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

Pull perf changes from Ingo Molnar:
 "Main changes:

  Kernel side changes:

   - Add SNB/IVB/HSW client uncore memory controller support (Stephane
     Eranian)

   - Fix various x86/P4 PMU driver bugs (Don Zickus)

  Tooling, user visible changes:

   - Add several futex 'perf bench' microbenchmarks (Davidlohr Bueso)

   - Speed up thread map generation (Don Zickus)

   - Introduce 'perf kvm --list-cmds' command line option for use by
     scripts (Ramkumar Ramachandra)

   - Print the evsel name in the annotate stdio output, prep to fix
     support outputting annotation for multiple events, not just for the
     first one (Arnaldo Carvalho de Melo)

   - Allow setting preferred callchain method in .perfconfig (Jiri Olsa)

   - Show in what binaries/modules 'perf probe's are set (Masami
     Hiramatsu)

   - Support distro-style debuginfo for uprobe in 'perf probe' (Masami
     Hiramatsu)

  Tooling, internal changes and fixes:

   - Use tid in mmap/mmap2 events to find maps (Don Zickus)

   - Record the reason for filtering an address_location (Namhyung Kim)

   - Apply all filters to an addr_location (Namhyung Kim)

   - Merge al->filtered with hist_entry->filtered in report/hists
     (Namhyung Kim)

   - Fix memory leak when synthesizing thread records (Namhyung Kim)

   - Use ui__has_annotation() in 'report' (Namhyung Kim)

   - hists browser refactorings to reuse code accross UIs (Namhyung Kim)

   - Add support for the new DWARF unwinder library in elfutils (Jiri
     Olsa)

   - Fix build race in the generation of bison files (Jiri Olsa)

   - Further streamline the feature detection display, trimming it a bit
     to show just the libraries detected, using VF=1 gets a more verbose
     output, showing the less interesting feature checks as well (Jiri
     Olsa).

   - Check compatible symtab type before loading dso (Namhyung Kim)

   - Check return value of filename__read_debuglink() (Stephane Eranian)

   - Move some hashing and fs related code from tools/perf/util/ to
     tools/lib/ so that it can be used by more tools/ living utilities
     (Borislav Petkov)

   - Prepare DWARF unwinding code for using an elfutils alternative
     unwinding library (Jiri Olsa)

   - Fix DWARF unwind max_stack processing (Jiri Olsa)

   - Add dwarf unwind 'perf test' entry (Jiri Olsa)

   - 'perf probe' improvements including memory leak fixes, sharing the
     intlist class with other tools, uprobes/kprobes code sharing and
     use of ref_reloc_sym (Masami Hiramatsu)

   - Shorten sample symbol resolving by adding cpumode to struct
     addr_location (Arnaldo Carvalho de Melo)

   - Fix synthesizing mmaps for threads (Don Zickus)

   - Fix invalid output on event group stdio report (Namhyung Kim)

   - Fixup header alignment in 'perf sched latency' output (Ramkumar
     Ramachandra)

   - Fix off-by-one error in 'perf timechart record' argv handling
     (Ramkumar Ramachandra)

  Tooling, cleanups:

   - Remove unused thread__find_map function (Jiri Olsa)

   - Remove unused simple_strtoul() function (Ramkumar Ramachandra)

  Tooling, documentation updates:

   - Update function names in debug messages (Ramkumar Ramachandra)

   - Update some code references in design.txt (Ramkumar Ramachandra)

   - Clarify load-latency information in the 'perf mem' docs (Andi
     Kleen)

   - Clarify x86 register naming in 'perf probe' docs (Andi Kleen)"

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (96 commits)
  perf tools: Remove unused simple_strtoul() function
  perf tools: Update some code references in design.txt
  perf evsel: Update function names in debug messages
  perf tools: Remove thread__find_map function
  perf annotate: Print the evsel name in the stdio output
  perf report: Use ui__has_annotation()
  perf tools: Fix memory leak when synthesizing thread records
  perf tools: Use tid in mmap/mmap2 events to find maps
  perf report: Merge al->filtered with hist_entry->filtered
  perf symbols: Apply all filters to an addr_location
  perf symbols: Record the reason for filