHA_MSR_PMON_BOX_FILTER_STATE;
+	if (fields & 0x4)
+		mask |= KNL_CHA_MSR_PMON_BOX_FILTER_OP;
+	return mask;
+}
+
+static struct event_constraint *
+knl_cha_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return __snbep_cbox_get_constraint(box, event, knl_cha_filter_mask);
+}
+
+static int knl_cha_hw_config(struct intel_uncore_box *box,
+			     struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct extra_reg *er;
+	int idx = 0;
+
+	for (er = knl_uncore_cha_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		idx |= er->idx;
+	}
+
+	if (idx) {
+		reg1->reg = HSWEP_C0_MSR_PMON_BOX_FILTER0 +
+			    KNL_CHA_MSR_OFFSET * box->pmu->pmu_idx;
+		reg1->config = event->attr.config1 & knl_cha_filter_mask(idx);
+
+		reg1->config |= KNL_CHA_MSR_PMON_BOX_FILTER_REMOTE_NODE;
+		reg1->config |= KNL_CHA_MSR_PMON_BOX_FILTER_LOCAL_NODE;
+		reg1->config |= KNL_CHA_MSR_PMON_BOX_FILTER_NNC;
+		reg1->idx = idx;
+	}
+	return 0;
+}
+
+static void hswep_cbox_enable_event(struct intel_uncore_box *box,
+				    struct perf_event *event);
+
+static struct intel_uncore_ops knl_uncore_cha_ops = {
+	.init_box		= snbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= hswep_cbox_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= knl_cha_hw_config,
+	.get_constraint		= knl_cha_get_constraint,
+	.put_constraint		= snbep_cbox_put_constraint,
+};
+
+static struct intel_uncore_type knl_uncore_cha = {
+	.name			= "cha",
+	.num_counters		= 4,
+	.num_boxes		= 38,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= HSWEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_C0_MSR_PMON_CTR0,
+	.event_mask		= KNL_CHA_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= KNL_CHA_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= knl_uncore_cha_constraints,
+	.ops			= &knl_uncore_cha_ops,
+	.format_group		= &knl_uncore_cha_format_group,
+};
+
+static struct attribute *knl_uncore_pcu_formats_attr[] = {
+	&format_attr_event2.attr,
+	&format_attr_use_occ_ctr.attr,
+	&format_attr_occ_sel.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh6.attr,
+	&format_attr_occ_invert.attr,
+	&format_attr_occ_edge_det.attr,
+	NULL,
+};
+
+static const struct attribute_group knl_uncore_pcu_format_group = {
+	.name = "format",
+	.attrs = knl_uncore_pcu_formats_attr,
+};
+
+static struct intel_uncore_type knl_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_PCU_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_PCU_MSR_PMON_CTL0,
+	.event_mask		= KNL_PCU_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_PCU_MSR_PMON_BOX_CTL,
+	.ops			= &snbep_uncore_msr_ops,
+	.format_group		= &knl_uncore_pcu_format_group,
+};
+
+static struct intel_uncore_type *knl_msr_uncores[] = {
+	&knl_uncore_ubox,
+	&knl_uncore_cha,
+	&knl_uncore_pcu,
+	NULL,
+};
+
+void knl_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = knl_msr_uncores;
+}
+
+static void knl_uncore_imc_enable_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	int box_ctl = uncore_pci_box_ctl(box);
+
+	pci_write_config_dword(pdev, box_ctl, 0);
+}
+
+static void knl_uncore_imc_enable_event(struct intel_uncore_box *box,
+					struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	if ((event->attr.config & SNBEP_PMON_CTL_EV_SEL_MASK)
+							== UNCORE_FIXED_EVENT)
+		pci_write_config_dword(pdev, hwc->config_base,
+				       hwc->config | KNL_PMON_FIXED_CTL_EN);
+	else
+		pci_write_config_dword(pdev, hwc->config_base,
+				       hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops knl_uncore_imc_ops = {
+	.init_box	= snbep_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= knl_uncore_imc_enable_box,
+	.read_counter	= snbep_uncore_pci_read_counter,
+	.enable_event	= knl_uncore_imc_enable_event,
+	.disable_event	= snbep_uncore_pci_disable_event,
+};
+
+static struct intel_uncore_type knl_uncore_imc_uclk = {
+	.name			= "imc_uclk",
+	.num_counters		= 4,
+	.num_boxes		= 2,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= KNL_UCLK_MSR_PMON_CTR0_LOW,
+	.event_ctl		= KNL_UCLK_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= KNL_UCLK_MSR_PMON_UCLK_FIXED_LOW,
+	.fixed_ctl		= KNL_UCLK_MSR_PMON_UCLK_FIXED_CTL,
+	.box_ctl		= KNL_UCLK_MSR_PMON_BOX_CTL,
+	.ops			= &knl_uncore_imc_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct intel_uncore_type knl_uncore_imc_dclk = {
+	.name			= "imc",
+	.num_counters		= 4,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= KNL_MC0_CH0_MSR_PMON_CTR0_LOW,
+	.event_ctl		= KNL_MC0_CH0_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= KNL_MC0_CH0_MSR_PMON_FIXED_LOW,
+	.fixed_ctl		= KNL_MC0_CH0_MSR_PMON_FIXED_CTL,
+	.box_ctl		= KNL_MC0_CH0_MSR_PMON_BOX_CTL,
+	.ops			= &knl_uncore_imc_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct intel_uncore_type knl_uncore_edc_uclk = {
+	.name			= "edc_uclk",
+	.num_counters		= 4,
+	.num_boxes		= 8,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= KNL_UCLK_MSR_PMON_CTR0_LOW,
+	.event_ctl		= KNL_UCLK_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= KNL_UCLK_MSR_PMON_UCLK_FIXED_LOW,
+	.fixed_ctl		= KNL_UCLK_MSR_PMON_UCLK_FIXED_CTL,
+	.box_ctl		= KNL_UCLK_MSR_PMON_BOX_CTL,
+	.ops			= &knl_uncore_imc_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct intel_uncore_type knl_uncore_edc_eclk = {
+	.name			= "edc_eclk",
+	.num_counters		= 4,
+	.num_boxes		= 8,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= KNL_EDC0_ECLK_MSR_PMON_CTR0_LOW,
+	.event_ctl		= KNL_EDC0_ECLK_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= KNL_EDC0_ECLK_MSR_PMON_ECLK_FIXED_LOW,
+	.fixed_ctl		= KNL_EDC0_ECLK_MSR_PMON_ECLK_FIXED_CTL,
+	.box_ctl		= KNL_EDC0_ECLK_MSR_PMON_BOX_CTL,
+	.ops			= &knl_uncore_imc_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct event_constraint knl_uncore_m2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type knl_uncore_m2pcie = {
+	.name		= "m2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.constraints	= knl_uncore_m2pcie_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct attribute *knl_uncore_irp_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_qor.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group knl_uncore_irp_format_group = {
+	.name = "format",
+	.attrs = knl_uncore_irp_formats_attr,
+};
+
+static struct intel_uncore_type knl_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCI_PMON_CTR0,
+	.event_ctl		= SNBEP_PCI_PMON_CTL0,
+	.event_mask		= KNL_IRP_PCI_PMON_RAW_EVENT_MASK,
+	.box_ctl		= KNL_IRP_PCI_PMON_BOX_CTL,
+	.ops			= &snbep_uncore_pci_ops,
+	.format_group		= &knl_uncore_irp_format_group,
+};
+
+enum {
+	KNL_PCI_UNCORE_MC_UCLK,
+	KNL_PCI_UNCORE_MC_DCLK,
+	KNL_PCI_UNCORE_EDC_UCLK,
+	KNL_PCI_UNCORE_EDC_ECLK,
+	KNL_PCI_UNCORE_M2PCIE,
+	KNL_PCI_UNCORE_IRP,
+};
+
+static struct intel_uncore_type *knl_pci_uncores[] = {
+	[KNL_PCI_UNCORE_MC_UCLK]	= &knl_uncore_imc_uclk,
+	[KNL_PCI_UNCORE_MC_DCLK]	= &knl_uncore_imc_dclk,
+	[KNL_PCI_UNCORE_EDC_UCLK]	= &knl_uncore_edc_uclk,
+	[KNL_PCI_UNCORE_EDC_ECLK]	= &knl_uncore_edc_eclk,
+	[KNL_PCI_UNCORE_M2PCIE]		= &knl_uncore_m2pcie,
+	[KNL_PCI_UNCORE_IRP]		= &knl_uncore_irp,
+	NULL,
+};
+
+/*
+ * KNL uses a common PCI device ID for multiple instances of an Uncore PMU
+ * device type. prior to KNL, each instance of a PMU device type had a unique
+ * device ID.
+ *
+ *	PCI Device ID	Uncore PMU Devices
+ *	----------------------------------
+ *	0x7841		MC0 UClk, MC1 UClk
+ *	0x7843		MC0 DClk CH 0, MC0 DClk CH 1, MC0 DClk CH 2,
+ *			MC1 DClk CH 0, MC1 DClk CH 1, MC1 DClk CH 2
+ *	0x7833		EDC0 UClk, EDC1 UClk, EDC2 UClk, EDC3 UClk,
+ *			EDC4 UClk, EDC5 UClk, EDC6 UClk, EDC7 UClk
+ *	0x7835		EDC0 EClk, EDC1 EClk, EDC2 EClk, EDC3 EClk,
+ *			EDC4 EClk, EDC5 EClk, EDC6 EClk, EDC7 EClk
+ *	0x7817		M2PCIe
+ *	0x7814		IRP
+*/
+
+static const struct pci_device_id knl_uncore_pci_ids[] = {
+	{ /* MC0 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7841),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(10, 0, KNL_PCI_UNCORE_MC_UCLK, 0),
+	},
+	{ /* MC1 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7841),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(11, 0, KNL_PCI_UNCORE_MC_UCLK, 1),
+	},
+	{ /* MC0 DClk CH 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(8, 2, KNL_PCI_UNCORE_MC_DCLK, 0),
+	},
+	{ /* MC0 DClk CH 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(8, 3, KNL_PCI_UNCORE_MC_DCLK, 1),
+	},
+	{ /* MC0 DClk CH 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(8, 4, KNL_PCI_UNCORE_MC_DCLK, 2),
+	},
+	{ /* MC1 DClk CH 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(9, 2, KNL_PCI_UNCORE_MC_DCLK, 3),
+	},
+	{ /* MC1 DClk CH 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(9, 3, KNL_PCI_UNCORE_MC_DCLK, 4),
+	},
+	{ /* MC1 DClk CH 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7843),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(9, 4, KNL_PCI_UNCORE_MC_DCLK, 5),
+	},
+	{ /* EDC0 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(15, 0, KNL_PCI_UNCORE_EDC_UCLK, 0),
+	},
+	{ /* EDC1 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(16, 0, KNL_PCI_UNCORE_EDC_UCLK, 1),
+	},
+	{ /* EDC2 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(17, 0, KNL_PCI_UNCORE_EDC_UCLK, 2),
+	},
+	{ /* EDC3 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(18, 0, KNL_PCI_UNCORE_EDC_UCLK, 3),
+	},
+	{ /* EDC4 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(19, 0, KNL_PCI_UNCORE_EDC_UCLK, 4),
+	},
+	{ /* EDC5 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(20, 0, KNL_PCI_UNCORE_EDC_UCLK, 5),
+	},
+	{ /* EDC6 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(21, 0, KNL_PCI_UNCORE_EDC_UCLK, 6),
+	},
+	{ /* EDC7 UClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7833),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(22, 0, KNL_PCI_UNCORE_EDC_UCLK, 7),
+	},
+	{ /* EDC0 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(24, 2, KNL_PCI_UNCORE_EDC_ECLK, 0),
+	},
+	{ /* EDC1 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(25, 2, KNL_PCI_UNCORE_EDC_ECLK, 1),
+	},
+	{ /* EDC2 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(26, 2, KNL_PCI_UNCORE_EDC_ECLK, 2),
+	},
+	{ /* EDC3 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(27, 2, KNL_PCI_UNCORE_EDC_ECLK, 3),
+	},
+	{ /* EDC4 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(28, 2, KNL_PCI_UNCORE_EDC_ECLK, 4),
+	},
+	{ /* EDC5 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(29, 2, KNL_PCI_UNCORE_EDC_ECLK, 5),
+	},
+	{ /* EDC6 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(30, 2, KNL_PCI_UNCORE_EDC_ECLK, 6),
+	},
+	{ /* EDC7 EClk */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7835),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(31, 2, KNL_PCI_UNCORE_EDC_ECLK, 7),
+	},
+	{ /* M2PCIe */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7817),
+		.driver_data = UNCORE_PCI_DEV_DATA(KNL_PCI_UNCORE_M2PCIE, 0),
+	},
+	{ /* IRP */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x7814),
+		.driver_data = UNCORE_PCI_DEV_DATA(KNL_PCI_UNCORE_IRP, 0),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver knl_uncore_pci_driver = {
+	.name		= "knl_uncore",
+	.id_table	= knl_uncore_pci_ids,
+};
+
+int knl_uncore_pci_init(void)
+{
+	int ret;
+
+	/* All KNL PCI based PMON units are on the same PCI bus except IRP */
+	ret = snb_pci2phy_map_init(0x7814); /* IRP */
+	if (ret)
+		return ret;
+	ret = snb_pci2phy_map_init(0x7817); /* M2PCIe */
+	if (ret)
+		return ret;
+	uncore_pci_uncores = knl_pci_uncores;
+	uncore_pci_driver = &knl_uncore_pci_driver;
+	return 0;
+}
+
+/* end of KNL uncore support */
+
+/* Haswell-EP uncore support */
+static struct attribute *hswep_uncore_ubox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh5.attr,
+	&format_attr_filter_tid2.attr,
+	&format_attr_filter_cid.attr,
+	NULL,
+};
+
+static const struct attribute_group hswep_uncore_ubox_format_group = {
+	.name = "format",
+	.attrs = hswep_uncore_ubox_formats_attr,
+};
+
+static int hswep_ubox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	reg1->reg = HSWEP_U_MSR_PMON_FILTER;
+	reg1->config = event->attr.config1 & HSWEP_U_MSR_PMON_BOX_FILTER_MASK;
+	reg1->idx = 0;
+	return 0;
+}
+
+static struct intel_uncore_ops hswep_uncore_ubox_ops = {
+	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= hswep_ubox_hw_config,
+	.get_constraint		= uncore_get_constraint,
+	.put_constraint		= uncore_put_constraint,
+};
+
+static struct intel_uncore_type hswep_uncore_ubox = {
+	.name			= "ubox",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 44,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_U_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_U_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &hswep_uncore_ubox_ops,
+	.format_group		= &hswep_uncore_ubox_format_group,
+};
+
+static struct attribute *hswep_uncore_cbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid3.attr,
+	&format_attr_filter_link2.attr,
+	&format_attr_filter_state3.attr,
+	&format_attr_filter_nid2.attr,
+	&format_attr_filter_opc2.attr,
+	&format_attr_filter_nc.attr,
+	&format_attr_filter_c6.attr,
+	&format_attr_filter_isoc.attr,
+	NULL,
+};
+
+static const struct attribute_group hswep_uncore_cbox_format_group = {
+	.name = "format",
+	.attrs = hswep_uncore_cbox_formats_attr,
+};
+
+static struct event_constraint hswep_uncore_cbox_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x01, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x09, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x38, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x3b, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x3e, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg hswep_uncore_cbox_extra_regs[] = {
+	SNBEP_CBO_EVENT_EXTRA_REG(SNBEP_CBO_PMON_CTL_TID_EN,
+				  SNBEP_CBO_PMON_CTL_TID_EN, 0x1),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0334, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0534, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0934, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x1134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4037, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4028, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4032, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4029, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4033, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x402A, 0x40ff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0135, 0xffff, 0x12),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4135, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4435, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4835, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5035, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4335, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a35, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8335, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2135, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8135, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4136, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4436, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4836, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4336, 0xffff, 0x18),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x4a36, 0xffff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8336, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x2136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x8136, 0xffff, 0x10),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x5036, 0xffff, 0x8),
+	EVENT_EXTRA_END
+};
+
+static u64 hswep_cbox_filter_mask(int fields)
+{
+	u64 mask = 0;
+	if (fields & 0x1)
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_TID;
+	if (fields & 0x2)
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_LINK;
+	if (fields & 0x4)
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_STATE;
+	if (fields & 0x8)
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_NID;
+	if (fields & 0x10) {
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_OPC;
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_NC;
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_C6;
+		mask |= HSWEP_CB0_MSR_PMON_BOX_FILTER_ISOC;
+	}
+	return mask;
+}
+
+static struct event_constraint *
+hswep_cbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return __snbep_cbox_get_constraint(box, event, hswep_cbox_filter_mask);
+}
+
+static int hswep_cbox_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct extra_reg *er;
+	int idx = 0;
+
+	for (er = hswep_uncore_cbox_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		idx |= er->idx;
+	}
+
+	if (idx) {
+		reg1->reg = HSWEP_C0_MSR_PMON_BOX_FILTER0 +
+			    HSWEP_CBO_MSR_OFFSET * box->pmu->pmu_idx;
+		reg1->config = event->attr.config1 & hswep_cbox_filter_mask(idx);
+		reg1->idx = idx;
+	}
+	return 0;
+}
+
+static void hswep_cbox_enable_event(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;
+
+	if (reg1->idx != EXTRA_REG_NONE) {
+		u64 filter = uncore_shared_reg_config(box, 0);
+		wrmsrl(reg1->reg, filter & 0xffffffff);
+		wrmsrl(reg1->reg + 1, filter >> 32);
+	}
+
+	wrmsrl(hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops hswep_uncore_cbox_ops = {
+	.init_box		= snbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= hswep_cbox_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= hswep_cbox_hw_config,
+	.get_constraint		= hswep_cbox_get_constraint,
+	.put_constraint		= snbep_cbox_put_constraint,
+};
+
+static struct intel_uncore_type hswep_uncore_cbox = {
+	.name			= "cbox",
+	.num_counters		= 4,
+	.num_boxes		= 18,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= HSWEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_C0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_CBO_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= hswep_uncore_cbox_constraints,
+	.ops			= &hswep_uncore_cbox_ops,
+	.format_group		= &hswep_uncore_cbox_format_group,
+};
+
+/*
+ * Write SBOX Initialization register bit by bit to avoid spurious #GPs
+ */
+static void hswep_uncore_sbox_msr_init_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+
+	if (msr) {
+		u64 init = SNBEP_PMON_BOX_CTL_INT;
+		u64 flags = 0;
+		int i;
+
+		for_each_set_bit(i, (unsigned long *)&init, 64) {
+			flags |= (1ULL << i);
+			wrmsrl(msr, flags);
+		}
+	}
+}
+
+static struct intel_uncore_ops hswep_uncore_sbox_msr_ops = {
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.init_box		= hswep_uncore_sbox_msr_init_box
+};
+
+static struct attribute *hswep_uncore_sbox_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group hswep_uncore_sbox_format_group = {
+	.name = "format",
+	.attrs = hswep_uncore_sbox_formats_attr,
+};
+
+static struct intel_uncore_type hswep_uncore_sbox = {
+	.name			= "sbox",
+	.num_counters		= 4,
+	.num_boxes		= 4,
+	.perf_ctr_bits		= 44,
+	.event_ctl		= HSWEP_S0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_S0_MSR_PMON_CTR0,
+	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_S0_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_SBOX_MSR_OFFSET,
+	.ops			= &hswep_uncore_sbox_msr_ops,
+	.format_group		= &hswep_uncore_sbox_format_group,
+};
+
+static int hswep_pcu_hw_config(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 ev_sel = hwc->config & SNBEP_PMON_CTL_EV_SEL_MASK;
+
+	if (ev_sel >= 0xb && ev_sel <= 0xe) {
+		reg1->reg = HSWEP_PCU_MSR_PMON_BOX_FILTER;
+		reg1->idx = ev_sel - 0xb;
+		reg1->config = event->attr.config1 & (0xff << reg1->idx);
+	}
+	return 0;
+}
+
+static struct intel_uncore_ops hswep_uncore_pcu_ops = {
+	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= hswep_pcu_hw_config,
+	.get_constraint		= snbep_pcu_get_constraint,
+	.put_constraint		= snbep_pcu_put_constraint,
+};
+
+static struct intel_uncore_type hswep_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_PCU_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_PCU_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PCU_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_PCU_MSR_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &hswep_uncore_pcu_ops,
+	.format_group		= &snbep_uncore_pcu_format_group,
+};
+
+static struct intel_uncore_type *hswep_msr_uncores[] = {
+	&hswep_uncore_ubox,
+	&hswep_uncore_cbox,
+	&hswep_uncore_sbox,
+	&hswep_uncore_pcu,
+	NULL,
+};
+
+#define HSWEP_PCU_DID			0x2fc0
+#define HSWEP_PCU_CAPID4_OFFET		0x94
+#define hswep_get_chop(_cap)		(((_cap) >> 6) & 0x3)
+
+static bool hswep_has_limit_sbox(unsigned int device)
+{
+	struct pci_dev *dev = pci_get_device(PCI_VENDOR_ID_INTEL, device, NULL);
+	u32 capid4;
+
+	if (!dev)
+		return false;
+
+	pci_read_config_dword(dev, HSWEP_PCU_CAPID4_OFFET, &capid4);
+	pci_dev_put(dev);
+	if (!hswep_get_chop(capid4))
+		return true;
+
+	return false;
+}
+
+void hswep_uncore_cpu_init(void)
+{
+	if (hswep_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		hswep_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+
+	/* Detect 6-8 core systems with only two SBOXes */
+	if (hswep_has_limit_sbox(HSWEP_PCU_DID))
+		hswep_uncore_sbox.num_boxes = 2;
+
+	uncore_msr_uncores = hswep_msr_uncores;
+}
+
+static struct intel_uncore_type hswep_uncore_ha = {
+	.name		= "ha",
+	.num_counters   = 4,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 48,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct uncore_event_desc hswep_uncore_imc_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks,      "event=0x00,umask=0x00"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read,  "event=0x04,umask=0x03"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.unit, "MiB"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write, "event=0x04,umask=0x0c"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.unit, "MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_type hswep_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 8,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTR,
+	.fixed_ctl	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTL,
+	.event_descs	= hswep_uncore_imc_events,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static unsigned hswep_uncore_irp_ctrs[] = {0xa0, 0xa8, 0xb0, 0xb8};
+
+static u64 hswep_uncore_irp_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 count = 0;
+
+	pci_read_config_dword(pdev, hswep_uncore_irp_ctrs[hwc->idx], (u32 *)&count);
+	pci_read_config_dword(pdev, hswep_uncore_irp_ctrs[hwc->idx] + 4, (u32 *)&count + 1);
+
+	return count;
+}
+
+static struct intel_uncore_ops hswep_uncore_irp_ops = {
+	.init_box	= snbep_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= ivbep_uncore_irp_disable_event,
+	.enable_event	= ivbep_uncore_irp_enable_event,
+	.read_counter	= hswep_uncore_irp_read_counter,
+};
+
+static struct intel_uncore_type hswep_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.ops			= &hswep_uncore_irp_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct intel_uncore_type hswep_uncore_qpi = {
+	.name			= "qpi",
+	.num_counters		= 4,
+	.num_boxes		= 3,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCI_PMON_CTR0,
+	.event_ctl		= SNBEP_PCI_PMON_CTL0,
+	.event_mask		= SNBEP_QPI_PCI_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &snbep_uncore_qpi_ops,
+	.format_group		= &snbep_uncore_qpi_format_group,
+};
+
+static struct event_constraint hswep_uncore_r2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x24, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x27, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x28, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x29, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2a, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x2b, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2c, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x32, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x35, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type hswep_uncore_r2pcie = {
+	.name		= "r2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.constraints	= hswep_uncore_r2pcie_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct event_constraint hswep_uncore_r3qpi_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x01, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x07, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x08, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x09, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x0a, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x0e, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x12, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x14, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x15, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x1f, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x20, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x21, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x22, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x28, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x29, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2c, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2e, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2f, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x31, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x32, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x37, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x38, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x39, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type hswep_uncore_r3qpi = {
+	.name		= "r3qpi",
+	.num_counters   = 3,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 44,
+	.constraints	= hswep_uncore_r3qpi_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+enum {
+	HSWEP_PCI_UNCORE_HA,
+	HSWEP_PCI_UNCORE_IMC,
+	HSWEP_PCI_UNCORE_IRP,
+	HSWEP_PCI_UNCORE_QPI,
+	HSWEP_PCI_UNCORE_R2PCIE,
+	HSWEP_PCI_UNCORE_R3QPI,
+};
+
+static struct intel_uncore_type *hswep_pci_uncores[] = {
+	[HSWEP_PCI_UNCORE_HA]	= &hswep_uncore_ha,
+	[HSWEP_PCI_UNCORE_IMC]	= &hswep_uncore_imc,
+	[HSWEP_PCI_UNCORE_IRP]	= &hswep_uncore_irp,
+	[HSWEP_PCI_UNCORE_QPI]	= &hswep_uncore_qpi,
+	[HSWEP_PCI_UNCORE_R2PCIE]	= &hswep_uncore_r2pcie,
+	[HSWEP_PCI_UNCORE_R3QPI]	= &hswep_uncore_r3qpi,
+	NULL,
+};
+
+static const struct pci_device_id hswep_uncore_pci_ids[] = {
+	{ /* Home Agent 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f30),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_HA, 0),
+	},
+	{ /* Home Agent 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f38),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_HA, 1),
+	},
+	{ /* MC0 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fb0),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 0),
+	},
+	{ /* MC0 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fb1),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 1),
+	},
+	{ /* MC0 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fb4),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 2),
+	},
+	{ /* MC0 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fb5),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 3),
+	},
+	{ /* MC1 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fd0),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 4),
+	},
+	{ /* MC1 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fd1),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 5),
+	},
+	{ /* MC1 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fd4),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 6),
+	},
+	{ /* MC1 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2fd5),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IMC, 7),
+	},
+	{ /* IRP */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f39),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_IRP, 0),
+	},
+	{ /* QPI0 Port 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f32),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_QPI, 0),
+	},
+	{ /* QPI0 Port 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f33),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_QPI, 1),
+	},
+	{ /* QPI1 Port 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f3a),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_QPI, 2),
+	},
+	{ /* R2PCIe */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f34),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_R2PCIE, 0),
+	},
+	{ /* R3QPI0 Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f36),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_R3QPI, 0),
+	},
+	{ /* R3QPI0 Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f37),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_R3QPI, 1),
+	},
+	{ /* R3QPI1 Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f3e),
+		.driver_data = UNCORE_PCI_DEV_DATA(HSWEP_PCI_UNCORE_R3QPI, 2),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f86),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT0_FILTER),
+	},
+	{ /* QPI Port 1 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2f96),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT1_FILTER),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver hswep_uncore_pci_driver = {
+	.name		= "hswep_uncore",
+	.id_table	= hswep_uncore_pci_ids,
+};
+
+int hswep_uncore_pci_init(void)
+{
+	int ret = snbep_pci2phy_map_init(0x2f1e, SNBEP_CPUNODEID, SNBEP_GIDNIDMAP, true);
+	if (ret)
+		return ret;
+	uncore_pci_uncores = hswep_pci_uncores;
+	uncore_pci_driver = &hswep_uncore_pci_driver;
+	return 0;
+}
+/* end of Haswell-EP uncore support */
+
+/* BDX uncore support */
+
+static struct intel_uncore_type bdx_uncore_ubox = {
+	.name			= "ubox",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_U_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_U_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &ivbep_uncore_ubox_format_group,
+};
+
+static struct event_constraint bdx_uncore_cbox_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x09, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x3e, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type bdx_uncore_cbox = {
+	.name			= "cbox",
+	.num_counters		= 4,
+	.num_boxes		= 24,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= HSWEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_C0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_CBO_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_CBO_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= bdx_uncore_cbox_constraints,
+	.ops			= &hswep_uncore_cbox_ops,
+	.format_group		= &hswep_uncore_cbox_format_group,
+};
+
+static struct intel_uncore_type bdx_uncore_sbox = {
+	.name			= "sbox",
+	.num_counters		= 4,
+	.num_boxes		= 4,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= HSWEP_S0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_S0_MSR_PMON_CTR0,
+	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_S0_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_SBOX_MSR_OFFSET,
+	.ops			= &hswep_uncore_sbox_msr_ops,
+	.format_group		= &hswep_uncore_sbox_format_group,
+};
+
+#define BDX_MSR_UNCORE_SBOX	3
+
+static struct intel_uncore_type *bdx_msr_uncores[] = {
+	&bdx_uncore_ubox,
+	&bdx_uncore_cbox,
+	&hswep_uncore_pcu,
+	&bdx_uncore_sbox,
+	NULL,
+};
+
+/* Bit 7 'Use Occupancy' is not available for counter 0 on BDX */
+static struct event_constraint bdx_uncore_pcu_constraints[] = {
+	EVENT_CONSTRAINT(0x80, 0xe, 0x80),
+	EVENT_CONSTRAINT_END
+};
+
+#define BDX_PCU_DID			0x6fc0
+
+void bdx_uncore_cpu_init(void)
+{
+	if (bdx_uncore_cbox.num_boxes > boot_cpu_data.x86_max_cores)
+		bdx_uncore_cbox.num_boxes = boot_cpu_data.x86_max_cores;
+	uncore_msr_uncores = bdx_msr_uncores;
+
+	/* Detect systems with no SBOXes */
+	if ((boot_cpu_data.x86_model == 86) || hswep_has_limit_sbox(BDX_PCU_DID))
+		uncore_msr_uncores[BDX_MSR_UNCORE_SBOX] = NULL;
+
+	hswep_uncore_pcu.constraints = bdx_uncore_pcu_constraints;
+}
+
+static struct intel_uncore_type bdx_uncore_ha = {
+	.name		= "ha",
+	.num_counters   = 4,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 48,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type bdx_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 8,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTR,
+	.fixed_ctl	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTL,
+	.event_descs	= hswep_uncore_imc_events,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct intel_uncore_type bdx_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.ops			= &hswep_uncore_irp_ops,
+	.format_group		= &snbep_uncore_format_group,
+};
+
+static struct intel_uncore_type bdx_uncore_qpi = {
+	.name			= "qpi",
+	.num_counters		= 4,
+	.num_boxes		= 3,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNBEP_PCI_PMON_CTR0,
+	.event_ctl		= SNBEP_PCI_PMON_CTL0,
+	.event_mask		= SNBEP_QPI_PCI_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNBEP_PCI_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &snbep_uncore_qpi_ops,
+	.format_group		= &snbep_uncore_qpi_format_group,
+};
+
+static struct event_constraint bdx_uncore_r2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x28, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2c, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type bdx_uncore_r2pcie = {
+	.name		= "r2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.constraints	= bdx_uncore_r2pcie_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+static struct event_constraint bdx_uncore_r3qpi_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x01, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x07, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x08, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x09, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x0a, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x0e, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x10, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x13, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x14, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x15, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x1f, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x20, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x21, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x22, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x25, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x26, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x28, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x29, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2c, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2e, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2f, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x33, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x34, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x37, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x38, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x39, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type bdx_uncore_r3qpi = {
+	.name		= "r3qpi",
+	.num_counters   = 3,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 48,
+	.constraints	= bdx_uncore_r3qpi_constraints,
+	SNBEP_UNCORE_PCI_COMMON_INIT(),
+};
+
+enum {
+	BDX_PCI_UNCORE_HA,
+	BDX_PCI_UNCORE_IMC,
+	BDX_PCI_UNCORE_IRP,
+	BDX_PCI_UNCORE_QPI,
+	BDX_PCI_UNCORE_R2PCIE,
+	BDX_PCI_UNCORE_R3QPI,
+};
+
+static struct intel_uncore_type *bdx_pci_uncores[] = {
+	[BDX_PCI_UNCORE_HA]	= &bdx_uncore_ha,
+	[BDX_PCI_UNCORE_IMC]	= &bdx_uncore_imc,
+	[BDX_PCI_UNCORE_IRP]	= &bdx_uncore_irp,
+	[BDX_PCI_UNCORE_QPI]	= &bdx_uncore_qpi,
+	[BDX_PCI_UNCORE_R2PCIE]	= &bdx_uncore_r2pcie,
+	[BDX_PCI_UNCORE_R3QPI]	= &bdx_uncore_r3qpi,
+	NULL,
+};
+
+static const struct pci_device_id bdx_uncore_pci_ids[] = {
+	{ /* Home Agent 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f30),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_HA, 0),
+	},
+	{ /* Home Agent 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f38),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_HA, 1),
+	},
+	{ /* MC0 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb0),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 0),
+	},
+	{ /* MC0 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb1),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 1),
+	},
+	{ /* MC0 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb4),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 2),
+	},
+	{ /* MC0 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fb5),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 3),
+	},
+	{ /* MC1 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fd0),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 4),
+	},
+	{ /* MC1 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fd1),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 5),
+	},
+	{ /* MC1 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fd4),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 6),
+	},
+	{ /* MC1 Channel 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6fd5),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IMC, 7),
+	},
+	{ /* IRP */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f39),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_IRP, 0),
+	},
+	{ /* QPI0 Port 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f32),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_QPI, 0),
+	},
+	{ /* QPI0 Port 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f33),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_QPI, 1),
+	},
+	{ /* QPI1 Port 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f3a),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_QPI, 2),
+	},
+	{ /* R2PCIe */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f34),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_R2PCIE, 0),
+	},
+	{ /* R3QPI0 Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f36),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_R3QPI, 0),
+	},
+	{ /* R3QPI0 Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f37),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_R3QPI, 1),
+	},
+	{ /* R3QPI1 Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f3e),
+		.driver_data = UNCORE_PCI_DEV_DATA(BDX_PCI_UNCORE_R3QPI, 2),
+	},
+	{ /* QPI Port 0 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f86),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT0_FILTER),
+	},
+	{ /* QPI Port 1 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f96),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   SNBEP_PCI_QPI_PORT1_FILTER),
+	},
+	{ /* QPI Port 2 filter  */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x6f46),
+		.driver_data = UNCORE_PCI_DEV_DATA(UNCORE_EXTRA_PCI_DEV,
+						   BDX_PCI_QPI_PORT2_FILTER),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver bdx_uncore_pci_driver = {
+	.name		= "bdx_uncore",
+	.id_table	= bdx_uncore_pci_ids,
+};
+
+int bdx_uncore_pci_init(void)
+{
+	int ret = snbep_pci2phy_map_init(0x6f1e, SNBEP_CPUNODEID, SNBEP_GIDNIDMAP, true);
+
+	if (ret)
+		return ret;
+	uncore_pci_uncores = bdx_pci_uncores;
+	uncore_pci_driver = &bdx_uncore_pci_driver;
+	return 0;
+}
+
+/* end of BDX uncore support */
+
+/* SKX uncore support */
+
+static struct intel_uncore_type skx_uncore_ubox = {
+	.name			= "ubox",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_U_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_U_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_U_MSR_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl		= HSWEP_U_MSR_PMON_UCLK_FIXED_CTL,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &ivbep_uncore_ubox_format_group,
+};
+
+static struct attribute *skx_uncore_cha_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid4.attr,
+	&format_attr_filter_state5.attr,
+	&format_attr_filter_rem.attr,
+	&format_attr_filter_loc.attr,
+	&format_attr_filter_nm.attr,
+	&format_attr_filter_all_op.attr,
+	&format_attr_filter_not_nm.attr,
+	&format_attr_filter_opc_0.attr,
+	&format_attr_filter_opc_1.attr,
+	&format_attr_filter_nc.attr,
+	&format_attr_filter_isoc.attr,
+	NULL,
+};
+
+static const struct attribute_group skx_uncore_chabox_format_group = {
+	.name = "format",
+	.attrs = skx_uncore_cha_formats_attr,
+};
+
+static struct event_constraint skx_uncore_chabox_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x11, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x36, 0x1),
+	EVENT_CONSTRAINT_END
+};
+
+static struct extra_reg skx_uncore_cha_extra_regs[] = {
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0334, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0534, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x0934, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x1134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x3134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x9134, 0xffff, 0x4),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x35, 0xff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x36, 0xff, 0x8),
+	SNBEP_CBO_EVENT_EXTRA_REG(0x38, 0xff, 0x3),
+	EVENT_EXTRA_END
+};
+
+static u64 skx_cha_filter_mask(int fields)
+{
+	u64 mask = 0;
+
+	if (fields & 0x1)
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_TID;
+	if (fields & 0x2)
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_LINK;
+	if (fields & 0x4)
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_STATE;
+	if (fields & 0x8) {
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_REM;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_LOC;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_ALL_OPC;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_NM;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_NOT_NM;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_OPC0;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_OPC1;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_NC;
+		mask |= SKX_CHA_MSR_PMON_BOX_FILTER_ISOC;
+	}
+	return mask;
+}
+
+static struct event_constraint *
+skx_cha_get_constraint(struct intel_uncore_box *box, struct perf_event *event)
+{
+	return __snbep_cbox_get_constraint(box, event, skx_cha_filter_mask);
+}
+
+static int skx_cha_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	struct extra_reg *er;
+	int idx = 0;
+	/* Any of the CHA events may be filtered by Thread/Core-ID.*/
+	if (event->hw.config & SNBEP_CBO_PMON_CTL_TID_EN)
+		idx = SKX_CHA_MSR_PMON_BOX_FILTER_TID;
+
+	for (er = skx_uncore_cha_extra_regs; er->msr; er++) {
+		if (er->event != (event->hw.config & er->config_mask))
+			continue;
+		idx |= er->idx;
+	}
+
+	if (idx) {
+		reg1->reg = HSWEP_C0_MSR_PMON_BOX_FILTER0 +
+			    HSWEP_CBO_MSR_OFFSET * box->pmu->pmu_idx;
+		reg1->config = event->attr.config1 & skx_cha_filter_mask(idx);
+		reg1->idx = idx;
+	}
+	return 0;
+}
+
+static struct intel_uncore_ops skx_uncore_chabox_ops = {
+	/* There is no frz_en for chabox ctl */
+	.init_box		= ivbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= hswep_cbox_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= skx_cha_hw_config,
+	.get_constraint		= skx_cha_get_constraint,
+	.put_constraint		= snbep_cbox_put_constraint,
+};
+
+static struct intel_uncore_type skx_uncore_chabox = {
+	.name			= "cha",
+	.num_counters		= 4,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= HSWEP_C0_MSR_PMON_CTL0,
+	.perf_ctr		= HSWEP_C0_MSR_PMON_CTR0,
+	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_C0_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_CBO_MSR_OFFSET,
+	.num_shared_regs	= 1,
+	.constraints		= skx_uncore_chabox_constraints,
+	.ops			= &skx_uncore_chabox_ops,
+	.format_group		= &skx_uncore_chabox_format_group,
+};
+
+static struct attribute *skx_uncore_iio_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh9.attr,
+	&format_attr_ch_mask.attr,
+	&format_attr_fc_mask.attr,
+	NULL,
+};
+
+static const struct attribute_group skx_uncore_iio_format_group = {
+	.name = "format",
+	.attrs = skx_uncore_iio_formats_attr,
+};
+
+static struct event_constraint skx_uncore_iio_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x83, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x88, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0x95, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xc0, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xc5, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xd4, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xd5, 0xc),
+	EVENT_CONSTRAINT_END
+};
+
+static void skx_iio_enable_event(struct intel_uncore_box *box,
+				 struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	wrmsrl(hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops skx_uncore_iio_ops = {
+	.init_box		= ivbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= skx_iio_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+};
+
+static inline u8 skx_iio_stack(struct intel_uncore_pmu *pmu, int die)
+{
+	return pmu->type->topology[die] >> (pmu->pmu_idx * BUS_NUM_STRIDE);
+}
+
+static umode_t
+pmu_iio_mapping_visible(struct kobject *kobj, struct attribute *attr,
+			 int die, int zero_bus_pmu)
+{
+	struct intel_uncore_pmu *pmu = dev_to_uncore_pmu(kobj_to_dev(kobj));
+
+	return (!skx_iio_stack(pmu, die) && pmu->pmu_idx != zero_bus_pmu) ? 0 : attr->mode;
+}
+
+static umode_t
+skx_iio_mapping_visible(struct kobject *kobj, struct attribute *attr, int die)
+{
+	/* Root bus 0x00 is valid only for pmu_idx = 0. */
+	return pmu_iio_mapping_visible(kobj, attr, die, 0);
+}
+
+static ssize_t skx_iio_mapping_show(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct pci_bus *bus = pci_find_next_bus(NULL);
+	struct intel_uncore_pmu *uncore_pmu = dev_to_uncore_pmu(dev);
+	struct dev_ext_attribute *ea = to_dev_ext_attribute(attr);
+	long die = (long)ea->var;
+
+	/*
+	 * Current implementation is for single segment configuration hence it's
+	 * safe to take the segment value from the first available root bus.
+	 */
+	return sprintf(buf, "%04x:%02x\n", pci_domain_nr(bus),
+					   skx_iio_stack(uncore_pmu, die));
+}
+
+static int skx_msr_cpu_bus_read(int cpu, u64 *topology)
+{
+	u64 msr_value;
+
+	if (rdmsrl_on_cpu(cpu, SKX_MSR_CPU_BUS_NUMBER, &msr_value) ||
+			!(msr_value & SKX_MSR_CPU_BUS_VALID_BIT))
+		return -ENXIO;
+
+	*topology = msr_value;
+
+	return 0;
+}
+
+static int die_to_cpu(int die)
+{
+	int res = 0, cpu, current_die;
+	/*
+	 * Using cpus_read_lock() to ensure cpu is not going down between
+	 * looking at cpu_online_mask.
+	 */
+	cpus_read_lock();
+	for_each_online_cpu(cpu) {
+		current_die = topology_logical_die_id(cpu);
+		if (current_die == die) {
+			res = cpu;
+			break;
+		}
+	}
+	cpus_read_unlock();
+	return res;
+}
+
+static int skx_iio_get_topology(struct intel_uncore_type *type)
+{
+	int i, ret;
+	struct pci_bus *bus = NULL;
+
+	/*
+	 * Verified single-segment environments only; disabled for multiple
+	 * segment topologies for now except VMD domains.
+	 * VMD domains start at 0x10000 to not clash with ACPI _SEG domains.
+	 */
+	while ((bus = pci_find_next_bus(bus))
+		&& (!pci_domain_nr(bus) || pci_domain_nr(bus) > 0xffff))
+		;
+	if (bus)
+		return -EPERM;
+
+	type->topology = kcalloc(uncore_max_dies(), sizeof(u64), GFP_KERNEL);
+	if (!type->topology)
+		return -ENOMEM;
+
+	for (i = 0; i < uncore_max_dies(); i++) {
+		ret = skx_msr_cpu_bus_read(die_to_cpu(i), &type->topology[i]);
+		if (ret) {
+			kfree(type->topology);
+			type->topology = NULL;
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+static struct attribute_group skx_iio_mapping_group = {
+	.is_visible	= skx_iio_mapping_visible,
+};
+
+static const struct attribute_group *skx_iio_attr_update[] = {
+	&skx_iio_mapping_group,
+	NULL,
+};
+
+static void pmu_clear_mapping_attr(const struct attribute_group **groups,
+				   struct attribute_group *ag)
+{
+	int i;
+
+	for (i = 0; groups[i]; i++) {
+		if (groups[i] == ag) {
+			for (i++; groups[i]; i++)
+				groups[i - 1] = groups[i];
+			groups[i - 1] = NULL;
+			break;
+		}
+	}
+}
+
+static int
+pmu_iio_set_mapping(struct intel_uncore_type *type, struct attribute_group *ag)
+{
+	char buf[64];
+	int ret;
+	long die = -1;
+	struct attribute **attrs = NULL;
+	struct dev_ext_attribute *eas = NULL;
+
+	ret = type->get_topology(type);
+	if (ret < 0)
+		goto clear_attr_update;
+
+	ret = -ENOMEM;
+
+	/* One more for NULL. */
+	attrs = kcalloc((uncore_max_dies() + 1), sizeof(*attrs), GFP_KERNEL);
+	if (!attrs)
+		goto clear_topology;
+
+	eas = kcalloc(uncore_max_dies(), sizeof(*eas), GFP_KERNEL);
+	if (!eas)
+		goto clear_attrs;
+
+	for (die = 0; die < uncore_max_dies(); die++) {
+		sprintf(buf, "die%ld", die);
+		sysfs_attr_init(&eas[die].attr.attr);
+		eas[die].attr.attr.name = kstrdup(buf, GFP_KERNEL);
+		if (!eas[die].attr.attr.name)
+			goto err;
+		eas[die].attr.attr.mode = 0444;
+		eas[die].attr.show = skx_iio_mapping_show;
+		eas[die].attr.store = NULL;
+		eas[die].var = (void *)die;
+		attrs[die] = &eas[die].attr.attr;
+	}
+	ag->attrs = attrs;
+
+	return 0;
+err:
+	for (; die >= 0; die--)
+		kfree(eas[die].attr.attr.name);
+	kfree(eas);
+clear_attrs:
+	kfree(attrs);
+clear_topology:
+	kfree(type->topology);
+clear_attr_update:
+	pmu_clear_mapping_attr(type->attr_update, ag);
+	return ret;
+}
+
+static int skx_iio_set_mapping(struct intel_uncore_type *type)
+{
+	return pmu_iio_set_mapping(type, &skx_iio_mapping_group);
+}
+
+static void skx_iio_cleanup_mapping(struct intel_uncore_type *type)
+{
+	struct attribute **attr = skx_iio_mapping_group.attrs;
+
+	if (!attr)
+		return;
+
+	for (; *attr; attr++)
+		kfree((*attr)->name);
+	kfree(attr_to_ext_attr(*skx_iio_mapping_group.attrs));
+	kfree(skx_iio_mapping_group.attrs);
+	skx_iio_mapping_group.attrs = NULL;
+	kfree(type->topology);
+}
+
+static struct intel_uncore_type skx_uncore_iio = {
+	.name			= "iio",
+	.num_counters		= 4,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= SKX_IIO0_MSR_PMON_CTL0,
+	.perf_ctr		= SKX_IIO0_MSR_PMON_CTR0,
+	.event_mask		= SKX_IIO_PMON_RAW_EVENT_MASK,
+	.event_mask_ext		= SKX_IIO_PMON_RAW_EVENT_MASK_EXT,
+	.box_ctl		= SKX_IIO0_MSR_PMON_BOX_CTL,
+	.msr_offset		= SKX_IIO_MSR_OFFSET,
+	.constraints		= skx_uncore_iio_constraints,
+	.ops			= &skx_uncore_iio_ops,
+	.format_group		= &skx_uncore_iio_format_group,
+	.attr_update		= skx_iio_attr_update,
+	.get_topology		= skx_iio_get_topology,
+	.set_mapping		= skx_iio_set_mapping,
+	.cleanup_mapping	= skx_iio_cleanup_mapping,
+};
+
+enum perf_uncore_iio_freerunning_type_id {
+	SKX_IIO_MSR_IOCLK			= 0,
+	SKX_IIO_MSR_BW				= 1,
+	SKX_IIO_MSR_UTIL			= 2,
+
+	SKX_IIO_FREERUNNING_TYPE_MAX,
+};
+
+
+static struct freerunning_counters skx_iio_freerunning[] = {
+	[SKX_IIO_MSR_IOCLK]	= { 0xa45, 0x1, 0x20, 1, 36 },
+	[SKX_IIO_MSR_BW]	= { 0xb00, 0x1, 0x10, 8, 36 },
+	[SKX_IIO_MSR_UTIL]	= { 0xb08, 0x1, 0x10, 8, 36 },
+};
+
+static struct uncore_event_desc skx_uncore_iio_freerunning_events[] = {
+	/* Free-Running IO CLOCKS Counter */
+	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
+	/* Free-Running IIO BANDWIDTH Counters */
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port0,		"event=0xff,umask=0x24"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port0.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port0.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port1,		"event=0xff,umask=0x25"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port1.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port1.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port2,		"event=0xff,umask=0x26"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port2.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port2.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port3,		"event=0xff,umask=0x27"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port3.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_out_port3.unit,	"MiB"),
+	/* Free-running IIO UTILIZATION Counters */
+	INTEL_UNCORE_EVENT_DESC(util_in_port0,		"event=0xff,umask=0x30"),
+	INTEL_UNCORE_EVENT_DESC(util_out_port0,		"event=0xff,umask=0x31"),
+	INTEL_UNCORE_EVENT_DESC(util_in_port1,		"event=0xff,umask=0x32"),
+	INTEL_UNCORE_EVENT_DESC(util_out_port1,		"event=0xff,umask=0x33"),
+	INTEL_UNCORE_EVENT_DESC(util_in_port2,		"event=0xff,umask=0x34"),
+	INTEL_UNCORE_EVENT_DESC(util_out_port2,		"event=0xff,umask=0x35"),
+	INTEL_UNCORE_EVENT_DESC(util_in_port3,		"event=0xff,umask=0x36"),
+	INTEL_UNCORE_EVENT_DESC(util_out_port3,		"event=0xff,umask=0x37"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_ops skx_uncore_iio_freerunning_ops = {
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= uncore_freerunning_hw_config,
+};
+
+static struct attribute *skx_uncore_iio_freerunning_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	NULL,
+};
+
+static const struct attribute_group skx_uncore_iio_freerunning_format_group = {
+	.name = "format",
+	.attrs = skx_uncore_iio_freerunning_formats_attr,
+};
+
+static struct intel_uncore_type skx_uncore_iio_free_running = {
+	.name			= "iio_free_running",
+	.num_counters		= 17,
+	.num_boxes		= 6,
+	.num_freerunning_types	= SKX_IIO_FREERUNNING_TYPE_MAX,
+	.freerunning		= skx_iio_freerunning,
+	.ops			= &skx_uncore_iio_freerunning_ops,
+	.event_descs		= skx_uncore_iio_freerunning_events,
+	.format_group		= &skx_uncore_iio_freerunning_format_group,
+};
+
+static struct attribute *skx_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group skx_uncore_format_group = {
+	.name = "format",
+	.attrs = skx_uncore_formats_attr,
+};
+
+static struct intel_uncore_type skx_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 2,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= SKX_IRP0_MSR_PMON_CTL0,
+	.perf_ctr		= SKX_IRP0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SKX_IRP0_MSR_PMON_BOX_CTL,
+	.msr_offset		= SKX_IRP_MSR_OFFSET,
+	.ops			= &skx_uncore_iio_ops,
+	.format_group		= &skx_uncore_format_group,
+};
+
+static struct attribute *skx_uncore_pcu_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_occ_invert.attr,
+	&format_attr_occ_edge_det.attr,
+	&format_attr_filter_band0.attr,
+	&format_attr_filter_band1.attr,
+	&format_attr_filter_band2.attr,
+	&format_attr_filter_band3.attr,
+	NULL,
+};
+
+static struct attribute_group skx_uncore_pcu_format_group = {
+	.name = "format",
+	.attrs = skx_uncore_pcu_formats_attr,
+};
+
+static struct intel_uncore_ops skx_uncore_pcu_ops = {
+	IVBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= hswep_pcu_hw_config,
+	.get_constraint		= snbep_pcu_get_constraint,
+	.put_constraint		= snbep_pcu_put_constraint,
+};
+
+static struct intel_uncore_type skx_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= HSWEP_PCU_MSR_PMON_CTR0,
+	.event_ctl		= HSWEP_PCU_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PCU_MSR_PMON_RAW_EVENT_MASK,
+	.box_ctl		= HSWEP_PCU_MSR_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &skx_uncore_pcu_ops,
+	.format_group		= &skx_uncore_pcu_format_group,
+};
+
+static struct intel_uncore_type *skx_msr_uncores[] = {
+	&skx_uncore_ubox,
+	&skx_uncore_chabox,
+	&skx_uncore_iio,
+	&skx_uncore_iio_free_running,
+	&skx_uncore_irp,
+	&skx_uncore_pcu,
+	NULL,
+};
+
+/*
+ * To determine the number of CHAs, it should read bits 27:0 in the CAPID6
+ * register which located at Device 30, Function 3, Offset 0x9C. PCI ID 0x2083.
+ */
+#define SKX_CAPID6		0x9c
+#define SKX_CHA_BIT_MASK	GENMASK(27, 0)
+
+static int skx_count_chabox(void)
+{
+	struct pci_dev *dev = NULL;
+	u32 val = 0;
+
+	dev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x2083, dev);
+	if (!dev)
+		goto out;
+
+	pci_read_config_dword(dev, SKX_CAPID6, &val);
+	val &= SKX_CHA_BIT_MASK;
+out:
+	pci_dev_put(dev);
+	return hweight32(val);
+}
+
+void skx_uncore_cpu_init(void)
+{
+	skx_uncore_chabox.num_boxes = skx_count_chabox();
+	uncore_msr_uncores = skx_msr_uncores;
+}
+
+static struct intel_uncore_type skx_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 6,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTR,
+	.fixed_ctl	= SNBEP_MC_CHy_PCI_PMON_FIXED_CTL,
+	.event_descs	= hswep_uncore_imc_events,
+	.perf_ctr	= SNBEP_PCI_PMON_CTR0,
+	.event_ctl	= SNBEP_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SNBEP_PCI_PMON_BOX_CTL,
+	.ops		= &ivbep_uncore_pci_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+static struct attribute *skx_upi_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask_ext.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group skx_upi_uncore_format_group = {
+	.name = "format",
+	.attrs = skx_upi_uncore_formats_attr,
+};
+
+static void skx_upi_uncore_pci_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+
+	__set_bit(UNCORE_BOX_FLAG_CTL_OFFS8, &box->flags);
+	pci_write_config_dword(pdev, SKX_UPI_PCI_PMON_BOX_CTL, IVBEP_PMON_BOX_CTL_INT);
+}
+
+static struct intel_uncore_ops skx_upi_uncore_pci_ops = {
+	.init_box	= skx_upi_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snbep_uncore_pci_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+};
+
+static struct intel_uncore_type skx_uncore_upi = {
+	.name		= "upi",
+	.num_counters   = 4,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= SKX_UPI_PCI_PMON_CTR0,
+	.event_ctl	= SKX_UPI_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext = SKX_UPI_CTL_UMASK_EXT,
+	.box_ctl	= SKX_UPI_PCI_PMON_BOX_CTL,
+	.ops		= &skx_upi_uncore_pci_ops,
+	.format_group	= &skx_upi_uncore_format_group,
+};
+
+static void skx_m2m_uncore_pci_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+
+	__set_bit(UNCORE_BOX_FLAG_CTL_OFFS8, &box->flags);
+	pci_write_config_dword(pdev, SKX_M2M_PCI_PMON_BOX_CTL, IVBEP_PMON_BOX_CTL_INT);
+}
+
+static struct intel_uncore_ops skx_m2m_uncore_pci_ops = {
+	.init_box	= skx_m2m_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snbep_uncore_pci_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+};
+
+static struct intel_uncore_type skx_uncore_m2m = {
+	.name		= "m2m",
+	.num_counters   = 4,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= SKX_M2M_PCI_PMON_CTR0,
+	.event_ctl	= SKX_M2M_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SKX_M2M_PCI_PMON_BOX_CTL,
+	.ops		= &skx_m2m_uncore_pci_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+static struct event_constraint skx_uncore_m2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type skx_uncore_m2pcie = {
+	.name		= "m2pcie",
+	.num_counters   = 4,
+	.num_boxes	= 4,
+	.perf_ctr_bits	= 48,
+	.constraints	= skx_uncore_m2pcie_constraints,
+	.perf_ctr	= SNBEP_PCI_PMON_CTR0,
+	.event_ctl	= SNBEP_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SNBEP_PCI_PMON_BOX_CTL,
+	.ops		= &ivbep_uncore_pci_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+static struct event_constraint skx_uncore_m3upi_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x1d, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x1e, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x40, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x4e, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x4f, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x50, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x51, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x52, 0x7),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type skx_uncore_m3upi = {
+	.name		= "m3upi",
+	.num_counters   = 3,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 48,
+	.constraints	= skx_uncore_m3upi_constraints,
+	.perf_ctr	= SNBEP_PCI_PMON_CTR0,
+	.event_ctl	= SNBEP_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SNBEP_PCI_PMON_BOX_CTL,
+	.ops		= &ivbep_uncore_pci_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+enum {
+	SKX_PCI_UNCORE_IMC,
+	SKX_PCI_UNCORE_M2M,
+	SKX_PCI_UNCORE_UPI,
+	SKX_PCI_UNCORE_M2PCIE,
+	SKX_PCI_UNCORE_M3UPI,
+};
+
+static struct intel_uncore_type *skx_pci_uncores[] = {
+	[SKX_PCI_UNCORE_IMC]	= &skx_uncore_imc,
+	[SKX_PCI_UNCORE_M2M]	= &skx_uncore_m2m,
+	[SKX_PCI_UNCORE_UPI]	= &skx_uncore_upi,
+	[SKX_PCI_UNCORE_M2PCIE]	= &skx_uncore_m2pcie,
+	[SKX_PCI_UNCORE_M3UPI]	= &skx_uncore_m3upi,
+	NULL,
+};
+
+static const struct pci_device_id skx_uncore_pci_ids[] = {
+	{ /* MC0 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2042),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(10, 2, SKX_PCI_UNCORE_IMC, 0),
+	},
+	{ /* MC0 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2046),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(10, 6, SKX_PCI_UNCORE_IMC, 1),
+	},
+	{ /* MC0 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x204a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(11, 2, SKX_PCI_UNCORE_IMC, 2),
+	},
+	{ /* MC1 Channel 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2042),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(12, 2, SKX_PCI_UNCORE_IMC, 3),
+	},
+	{ /* MC1 Channel 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2046),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(12, 6, SKX_PCI_UNCORE_IMC, 4),
+	},
+	{ /* MC1 Channel 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x204a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(13, 2, SKX_PCI_UNCORE_IMC, 5),
+	},
+	{ /* M2M0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2066),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(8, 0, SKX_PCI_UNCORE_M2M, 0),
+	},
+	{ /* M2M1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2066),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(9, 0, SKX_PCI_UNCORE_M2M, 1),
+	},
+	{ /* UPI0 Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2058),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(14, 0, SKX_PCI_UNCORE_UPI, 0),
+	},
+	{ /* UPI0 Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2058),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(15, 0, SKX_PCI_UNCORE_UPI, 1),
+	},
+	{ /* UPI1 Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2058),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(16, 0, SKX_PCI_UNCORE_UPI, 2),
+	},
+	{ /* M2PCIe 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2088),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(21, 1, SKX_PCI_UNCORE_M2PCIE, 0),
+	},
+	{ /* M2PCIe 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2088),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(22, 1, SKX_PCI_UNCORE_M2PCIE, 1),
+	},
+	{ /* M2PCIe 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2088),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(23, 1, SKX_PCI_UNCORE_M2PCIE, 2),
+	},
+	{ /* M2PCIe 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x2088),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(21, 5, SKX_PCI_UNCORE_M2PCIE, 3),
+	},
+	{ /* M3UPI0 Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x204D),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(18, 1, SKX_PCI_UNCORE_M3UPI, 0),
+	},
+	{ /* M3UPI0 Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x204E),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(18, 2, SKX_PCI_UNCORE_M3UPI, 1),
+	},
+	{ /* M3UPI1 Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x204D),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(18, 5, SKX_PCI_UNCORE_M3UPI, 2),
+	},
+	{ /* end: all zeroes */ }
+};
+
+
+static struct pci_driver skx_uncore_pci_driver = {
+	.name		= "skx_uncore",
+	.id_table	= skx_uncore_pci_ids,
+};
+
+int skx_uncore_pci_init(void)
+{
+	/* need to double check pci address */
+	int ret = snbep_pci2phy_map_init(0x2014, SKX_CPUNODEID, SKX_GIDNIDMAP, false);
+
+	if (ret)
+		return ret;
+
+	uncore_pci_uncores = skx_pci_uncores;
+	uncore_pci_driver = &skx_uncore_pci_driver;
+	return 0;
+}
+
+/* end of SKX uncore support */
+
+/* SNR uncore support */
+
+static struct intel_uncore_type snr_uncore_ubox = {
+	.name			= "ubox",
+	.num_counters		= 2,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.fixed_ctr_bits		= 48,
+	.perf_ctr		= SNR_U_MSR_PMON_CTR0,
+	.event_ctl		= SNR_U_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.fixed_ctr		= SNR_U_MSR_PMON_UCLK_FIXED_CTR,
+	.fixed_ctl		= SNR_U_MSR_PMON_UCLK_FIXED_CTL,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &ivbep_uncore_format_group,
+};
+
+static struct attribute *snr_uncore_cha_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask_ext2.attr,
+	&format_attr_edge.attr,
+	&format_attr_tid_en.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	&format_attr_filter_tid5.attr,
+	NULL,
+};
+static const struct attribute_group snr_uncore_chabox_format_group = {
+	.name = "format",
+	.attrs = snr_uncore_cha_formats_attr,
+};
+
+static int snr_cha_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+
+	reg1->reg = SNR_C0_MSR_PMON_BOX_FILTER0 +
+		    box->pmu->type->msr_offset * box->pmu->pmu_idx;
+	reg1->config = event->attr.config1 & SKX_CHA_MSR_PMON_BOX_FILTER_TID;
+	reg1->idx = 0;
+
+	return 0;
+}
+
+static void snr_cha_enable_event(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;
+
+	if (reg1->idx != EXTRA_REG_NONE)
+		wrmsrl(reg1->reg, reg1->config);
+
+	wrmsrl(hwc->config_base, hwc->config | SNBEP_PMON_CTL_EN);
+}
+
+static struct intel_uncore_ops snr_uncore_chabox_ops = {
+	.init_box		= ivbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= snr_cha_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= snr_cha_hw_config,
+};
+
+static struct intel_uncore_type snr_uncore_chabox = {
+	.name			= "cha",
+	.num_counters		= 4,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= SNR_CHA_MSR_PMON_CTL0,
+	.perf_ctr		= SNR_CHA_MSR_PMON_CTR0,
+	.box_ctl		= SNR_CHA_MSR_PMON_BOX_CTL,
+	.msr_offset		= HSWEP_CBO_MSR_OFFSET,
+	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
+	.event_mask_ext		= SNR_CHA_RAW_EVENT_MASK_EXT,
+	.ops			= &snr_uncore_chabox_ops,
+	.format_group		= &snr_uncore_chabox_format_group,
+};
+
+static struct attribute *snr_uncore_iio_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh9.attr,
+	&format_attr_ch_mask2.attr,
+	&format_attr_fc_mask2.attr,
+	NULL,
+};
+
+static const struct attribute_group snr_uncore_iio_format_group = {
+	.name = "format",
+	.attrs = snr_uncore_iio_formats_attr,
+};
+
+static struct intel_uncore_type snr_uncore_iio = {
+	.name			= "iio",
+	.num_counters		= 4,
+	.num_boxes		= 5,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= SNR_IIO_MSR_PMON_CTL0,
+	.perf_ctr		= SNR_IIO_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext		= SNR_IIO_PMON_RAW_EVENT_MASK_EXT,
+	.box_ctl		= SNR_IIO_MSR_PMON_BOX_CTL,
+	.msr_offset		= SNR_IIO_MSR_OFFSET,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &snr_uncore_iio_format_group,
+};
+
+static struct intel_uncore_type snr_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 2,
+	.num_boxes		= 5,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= SNR_IRP0_MSR_PMON_CTL0,
+	.perf_ctr		= SNR_IRP0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNR_IRP0_MSR_PMON_BOX_CTL,
+	.msr_offset		= SNR_IRP_MSR_OFFSET,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &ivbep_uncore_format_group,
+};
+
+static struct intel_uncore_type snr_uncore_m2pcie = {
+	.name		= "m2pcie",
+	.num_counters	= 4,
+	.num_boxes	= 5,
+	.perf_ctr_bits	= 48,
+	.event_ctl	= SNR_M2PCIE_MSR_PMON_CTL0,
+	.perf_ctr	= SNR_M2PCIE_MSR_PMON_CTR0,
+	.box_ctl	= SNR_M2PCIE_MSR_PMON_BOX_CTL,
+	.msr_offset	= SNR_M2PCIE_MSR_OFFSET,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.ops		= &ivbep_uncore_msr_ops,
+	.format_group	= &ivbep_uncore_format_group,
+};
+
+static int snr_pcu_hw_config(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 ev_sel = hwc->config & SNBEP_PMON_CTL_EV_SEL_MASK;
+
+	if (ev_sel >= 0xb && ev_sel <= 0xe) {
+		reg1->reg = SNR_PCU_MSR_PMON_BOX_FILTER;
+		reg1->idx = ev_sel - 0xb;
+		reg1->config = event->attr.config1 & (0xff << reg1->idx);
+	}
+	return 0;
+}
+
+static struct intel_uncore_ops snr_uncore_pcu_ops = {
+	IVBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.hw_config		= snr_pcu_hw_config,
+	.get_constraint		= snbep_pcu_get_constraint,
+	.put_constraint		= snbep_pcu_put_constraint,
+};
+
+static struct intel_uncore_type snr_uncore_pcu = {
+	.name			= "pcu",
+	.num_counters		= 4,
+	.num_boxes		= 1,
+	.perf_ctr_bits		= 48,
+	.perf_ctr		= SNR_PCU_MSR_PMON_CTR0,
+	.event_ctl		= SNR_PCU_MSR_PMON_CTL0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= SNR_PCU_MSR_PMON_BOX_CTL,
+	.num_shared_regs	= 1,
+	.ops			= &snr_uncore_pcu_ops,
+	.format_group		= &skx_uncore_pcu_format_group,
+};
+
+enum perf_uncore_snr_iio_freerunning_type_id {
+	SNR_IIO_MSR_IOCLK,
+	SNR_IIO_MSR_BW_IN,
+
+	SNR_IIO_FREERUNNING_TYPE_MAX,
+};
+
+static struct freerunning_counters snr_iio_freerunning[] = {
+	[SNR_IIO_MSR_IOCLK]	= { 0x1eac, 0x1, 0x10, 1, 48 },
+	[SNR_IIO_MSR_BW_IN]	= { 0x1f00, 0x1, 0x10, 8, 48 },
+};
+
+static struct uncore_event_desc snr_uncore_iio_freerunning_events[] = {
+	/* Free-Running IIO CLOCKS Counter */
+	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
+	/* Free-Running IIO BANDWIDTH IN Counters */
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4,		"event=0xff,umask=0x24"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5,		"event=0xff,umask=0x25"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6,		"event=0xff,umask=0x26"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7,		"event=0xff,umask=0x27"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit,	"MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_type snr_uncore_iio_free_running = {
+	.name			= "iio_free_running",
+	.num_counters		= 9,
+	.num_boxes		= 5,
+	.num_freerunning_types	= SNR_IIO_FREERUNNING_TYPE_MAX,
+	.freerunning		= snr_iio_freerunning,
+	.ops			= &skx_uncore_iio_freerunning_ops,
+	.event_descs		= snr_uncore_iio_freerunning_events,
+	.format_group		= &skx_uncore_iio_freerunning_format_group,
+};
+
+static struct intel_uncore_type *snr_msr_uncores[] = {
+	&snr_uncore_ubox,
+	&snr_uncore_chabox,
+	&snr_uncore_iio,
+	&snr_uncore_irp,
+	&snr_uncore_m2pcie,
+	&snr_uncore_pcu,
+	&snr_uncore_iio_free_running,
+	NULL,
+};
+
+void snr_uncore_cpu_init(void)
+{
+	uncore_msr_uncores = snr_msr_uncores;
+}
+
+static void snr_m2m_uncore_pci_init_box(struct intel_uncore_box *box)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	int box_ctl = uncore_pci_box_ctl(box);
+
+	__set_bit(UNCORE_BOX_FLAG_CTL_OFFS8, &box->flags);
+	pci_write_config_dword(pdev, box_ctl, IVBEP_PMON_BOX_CTL_INT);
+}
+
+static struct intel_uncore_ops snr_m2m_uncore_pci_ops = {
+	.init_box	= snr_m2m_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snbep_uncore_pci_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+};
+
+static struct attribute *snr_m2m_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask_ext3.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group snr_m2m_uncore_format_group = {
+	.name = "format",
+	.attrs = snr_m2m_uncore_formats_attr,
+};
+
+static struct intel_uncore_type snr_uncore_m2m = {
+	.name		= "m2m",
+	.num_counters   = 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= SNR_M2M_PCI_PMON_CTR0,
+	.event_ctl	= SNR_M2M_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext	= SNR_M2M_PCI_PMON_UMASK_EXT,
+	.box_ctl	= SNR_M2M_PCI_PMON_BOX_CTL,
+	.ops		= &snr_m2m_uncore_pci_ops,
+	.format_group	= &snr_m2m_uncore_format_group,
+};
+
+static void snr_uncore_pci_enable_event(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+
+	pci_write_config_dword(pdev, hwc->config_base, (u32)(hwc->config | SNBEP_PMON_CTL_EN));
+	pci_write_config_dword(pdev, hwc->config_base + 4, (u32)(hwc->config >> 32));
+}
+
+static struct intel_uncore_ops snr_pcie3_uncore_pci_ops = {
+	.init_box	= snr_m2m_uncore_pci_init_box,
+	.disable_box	= snbep_uncore_pci_disable_box,
+	.enable_box	= snbep_uncore_pci_enable_box,
+	.disable_event	= snbep_uncore_pci_disable_event,
+	.enable_event	= snr_uncore_pci_enable_event,
+	.read_counter	= snbep_uncore_pci_read_counter,
+};
+
+static struct intel_uncore_type snr_uncore_pcie3 = {
+	.name		= "pcie3",
+	.num_counters	= 4,
+	.num_boxes	= 1,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= SNR_PCIE3_PCI_PMON_CTR0,
+	.event_ctl	= SNR_PCIE3_PCI_PMON_CTL0,
+	.event_mask	= SKX_IIO_PMON_RAW_EVENT_MASK,
+	.event_mask_ext	= SKX_IIO_PMON_RAW_EVENT_MASK_EXT,
+	.box_ctl	= SNR_PCIE3_PCI_PMON_BOX_CTL,
+	.ops		= &snr_pcie3_uncore_pci_ops,
+	.format_group	= &skx_uncore_iio_format_group,
+};
+
+enum {
+	SNR_PCI_UNCORE_M2M,
+	SNR_PCI_UNCORE_PCIE3,
+};
+
+static struct intel_uncore_type *snr_pci_uncores[] = {
+	[SNR_PCI_UNCORE_M2M]		= &snr_uncore_m2m,
+	[SNR_PCI_UNCORE_PCIE3]		= &snr_uncore_pcie3,
+	NULL,
+};
+
+static const struct pci_device_id snr_uncore_pci_ids[] = {
+	{ /* M2M */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x344a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(12, 0, SNR_PCI_UNCORE_M2M, 0),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver snr_uncore_pci_driver = {
+	.name		= "snr_uncore",
+	.id_table	= snr_uncore_pci_ids,
+};
+
+static const struct pci_device_id snr_uncore_pci_sub_ids[] = {
+	{ /* PCIe3 RP */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x334a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(4, 0, SNR_PCI_UNCORE_PCIE3, 0),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver snr_uncore_pci_sub_driver = {
+	.name		= "snr_uncore_sub",
+	.id_table	= snr_uncore_pci_sub_ids,
+};
+
+int snr_uncore_pci_init(void)
+{
+	/* SNR UBOX DID */
+	int ret = snbep_pci2phy_map_init(0x3460, SKX_CPUNODEID,
+					 SKX_GIDNIDMAP, true);
+
+	if (ret)
+		return ret;
+
+	uncore_pci_uncores = snr_pci_uncores;
+	uncore_pci_driver = &snr_uncore_pci_driver;
+	uncore_pci_sub_driver = &snr_uncore_pci_sub_driver;
+	return 0;
+}
+
+static struct pci_dev *snr_uncore_get_mc_dev(int id)
+{
+	struct pci_dev *mc_dev = NULL;
+	int phys_id, pkg;
+
+	while (1) {
+		mc_dev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x3451, mc_dev);
+		if (!mc_dev)
+			break;
+		phys_id = uncore_pcibus_to_physid(mc_dev->bus);
+		if (phys_id < 0)
+			continue;
+		pkg = topology_phys_to_logical_pkg(phys_id);
+		if (pkg < 0)
+			continue;
+		else if (pkg == id)
+			break;
+	}
+	return mc_dev;
+}
+
+static void __snr_uncore_mmio_init_box(struct intel_uncore_box *box,
+				       unsigned int box_ctl, int mem_offset)
+{
+	struct pci_dev *pdev = snr_uncore_get_mc_dev(box->dieid);
+	struct intel_uncore_type *type = box->pmu->type;
+	resource_size_t addr;
+	u32 pci_dword;
+
+	if (!pdev)
+		return;
+
+	pci_read_config_dword(pdev, SNR_IMC_MMIO_BASE_OFFSET, &pci_dword);
+	addr = ((resource_size_t)pci_dword & SNR_IMC_MMIO_BASE_MASK) << 23;
+
+	pci_read_config_dword(pdev, mem_offset, &pci_dword);
+	addr |= (pci_dword & SNR_IMC_MMIO_MEM0_MASK) << 12;
+
+	addr += box_ctl;
+
+	pci_dev_put(pdev);
+
+	box->io_addr = ioremap(addr, type->mmio_map_size);
+	if (!box->io_addr) {
+		pr_warn("perf uncore: Failed to ioremap for %s.\n", type->name);
+		return;
+	}
+
+	writel(IVBEP_PMON_BOX_CTL_INT, box->io_addr);
+}
+
+static void snr_uncore_mmio_init_box(struct intel_uncore_box *box)
+{
+	__snr_uncore_mmio_init_box(box, uncore_mmio_box_ctl(box),
+				   SNR_IMC_MMIO_MEM0_OFFSET);
+}
+
+static void snr_uncore_mmio_disable_box(struct intel_uncore_box *box)
+{
+	u32 config;
+
+	if (!box->io_addr)
+		return;
+
+	config = readl(box->io_addr);
+	config |= SNBEP_PMON_BOX_CTL_FRZ;
+	writel(config, box->io_addr);
+}
+
+static void snr_uncore_mmio_enable_box(struct intel_uncore_box *box)
+{
+	u32 config;
+
+	if (!box->io_addr)
+		return;
+
+	config = readl(box->io_addr);
+	config &= ~SNBEP_PMON_BOX_CTL_FRZ;
+	writel(config, box->io_addr);
+}
+
+static void snr_uncore_mmio_enable_event(struct intel_uncore_box *box,
+					   struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!box->io_addr)
+		return;
+
+	if (!uncore_mmio_is_valid_offset(box, hwc->config_base))
+		return;
+
+	writel(hwc->config | SNBEP_PMON_CTL_EN,
+	       box->io_addr + hwc->config_base);
+}
+
+static void snr_uncore_mmio_disable_event(struct intel_uncore_box *box,
+					    struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!box->io_addr)
+		return;
+
+	if (!uncore_mmio_is_valid_offset(box, hwc->config_base))
+		return;
+
+	writel(hwc->config, box->io_addr + hwc->config_base);
+}
+
+static struct intel_uncore_ops snr_uncore_mmio_ops = {
+	.init_box	= snr_uncore_mmio_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.disable_box	= snr_uncore_mmio_disable_box,
+	.enable_box	= snr_uncore_mmio_enable_box,
+	.disable_event	= snr_uncore_mmio_disable_event,
+	.enable_event	= snr_uncore_mmio_enable_event,
+	.read_counter	= uncore_mmio_read_counter,
+};
+
+static struct uncore_event_desc snr_uncore_imc_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks,      "event=0x00,umask=0x00"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read,  "event=0x04,umask=0x0f"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_read.unit, "MiB"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write, "event=0x04,umask=0x30"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.scale, "6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(cas_count_write.unit, "MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_type snr_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 2,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNR_IMC_MMIO_PMON_FIXED_CTR,
+	.fixed_ctl	= SNR_IMC_MMIO_PMON_FIXED_CTL,
+	.event_descs	= snr_uncore_imc_events,
+	.perf_ctr	= SNR_IMC_MMIO_PMON_CTR0,
+	.event_ctl	= SNR_IMC_MMIO_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SNR_IMC_MMIO_PMON_BOX_CTL,
+	.mmio_offset	= SNR_IMC_MMIO_OFFSET,
+	.mmio_map_size	= SNR_IMC_MMIO_SIZE,
+	.ops		= &snr_uncore_mmio_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+enum perf_uncore_snr_imc_freerunning_type_id {
+	SNR_IMC_DCLK,
+	SNR_IMC_DDR,
+
+	SNR_IMC_FREERUNNING_TYPE_MAX,
+};
+
+static struct freerunning_counters snr_imc_freerunning[] = {
+	[SNR_IMC_DCLK]	= { 0x22b0, 0x0, 0, 1, 48 },
+	[SNR_IMC_DDR]	= { 0x2290, 0x8, 0, 2, 48 },
+};
+
+static struct uncore_event_desc snr_uncore_imc_freerunning_events[] = {
+	INTEL_UNCORE_EVENT_DESC(dclk,		"event=0xff,umask=0x10"),
+
+	INTEL_UNCORE_EVENT_DESC(read,		"event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(read.scale,	"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(read.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(write,		"event=0xff,umask=0x21"),
+	INTEL_UNCORE_EVENT_DESC(write.scale,	"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(write.unit,	"MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_ops snr_uncore_imc_freerunning_ops = {
+	.init_box	= snr_uncore_mmio_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.read_counter	= uncore_mmio_read_counter,
+	.hw_config	= uncore_freerunning_hw_config,
+};
+
+static struct intel_uncore_type snr_uncore_imc_free_running = {
+	.name			= "imc_free_running",
+	.num_counters		= 3,
+	.num_boxes		= 1,
+	.num_freerunning_types	= SNR_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= SNR_IMC_MMIO_SIZE,
+	.freerunning		= snr_imc_freerunning,
+	.ops			= &snr_uncore_imc_freerunning_ops,
+	.event_descs		= snr_uncore_imc_freerunning_events,
+	.format_group		= &skx_uncore_iio_freerunning_format_group,
+};
+
+static struct intel_uncore_type *snr_mmio_uncores[] = {
+	&snr_uncore_imc,
+	&snr_uncore_imc_free_running,
+	NULL,
+};
+
+void snr_uncore_mmio_init(void)
+{
+	uncore_mmio_uncores = snr_mmio_uncores;
+}
+
+/* end of SNR uncore support */
+
+/* ICX uncore support */
+
+static unsigned icx_cha_msr_offsets[] = {
+	0x2a0, 0x2ae, 0x2bc, 0x2ca, 0x2d8, 0x2e6, 0x2f4, 0x302, 0x310,
+	0x31e, 0x32c, 0x33a, 0x348, 0x356, 0x364, 0x372, 0x380, 0x38e,
+	0x3aa, 0x3b8, 0x3c6, 0x3d4, 0x3e2, 0x3f0, 0x3fe, 0x40c, 0x41a,
+	0x428, 0x436, 0x444, 0x452, 0x460, 0x46e, 0x47c, 0x0,   0xe,
+	0x1c,  0x2a,  0x38,  0x46,
+};
+
+static int icx_cha_hw_config(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event_extra *reg1 = &event->hw.extra_reg;
+	bool tie_en = !!(event->hw.config & SNBEP_CBO_PMON_CTL_TID_EN);
+
+	if (tie_en) {
+		reg1->reg = ICX_C34_MSR_PMON_BOX_FILTER0 +
+			    icx_cha_msr_offsets[box->pmu->pmu_idx];
+		reg1->config = event->attr.config1 & SKX_CHA_MSR_PMON_BOX_FILTER_TID;
+		reg1->idx = 0;
+	}
+
+	return 0;
+}
+
+static struct intel_uncore_ops icx_uncore_chabox_ops = {
+	.init_box		= ivbep_uncore_msr_init_box,
+	.disable_box		= snbep_uncore_msr_disable_box,
+	.enable_box		= snbep_uncore_msr_enable_box,
+	.disable_event		= snbep_uncore_msr_disable_event,
+	.enable_event		= snr_cha_enable_event,
+	.read_counter		= uncore_msr_read_counter,
+	.hw_config		= icx_cha_hw_config,
+};
+
+static struct intel_uncore_type icx_uncore_chabox = {
+	.name			= "cha",
+	.num_counters		= 4,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= ICX_C34_MSR_PMON_CTL0,
+	.perf_ctr		= ICX_C34_MSR_PMON_CTR0,
+	.box_ctl		= ICX_C34_MSR_PMON_BOX_CTL,
+	.msr_offsets		= icx_cha_msr_offsets,
+	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
+	.event_mask_ext		= SNR_CHA_RAW_EVENT_MASK_EXT,
+	.constraints		= skx_uncore_chabox_constraints,
+	.ops			= &icx_uncore_chabox_ops,
+	.format_group		= &snr_uncore_chabox_format_group,
+};
+
+static unsigned icx_msr_offsets[] = {
+	0x0, 0x20, 0x40, 0x90, 0xb0, 0xd0,
+};
+
+static struct event_constraint icx_uncore_iio_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x02, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x03, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x83, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x88, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xc0, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xc5, 0xc),
+	UNCORE_EVENT_CONSTRAINT(0xd5, 0xc),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type icx_uncore_iio = {
+	.name			= "iio",
+	.num_counters		= 4,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= ICX_IIO_MSR_PMON_CTL0,
+	.perf_ctr		= ICX_IIO_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext		= SNR_IIO_PMON_RAW_EVENT_MASK_EXT,
+	.box_ctl		= ICX_IIO_MSR_PMON_BOX_CTL,
+	.msr_offsets		= icx_msr_offsets,
+	.constraints		= icx_uncore_iio_constraints,
+	.ops			= &skx_uncore_iio_ops,
+	.format_group		= &snr_uncore_iio_format_group,
+};
+
+static struct intel_uncore_type icx_uncore_irp = {
+	.name			= "irp",
+	.num_counters		= 2,
+	.num_boxes		= 6,
+	.perf_ctr_bits		= 48,
+	.event_ctl		= ICX_IRP0_MSR_PMON_CTL0,
+	.perf_ctr		= ICX_IRP0_MSR_PMON_CTR0,
+	.event_mask		= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl		= ICX_IRP0_MSR_PMON_BOX_CTL,
+	.msr_offsets		= icx_msr_offsets,
+	.ops			= &ivbep_uncore_msr_ops,
+	.format_group		= &ivbep_uncore_format_group,
+};
+
+static struct event_constraint icx_uncore_m2pcie_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x14, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x23, 0x3),
+	UNCORE_EVENT_CONSTRAINT(0x2d, 0x3),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type icx_uncore_m2pcie = {
+	.name		= "m2pcie",
+	.num_counters	= 4,
+	.num_boxes	= 6,
+	.perf_ctr_bits	= 48,
+	.event_ctl	= ICX_M2PCIE_MSR_PMON_CTL0,
+	.perf_ctr	= ICX_M2PCIE_MSR_PMON_CTR0,
+	.box_ctl	= ICX_M2PCIE_MSR_PMON_BOX_CTL,
+	.msr_offsets	= icx_msr_offsets,
+	.constraints	= icx_uncore_m2pcie_constraints,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.ops		= &ivbep_uncore_msr_ops,
+	.format_group	= &ivbep_uncore_format_group,
+};
+
+enum perf_uncore_icx_iio_freerunning_type_id {
+	ICX_IIO_MSR_IOCLK,
+	ICX_IIO_MSR_BW_IN,
+
+	ICX_IIO_FREERUNNING_TYPE_MAX,
+};
+
+static unsigned icx_iio_clk_freerunning_box_offsets[] = {
+	0x0, 0x20, 0x40, 0x90, 0xb0, 0xd0,
+};
+
+static unsigned icx_iio_bw_freerunning_box_offsets[] = {
+	0x0, 0x10, 0x20, 0x90, 0xa0, 0xb0,
+};
+
+static struct freerunning_counters icx_iio_freerunning[] = {
+	[ICX_IIO_MSR_IOCLK]	= { 0xa55, 0x1, 0x20, 1, 48, icx_iio_clk_freerunning_box_offsets },
+	[ICX_IIO_MSR_BW_IN]	= { 0xaa0, 0x1, 0x10, 8, 48, icx_iio_bw_freerunning_box_offsets },
+};
+
+static struct uncore_event_desc icx_uncore_iio_freerunning_events[] = {
+	/* Free-Running IIO CLOCKS Counter */
+	INTEL_UNCORE_EVENT_DESC(ioclk,			"event=0xff,umask=0x10"),
+	/* Free-Running IIO BANDWIDTH IN Counters */
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0,		"event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port0.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1,		"event=0xff,umask=0x21"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port1.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2,		"event=0xff,umask=0x22"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port2.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3,		"event=0xff,umask=0x23"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port3.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4,		"event=0xff,umask=0x24"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port4.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5,		"event=0xff,umask=0x25"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port5.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6,		"event=0xff,umask=0x26"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port6.unit,	"MiB"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7,		"event=0xff,umask=0x27"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7.scale,	"3.814697266e-6"),
+	INTEL_UNCORE_EVENT_DESC(bw_in_port7.unit,	"MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static struct intel_uncore_type icx_uncore_iio_free_running = {
+	.name			= "iio_free_running",
+	.num_counters		= 9,
+	.num_boxes		= 6,
+	.num_freerunning_types	= ICX_IIO_FREERUNNING_TYPE_MAX,
+	.freerunning		= icx_iio_freerunning,
+	.ops			= &skx_uncore_iio_freerunning_ops,
+	.event_descs		= icx_uncore_iio_freerunning_events,
+	.format_group		= &skx_uncore_iio_freerunning_format_group,
+};
+
+static struct intel_uncore_type *icx_msr_uncores[] = {
+	&skx_uncore_ubox,
+	&icx_uncore_chabox,
+	&icx_uncore_iio,
+	&icx_uncore_irp,
+	&icx_uncore_m2pcie,
+	&skx_uncore_pcu,
+	&icx_uncore_iio_free_running,
+	NULL,
+};
+
+/*
+ * To determine the number of CHAs, it should read CAPID6(Low) and CAPID7 (High)
+ * registers which located at Device 30, Function 3
+ */
+#define ICX_CAPID6		0x9c
+#define ICX_CAPID7		0xa0
+
+static u64 icx_count_chabox(void)
+{
+	struct pci_dev *dev = NULL;
+	u64 caps = 0;
+
+	dev = pci_get_device(PCI_VENDOR_ID_INTEL, 0x345b, dev);
+	if (!dev)
+		goto out;
+
+	pci_read_config_dword(dev, ICX_CAPID6, (u32 *)&caps);
+	pci_read_config_dword(dev, ICX_CAPID7, (u32 *)&caps + 1);
+out:
+	pci_dev_put(dev);
+	return hweight64(caps);
+}
+
+void icx_uncore_cpu_init(void)
+{
+	u64 num_boxes = icx_count_chabox();
+
+	if (WARN_ON(num_boxes > ARRAY_SIZE(icx_cha_msr_offsets)))
+		return;
+	icx_uncore_chabox.num_boxes = num_boxes;
+	uncore_msr_uncores = icx_msr_uncores;
+}
+
+static struct intel_uncore_type icx_uncore_m2m = {
+	.name		= "m2m",
+	.num_counters   = 4,
+	.num_boxes	= 4,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= SNR_M2M_PCI_PMON_CTR0,
+	.event_ctl	= SNR_M2M_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext	= SNR_M2M_PCI_PMON_UMASK_EXT,
+	.box_ctl	= SNR_M2M_PCI_PMON_BOX_CTL,
+	.ops		= &snr_m2m_uncore_pci_ops,
+	.format_group	= &snr_m2m_uncore_format_group,
+};
+
+static struct attribute *icx_upi_uncore_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask_ext4.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_thresh8.attr,
+	NULL,
+};
+
+static const struct attribute_group icx_upi_uncore_format_group = {
+	.name = "format",
+	.attrs = icx_upi_uncore_formats_attr,
+};
+
+static struct intel_uncore_type icx_uncore_upi = {
+	.name		= "upi",
+	.num_counters   = 4,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= ICX_UPI_PCI_PMON_CTR0,
+	.event_ctl	= ICX_UPI_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.event_mask_ext = ICX_UPI_CTL_UMASK_EXT,
+	.box_ctl	= ICX_UPI_PCI_PMON_BOX_CTL,
+	.ops		= &skx_upi_uncore_pci_ops,
+	.format_group	= &icx_upi_uncore_format_group,
+};
+
+static struct event_constraint icx_uncore_m3upi_constraints[] = {
+	UNCORE_EVENT_CONSTRAINT(0x1c, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x1d, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x1e, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x1f, 0x1),
+	UNCORE_EVENT_CONSTRAINT(0x40, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x4e, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x4f, 0x7),
+	UNCORE_EVENT_CONSTRAINT(0x50, 0x7),
+	EVENT_CONSTRAINT_END
+};
+
+static struct intel_uncore_type icx_uncore_m3upi = {
+	.name		= "m3upi",
+	.num_counters   = 4,
+	.num_boxes	= 3,
+	.perf_ctr_bits	= 48,
+	.perf_ctr	= ICX_M3UPI_PCI_PMON_CTR0,
+	.event_ctl	= ICX_M3UPI_PCI_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= ICX_M3UPI_PCI_PMON_BOX_CTL,
+	.constraints	= icx_uncore_m3upi_constraints,
+	.ops		= &ivbep_uncore_pci_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+enum {
+	ICX_PCI_UNCORE_M2M,
+	ICX_PCI_UNCORE_UPI,
+	ICX_PCI_UNCORE_M3UPI,
+};
+
+static struct intel_uncore_type *icx_pci_uncores[] = {
+	[ICX_PCI_UNCORE_M2M]		= &icx_uncore_m2m,
+	[ICX_PCI_UNCORE_UPI]		= &icx_uncore_upi,
+	[ICX_PCI_UNCORE_M3UPI]		= &icx_uncore_m3upi,
+	NULL,
+};
+
+static const struct pci_device_id icx_uncore_pci_ids[] = {
+	{ /* M2M 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x344a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(12, 0, ICX_PCI_UNCORE_M2M, 0),
+	},
+	{ /* M2M 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x344a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(13, 0, ICX_PCI_UNCORE_M2M, 1),
+	},
+	{ /* M2M 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x344a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(14, 0, ICX_PCI_UNCORE_M2M, 2),
+	},
+	{ /* M2M 3 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x344a),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(15, 0, ICX_PCI_UNCORE_M2M, 3),
+	},
+	{ /* UPI Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3441),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(2, 1, ICX_PCI_UNCORE_UPI, 0),
+	},
+	{ /* UPI Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3441),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(3, 1, ICX_PCI_UNCORE_UPI, 1),
+	},
+	{ /* UPI Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3441),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(4, 1, ICX_PCI_UNCORE_UPI, 2),
+	},
+	{ /* M3UPI Link 0 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3446),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(5, 1, ICX_PCI_UNCORE_M3UPI, 0),
+	},
+	{ /* M3UPI Link 1 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3446),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(6, 1, ICX_PCI_UNCORE_M3UPI, 1),
+	},
+	{ /* M3UPI Link 2 */
+		PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x3446),
+		.driver_data = UNCORE_PCI_DEV_FULL_DATA(7, 1, ICX_PCI_UNCORE_M3UPI, 2),
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_driver icx_uncore_pci_driver = {
+	.name		= "icx_uncore",
+	.id_table	= icx_uncore_pci_ids,
+};
+
+int icx_uncore_pci_init(void)
+{
+	/* ICX UBOX DID */
+	int ret = snbep_pci2phy_map_init(0x3450, SKX_CPUNODEID,
+					 SKX_GIDNIDMAP, true);
+
+	if (ret)
+		return ret;
+
+	uncore_pci_uncores = icx_pci_uncores;
+	uncore_pci_driver = &icx_uncore_pci_driver;
+	return 0;
+}
+
+static void icx_uncore_imc_init_box(struct intel_uncore_box *box)
+{
+	unsigned int box_ctl = box->pmu->type->box_ctl +
+			       box->pmu->type->mmio_offset * (box->pmu->pmu_idx % ICX_NUMBER_IMC_CHN);
+	int mem_offset = (box->pmu->pmu_idx / ICX_NUMBER_IMC_CHN) * ICX_IMC_MEM_STRIDE +
+			 SNR_IMC_MMIO_MEM0_OFFSET;
+
+	__snr_uncore_mmio_init_box(box, box_ctl, mem_offset);
+}
+
+static struct intel_uncore_ops icx_uncore_mmio_ops = {
+	.init_box	= icx_uncore_imc_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.disable_box	= snr_uncore_mmio_disable_box,
+	.enable_box	= snr_uncore_mmio_enable_box,
+	.disable_event	= snr_uncore_mmio_disable_event,
+	.enable_event	= snr_uncore_mmio_enable_event,
+	.read_counter	= uncore_mmio_read_counter,
+};
+
+static struct intel_uncore_type icx_uncore_imc = {
+	.name		= "imc",
+	.num_counters   = 4,
+	.num_boxes	= 12,
+	.perf_ctr_bits	= 48,
+	.fixed_ctr_bits	= 48,
+	.fixed_ctr	= SNR_IMC_MMIO_PMON_FIXED_CTR,
+	.fixed_ctl	= SNR_IMC_MMIO_PMON_FIXED_CTL,
+	.event_descs	= snr_uncore_imc_events,
+	.perf_ctr	= SNR_IMC_MMIO_PMON_CTR0,
+	.event_ctl	= SNR_IMC_MMIO_PMON_CTL0,
+	.event_mask	= SNBEP_PMON_RAW_EVENT_MASK,
+	.box_ctl	= SNR_IMC_MMIO_PMON_BOX_CTL,
+	.mmio_offset	= SNR_IMC_MMIO_OFFSET,
+	.mmio_map_size	= SNR_IMC_MMIO_SIZE,
+	.ops		= &icx_uncore_mmio_ops,
+	.format_group	= &skx_uncore_format_group,
+};
+
+enum perf_uncore_icx_imc_freerunning_type_id {
+	ICX_IMC_DCLK,
+	ICX_IMC_DDR,
+	ICX_IMC_DDRT,
+
+	ICX_IMC_FREERUNNING_TYPE_MAX,
+};
+
+static struct freerunning_counters icx_imc_freerunning[] = {
+	[ICX_IMC_DCLK]	= { 0x22b0, 0x0, 0, 1, 48 },
+	[ICX_IMC_DDR]	= { 0x2290, 0x8, 0, 2, 48 },
+	[ICX_IMC_DDRT]	= { 0x22a0, 0x8, 0, 2, 48 },
+};
+
+static struct uncore_event_desc icx_uncore_imc_freerunning_events[] = {
+	INTEL_UNCORE_EVENT_DESC(dclk,			"event=0xff,umask=0x10"),
+
+	INTEL_UNCORE_EVENT_DESC(read,			"event=0xff,umask=0x20"),
+	INTEL_UNCORE_EVENT_DESC(read.scale,		"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(read.unit,		"MiB"),
+	INTEL_UNCORE_EVENT_DESC(write,			"event=0xff,umask=0x21"),
+	INTEL_UNCORE_EVENT_DESC(write.scale,		"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(write.unit,		"MiB"),
+
+	INTEL_UNCORE_EVENT_DESC(ddrt_read,		"event=0xff,umask=0x30"),
+	INTEL_UNCORE_EVENT_DESC(ddrt_read.scale,	"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(ddrt_read.unit,		"MiB"),
+	INTEL_UNCORE_EVENT_DESC(ddrt_write,		"event=0xff,umask=0x31"),
+	INTEL_UNCORE_EVENT_DESC(ddrt_write.scale,	"6.103515625e-5"),
+	INTEL_UNCORE_EVENT_DESC(ddrt_write.unit,	"MiB"),
+	{ /* end: all zeroes */ },
+};
+
+static void icx_uncore_imc_freerunning_init_box(struct intel_uncore_box *box)
+{
+	int mem_offset = box->pmu->pmu_idx * ICX_IMC_MEM_STRIDE +
+			 SNR_IMC_MMIO_MEM0_OFFSET;
+
+	__snr_uncore_mmio_init_box(box, uncore_mmio_box_ctl(box), mem_offset);
+}
+
+static struct intel_uncore_ops icx_uncore_imc_freerunning_ops = {
+	.init_box	= icx_uncore_imc_freerunning_init_box,
+	.exit_box	= uncore_mmio_exit_box,
+	.read_counter	= uncore_mmio_read_counter,
+	.hw_config	= uncore_freerunning_hw_config,
+};
+
+static struct intel_uncore_type icx_uncore_imc_free_running = {
+	.name			= "imc_free_running",
+	.num_counters		= 5,
+	.num_boxes		= 4,
+	.num_freerunning_types	= ICX_IMC_FREERUNNING_TYPE_MAX,
+	.mmio_map_size		= SNR_IMC_MMIO_SIZE,
+	.freerunning		= icx_imc_freerunning,
+	.ops			= &icx_uncore_imc_freerunning_ops,
+	.event_descs		= icx_uncore_imc_freerunning_events,
+	.format_group		= &skx_uncore_iio_freerunning_format_group,
+};
+
+static struct intel_uncore_type *icx_mmio_uncores[] = {
+	&icx_uncore_imc,
+	&icx_uncore_imc_free_running,
+	NULL,
+};
+
+void icx_uncore_mmio_init(void)
+{
+	uncore_mmio_uncores = icx_mmio_uncores;
+}
+
+/* end of ICX uncore support */
diff --git a/arch/x86/events/msr.c b/arch/x86/events/msr.c
new file mode 100644
index 000000000..ca8ce64df
--- /dev/null
+++ b/arch/x86/events/msr.c
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/perf_event.h>
+#include <linux/sysfs.h>
+#include <linux/nospec.h>
+#include <asm/intel-family.h>
+#include "probe.h"
+
+enum perf_msr_id {
+	PERF_MSR_TSC			= 0,
+	PERF_MSR_APERF			= 1,
+	PERF_MSR_MPERF			= 2,
+	PERF_MSR_PPERF			= 3,
+	PERF_MSR_SMI			= 4,
+	PERF_MSR_PTSC			= 5,
+	PERF_MSR_IRPERF			= 6,
+	PERF_MSR_THERM			= 7,
+	PERF_MSR_EVENT_MAX,
+};
+
+static bool test_aperfmperf(int idx, void *data)
+{
+	return boot_cpu_has(X86_FEATURE_APERFMPERF);
+}
+
+static bool test_ptsc(int idx, void *data)
+{
+	return boot_cpu_has(X86_FEATURE_PTSC);
+}
+
+static bool test_irperf(int idx, void *data)
+{
+	return boot_cpu_has(X86_FEATURE_IRPERF);
+}
+
+static bool test_therm_status(int idx, void *data)
+{
+	return boot_cpu_has(X86_FEATURE_DTHERM);
+}
+
+static bool test_intel(int idx, void *data)
+{
+	if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL ||
+	    boot_cpu_data.x86 != 6)
+		return false;
+
+	switch (boot_cpu_data.x86_model) {
+	case INTEL_FAM6_NEHALEM:
+	case INTEL_FAM6_NEHALEM_G:
+	case INTEL_FAM6_NEHALEM_EP:
+	case INTEL_FAM6_NEHALEM_EX:
+
+	case INTEL_FAM6_WESTMERE:
+	case INTEL_FAM6_WESTMERE_EP:
+	case INTEL_FAM6_WESTMERE_EX:
+
+	case INTEL_FAM6_SANDYBRIDGE:
+	case INTEL_FAM6_SANDYBRIDGE_X:
+
+	case INTEL_FAM6_IVYBRIDGE:
+	case INTEL_FAM6_IVYBRIDGE_X:
+
+	case INTEL_FAM6_HASWELL:
+	case INTEL_FAM6_HASWELL_X:
+	case INTEL_FAM6_HASWELL_L:
+	case INTEL_FAM6_HASWELL_G:
+
+	case INTEL_FAM6_BROADWELL:
+	case INTEL_FAM6_BROADWELL_D:
+	case INTEL_FAM6_BROADWELL_G:
+	case INTEL_FAM6_BROADWELL_X:
+	case INTEL_FAM6_SAPPHIRERAPIDS_X:
+
+	case INTEL_FAM6_ATOM_SILVERMONT:
+	case INTEL_FAM6_ATOM_SILVERMONT_D:
+	case INTEL_FAM6_ATOM_AIRMONT:
+
+	case INTEL_FAM6_ATOM_GOLDMONT:
+	case INTEL_FAM6_ATOM_GOLDMONT_D:
+	case INTEL_FAM6_ATOM_GOLDMONT_PLUS:
+	case INTEL_FAM6_ATOM_TREMONT_D:
+	case INTEL_FAM6_ATOM_TREMONT:
+	case INTEL_FAM6_ATOM_TREMONT_L:
+
+	case INTEL_FAM6_XEON_PHI_KNL:
+	case INTEL_FAM6_XEON_PHI_KNM:
+		if (idx == PERF_MSR_SMI)
+			return true;
+		break;
+
+	case INTEL_FAM6_SKYLAKE_L:
+	case INTEL_FAM6_SKYLAKE:
+	case INTEL_FAM6_SKYLAKE_X:
+	case INTEL_FAM6_KABYLAKE_L:
+	case INTEL_FAM6_KABYLAKE:
+	case INTEL_FAM6_COMETLAKE_L:
+	case INTEL_FAM6_COMETLAKE:
+	case INTEL_FAM6_ICELAKE_L:
+	case INTEL_FAM6_ICELAKE:
+	case INTEL_FAM6_ICELAKE_X:
+	case INTEL_FAM6_ICELAKE_D:
+	case INTEL_FAM6_TIGERLAKE_L:
+	case INTEL_FAM6_TIGERLAKE:
+		if (idx == PERF_MSR_SMI || idx == PERF_MSR_PPERF)
+			return true;
+		break;
+	}
+
+	return false;
+}
+
+PMU_EVENT_ATTR_STRING(tsc,				attr_tsc,		"event=0x00"	);
+PMU_EVENT_ATTR_STRING(aperf,				attr_aperf,		"event=0x01"	);
+PMU_EVENT_ATTR_STRING(mperf,				attr_mperf,		"event=0x02"	);
+PMU_EVENT_ATTR_STRING(pperf,				attr_pperf,		"event=0x03"	);
+PMU_EVENT_ATTR_STRING(smi,				attr_smi,		"event=0x04"	);
+PMU_EVENT_ATTR_STRING(ptsc,				attr_ptsc,		"event=0x05"	);
+PMU_EVENT_ATTR_STRING(irperf,				attr_irperf,		"event=0x06"	);
+PMU_EVENT_ATTR_STRING(cpu_thermal_margin,		attr_therm,		"event=0x07"	);
+PMU_EVENT_ATTR_STRING(cpu_thermal_margin.snapshot,	attr_therm_snap,	"1"		);
+PMU_EVENT_ATTR_STRING(cpu_thermal_margin.unit,		attr_therm_unit,	"C"		);
+
+static unsigned long msr_mask;
+
+PMU_EVENT_GROUP(events, aperf);
+PMU_EVENT_GROUP(events, mperf);
+PMU_EVENT_GROUP(events, pperf);
+PMU_EVENT_GROUP(events, smi);
+PMU_EVENT_GROUP(events, ptsc);
+PMU_EVENT_GROUP(events, irperf);
+
+static struct attribute *attrs_therm[] = {
+	&attr_therm.attr.attr,
+	&attr_therm_snap.attr.attr,
+	&attr_therm_unit.attr.attr,
+	NULL,
+};
+
+static struct attribute_group group_therm = {
+	.name  = "events",
+	.attrs = attrs_therm,
+};
+
+static struct perf_msr msr[] = {
+	[PERF_MSR_TSC]		= { .no_check = true,								},
+	[PERF_MSR_APERF]	= { MSR_IA32_APERF,		&group_aperf,		test_aperfmperf,	},
+	[PERF_MSR_MPERF]	= { MSR_IA32_MPERF,		&group_mperf,		test_aperfmperf,	},
+	[PERF_MSR_PPERF]	= { MSR_PPERF,			&group_pperf,		test_intel,		},
+	[PERF_MSR_SMI]		= { MSR_SMI_COUNT,		&group_smi,		test_intel,		},
+	[PERF_MSR_PTSC]		= { MSR_F15H_PTSC,		&group_ptsc,		test_ptsc,		},
+	[PERF_MSR_IRPERF]	= { MSR_F17H_IRPERF,		&group_irperf,		test_irperf,		},
+	[PERF_MSR_THERM]	= { MSR_IA32_THERM_STATUS,	&group_therm,		test_therm_status,	},
+};
+
+static struct attribute *events_attrs[] = {
+	&attr_tsc.attr.attr,
+	NULL,
+};
+
+static struct attribute_group events_attr_group = {
+	.name = "events",
+	.attrs = events_attrs,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-63");
+static struct attribute *format_attrs[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+static struct attribute_group format_attr_group = {
+	.name = "format",
+	.attrs = format_attrs,
+};
+
+static const struct attribute_group *attr_groups[] = {
+	&events_attr_group,
+	&format_attr_group,
+	NULL,
+};
+
+static const struct attribute_group *attr_update[] = {
+	&group_aperf,
+	&group_mperf,
+	&group_pperf,
+	&group_smi,
+	&group_ptsc,
+	&group_irperf,
+	&group_therm,
+	NULL,
+};
+
+static int msr_event_init(struct perf_event *event)
+{
+	u64 cfg = event->attr.config;
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	/* unsupported modes and filters */
+	if (event->attr.sample_period) /* no sampling */
+		return -EINVAL;
+
+	if (cfg >= PERF_MSR_EVENT_MAX)
+		return -EINVAL;
+
+	cfg = array_index_nospec((unsigned long)cfg, PERF_MSR_EVENT_MAX);
+
+	if (!(msr_mask & (1 << cfg)))
+		return -EINVAL;
+
+	event->hw.idx		= -1;
+	event->hw.event_base	= msr[cfg].msr;
+	event->hw.config	= cfg;
+
+	return 0;
+}
+
+static inline u64 msr_read_counter(struct perf_event *event)
+{
+	u64 now;
+
+	if (event->hw.event_base)
+		rdmsrl(event->hw.event_base, now);
+	else
+		now = rdtsc_ordered();
+
+	return now;
+}
+
+static void msr_event_update(struct perf_event *event)
+{
+	u64 prev, now;
+	s64 delta;
+
+	/* Careful, an NMI might modify the previous event value: */
+again:
+	prev = local64_read(&event->hw.prev_count);
+	now = msr_read_counter(event);
+
+	if (local64_cmpxchg(&event->hw.prev_count, prev, now) != prev)
+		goto again;
+
+	delta = now - prev;
+	if (unlikely(event->hw.event_base == MSR_SMI_COUNT)) {
+		delta = sign_extend64(delta, 31);
+		local64_add(delta, &event->count);
+	} else if (unlikely(event->hw.event_base == MSR_IA32_THERM_STATUS)) {
+		/* If valid, extract digital readout, otherwise set to -1: */
+		now = now & (1ULL << 31) ? (now >> 16) & 0x3f :  -1;
+		local64_set(&event->count, now);
+	} else {
+		local64_add(delta, &event->count);
+	}
+}
+
+static void msr_event_start(struct perf_event *event, int flags)
+{
+	u64 now = msr_read_counter(event);
+
+	local64_set(&event->hw.prev_count, now);
+}
+
+static void msr_event_stop(struct perf_event *event, int flags)
+{
+	msr_event_update(event);
+}
+
+static void msr_event_del(struct perf_event *event, int flags)
+{
+	msr_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int msr_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		msr_event_start(event, flags);
+
+	return 0;
+}
+
+static struct pmu pmu_msr = {
+	.task_ctx_nr	= perf_sw_context,
+	.attr_groups	= attr_groups,
+	.event_init	= msr_event_init,
+	.add		= msr_event_add,
+	.del		= msr_event_del,
+	.start		= msr_event_start,
+	.stop		= msr_event_stop,
+	.read		= msr_event_update,
+	.capabilities	= PERF_PMU_CAP_NO_INTERRUPT | PERF_PMU_CAP_NO_EXCLUDE,
+	.attr_update	= attr_update,
+};
+
+static int __init msr_init(void)
+{
+	if (!boot_cpu_has(X86_FEATURE_TSC)) {
+		pr_cont("no MSR PMU driver.\n");
+		return 0;
+	}
+
+	msr_mask = perf_msr_probe(msr, PERF_MSR_EVENT_MAX, true, NULL);
+
+	perf_pmu_register(&pmu_msr, "msr", -1);
+
+	return 0;
+}
+device_initcall(msr_init);
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
new file mode 100644
index 000000000..0c3b8fa7e
--- /dev/null
+++ b/arch/x86/events/perf_event.h
@@ -0,0 +1,1306 @@
+/*
+ * Performance events x86 architecture header
+ *
+ *  Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
+ *  Copyright (C) 2009 Jaswinder Singh Rajput
+ *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra
+ *  Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
+ *  Copyright (C) 2009 Google, Inc., Stephane Eranian
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+
+#include <asm/intel_ds.h>
+
+/* To enable MSR tracing please use the generic trace points. */
+
+/*
+ *          |   NHM/WSM    |      SNB     |
+ * register -------------------------------
+ *          |  HT  | no HT |  HT  | no HT |
+ *-----------------------------------------
+ * offcore  | core | core  | cpu  | core  |
+ * lbr_sel  | core | core  | cpu  | core  |
+ * ld_lat   | cpu  | core  | cpu  | core  |
+ *-----------------------------------------
+ *
+ * Given that there is a small number of shared regs,
+ * we can pre-allocate their slot in the per-cpu
+ * per-core reg tables.
+ */
+enum extra_reg_type {
+	EXTRA_REG_NONE  = -1,	/* not used */
+
+	EXTRA_REG_RSP_0 = 0,	/* offcore_response_0 */
+	EXTRA_REG_RSP_1 = 1,	/* offcore_response_1 */
+	EXTRA_REG_LBR   = 2,	/* lbr_select */
+	EXTRA_REG_LDLAT = 3,	/* ld_lat_threshold */
+	EXTRA_REG_FE    = 4,    /* fe_* */
+
+	EXTRA_REG_MAX		/* number of entries needed */
+};
+
+struct event_constraint {
+	union {
+		unsigned long	idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+		u64		idxmsk64;
+	};
+	u64		code;
+	u64		cmask;
+	int		weight;
+	int		overlap;
+	int		flags;
+	unsigned int	size;
+};
+
+static inline bool constraint_match(struct event_constraint *c, u64 ecode)
+{
+	return ((ecode & c->cmask) - c->code) <= (u64)c->size;
+}
+
+/*
+ * struct hw_perf_event.flags flags
+ */
+#define PERF_X86_EVENT_PEBS_LDLAT	0x0001 /* ld+ldlat data address sampling */
+#define PERF_X86_EVENT_PEBS_ST		0x0002 /* st data address sampling */
+#define PERF_X86_EVENT_PEBS_ST_HSW	0x0004 /* haswell style datala, store */
+#define PERF_X86_EVENT_PEBS_LD_HSW	0x0008 /* haswell style datala, load */
+#define PERF_X86_EVENT_PEBS_NA_HSW	0x0010 /* haswell style datala, unknown */
+#define PERF_X86_EVENT_EXCL		0x0020 /* HT exclusivity on counter */
+#define PERF_X86_EVENT_DYNAMIC		0x0040 /* dynamic alloc'd constraint */
+#define PERF_X86_EVENT_RDPMC_ALLOWED	0x0080 /* grant rdpmc permission */
+#define PERF_X86_EVENT_EXCL_ACCT	0x0100 /* accounted EXCL event */
+#define PERF_X86_EVENT_AUTO_RELOAD	0x0200 /* use PEBS auto-reload */
+#define PERF_X86_EVENT_LARGE_PEBS	0x0400 /* use large PEBS */
+#define PERF_X86_EVENT_PEBS_VIA_PT	0x0800 /* use PT buffer for PEBS */
+#define PERF_X86_EVENT_PAIR		0x1000 /* Large Increment per Cycle */
+#define PERF_X86_EVENT_LBR_SELECT	0x2000 /* Save/Restore MSR_LBR_SELECT */
+#define PERF_X86_EVENT_TOPDOWN		0x4000 /* Count Topdown slots/metrics events */
+
+static inline bool is_topdown_count(struct perf_event *event)
+{
+	return event->hw.flags & PERF_X86_EVENT_TOPDOWN;
+}
+
+static inline bool is_metric_event(struct perf_event *event)
+{
+	u64 config = event->attr.config;
+
+	return ((config & ARCH_PERFMON_EVENTSEL_EVENT) == 0) &&
+		((config & INTEL_ARCH_EVENT_MASK) >= INTEL_TD_METRIC_RETIRING)  &&
+		((config & INTEL_ARCH_EVENT_MASK) <= INTEL_TD_METRIC_MAX);
+}
+
+static inline bool is_slots_event(struct perf_event *event)
+{
+	return (event->attr.config & INTEL_ARCH_EVENT_MASK) == INTEL_TD_SLOTS;
+}
+
+static inline bool is_topdown_event(struct perf_event *event)
+{
+	return is_metric_event(event) || is_slots_event(event);
+}
+
+struct amd_nb {
+	int nb_id;  /* NorthBridge id */
+	int refcnt; /* reference count */
+	struct perf_event *owners[X86_PMC_IDX_MAX];
+	struct event_constraint event_constraints[X86_PMC_IDX_MAX];
+};
+
+#define PEBS_COUNTER_MASK	((1ULL << MAX_PEBS_EVENTS) - 1)
+#define PEBS_PMI_AFTER_EACH_RECORD BIT_ULL(60)
+#define PEBS_OUTPUT_OFFSET	61
+#define PEBS_OUTPUT_MASK	(3ull << PEBS_OUTPUT_OFFSET)
+#define PEBS_OUTPUT_PT		(1ull << PEBS_OUTPUT_OFFSET)
+#define PEBS_VIA_PT_MASK	(PEBS_OUTPUT_PT | PEBS_PMI_AFTER_EACH_RECORD)
+
+/*
+ * Flags PEBS can handle without an PMI.
+ *
+ * TID can only be handled by flushing at context switch.
+ * REGS_USER can be handled for events limited to ring 3.
+ *
+ */
+#define LARGE_PEBS_FLAGS \
+	(PERF_SAMPLE_IP | PERF_SAMPLE_TID | PERF_SAMPLE_ADDR | \
+	PERF_SAMPLE_ID | PERF_SAMPLE_CPU | PERF_SAMPLE_STREAM_ID | \
+	PERF_SAMPLE_DATA_SRC | PERF_SAMPLE_IDENTIFIER | \
+	PERF_SAMPLE_TRANSACTION | PERF_SAMPLE_PHYS_ADDR | \
+	PERF_SAMPLE_REGS_INTR | PERF_SAMPLE_REGS_USER | \
+	PERF_SAMPLE_PERIOD)
+
+#define PEBS_GP_REGS			\
+	((1ULL << PERF_REG_X86_AX)    | \
+	 (1ULL << PERF_REG_X86_BX)    | \
+	 (1ULL << PERF_REG_X86_CX)    | \
+	 (1ULL << PERF_REG_X86_DX)    | \
+	 (1ULL << PERF_REG_X86_DI)    | \
+	 (1ULL << PERF_REG_X86_SI)    | \
+	 (1ULL << PERF_REG_X86_SP)    | \
+	 (1ULL << PERF_REG_X86_BP)    | \
+	 (1ULL << PERF_REG_X86_IP)    | \
+	 (1ULL << PERF_REG_X86_FLAGS) | \
+	 (1ULL << PERF_REG_X86_R8)    | \
+	 (1ULL << PERF_REG_X86_R9)    | \
+	 (1ULL << PERF_REG_X86_R10)   | \
+	 (1ULL << PERF_REG_X86_R11)   | \
+	 (1ULL << PERF_REG_X86_R12)   | \
+	 (1ULL << PERF_REG_X86_R13)   | \
+	 (1ULL << PERF_REG_X86_R14)   | \
+	 (1ULL << PERF_REG_X86_R15))
+
+/*
+ * Per register state.
+ */
+struct er_account {
+	raw_spinlock_t      lock;	/* per-core: protect structure */
+	u64                 config;	/* extra MSR config */
+	u64                 reg;	/* extra MSR number */
+	atomic_t            ref;	/* reference count */
+};
+
+/*
+ * Per core/cpu state
+ *
+ * Used to coordinate shared registers between HT threads or
+ * among events on a single PMU.
+ */
+struct intel_shared_regs {
+	struct er_account       regs[EXTRA_REG_MAX];
+	int                     refcnt;		/* per-core: #HT threads */
+	unsigned                core_id;	/* per-core: core id */
+};
+
+enum intel_excl_state_type {
+	INTEL_EXCL_UNUSED    = 0, /* counter is unused */
+	INTEL_EXCL_SHARED    = 1, /* counter can be used by both threads */
+	INTEL_EXCL_EXCLUSIVE = 2, /* counter can be used by one thread only */
+};
+
+struct intel_excl_states {
+	enum intel_excl_state_type state[X86_PMC_IDX_MAX];
+	bool sched_started; /* true if scheduling has started */
+};
+
+struct intel_excl_cntrs {
+	raw_spinlock_t	lock;
+
+	struct intel_excl_states states[2];
+
+	union {
+		u16	has_exclusive[2];
+		u32	exclusive_present;
+	};
+
+	int		refcnt;		/* per-core: #HT threads */
+	unsigned	core_id;	/* per-core: core id */
+};
+
+struct x86_perf_task_context;
+#define MAX_LBR_ENTRIES		32
+
+enum {
+	LBR_FORMAT_32		= 0x00,
+	LBR_FORMAT_LIP		= 0x01,
+	LBR_FORMAT_EIP		= 0x02,
+	LBR_FORMAT_EIP_FLAGS	= 0x03,
+	LBR_FORMAT_EIP_FLAGS2	= 0x04,
+	LBR_FORMAT_INFO		= 0x05,
+	LBR_FORMAT_TIME		= 0x06,
+	LBR_FORMAT_MAX_KNOWN    = LBR_FORMAT_TIME,
+};
+
+enum {
+	X86_PERF_KFREE_SHARED = 0,
+	X86_PERF_KFREE_EXCL   = 1,
+	X86_PERF_KFREE_MAX
+};
+
+struct cpu_hw_events {
+	/*
+	 * Generic x86 PMC bits
+	 */
+	struct perf_event	*events[X86_PMC_IDX_MAX]; /* in counter order */
+	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long		running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	int			enabled;
+
+	int			n_events; /* the # of events in the below arrays */
+	int			n_added;  /* the # last events in the below arrays;
+					     they've never been enabled yet */
+	int			n_txn;    /* the # last events in the below arrays;
+					     added in the current transaction */
+	int			n_txn_pair;
+	int			n_txn_metric;
+	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
+	u64			tags[X86_PMC_IDX_MAX];
+
+	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+	struct event_constraint	*event_constraint[X86_PMC_IDX_MAX];
+
+	int			n_excl; /* the number of exclusive events */
+
+	unsigned int		txn_flags;
+	int			is_fake;
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	struct debug_store	*ds;
+	void			*ds_pebs_vaddr;
+	void			*ds_bts_vaddr;
+	u64			pebs_enabled;
+	int			n_pebs;
+	int			n_large_pebs;
+	int			n_pebs_via_pt;
+	int			pebs_output;
+
+	/* Current super set of events hardware configuration */
+	u64			pebs_data_cfg;
+	u64			active_pebs_data_cfg;
+	int			pebs_record_size;
+
+	/*
+	 * Intel LBR bits
+	 */
+	int				lbr_users;
+	int				lbr_pebs_users;
+	struct perf_branch_stack	lbr_stack;
+	struct perf_branch_entry	lbr_entries[MAX_LBR_ENTRIES];
+	union {
+		struct er_account		*lbr_sel;
+		struct er_account		*lbr_ctl;
+	};
+	u64				br_sel;
+	void				*last_task_ctx;
+	int				last_log_id;
+	int				lbr_select;
+	void				*lbr_xsave;
+
+	/*
+	 * Intel host/guest exclude bits
+	 */
+	u64				intel_ctrl_guest_mask;
+	u64				intel_ctrl_host_mask;
+	struct perf_guest_switch_msr	guest_switch_msrs[X86_PMC_IDX_MAX];
+
+	/*
+	 * Intel checkpoint mask
+	 */
+	u64				intel_cp_status;
+
+	/*
+	 * manage shared (per-core, per-cpu) registers
+	 * used on Intel NHM/WSM/SNB
+	 */
+	struct intel_shared_regs	*shared_regs;
+	/*
+	 * manage exclusive counter access between hyperthread
+	 */
+	struct event_constraint *constraint_list; /* in enable order */
+	struct intel_excl_cntrs		*excl_cntrs;
+	int excl_thread_id; /* 0 or 1 */
+
+	/*
+	 * SKL TSX_FORCE_ABORT shadow
+	 */
+	u64				tfa_shadow;
+
+	/*
+	 * Perf Metrics
+	 */
+	/* number of accepted metrics events */
+	int				n_metric;
+
+	/*
+	 * AMD specific bits
+	 */
+	struct amd_nb			*amd_nb;
+	/* Inverted mask of bits to clear in the perf_ctr ctrl registers */
+	u64				perf_ctr_virt_mask;
+	int				n_pair; /* Large increment events */
+
+	void				*kfree_on_online[X86_PERF_KFREE_MAX];
+
+	struct pmu			*pmu;
+};
+
+#define __EVENT_CONSTRAINT_RANGE(c, e, n, m, w, o, f) {	\
+	{ .idxmsk64 = (n) },		\
+	.code = (c),			\
+	.size = (e) - (c),		\
+	.cmask = (m),			\
+	.weight = (w),			\
+	.overlap = (o),			\
+	.flags = f,			\
+}
+
+#define __EVENT_CONSTRAINT(c, n, m, w, o, f) \
+	__EVENT_CONSTRAINT_RANGE(c, c, n, m, w, o, f)
+
+#define EVENT_CONSTRAINT(c, n, m)	\
+	__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n), 0, 0)
+
+/*
+ * The constraint_match() function only works for 'simple' event codes
+ * and not for extended (AMD64_EVENTSEL_EVENT) events codes.
+ */
+#define EVENT_CONSTRAINT_RANGE(c, e, n, m) \
+	__EVENT_CONSTRAINT_RANGE(c, e, n, m, HWEIGHT(n), 0, 0)
+
+#define INTEL_EXCLEVT_CONSTRAINT(c, n)	\
+	__EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT, HWEIGHT(n),\
+			   0, PERF_X86_EVENT_EXCL)
+
+/*
+ * The overlap flag marks event constraints with overlapping counter
+ * masks. This is the case if the counter mask of such an event is not
+ * a subset of any other counter mask of a constraint with an equal or
+ * higher weight, e.g.:
+ *
+ *  c_overlaps = EVENT_CONSTRAINT_OVERLAP(0, 0x09, 0);
+ *  c_another1 = EVENT_CONSTRAINT(0, 0x07, 0);
+ *  c_another2 = EVENT_CONSTRAINT(0, 0x38, 0);
+ *
+ * The event scheduler may not select the correct counter in the first
+ * cycle because it needs to know which subsequent events will be
+ * scheduled. It may fail to schedule the events then. So we set the
+ * overlap flag for such constraints to give the scheduler a hint which
+ * events to select for counter rescheduling.
+ *
+ * Care must be taken as the rescheduling algorithm is O(n!) which
+ * will increase scheduling cycles for an over-committed system
+ * dramatically.  The number of such EVENT_CONSTRAINT_OVERLAP() macros
+ * and its counter masks must be kept at a minimum.
+ */
+#define EVENT_CONSTRAINT_OVERLAP(c, n, m)	\
+	__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n), 1, 0)
+
+/*
+ * Constraint on the Event code.
+ */
+#define INTEL_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
+
+/*
+ * Constraint on a range of Event codes
+ */
+#define INTEL_EVENT_CONSTRAINT_RANGE(c, e, n)			\
+	EVENT_CONSTRAINT_RANGE(c, e, n, ARCH_PERFMON_EVENTSEL_EVENT)
+
+/*
+ * Constraint on the Event code + UMask + fixed-mask
+ *
+ * filter mask to validate fixed counter events.
+ * the following filters disqualify for fixed counters:
+ *  - inv
+ *  - edge
+ *  - cnt-mask
+ *  - in_tx
+ *  - in_tx_checkpointed
+ *  The other filters are supported by fixed counters.
+ *  The any-thread option is supported starting with v3.
+ */
+#define FIXED_EVENT_FLAGS (X86_RAW_EVENT_MASK|HSW_IN_TX|HSW_IN_TX_CHECKPOINTED)
+#define FIXED_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, (1ULL << (32+n)), FIXED_EVENT_FLAGS)
+
+/*
+ * The special metric counters do not actually exist. They are calculated from
+ * the combination of the FxCtr3 + MSR_PERF_METRICS.
+ *
+ * The special metric counters are mapped to a dummy offset for the scheduler.
+ * The sharing between multiple users of the same metric without multiplexing
+ * is not allowed, even though the hardware supports that in principle.
+ */
+
+#define METRIC_EVENT_CONSTRAINT(c, n)					\
+	EVENT_CONSTRAINT(c, (1ULL << (INTEL_PMC_IDX_METRIC_BASE + n)),	\
+			 INTEL_ARCH_EVENT_MASK)
+
+/*
+ * Constraint on the Event code + UMask
+ */
+#define INTEL_UEVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
+
+/* Constraint on specific umask bit only + event */
+#define INTEL_UBIT_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT|(c))
+
+/* Like UEVENT_CONSTRAINT, but match flags too */
+#define INTEL_FLAGS_UEVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS)
+
+#define INTEL_EXCLUEVT_CONSTRAINT(c, n)	\
+	__EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK, \
+			   HWEIGHT(n), 0, PERF_X86_EVENT_EXCL)
+
+#define INTEL_PLD_CONSTRAINT(c, n)	\
+	__EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			   HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_LDLAT)
+
+#define INTEL_PST_CONSTRAINT(c, n)	\
+	__EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST)
+
+/* Event constraint, but match on all event flags too. */
+#define INTEL_FLAGS_EVENT_CONSTRAINT(c, n) \
+	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS)
+
+#define INTEL_FLAGS_EVENT_CONSTRAINT_RANGE(c, e, n)			\
+	EVENT_CONSTRAINT_RANGE(c, e, n, ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS)
+
+/* Check only flags, but allow all event/umask */
+#define INTEL_ALL_EVENT_CONSTRAINT(code, n)	\
+	EVENT_CONSTRAINT(code, n, X86_ALL_EVENT_FLAGS)
+
+/* Check flags and event code, and set the HSW store flag */
+#define INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_ST(code, n) \
+	__EVENT_CONSTRAINT(code, n, 			\
+			  ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST_HSW)
+
+/* Check flags and event code, and set the HSW load flag */
+#define INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD(code, n) \
+	__EVENT_CONSTRAINT(code, n,			\
+			  ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_LD_HSW)
+
+#define INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_LD_RANGE(code, end, n) \
+	__EVENT_CONSTRAINT_RANGE(code, end, n,				\
+			  ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_LD_HSW)
+
+#define INTEL_FLAGS_EVENT_CONSTRAINT_DATALA_XLD(code, n) \
+	__EVENT_CONSTRAINT(code, n,			\
+			  ARCH_PERFMON_EVENTSEL_EVENT|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, \
+			  PERF_X86_EVENT_PEBS_LD_HSW|PERF_X86_EVENT_EXCL)
+
+/* Check flags and event code/umask, and set the HSW store flag */
+#define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_ST(code, n) \
+	__EVENT_CONSTRAINT(code, n, 			\
+			  INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_ST_HSW)
+
+#define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XST(code, n) \
+	__EVENT_CONSTRAINT(code, n,			\
+			  INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, \
+			  PERF_X86_EVENT_PEBS_ST_HSW|PERF_X86_EVENT_EXCL)
+
+/* Check flags and event code/umask, and set the HSW load flag */
+#define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_LD(code, n) \
+	__EVENT_CONSTRAINT(code, n, 			\
+			  INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_LD_HSW)
+
+#define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_XLD(code, n) \
+	__EVENT_CONSTRAINT(code, n,			\
+			  INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, \
+			  PERF_X86_EVENT_PEBS_LD_HSW|PERF_X86_EVENT_EXCL)
+
+/* Check flags and event code/umask, and set the HSW N/A flag */
+#define INTEL_FLAGS_UEVENT_CONSTRAINT_DATALA_NA(code, n) \
+	__EVENT_CONSTRAINT(code, n, 			\
+			  INTEL_ARCH_EVENT_MASK|X86_ALL_EVENT_FLAGS, \
+			  HWEIGHT(n), 0, PERF_X86_EVENT_PEBS_NA_HSW)
+
+
+/*
+ * We define the end marker as having a weight of -1
+ * to enable blacklisting of events using a counter bitmask
+ * of zero and thus a weight of zero.
+ * The end marker has a weight that cannot possibly be
+ * obtained from counting the bits in the bitmask.
+ */
+#define EVENT_CONSTRAINT_END { .weight = -1 }
+
+/*
+ * Check for end marker with weight == -1
+ */
+#define for_each_event_constraint(e, c)	\
+	for ((e) = (c); (e)->weight != -1; (e)++)
+
+/*
+ * Extra registers for specific events.
+ *
+ * Some events need large masks and require external MSRs.
+ * Those extra MSRs end up being shared for all events on
+ * a PMU and sometimes between PMU of sibling HT threads.
+ * In either case, the kernel needs to handle conflicting
+ * accesses to those extra, shared, regs. The data structure
+ * to manage those registers is stored in cpu_hw_event.
+ */
+struct extra_reg {
+	unsigned int		event;
+	unsigned int		msr;
+	u64			config_mask;
+	u64			valid_mask;
+	int			idx;  /* per_xxx->regs[] reg index */
+	bool			extra_msr_access;
+};
+
+#define EVENT_EXTRA_REG(e, ms, m, vm, i) {	\
+	.event = (e),			\
+	.msr = (ms),			\
+	.config_mask = (m),		\
+	.valid_mask = (vm),		\
+	.idx = EXTRA_REG_##i,		\
+	.extra_msr_access = true,	\
+	}
+
+#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx)	\
+	EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
+
+#define INTEL_UEVENT_EXTRA_REG(event, msr, vm, idx) \
+	EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT | \
+			ARCH_PERFMON_EVENTSEL_UMASK, vm, idx)
+
+#define INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(c) \
+	INTEL_UEVENT_EXTRA_REG(c, \
+			       MSR_PEBS_LD_LAT_THRESHOLD, \
+			       0xffff, \
+			       LDLAT)
+
+#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
+
+union perf_capabilities {
+	struct {
+		u64	lbr_format:6;
+		u64	pebs_trap:1;
+		u64	pebs_arch_reg:1;
+		u64	pebs_format:4;
+		u64	smm_freeze:1;
+		/*
+		 * PMU supports separate counter range for writing
+		 * values > 32bit.
+		 */
+		u64	full_width_write:1;
+		u64     pebs_baseline:1;
+		u64	perf_metrics:1;
+		u64	pebs_output_pt_available:1;
+		u64	anythread_deprecated:1;
+	};
+	u64	capabilities;
+};
+
+struct x86_pmu_quirk {
+	struct x86_pmu_quirk *next;
+	void (*func)(void);
+};
+
+union x86_pmu_config {
+	struct {
+		u64 event:8,
+		    umask:8,
+		    usr:1,
+		    os:1,
+		    edge:1,
+		    pc:1,
+		    interrupt:1,
+		    __reserved1:1,
+		    en:1,
+		    inv:1,
+		    cmask:8,
+		    event2:4,
+		    __reserved2:4,
+		    go:1,
+		    ho:1;
+	} bits;
+	u64 value;
+};
+
+#define X86_CONFIG(args...) ((union x86_pmu_config){.bits = {args}}).value
+
+enum {
+	x86_lbr_exclusive_lbr,
+	x86_lbr_exclusive_bts,
+	x86_lbr_exclusive_pt,
+	x86_lbr_exclusive_max,
+};
+
+/*
+ * struct x86_pmu - generic x86 pmu
+ */
+struct x86_pmu {
+	/*
+	 * Generic x86 PMC bits
+	 */
+	const char	*name;
+	int		version;
+	int		(*handle_irq)(struct pt_regs *);
+	void		(*disable_all)(void);
+	void		(*enable_all)(int added);
+	void		(*enable)(struct perf_event *);
+	void		(*disable)(struct perf_event *);
+	void		(*add)(struct perf_event *);
+	void		(*del)(struct perf_event *);
+	void		(*read)(struct perf_event *event);
+	int		(*hw_config)(struct perf_event *event);
+	int		(*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
+	unsigned	eventsel;
+	unsigned	perfctr;
+	int		(*addr_offset)(int index, bool eventsel);
+	int		(*rdpmc_index)(int index);
+	u64		(*event_map)(int);
+	int		max_events;
+	int		num_counters;
+	int		num_counters_fixed;
+	int		cntval_bits;
+	u64		cntval_mask;
+	union {
+			unsigned long events_maskl;
+			unsigned long events_mask[BITS_TO_LONGS(ARCH_PERFMON_EVENTS_COUNT)];
+	};
+	int		events_mask_len;
+	int		apic;
+	u64		max_period;
+	struct event_constraint *
+			(*get_event_constraints)(struct cpu_hw_events *cpuc,
+						 int idx,
+						 struct perf_event *event);
+
+	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
+						 struct perf_event *event);
+
+	void		(*start_scheduling)(struct cpu_hw_events *cpuc);
+
+	void		(*commit_scheduling)(struct cpu_hw_events *cpuc, int idx, int cntr);
+
+	void		(*stop_scheduling)(struct cpu_hw_events *cpuc);
+
+	struct event_constraint *event_constraints;
+	struct x86_pmu_quirk *quirks;
+	int		perfctr_second_write;
+	u64		(*limit_period)(struct perf_event *event, u64 l);
+
+	/* PMI handler bits */
+	unsigned int	late_ack		:1,
+			enabled_ack		:1,
+			counter_freezing	:1;
+	/*
+	 * sysfs attrs
+	 */
+	int		attr_rdpmc_broken;
+	int		attr_rdpmc;
+	struct attribute **format_attrs;
+
+	ssize_t		(*events_sysfs_show)(char *page, u64 config);
+	const struct attribute_group **attr_update;
+
+	unsigned long	attr_freeze_on_smi;
+
+	/*
+	 * CPU Hotplug hooks
+	 */
+	int		(*cpu_prepare)(int cpu);
+	void		(*cpu_starting)(int cpu);
+	void		(*cpu_dying)(int cpu);
+	void		(*cpu_dead)(int cpu);
+
+	void		(*check_microcode)(void);
+	void		(*sched_task)(struct perf_event_context *ctx,
+				      bool sched_in);
+
+	/*
+	 * Intel Arch Perfmon v2+
+	 */
+	u64			intel_ctrl;
+	union perf_capabilities intel_cap;
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	unsigned int	bts			:1,
+			bts_active		:1,
+			pebs			:1,
+			pebs_active		:1,
+			pebs_broken		:1,
+			pebs_prec_dist		:1,
+			pebs_no_tlb		:1,
+			pebs_no_isolation	:1;
+	int		pebs_record_size;
+	int		pebs_buffer_size;
+	int		max_pebs_events;
+	void		(*drain_pebs)(struct pt_regs *regs, struct perf_sample_data *data);
+	struct event_constraint *pebs_constraints;
+	void		(*pebs_aliases)(struct perf_event *event);
+	unsigned long	large_pebs_flags;
+	u64		rtm_abort_event;
+
+	/*
+	 * Intel LBR
+	 */
+	unsigned int	lbr_tos, lbr_from, lbr_to,
+			lbr_info, lbr_nr;	   /* LBR base regs and size */
+	union {
+		u64	lbr_sel_mask;		   /* LBR_SELECT valid bits */
+		u64	lbr_ctl_mask;		   /* LBR_CTL valid bits */
+	};
+	union {
+		const int	*lbr_sel_map;	   /* lbr_select mappings */
+		int		*lbr_ctl_map;	   /* LBR_CTL mappings */
+	};
+	bool		lbr_double_abort;	   /* duplicated lbr aborts */
+	bool		lbr_pt_coexist;		   /* (LBR|BTS) may coexist with PT */
+
+	/*
+	 * Intel Architectural LBR CPUID Enumeration
+	 */
+	unsigned int	lbr_depth_mask:8;
+	unsigned int	lbr_deep_c_reset:1;
+	unsigned int	lbr_lip:1;
+	unsigned int	lbr_cpl:1;
+	unsigned int	lbr_filter:1;
+	unsigned int	lbr_call_stack:1;
+	unsigned int	lbr_mispred:1;
+	unsigned int	lbr_timed_lbr:1;
+	unsigned int	lbr_br_type:1;
+
+	void		(*lbr_reset)(void);
+	void		(*lbr_read)(struct cpu_hw_events *cpuc);
+	void		(*lbr_save)(void *ctx);
+	void		(*lbr_restore)(void *ctx);
+
+	/*
+	 * Intel PT/LBR/BTS are exclusive
+	 */
+	atomic_t	lbr_exclusive[x86_lbr_exclusive_max];
+
+	/*
+	 * Intel perf metrics
+	 */
+	u64		(*update_topdown_event)(struct perf_event *event);
+	int		(*set_topdown_event_period)(struct perf_event *event);
+
+	/*
+	 * perf task context (i.e. struct perf_event_context::task_ctx_data)
+	 * switch helper to bridge calls from perf/core to perf/x86.
+	 * See struct pmu::swap_task_ctx() usage for examples;
+	 */
+	void		(*swap_task_ctx)(struct perf_event_context *prev,
+					 struct perf_event_context *next);
+
+	/*
+	 * AMD bits
+	 */
+	unsigned int	amd_nb_constraints : 1;
+	u64		perf_ctr_pair_en;
+
+	/*
+	 * Extra registers for events
+	 */
+	struct extra_reg *extra_regs;
+	unsigned int flags;
+
+	/*
+	 * Intel host/guest support (KVM)
+	 */
+	struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr);
+
+	/*
+	 * Check period value for PERF_EVENT_IOC_PERIOD ioctl.
+	 */
+	int (*check_period) (struct perf_event *event, u64 period);
+
+	int (*aux_output_match) (struct perf_event *event);
+};
+
+struct x86_perf_task_context_opt {
+	int lbr_callstack_users;
+	int lbr_stack_state;
+	int log_id;
+};
+
+struct x86_perf_task_context {
+	u64 lbr_sel;
+	int tos;
+	int valid_lbrs;
+	struct x86_perf_task_context_opt opt;
+	struct lbr_entry lbr[MAX_LBR_ENTRIES];
+};
+
+struct x86_perf_task_context_arch_lbr {
+	struct x86_perf_task_context_opt opt;
+	struct lbr_entry entries[];
+};
+
+/*
+ * Add padding to guarantee the 64-byte alignment of the state buffer.
+ *
+ * The structure is dynamically allocated. The size of the LBR state may vary
+ * based on the number of LBR registers.
+ *
+ * Do not put anything after the LBR state.
+ */
+struct x86_perf_task_context_arch_lbr_xsave {
+	struct x86_perf_task_context_opt		opt;
+
+	union {
+		struct xregs_state			xsave;
+		struct {
+			struct fxregs_state		i387;
+			struct xstate_header		header;
+			struct arch_lbr_state		lbr;
+		} __attribute__ ((packed, aligned (XSAVE_ALIGNMENT)));
+	};
+};
+
+#define x86_add_quirk(func_)						\
+do {									\
+	static struct x86_pmu_quirk __quirk __initdata = {		\
+		.func = func_,						\
+	};								\
+	__quirk.next = x86_pmu.quirks;					\
+	x86_pmu.quirks = &__quirk;					\
+} while (0)
+
+/*
+ * x86_pmu flags
+ */
+#define PMU_FL_NO_HT_SHARING	0x1 /* no hyper-threading resource sharing */
+#define PMU_FL_HAS_RSP_1	0x2 /* has 2 equivalent offcore_rsp regs   */
+#define PMU_FL_EXCL_CNTRS	0x4 /* has exclusive counter requirements  */
+#define PMU_FL_EXCL_ENABLED	0x8 /* exclusive counter active */
+#define PMU_FL_PEBS_ALL		0x10 /* all events are valid PEBS events */
+#define PMU_FL_TFA		0x20 /* deal with TSX force abort */
+#define PMU_FL_PAIR		0x40 /* merge counters for large incr. events */
+
+#define EVENT_VAR(_id)  event_attr_##_id
+#define EVENT_PTR(_id) &event_attr_##_id.attr.attr
+
+#define EVENT_ATTR(_name, _id)						\
+static struct perf_pmu_events_attr EVENT_VAR(_id) = {			\
+	.attr		= __ATTR(_name, 0444, events_sysfs_show, NULL),	\
+	.id		= PERF_COUNT_HW_##_id,				\
+	.event_str	= NULL,						\
+};
+
+#define EVENT_ATTR_STR(_name, v, str)					\
+static struct perf_pmu_events_attr event_attr_##v = {			\
+	.attr		= __ATTR(_name, 0444, events_sysfs_show, NULL),	\
+	.id		= 0,						\
+	.event_str	= str,						\
+};
+
+#define EVENT_ATTR_STR_HT(_name, v, noht, ht)				\
+static struct perf_pmu_events_ht_attr event_attr_##v = {		\
+	.attr		= __ATTR(_name, 0444, events_ht_sysfs_show, NULL),\
+	.id		= 0,						\
+	.event_str_noht	= noht,						\
+	.event_str_ht	= ht,						\
+}
+
+struct pmu *x86_get_pmu(unsigned int cpu);
+extern struct x86_pmu x86_pmu __read_mostly;
+
+static __always_inline struct x86_perf_task_context_opt *task_context_opt(void *ctx)
+{
+	if (static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return &((struct x86_perf_task_context_arch_lbr *)ctx)->opt;
+
+	return &((struct x86_perf_task_context *)ctx)->opt;
+}
+
+static inline bool x86_pmu_has_lbr_callstack(void)
+{
+	return  x86_pmu.lbr_sel_map &&
+		x86_pmu.lbr_sel_map[PERF_SAMPLE_BRANCH_CALL_STACK_SHIFT] > 0;
+}
+
+DECLARE_PER_CPU(struct cpu_hw_events, cpu_hw_events);
+
+int x86_perf_event_set_period(struct perf_event *event);
+
+/*
+ * Generalized hw caching related hw_event table, filled
+ * in on a per model basis. A value of 0 means
+ * 'not supported', -1 means 'hw_event makes no sense on
+ * this CPU', any other value means the raw hw_event
+ * ID.
+ */
+
+#define C(x) PERF_COUNT_HW_CACHE_##x
+
+extern u64 __read_mostly hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+extern u64 __read_mostly hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+
+u64 x86_perf_event_update(struct perf_event *event);
+
+static inline unsigned int x86_pmu_config_addr(int index)
+{
+	return x86_pmu.eventsel + (x86_pmu.addr_offset ?
+				   x86_pmu.addr_offset(index, true) : index);
+}
+
+static inline unsigned int x86_pmu_event_addr(int index)
+{
+	return x86_pmu.perfctr + (x86_pmu.addr_offset ?
+				  x86_pmu.addr_offset(index, false) : index);
+}
+
+static inline int x86_pmu_rdpmc_index(int index)
+{
+	return x86_pmu.rdpmc_index ? x86_pmu.rdpmc_index(index) : index;
+}
+
+int x86_add_exclusive(unsigned int what);
+
+void x86_del_exclusive(unsigned int what);
+
+int x86_reserve_hardware(void);
+
+void x86_release_hardware(void);
+
+int x86_pmu_max_precise(void);
+
+void hw_perf_lbr_event_destroy(struct perf_event *event);
+
+int x86_setup_perfctr(struct perf_event *event);
+
+int x86_pmu_hw_config(struct perf_event *event);
+
+void x86_pmu_disable_all(void);
+
+static inline bool is_counter_pair(struct hw_perf_event *hwc)
+{
+	return hwc->flags & PERF_X86_EVENT_PAIR;
+}
+
+static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
+					  u64 enable_mask)
+{
+	u64 disable_mask = __this_cpu_read(cpu_hw_events.perf_ctr_virt_mask);
+
+	if (hwc->extra_reg.reg)
+		wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
+
+	/*
+	 * Add enabled Merge event on next counter
+	 * if large increment event being enabled on this counter
+	 */
+	if (is_counter_pair(hwc))
+		wrmsrl(x86_pmu_config_addr(hwc->idx + 1), x86_pmu.perf_ctr_pair_en);
+
+	wrmsrl(hwc->config_base, (hwc->config | enable_mask) & ~disable_mask);
+}
+
+void x86_pmu_enable_all(int added);
+
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int gpmax, int *assign);
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
+
+void x86_pmu_stop(struct perf_event *event, int flags);
+
+static inline void x86_pmu_disable_event(struct perf_event *event)
+{
+	u64 disable_mask = __this_cpu_read(cpu_hw_events.perf_ctr_virt_mask);
+	struct hw_perf_event *hwc = &event->hw;
+
+	wrmsrl(hwc->config_base, hwc->config & ~disable_mask);
+
+	if (is_counter_pair(hwc))
+		wrmsrl(x86_pmu_config_addr(hwc->idx + 1), 0);
+}
+
+void x86_pmu_enable_event(struct perf_event *event);
+
+int x86_pmu_handle_irq(struct pt_regs *regs);
+
+extern struct event_constraint emptyconstraint;
+
+extern struct event_constraint unconstrained;
+
+static inline bool kernel_ip(unsigned long ip)
+{
+#ifdef CONFIG_X86_32
+	return ip > PAGE_OFFSET;
+#else
+	return (long)ip < 0;
+#endif
+}
+
+/*
+ * Not all PMUs provide the right context information to place the reported IP
+ * into full context. Specifically segment registers are typically not
+ * supplied.
+ *
+ * Assuming the address is a linear address (it is for IBS), we fake the CS and
+ * vm86 mode using the known zero-based code segment and 'fix up' the registers
+ * to reflect this.
+ *
+ * Intel PEBS/LBR appear to typically provide the effective address, nothing
+ * much we can do about that but pray and treat it like a linear address.
+ */
+static inline void set_linear_ip(struct pt_regs *regs, unsigned long ip)
+{
+	regs->cs = kernel_ip(ip) ? __KERNEL_CS : __USER_CS;
+	if (regs->flags & X86_VM_MASK)
+		regs->flags ^= (PERF_EFLAGS_VM | X86_VM_MASK);
+	regs->ip = ip;
+}
+
+ssize_t x86_event_sysfs_show(char *page, u64 config, u64 event);
+ssize_t intel_event_sysfs_show(char *page, u64 config);
+
+ssize_t events_sysfs_show(struct device *dev, struct device_attribute *attr,
+			  char *page);
+ssize_t events_ht_sysfs_show(struct device *dev, struct device_attribute *attr,
+			  char *page);
+
+#ifdef CONFIG_CPU_SUP_AMD
+
+int amd_pmu_init(void);
+
+#else /* CONFIG_CPU_SUP_AMD */
+
+static inline int amd_pmu_init(void)
+{
+	return 0;
+}
+
+#endif /* CONFIG_CPU_SUP_AMD */
+
+static inline int is_pebs_pt(struct perf_event *event)
+{
+	return !!(event->hw.flags & PERF_X86_EVENT_PEBS_VIA_PT);
+}
+
+#ifdef CONFIG_CPU_SUP_INTEL
+
+static inline bool intel_pmu_has_bts_period(struct perf_event *event, u64 period)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned int hw_event, bts_event;
+
+	if (event->attr.freq)
+		return false;
+
+	hw_event = hwc->config & INTEL_ARCH_EVENT_MASK;
+	bts_event = x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS);
+
+	return hw_event == bts_event && period == 1;
+}
+
+static inline bool intel_pmu_has_bts(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	return intel_pmu_has_bts_period(event, hwc->sample_period);
+}
+
+int intel_pmu_save_and_restart(struct perf_event *event);
+
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			  struct perf_event *event);
+
+extern int intel_cpuc_prepare(struct cpu_hw_events *cpuc, int cpu);
+extern void intel_cpuc_finish(struct cpu_hw_events *cpuc);
+
+int intel_pmu_init(void);
+
+void init_debug_store_on_cpu(int cpu);
+
+void fini_debug_store_on_cpu(int cpu);
+
+void release_ds_buffers(void);
+
+void reserve_ds_buffers(void);
+
+void release_lbr_buffers(void);
+
+void reserve_lbr_buffers(void);
+
+extern struct event_constraint bts_constraint;
+extern struct event_constraint vlbr_constraint;
+
+void intel_pmu_enable_bts(u64 config);
+
+void intel_pmu_disable_bts(void);
+
+int intel_pmu_drain_bts_buffer(void);
+
+extern struct event_constraint intel_core2_pebs_event_constraints[];
+
+extern struct event_constraint intel_atom_pebs_event_constraints[];
+
+extern struct event_constraint intel_slm_pebs_event_constraints[];
+
+extern struct event_constraint intel_glm_pebs_event_constraints[];
+
+extern struct event_constraint intel_glp_pebs_event_constraints[];
+
+extern struct event_constraint intel_nehalem_pebs_event_constraints[];
+
+extern struct event_constraint intel_westmere_pebs_event_constraints[];
+
+extern struct event_constraint intel_snb_pebs_event_constraints[];
+
+extern struct event_constraint intel_ivb_pebs_event_constraints[];
+
+extern struct event_constraint intel_hsw_pebs_event_constraints[];
+
+extern struct event_constraint intel_bdw_pebs_event_constraints[];
+
+extern struct event_constraint intel_skl_pebs_event_constraints[];
+
+extern struct event_constraint intel_icl_pebs_event_constraints[];
+
+struct event_constraint *intel_pebs_constraints(struct perf_event *event);
+
+void intel_pmu_pebs_add(struct perf_event *event);
+
+void intel_pmu_pebs_del(struct perf_event *event);
+
+void intel_pmu_pebs_enable(struct perf_event *event);
+
+void intel_pmu_pebs_disable(struct perf_event *event);
+
+void intel_pmu_pebs_enable_all(void);
+
+void intel_pmu_pebs_disable_all(void);
+
+void intel_pmu_pebs_sched_task(struct perf_event_context *ctx, bool sched_in);
+
+void intel_pmu_auto_reload_read(struct perf_event *event);
+
+void intel_pmu_store_pebs_lbrs(struct lbr_entry *lbr);
+
+void intel_ds_init(void);
+
+void intel_pmu_lbr_swap_task_ctx(struct perf_event_context *prev,
+				 struct perf_event_context *next);
+
+void intel_pmu_lbr_sched_task(struct perf_event_context *ctx, bool sched_in);
+
+u64 lbr_from_signext_quirk_wr(u64 val);
+
+void intel_pmu_lbr_reset(void);
+
+void intel_pmu_lbr_reset_32(void);
+
+void intel_pmu_lbr_reset_64(void);
+
+void intel_pmu_lbr_add(struct perf_event *event);
+
+void intel_pmu_lbr_del(struct perf_event *event);
+
+void intel_pmu_lbr_enable_all(bool pmi);
+
+void intel_pmu_lbr_disable_all(void);
+
+void intel_pmu_lbr_read(void);
+
+void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc);
+
+void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc);
+
+void intel_pmu_lbr_save(void *ctx);
+
+void intel_pmu_lbr_restore(void *ctx);
+
+void intel_pmu_lbr_init_core(void);
+
+void intel_pmu_lbr_init_nhm(void);
+
+void intel_pmu_lbr_init_atom(void);
+
+void intel_pmu_lbr_init_slm(void);
+
+void intel_pmu_lbr_init_snb(void);
+
+void intel_pmu_lbr_init_hsw(void);
+
+void intel_pmu_lbr_init_skl(void);
+
+void intel_pmu_lbr_init_knl(void);
+
+void intel_pmu_arch_lbr_init(void);
+
+void intel_pmu_pebs_data_source_nhm(void);
+
+void intel_pmu_pebs_data_source_skl(bool pmem);
+
+int intel_pmu_setup_lbr_filter(struct perf_event *event);
+
+void intel_pt_interrupt(void);
+
+int intel_bts_interrupt(void);
+
+void intel_bts_enable_local(void);
+
+void intel_bts_disable_local(void);
+
+int p4_pmu_init(void);
+
+int p6_pmu_init(void);
+
+int knc_pmu_init(void);
+
+static inline int is_ht_workaround_enabled(void)
+{
+	return !!(x86_pmu.flags & PMU_FL_EXCL_ENABLED);
+}
+
+#else /* CONFIG_CPU_SUP_INTEL */
+
+static inline void reserve_ds_buffers(void)
+{
+}
+
+static inline void release_ds_buffers(void)
+{
+}
+
+static inline void release_lbr_buffers(void)
+{
+}
+
+static inline void reserve_lbr_buffers(void)
+{
+}
+
+static inline int intel_pmu_init(void)
+{
+	return 0;
+}
+
+static inline int intel_cpuc_prepare(struct cpu_hw_events *cpuc, int cpu)
+{
+	return 0;
+}
+
+static inline void intel_cpuc_finish(struct cpu_hw_events *cpuc)
+{
+}
+
+static inline int is_ht_workaround_enabled(void)
+{
+	return 0;
+}
+#endif /* CONFIG_CPU_SUP_INTEL */
+
+#if ((defined CONFIG_CPU_SUP_CENTAUR) || (defined CONFIG_CPU_SUP_ZHAOXIN))
+int zhaoxin_pmu_init(void);
+#else
+static inline int zhaoxin_pmu_init(void)
+{
+	return 0;
+}
+#endif /*CONFIG_CPU_SUP_CENTAUR or CONFIG_CPU_SUP_ZHAOXIN*/
diff --git a/arch/x86/events/probe.c b/arch/x86/events/probe.c
new file mode 100644
index 000000000..136a1e847
--- /dev/null
+++ b/arch/x86/events/probe.c
@@ -0,0 +1,58 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/bits.h>
+#include "probe.h"
+
+static umode_t
+not_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return 0;
+}
+
+/*
+ * Accepts msr[] array with non populated entries as long as either
+ * msr[i].msr is 0 or msr[i].grp is NULL. Note that the default sysfs
+ * visibility is visible when group->is_visible callback is set.
+ */
+unsigned long
+perf_msr_probe(struct perf_msr *msr, int cnt, bool zero, void *data)
+{
+	unsigned long avail = 0;
+	unsigned int bit;
+	u64 val;
+
+	if (cnt >= BITS_PER_LONG)
+		return 0;
+
+	for (bit = 0; bit < cnt; bit++) {
+		if (!msr[bit].no_check) {
+			struct attribute_group *grp = msr[bit].grp;
+
+			/* skip entry with no group */
+			if (!grp)
+				continue;
+
+			grp->is_visible = not_visible;
+
+			/* skip unpopulated entry */
+			if (!msr[bit].msr)
+				continue;
+
+			if (msr[bit].test && !msr[bit].test(bit, data))
+				continue;
+			/* Virt sucks; you cannot tell if a R/O MSR is present :/ */
+			if (rdmsrl_safe(msr[bit].msr, &val))
+				continue;
+			/* Disable zero counters if requested. */
+			if (!zero && !val)
+				continue;
+
+			grp->is_visible = NULL;
+		}
+		avail |= BIT(bit);
+	}
+
+	return avail;
+}
+EXPORT_SYMBOL_GPL(perf_msr_probe);
diff --git a/arch/x86/events/probe.h b/arch/x86/events/probe.h
new file mode 100644
index 000000000..4c8e0afc5
--- /dev/null
+++ b/arch/x86/events/probe.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ARCH_X86_EVENTS_PROBE_H__
+#define __ARCH_X86_EVENTS_PROBE_H__
+#include <linux/sysfs.h>
+
+struct perf_msr {
+	u64			  msr;
+	struct attribute_group	 *grp;
+	bool			(*test)(int idx, void *data);
+	bool			  no_check;
+};
+
+unsigned long
+perf_msr_probe(struct perf_msr *msr, int cnt, bool no_zero, void *data);
+
+#define __PMU_EVENT_GROUP(_name)			\
+static struct attribute *attrs_##_name[] = {		\
+	&attr_##_name.attr.attr,			\
+	NULL,						\
+}
+
+#define PMU_EVENT_GROUP(_grp, _name)			\
+__PMU_EVENT_GROUP(_name);				\
+static struct attribute_group group_##_name = {		\
+	.name  = #_grp,					\
+	.attrs = attrs_##_name,				\
+}
+
+#endif /* __ARCH_X86_EVENTS_PROBE_H__ */
diff --git a/arch/x86/events/rapl.c b/arch/x86/events/rapl.c
new file mode 100644
index 000000000..7dbbeaacd
--- /dev/null
+++ b/arch/x86/events/rapl.c
@@ -0,0 +1,867 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Support Intel/AMD RAPL energy consumption counters
+ * Copyright (C) 2013 Google, Inc., Stephane Eranian
+ *
+ * Intel RAPL interface is specified in the IA-32 Manual Vol3b
+ * section 14.7.1 (September 2013)
+ *
+ * AMD RAPL interface for Fam17h is described in the public PPR:
+ * https://bugzilla.kernel.org/show_bug.cgi?id=206537
+ *
+ * RAPL provides more controls than just reporting energy consumption
+ * however here we only expose the 3 energy consumption free running
+ * counters (pp0, pkg, dram).
+ *
+ * Each of those counters increments in a power unit defined by the
+ * RAPL_POWER_UNIT MSR. On SandyBridge, this unit is 1/(2^16) Joules
+ * but it can vary.
+ *
+ * Counter to rapl events mappings:
+ *
+ *  pp0 counter: consumption of all physical cores (power plane 0)
+ * 	  event: rapl_energy_cores
+ *    perf code: 0x1
+ *
+ *  pkg counter: consumption of the whole processor package
+ *	  event: rapl_energy_pkg
+ *    perf code: 0x2
+ *
+ * dram counter: consumption of the dram domain (servers only)
+ *	  event: rapl_energy_dram
+ *    perf code: 0x3
+ *
+ * gpu counter: consumption of the builtin-gpu domain (client only)
+ *	  event: rapl_energy_gpu
+ *    perf code: 0x4
+ *
+ *  psys counter: consumption of the builtin-psys domain (client only)
+ *	  event: rapl_energy_psys
+ *    perf code: 0x5
+ *
+ * We manage those counters as free running (read-only). They may be
+ * use simultaneously by other tools, such as turbostat.
+ *
+ * The events only support system-wide mode counting. There is no
+ * sampling support because it does not make sense and is not
+ * supported by the RAPL hardware.
+ *
+ * Because we want to avoid floating-point operations in the kernel,
+ * the events are all reported in fixed point arithmetic (32.32).
+ * Tools must adjust the counts to convert them to Watts using
+ * the duration of the measurement. Tools may use a function such as
+ * ldexp(raw_count, -32);
+ */
+
+#define pr_fmt(fmt) "RAPL PMU: " fmt
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/perf_event.h>
+#include <linux/nospec.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include "perf_event.h"
+#include "probe.h"
+
+MODULE_LICENSE("GPL");
+
+/*
+ * RAPL energy status counters
+ */
+enum perf_rapl_events {
+	PERF_RAPL_PP0 = 0,		/* all cores */
+	PERF_RAPL_PKG,			/* entire package */
+	PERF_RAPL_RAM,			/* DRAM */
+	PERF_RAPL_PP1,			/* gpu */
+	PERF_RAPL_PSYS,			/* psys */
+
+	PERF_RAPL_MAX,
+	NR_RAPL_DOMAINS = PERF_RAPL_MAX,
+};
+
+static const char *const rapl_domain_names[NR_RAPL_DOMAINS] __initconst = {
+	"pp0-core",
+	"package",
+	"dram",
+	"pp1-gpu",
+	"psys",
+};
+
+/*
+ * event code: LSB 8 bits, passed in attr->config
+ * any other bit is reserved
+ */
+#define RAPL_EVENT_MASK	0xFFULL
+#define RAPL_CNTR_WIDTH 32
+
+#define RAPL_EVENT_ATTR_STR(_name, v, str)					\
+static struct perf_pmu_events_attr event_attr_##v = {				\
+	.attr		= __ATTR(_name, 0444, perf_event_sysfs_show, NULL),	\
+	.id		= 0,							\
+	.event_str	= str,							\
+};
+
+struct rapl_pmu {
+	raw_spinlock_t		lock;
+	int			n_active;
+	int			cpu;
+	struct list_head	active_list;
+	struct pmu		*pmu;
+	ktime_t			timer_interval;
+	struct hrtimer		hrtimer;
+};
+
+struct rapl_pmus {
+	struct pmu		pmu;
+	unsigned int		maxdie;
+	struct rapl_pmu		*pmus[];
+};
+
+enum rapl_unit_quirk {
+	RAPL_UNIT_QUIRK_NONE,
+	RAPL_UNIT_QUIRK_INTEL_HSW,
+	RAPL_UNIT_QUIRK_INTEL_SPR,
+};
+
+struct rapl_model {
+	struct perf_msr *rapl_msrs;
+	unsigned long	events;
+	unsigned int	msr_power_unit;
+	enum rapl_unit_quirk	unit_quirk;
+};
+
+ /* 1/2^hw_unit Joule */
+static int rapl_hw_unit[NR_RAPL_DOMAINS] __read_mostly;
+static struct rapl_pmus *rapl_pmus;
+static cpumask_t rapl_cpu_mask;
+static unsigned int rapl_cntr_mask;
+static u64 rapl_timer_ms;
+static struct perf_msr *rapl_msrs;
+
+static inline struct rapl_pmu *cpu_to_rapl_pmu(unsigned int cpu)
+{
+	unsigned int dieid = topology_logical_die_id(cpu);
+
+	/*
+	 * The unsigned check also catches the '-1' return value for non
+	 * existent mappings in the topology map.
+	 */
+	return dieid < rapl_pmus->maxdie ? rapl_pmus->pmus[dieid] : NULL;
+}
+
+static inline u64 rapl_read_counter(struct perf_event *event)
+{
+	u64 raw;
+	rdmsrl(event->hw.event_base, raw);
+	return raw;
+}
+
+static inline u64 rapl_scale(u64 v, int cfg)
+{
+	if (cfg > NR_RAPL_DOMAINS) {
+		pr_warn("Invalid domain %d, failed to scale data\n", cfg);
+		return v;
+	}
+	/*
+	 * scale delta to smallest unit (1/2^32)
+	 * users must then scale back: count * 1/(1e9*2^32) to get Joules
+	 * or use ldexp(count, -32).
+	 * Watts = Joules/Time delta
+	 */
+	return v << (32 - rapl_hw_unit[cfg - 1]);
+}
+
+static u64 rapl_event_update(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	u64 prev_raw_count, new_raw_count;
+	s64 delta, sdelta;
+	int shift = RAPL_CNTR_WIDTH;
+
+again:
+	prev_raw_count = local64_read(&hwc->prev_count);
+	rdmsrl(event->hw.event_base, new_raw_count);
+
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+			    new_raw_count) != prev_raw_count) {
+		cpu_relax();
+		goto again;
+	}
+
+	/*
+	 * Now we have the new raw value and have updated the prev
+	 * timestamp already. We can now calculate the elapsed delta
+	 * (event-)time and add that to the generic event.
+	 *
+	 * Careful, not all hw sign-extends above the physical width
+	 * of the count.
+	 */
+	delta = (new_raw_count << shift) - (prev_raw_count << shift);
+	delta >>= shift;
+
+	sdelta = rapl_scale(delta, event->hw.config);
+
+	local64_add(sdelta, &event->count);
+
+	return new_raw_count;
+}
+
+static void rapl_start_hrtimer(struct rapl_pmu *pmu)
+{
+       hrtimer_start(&pmu->hrtimer, pmu->timer_interval,
+		     HRTIMER_MODE_REL_PINNED);
+}
+
+static enum hrtimer_restart rapl_hrtimer_handle(struct hrtimer *hrtimer)
+{
+	struct rapl_pmu *pmu = container_of(hrtimer, struct rapl_pmu, hrtimer);
+	struct perf_event *event;
+	unsigned long flags;
+
+	if (!pmu->n_active)
+		return HRTIMER_NORESTART;
+
+	raw_spin_lock_irqsave(&pmu->lock, flags);
+
+	list_for_each_entry(event, &pmu->active_list, active_entry)
+		rapl_event_update(event);
+
+	raw_spin_unlock_irqrestore(&pmu->lock, flags);
+
+	hrtimer_forward_now(hrtimer, pmu->timer_interval);
+
+	return HRTIMER_RESTART;
+}
+
+static void rapl_hrtimer_init(struct rapl_pmu *pmu)
+{
+	struct hrtimer *hr = &pmu->hrtimer;
+
+	hrtimer_init(hr, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hr->function = rapl_hrtimer_handle;
+}
+
+static void __rapl_pmu_event_start(struct rapl_pmu *pmu,
+				   struct perf_event *event)
+{
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	event->hw.state = 0;
+
+	list_add_tail(&event->active_entry, &pmu->active_list);
+
+	local64_set(&event->hw.prev_count, rapl_read_counter(event));
+
+	pmu->n_active++;
+	if (pmu->n_active == 1)
+		rapl_start_hrtimer(pmu);
+}
+
+static void rapl_pmu_event_start(struct perf_event *event, int mode)
+{
+	struct rapl_pmu *pmu = event->pmu_private;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&pmu->lock, flags);
+	__rapl_pmu_event_start(pmu, event);
+	raw_spin_unlock_irqrestore(&pmu->lock, flags);
+}
+
+static void rapl_pmu_event_stop(struct perf_event *event, int mode)
+{
+	struct rapl_pmu *pmu = event->pmu_private;
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&pmu->lock, flags);
+
+	/* mark event as deactivated and stopped */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		WARN_ON_ONCE(pmu->n_active <= 0);
+		pmu->n_active--;
+		if (pmu->n_active == 0)
+			hrtimer_cancel(&pmu->hrtimer);
+
+		list_del(&event->active_entry);
+
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	/* check if update of sw counter is necessary */
+	if ((mode & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		rapl_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+
+	raw_spin_unlock_irqrestore(&pmu->lock, flags);
+}
+
+static int rapl_pmu_event_add(struct perf_event *event, int mode)
+{
+	struct rapl_pmu *pmu = event->pmu_private;
+	struct hw_perf_event *hwc = &event->hw;
+	unsigned long flags;
+
+	raw_spin_lock_irqsave(&pmu->lock, flags);
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+
+	if (mode & PERF_EF_START)
+		__rapl_pmu_event_start(pmu, event);
+
+	raw_spin_unlock_irqrestore(&pmu->lock, flags);
+
+	return 0;
+}
+
+static void rapl_pmu_event_del(struct perf_event *event, int flags)
+{
+	rapl_pmu_event_stop(event, PERF_EF_UPDATE);
+}
+
+static int rapl_pmu_event_init(struct perf_event *event)
+{
+	u64 cfg = event->attr.config & RAPL_EVENT_MASK;
+	int bit, ret = 0;
+	struct rapl_pmu *pmu;
+
+	/* only look at RAPL events */
+	if (event->attr.type != rapl_pmus->pmu.type)
+		return -ENOENT;
+
+	/* check only supported bits are set */
+	if (event->attr.config & ~RAPL_EVENT_MASK)
+		return -EINVAL;
+
+	if (event->cpu < 0)
+		return -EINVAL;
+
+	event->event_caps |= PERF_EV_CAP_READ_ACTIVE_PKG;
+
+	if (!cfg || cfg >= NR_RAPL_DOMAINS + 1)
+		return -EINVAL;
+
+	cfg = array_index_nospec((long)cfg, NR_RAPL_DOMAINS + 1);
+	bit = cfg - 1;
+
+	/* check event supported */
+	if (!(rapl_cntr_mask & (1 << bit)))
+		return -EINVAL;
+
+	/* unsupported modes and filters */
+	if (event->attr.sample_period) /* no sampling */
+		return -EINVAL;
+
+	/* must be done before validate_group */
+	pmu = cpu_to_rapl_pmu(event->cpu);
+	if (!pmu)
+		return -EINVAL;
+	event->cpu = pmu->cpu;
+	event->pmu_private = pmu;
+	event->hw.event_base = rapl_msrs[bit].msr;
+	event->hw.config = cfg;
+	event->hw.idx = bit;
+
+	return ret;
+}
+
+static void rapl_pmu_event_read(struct perf_event *event)
+{
+	rapl_event_update(event);
+}
+
+static ssize_t rapl_get_attr_cpumask(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	return cpumap_print_to_pagebuf(true, buf, &rapl_cpu_mask);
+}
+
+static DEVICE_ATTR(cpumask, S_IRUGO, rapl_get_attr_cpumask, NULL);
+
+static struct attribute *rapl_pmu_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group rapl_pmu_attr_group = {
+	.attrs = rapl_pmu_attrs,
+};
+
+RAPL_EVENT_ATTR_STR(energy-cores, rapl_cores, "event=0x01");
+RAPL_EVENT_ATTR_STR(energy-pkg  ,   rapl_pkg, "event=0x02");
+RAPL_EVENT_ATTR_STR(energy-ram  ,   rapl_ram, "event=0x03");
+RAPL_EVENT_ATTR_STR(energy-gpu  ,   rapl_gpu, "event=0x04");
+RAPL_EVENT_ATTR_STR(energy-psys,   rapl_psys, "event=0x05");
+
+RAPL_EVENT_ATTR_STR(energy-cores.unit, rapl_cores_unit, "Joules");
+RAPL_EVENT_ATTR_STR(energy-pkg.unit  ,   rapl_pkg_unit, "Joules");
+RAPL_EVENT_ATTR_STR(energy-ram.unit  ,   rapl_ram_unit, "Joules");
+RAPL_EVENT_ATTR_STR(energy-gpu.unit  ,   rapl_gpu_unit, "Joules");
+RAPL_EVENT_ATTR_STR(energy-psys.unit,   rapl_psys_unit, "Joules");
+
+/*
+ * we compute in 0.23 nJ increments regardless of MSR
+ */
+RAPL_EVENT_ATTR_STR(energy-cores.scale, rapl_cores_scale, "2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-pkg.scale,     rapl_pkg_scale, "2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-ram.scale,     rapl_ram_scale, "2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-gpu.scale,     rapl_gpu_scale, "2.3283064365386962890625e-10");
+RAPL_EVENT_ATTR_STR(energy-psys.scale,   rapl_psys_scale, "2.3283064365386962890625e-10");
+
+/*
+ * There are no default events, but we need to create
+ * "events" group (with empty attrs) before updating
+ * it with detected events.
+ */
+static struct attribute *attrs_empty[] = {
+	NULL,
+};
+
+static struct attribute_group rapl_pmu_events_group = {
+	.name = "events",
+	.attrs = attrs_empty,
+};
+
+PMU_FORMAT_ATTR(event, "config:0-7");
+static struct attribute *rapl_formats_attr[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+
+static struct attribute_group rapl_pmu_format_group = {
+	.name = "format",
+	.attrs = rapl_formats_attr,
+};
+
+static const struct attribute_group *rapl_attr_groups[] = {
+	&rapl_pmu_attr_group,
+	&rapl_pmu_format_group,
+	&rapl_pmu_events_group,
+	NULL,
+};
+
+static struct attribute *rapl_events_cores[] = {
+	EVENT_PTR(rapl_cores),
+	EVENT_PTR(rapl_cores_unit),
+	EVENT_PTR(rapl_cores_scale),
+	NULL,
+};
+
+static umode_t
+rapl_not_visible(struct kobject *kobj, struct attribute *attr, int i)
+{
+	return 0;
+}
+
+static struct attribute_group rapl_events_cores_group = {
+	.name  = "events",
+	.attrs = rapl_events_cores,
+	.is_visible = rapl_not_visible,
+};
+
+static struct attribute *rapl_events_pkg[] = {
+	EVENT_PTR(rapl_pkg),
+	EVENT_PTR(rapl_pkg_unit),
+	EVENT_PTR(rapl_pkg_scale),
+	NULL,
+};
+
+static struct attribute_group rapl_events_pkg_group = {
+	.name  = "events",
+	.attrs = rapl_events_pkg,
+	.is_visible = rapl_not_visible,
+};
+
+static struct attribute *rapl_events_ram[] = {
+	EVENT_PTR(rapl_ram),
+	EVENT_PTR(rapl_ram_unit),
+	EVENT_PTR(rapl_ram_scale),
+	NULL,
+};
+
+static struct attribute_group rapl_events_ram_group = {
+	.name  = "events",
+	.attrs = rapl_events_ram,
+	.is_visible = rapl_not_visible,
+};
+
+static struct attribute *rapl_events_gpu[] = {
+	EVENT_PTR(rapl_gpu),
+	EVENT_PTR(rapl_gpu_unit),
+	EVENT_PTR(rapl_gpu_scale),
+	NULL,
+};
+
+static struct attribute_group rapl_events_gpu_group = {
+	.name  = "events",
+	.attrs = rapl_events_gpu,
+	.is_visible = rapl_not_visible,
+};
+
+static struct attribute *rapl_events_psys[] = {
+	EVENT_PTR(rapl_psys),
+	EVENT_PTR(rapl_psys_unit),
+	EVENT_PTR(rapl_psys_scale),
+	NULL,
+};
+
+static struct attribute_group rapl_events_psys_group = {
+	.name  = "events",
+	.attrs = rapl_events_psys,
+	.is_visible = rapl_not_visible,
+};
+
+static bool test_msr(int idx, void *data)
+{
+	return test_bit(idx, (unsigned long *) data);
+}
+
+static struct perf_msr intel_rapl_msrs[] = {
+	[PERF_RAPL_PP0]  = { MSR_PP0_ENERGY_STATUS,      &rapl_events_cores_group, test_msr },
+	[PERF_RAPL_PKG]  = { MSR_PKG_ENERGY_STATUS,      &rapl_events_pkg_group,   test_msr },
+	[PERF_RAPL_RAM]  = { MSR_DRAM_ENERGY_STATUS,     &rapl_events_ram_group,   test_msr },
+	[PERF_RAPL_PP1]  = { MSR_PP1_ENERGY_STATUS,      &rapl_events_gpu_group,   test_msr },
+	[PERF_RAPL_PSYS] = { MSR_PLATFORM_ENERGY_STATUS, &rapl_events_psys_group,  test_msr },
+};
+
+/*
+ * Force to PERF_RAPL_MAX size due to:
+ * - perf_msr_probe(PERF_RAPL_MAX)
+ * - want to use same event codes across both architectures
+ */
+static struct perf_msr amd_rapl_msrs[PERF_RAPL_MAX] = {
+	[PERF_RAPL_PKG]  = { MSR_AMD_PKG_ENERGY_STATUS,  &rapl_events_pkg_group,   test_msr },
+};
+
+
+static int rapl_cpu_offline(unsigned int cpu)
+{
+	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);
+	int target;
+
+	/* Check if exiting cpu is used for collecting rapl events */
+	if (!cpumask_test_and_clear_cpu(cpu, &rapl_cpu_mask))
+		return 0;
+
+	pmu->cpu = -1;
+	/* Find a new cpu to collect rapl events */
+	target = cpumask_any_but(topology_die_cpumask(cpu), cpu);
+
+	/* Migrate rapl events to the new target */
+	if (target < nr_cpu_ids) {
+		cpumask_set_cpu(target, &rapl_cpu_mask);
+		pmu->cpu = target;
+		perf_pmu_migrate_context(pmu->pmu, cpu, target);
+	}
+	return 0;
+}
+
+static int rapl_cpu_online(unsigned int cpu)
+{
+	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);
+	int target;
+
+	if (!pmu) {
+		pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu));
+		if (!pmu)
+			return -ENOMEM;
+
+		raw_spin_lock_init(&pmu->lock);
+		INIT_LIST_HEAD(&pmu->active_list);
+		pmu->pmu = &rapl_pmus->pmu;
+		pmu->timer_interval = ms_to_ktime(rapl_timer_ms);
+		rapl_hrtimer_init(pmu);
+
+		rapl_pmus->pmus[topology_logical_die_id(cpu)] = pmu;
+	}
+
+	/*
+	 * Check if there is an online cpu in the package which collects rapl
+	 * events already.
+	 */
+	target = cpumask_any_and(&rapl_cpu_mask, topology_die_cpumask(cpu));
+	if (target < nr_cpu_ids)
+		return 0;
+
+	cpumask_set_cpu(cpu, &rapl_cpu_mask);
+	pmu->cpu = cpu;
+	return 0;
+}
+
+static int rapl_check_hw_unit(struct rapl_model *rm)
+{
+	u64 msr_rapl_power_unit_bits;
+	int i;
+
+	/* protect rdmsrl() to handle virtualization */
+	if (rdmsrl_safe(rm->msr_power_unit, &msr_rapl_power_unit_bits))
+		return -1;
+	for (i = 0; i < NR_RAPL_DOMAINS; i++)
+		rapl_hw_unit[i] = (msr_rapl_power_unit_bits >> 8) & 0x1FULL;
+
+	switch (rm->unit_quirk) {
+	/*
+	 * DRAM domain on HSW server and KNL has fixed energy unit which can be
+	 * different than the unit from power unit MSR. See
+	 * "Intel Xeon Processor E5-1600 and E5-2600 v3 Product Families, V2
+	 * of 2. Datasheet, September 2014, Reference Number: 330784-001 "
+	 */
+	case RAPL_UNIT_QUIRK_INTEL_HSW:
+		rapl_hw_unit[PERF_RAPL_RAM] = 16;
+		break;
+	/*
+	 * SPR shares the same DRAM domain energy unit as HSW, plus it
+	 * also has a fixed energy unit for Psys domain.
+	 */
+	case RAPL_UNIT_QUIRK_INTEL_SPR:
+		rapl_hw_unit[PERF_RAPL_RAM] = 16;
+		rapl_hw_unit[PERF_RAPL_PSYS] = 0;
+		break;
+	default:
+		break;
+	}
+
+
+	/*
+	 * Calculate the timer rate:
+	 * Use reference of 200W for scaling the timeout to avoid counter
+	 * overflows. 200W = 200 Joules/sec
+	 * Divide interval by 2 to avoid lockstep (2 * 100)
+	 * if hw unit is 32, then we use 2 ms 1/200/2
+	 */
+	rapl_timer_ms = 2;
+	if (rapl_hw_unit[0] < 32) {
+		rapl_timer_ms = (1000 / (2 * 100));
+		rapl_timer_ms *= (1ULL << (32 - rapl_hw_unit[0] - 1));
+	}
+	return 0;
+}
+
+static void __init rapl_advertise(void)
+{
+	int i;
+
+	pr_info("API unit is 2^-32 Joules, %d fixed counters, %llu ms ovfl timer\n",
+		hweight32(rapl_cntr_mask), rapl_timer_ms);
+
+	for (i = 0; i < NR_RAPL_DOMAINS; i++) {
+		if (rapl_cntr_mask & (1 << i)) {
+			pr_info("hw unit of domain %s 2^-%d Joules\n",
+				rapl_domain_names[i], rapl_hw_unit[i]);
+		}
+	}
+}
+
+static void cleanup_rapl_pmus(void)
+{
+	int i;
+
+	for (i = 0; i < rapl_pmus->maxdie; i++)
+		kfree(rapl_pmus->pmus[i]);
+	kfree(rapl_pmus);
+}
+
+static const struct attribute_group *rapl_attr_update[] = {
+	&rapl_events_cores_group,
+	&rapl_events_pkg_group,
+	&rapl_events_ram_group,
+	&rapl_events_gpu_group,
+	&rapl_events_psys_group,
+	NULL,
+};
+
+static int __init init_rapl_pmus(void)
+{
+	int maxdie = topology_max_packages() * topology_max_die_per_package();
+	size_t size;
+
+	size = sizeof(*rapl_pmus) + maxdie * sizeof(struct rapl_pmu *);
+	rapl_pmus = kzalloc(size, GFP_KERNEL);
+	if (!rapl_pmus)
+		return -ENOMEM;
+
+	rapl_pmus->maxdie		= maxdie;
+	rapl_pmus->pmu.attr_groups	= rapl_attr_groups;
+	rapl_pmus->pmu.attr_update	= rapl_attr_update;
+	rapl_pmus->pmu.task_ctx_nr	= perf_invalid_context;
+	rapl_pmus->pmu.event_init	= rapl_pmu_event_init;
+	rapl_pmus->pmu.add		= rapl_pmu_event_add;
+	rapl_pmus->pmu.del		= rapl_pmu_event_del;
+	rapl_pmus->pmu.start		= rapl_pmu_event_start;
+	rapl_pmus->pmu.stop		= rapl_pmu_event_stop;
+	rapl_pmus->pmu.read		= rapl_pmu_event_read;
+	rapl_pmus->pmu.module		= THIS_MODULE;
+	rapl_pmus->pmu.capabilities	= PERF_PMU_CAP_NO_EXCLUDE;
+	return 0;
+}
+
+static struct rapl_model model_snb = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_PP1),
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_snbep = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM),
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_hsw = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM) |
+			  BIT(PERF_RAPL_PP1),
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_hsx = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM),
+	.unit_quirk	= RAPL_UNIT_QUIRK_INTEL_HSW,
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_knl = {
+	.events		= BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM),
+	.unit_quirk	= RAPL_UNIT_QUIRK_INTEL_HSW,
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_skl = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM) |
+			  BIT(PERF_RAPL_PP1) |
+			  BIT(PERF_RAPL_PSYS),
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_spr = {
+	.events		= BIT(PERF_RAPL_PP0) |
+			  BIT(PERF_RAPL_PKG) |
+			  BIT(PERF_RAPL_RAM) |
+			  BIT(PERF_RAPL_PSYS),
+	.unit_quirk	= RAPL_UNIT_QUIRK_INTEL_SPR,
+	.msr_power_unit = MSR_RAPL_POWER_UNIT,
+	.rapl_msrs      = intel_rapl_msrs,
+};
+
+static struct rapl_model model_amd_fam17h = {
+	.events		= BIT(PERF_RAPL_PKG),
+	.msr_power_unit = MSR_AMD_RAPL_POWER_UNIT,
+	.rapl_msrs      = amd_rapl_msrs,
+};
+
+static const struct x86_cpu_id rapl_model_match[] __initconst = {
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE,		&model_snb),
+	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE_X,	&model_snbep),
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE,		&model_snb),
+	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE_X,		&model_snbep),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL,		&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_X,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_L,		&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_G,		&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL,		&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_G,		&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_X,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_D,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL,	&model_knl),
+	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM,	&model_knl),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(CANNONLAKE_L,	&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT,	&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D,	&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_PLUS,	&model_hsw),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_L,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X,		&model_hsx),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE_L,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(COMETLAKE,		&model_skl),
+	X86_MATCH_INTEL_FAM6_MODEL(SAPPHIRERAPIDS_X,	&model_spr),
+	X86_MATCH_VENDOR_FAM(AMD,	0x17,		&model_amd_fam17h),
+	X86_MATCH_VENDOR_FAM(HYGON,	0x18,		&model_amd_fam17h),
+	X86_MATCH_VENDOR_FAM(AMD,	0x19,		&model_amd_fam17h),
+	{},
+};
+MODULE_DEVICE_TABLE(x86cpu, rapl_model_match);
+
+static int __init rapl_pmu_init(void)
+{
+	const struct x86_cpu_id *id;
+	struct rapl_model *rm;
+	int ret;
+
+	id = x86_match_cpu(rapl_model_match);
+	if (!id)
+		return -ENODEV;
+
+	rm = (struct rapl_model *) id->driver_data;
+
+	rapl_msrs = rm->rapl_msrs;
+
+	rapl_cntr_mask = perf_msr_probe(rapl_msrs, PERF_RAPL_MAX,
+					false, (void *) &rm->events);
+
+	ret = rapl_check_hw_unit(rm);
+	if (ret)
+		return ret;
+
+	ret = init_rapl_pmus();
+	if (ret)
+		return ret;
+
+	/*
+	 * Install callbacks. Core will call them for each online cpu.
+	 */
+	ret = cpuhp_setup_state(CPUHP_AP_PERF_X86_RAPL_ONLINE,
+				"perf/x86/rapl:online",
+				rapl_cpu_online, rapl_cpu_offline);
+	if (ret)
+		goto out;
+
+	ret = perf_pmu_register(&rapl_pmus->pmu, "power", -1);
+	if (ret)
+		goto out1;
+
+	rapl_advertise();
+	return 0;
+
+out1:
+	cpuhp_remove_state(CPUHP_AP_PERF_X86_RAPL_ONLINE);
+out:
+	pr_warn("Initialization failed (%d), disabled\n", ret);
+	cleanup_rapl_pmus();
+	return ret;
+}
+module_init(rapl_pmu_init);
+
+static void __exit intel_rapl_exit(void)
+{
+	cpuhp_remove_state_nocalls(CPUHP_AP_PERF_X86_RAPL_ONLINE);
+	perf_pmu_unregister(&rapl_pmus->pmu);
+	cleanup_rapl_pmus();
+}
+module_exit(intel_rapl_exit);
diff --git a/arch/x86/events/zhaoxin/Makefile b/arch/x86/events/zhaoxin/Makefile
new file mode 100644
index 000000000..642c1174d
--- /dev/null
+++ b/arch/x86/events/zhaoxin/Makefile
@@ -0,0 +1,2 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-y	+= core.o
diff --git a/arch/x86/events/zhaoxin/core.c b/arch/x86/events/zhaoxin/core.c
new file mode 100644
index 000000000..e92734696
--- /dev/null
+++ b/arch/x86/events/zhaoxin/core.c
@@ -0,0 +1,619 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Zhaoxin PMU; like Intel Architectural PerfMon-v2
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <linux/nmi.h>
+
+#include <asm/cpufeature.h>
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "../perf_event.h"
+
+/*
+ * Zhaoxin PerfMon, used on zxc and later.
+ */
+static u64 zx_pmon_event_map[PERF_COUNT_HW_MAX] __read_mostly = {
+
+	[PERF_COUNT_HW_CPU_CYCLES]        = 0x0082,
+	[PERF_COUNT_HW_INSTRUCTIONS]      = 0x00c0,
+	[PERF_COUNT_HW_CACHE_REFERENCES]  = 0x0515,
+	[PERF_COUNT_HW_CACHE_MISSES]      = 0x051a,
+	[PERF_COUNT_HW_BUS_CYCLES]        = 0x0083,
+};
+
+static struct event_constraint zxc_event_constraints[] __read_mostly = {
+
+	FIXED_EVENT_CONSTRAINT(0x0082, 1), /* unhalted core clock cycles */
+	EVENT_CONSTRAINT_END
+};
+
+static struct event_constraint zxd_event_constraints[] __read_mostly = {
+
+	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* retired instructions */
+	FIXED_EVENT_CONSTRAINT(0x0082, 1), /* unhalted core clock cycles */
+	FIXED_EVENT_CONSTRAINT(0x0083, 2), /* unhalted bus clock cycles */
+	EVENT_CONSTRAINT_END
+};
+
+static __initconst const u64 zxd_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)] = 0x0042,
+		[C(RESULT_MISS)] = 0x0538,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0x0043,
+		[C(RESULT_MISS)] = 0x0562,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0300,
+		[C(RESULT_MISS)] = 0x0301,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0x030a,
+		[C(RESULT_MISS)] = 0x030b,
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0042,
+		[C(RESULT_MISS)] = 0x052c,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0x0043,
+		[C(RESULT_MISS)] = 0x0530,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0x0564,
+		[C(RESULT_MISS)] = 0x0565,
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x00c0,
+		[C(RESULT_MISS)] = 0x0534,
+	},
+	[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)] = 0x0700,
+		[C(RESULT_MISS)] = 0x0709,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+[C(NODE)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+};
+
+static __initconst const u64 zxe_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)] = 0x0568,
+		[C(RESULT_MISS)] = 0x054b,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0x0669,
+		[C(RESULT_MISS)] = 0x0562,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+[C(L1I)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0300,
+		[C(RESULT_MISS)] = 0x0301,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0x030a,
+		[C(RESULT_MISS)] = 0x030b,
+	},
+},
+[C(LL)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0,
+		[C(RESULT_MISS)] = 0x0,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0x0,
+		[C(RESULT_MISS)] = 0x0,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0x0,
+		[C(RESULT_MISS)] = 0x0,
+	},
+},
+[C(DTLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x0568,
+		[C(RESULT_MISS)] = 0x052c,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = 0x0669,
+		[C(RESULT_MISS)] = 0x0530,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = 0x0564,
+		[C(RESULT_MISS)] = 0x0565,
+	},
+},
+[C(ITLB)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = 0x00c0,
+		[C(RESULT_MISS)] = 0x0534,
+	},
+	[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)] = 0x0028,
+		[C(RESULT_MISS)] = 0x0029,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+[C(NODE)] = {
+	[C(OP_READ)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_WRITE)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+	[C(OP_PREFETCH)] = {
+		[C(RESULT_ACCESS)] = -1,
+		[C(RESULT_MISS)] = -1,
+	},
+},
+};
+
+static void zhaoxin_pmu_disable_all(void)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
+}
+
+static void zhaoxin_pmu_enable_all(int added)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+}
+
+static inline u64 zhaoxin_pmu_get_status(void)
+{
+	u64 status;
+
+	rdmsrl(MSR_CORE_PERF_GLOBAL_STATUS, status);
+
+	return status;
+}
+
+static inline void zhaoxin_pmu_ack_status(u64 ack)
+{
+	wrmsrl(MSR_CORE_PERF_GLOBAL_OVF_CTRL, ack);
+}
+
+static inline void zxc_pmu_ack_status(u64 ack)
+{
+	/*
+	 * ZXC needs global control enabled in order to clear status bits.
+	 */
+	zhaoxin_pmu_enable_all(0);
+	zhaoxin_pmu_ack_status(ack);
+	zhaoxin_pmu_disable_all();
+}
+
+static void zhaoxin_pmu_disable_fixed(struct hw_perf_event *hwc)
+{
+	int idx = hwc->idx - INTEL_PMC_IDX_FIXED;
+	u64 ctrl_val, mask;
+
+	mask = 0xfULL << (idx * 4);
+
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static void zhaoxin_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
+		zhaoxin_pmu_disable_fixed(hwc);
+		return;
+	}
+
+	x86_pmu_disable_event(event);
+}
+
+static void zhaoxin_pmu_enable_fixed(struct hw_perf_event *hwc)
+{
+	int idx = hwc->idx - INTEL_PMC_IDX_FIXED;
+	u64 ctrl_val, bits, mask;
+
+	/*
+	 * Enable IRQ generation (0x8),
+	 * and enable ring-3 counting (0x2) and ring-0 counting (0x1)
+	 * if requested:
+	 */
+	bits = 0x8ULL;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_USR)
+		bits |= 0x2;
+	if (hwc->config & ARCH_PERFMON_EVENTSEL_OS)
+		bits |= 0x1;
+
+	bits <<= (idx * 4);
+	mask = 0xfULL << (idx * 4);
+
+	rdmsrl(hwc->config_base, ctrl_val);
+	ctrl_val &= ~mask;
+	ctrl_val |= bits;
+	wrmsrl(hwc->config_base, ctrl_val);
+}
+
+static void zhaoxin_pmu_enable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
+		zhaoxin_pmu_enable_fixed(hwc);
+		return;
+	}
+
+	__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+}
+
+/*
+ * This handler is triggered by the local APIC, so the APIC IRQ handling
+ * rules apply:
+ */
+static int zhaoxin_pmu_handle_irq(struct pt_regs *regs)
+{
+	struct perf_sample_data data;
+	struct cpu_hw_events *cpuc;
+	int handled = 0;
+	u64 status;
+	int bit;
+
+	cpuc = this_cpu_ptr(&cpu_hw_events);
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	zhaoxin_pmu_disable_all();
+	status = zhaoxin_pmu_get_status();
+	if (!status)
+		goto done;
+
+again:
+	if (x86_pmu.enabled_ack)
+		zxc_pmu_ack_status(status);
+	else
+		zhaoxin_pmu_ack_status(status);
+
+	inc_irq_stat(apic_perf_irqs);
+
+	/*
+	 * CondChgd bit 63 doesn't mean any overflow status. Ignore
+	 * and clear the bit.
+	 */
+	if (__test_and_clear_bit(63, (unsigned long *)&status)) {
+		if (!status)
+			goto done;
+	}
+
+	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;
+
+		x86_perf_event_update(event);
+		perf_sample_data_init(&data, 0, event->hw.last_period);
+
+		if (!x86_perf_event_set_period(event))
+			continue;
+
+		if (perf_event_overflow(event, &data, regs))
+			x86_pmu_stop(event, 0);
+	}
+
+	/*
+	 * Repeat if there is more work to be done:
+	 */
+	status = zhaoxin_pmu_get_status();
+	if (status)
+		goto again;
+
+done:
+	zhaoxin_pmu_enable_all(0);
+	return handled;
+}
+
+static u64 zhaoxin_pmu_event_map(int hw_event)
+{
+	return zx_pmon_event_map[hw_event];
+}
+
+static struct event_constraint *
+zhaoxin_get_event_constraints(struct cpu_hw_events *cpuc, int idx,
+			struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	if (x86_pmu.event_constraints) {
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			if ((event->hw.config & c->cmask) == c->code)
+				return c;
+		}
+	}
+
+	return &unconstrained;
+}
+
+PMU_FORMAT_ATTR(event,	"config:0-7");
+PMU_FORMAT_ATTR(umask,	"config:8-15");
+PMU_FORMAT_ATTR(edge,	"config:18");
+PMU_FORMAT_ATTR(inv,	"config:23");
+PMU_FORMAT_ATTR(cmask,	"config:24-31");
+
+static struct attribute *zx_arch_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+static ssize_t zhaoxin_event_sysfs_show(char *page, u64 config)
+{
+	u64 event = (config & ARCH_PERFMON_EVENTSEL_EVENT);
+
+	return x86_event_sysfs_show(page, config, event);
+}
+
+static const struct x86_pmu zhaoxin_pmu __initconst = {
+	.name			= "zhaoxin",
+	.handle_irq		= zhaoxin_pmu_handle_irq,
+	.disable_all		= zhaoxin_pmu_disable_all,
+	.enable_all		= zhaoxin_pmu_enable_all,
+	.enable			= zhaoxin_pmu_enable_event,
+	.disable		= zhaoxin_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_ARCH_PERFMON_EVENTSEL0,
+	.perfctr		= MSR_ARCH_PERFMON_PERFCTR0,
+	.event_map		= zhaoxin_pmu_event_map,
+	.max_events		= ARRAY_SIZE(zx_pmon_event_map),
+	.apic			= 1,
+	/*
+	 * For zxd/zxe, read/write operation for PMCx MSR is 48 bits.
+	 */
+	.max_period		= (1ULL << 47) - 1,
+	.get_event_constraints	= zhaoxin_get_event_constraints,
+
+	.format_attrs		= zx_arch_formats_attr,
+	.events_sysfs_show	= zhaoxin_event_sysfs_show,
+};
+
+static const struct { int id; char *name; } zx_arch_events_map[] __initconst = {
+	{ PERF_COUNT_HW_CPU_CYCLES, "cpu cycles" },
+	{ PERF_COUNT_HW_INSTRUCTIONS, "instructions" },
+	{ PERF_COUNT_HW_BUS_CYCLES, "bus cycles" },
+	{ PERF_COUNT_HW_CACHE_REFERENCES, "cache references" },
+	{ PERF_COUNT_HW_CACHE_MISSES, "cache misses" },
+	{ PERF_COUNT_HW_BRANCH_INSTRUCTIONS, "branch instructions" },
+	{ PERF_COUNT_HW_BRANCH_MISSES, "branch misses" },
+};
+
+static __init void zhaoxin_arch_events_quirk(void)
+{
+	int bit;
+
+	/* disable event that reported as not presend by cpuid */
+	for_each_set_bit(bit, x86_pmu.events_mask, ARRAY_SIZE(zx_arch_events_map)) {
+		zx_pmon_event_map[zx_arch_events_map[bit].id] = 0;
+		pr_warn("CPUID marked event: \'%s\' unavailable\n",
+			zx_arch_events_map[bit].name);
+	}
+}
+
+__init int zhaoxin_pmu_init(void)
+{
+	union cpuid10_edx edx;
+	union cpuid10_eax eax;
+	union cpuid10_ebx ebx;
+	struct event_constraint *c;
+	unsigned int unused;
+	int version;
+
+	pr_info("Welcome to zhaoxin pmu!\n");
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * hw_event or not.
+	 */
+	cpuid(10, &eax.full, &ebx.full, &unused, &edx.full);
+
+	if (eax.split.mask_length < ARCH_PERFMON_EVENTS_COUNT - 1)
+		return -ENODEV;
+
+	version = eax.split.version_id;
+	if (version != 2)
+		return -ENODEV;
+
+	x86_pmu = zhaoxin_pmu;
+	pr_info("Version check pass!\n");
+
+	x86_pmu.version			= version;
+	x86_pmu.num_counters		= eax.split.num_counters;
+	x86_pmu.cntval_bits		= eax.split.bit_width;
+	x86_pmu.cntval_mask		= (1ULL << eax.split.bit_width) - 1;
+	x86_pmu.events_maskl		= ebx.full;
+	x86_pmu.events_mask_len		= eax.split.mask_length;
+
+	x86_pmu.num_counters_fixed = edx.split.num_counters_fixed;
+	x86_add_quirk(zhaoxin_arch_events_quirk);
+
+	switch (boot_cpu_data.x86) {
+	case 0x06:
+		/*
+		 * Support Zhaoxin CPU from ZXC series, exclude Nano series through FMS.
+		 * Nano FMS: Family=6, Model=F, Stepping=[0-A][C-D]
+		 * ZXC FMS: Family=6, Model=F, Stepping=E-F OR Family=6, Model=0x19, Stepping=0-3
+		 */
+		if ((boot_cpu_data.x86_model == 0x0f && boot_cpu_data.x86_stepping >= 0x0e) ||
+			boot_cpu_data.x86_model == 0x19) {
+
+			x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
+
+			/* Clearing status works only if the global control is enable on zxc. */
+			x86_pmu.enabled_ack = 1;
+
+			x86_pmu.event_constraints = zxc_event_constraints;
+			zx_pmon_event_map[PERF_COUNT_HW_INSTRUCTIONS] = 0;
+			zx_pmon_event_map[PERF_COUNT_HW_CACHE_REFERENCES] = 0;
+			zx_pmon_event_map[PERF_COUNT_HW_CACHE_MISSES] = 0;
+			zx_pmon_event_map[PERF_COUNT_HW_BUS_CYCLES] = 0;
+
+			pr_cont("ZXC events, ");
+			break;
+		}
+		return -ENODEV;
+
+	case 0x07:
+		zx_pmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event = 0x01, .umask = 0x01, .inv = 0x01, .cmask = 0x01);
+
+		zx_pmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] =
+			X86_CONFIG(.event = 0x0f, .umask = 0x04, .inv = 0, .cmask = 0);
+
+		switch (boot_cpu_data.x86_model) {
+		case 0x1b:
+			memcpy(hw_cache_event_ids, zxd_hw_cache_event_ids,
+			       sizeof(hw_cache_event_ids));
+
+			x86_pmu.event_constraints = zxd_event_constraints;
+
+			zx_pmon_event_map[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x0700;
+			zx_pmon_event_map[PERF_COUNT_HW_BRANCH_MISSES] = 0x0709;
+
+			pr_cont("ZXD events, ");
+			break;
+		case 0x3b:
+			memcpy(hw_cache_event_ids, zxe_hw_cache_event_ids,
+			       sizeof(hw_cache_event_ids));
+
+			x86_pmu.event_constraints = zxd_event_constraints;
+
+			zx_pmon_event_map[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x0028;
+			zx_pmon_event_map[PERF_COUNT_HW_BRANCH_MISSES] = 0x0029;
+
+			pr_cont("ZXE events, ");
+			break;
+		default:
+			return -ENODEV;
+		}
+		break;
+
+	default:
+		return -ENODEV;
+	}
+
+	x86_pmu.intel_ctrl = (1 << (x86_pmu.num_counters)) - 1;
+	x86_pmu.intel_ctrl |= ((1LL << x86_pmu.num_counters_fixed)-1) << INTEL_PMC_IDX_FIXED;
+
+	if (x86_pmu.event_constraints) {
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
+			c->weight += x86_pmu.num_counters;
+		}
+	}
+
+	return 0;
+}
+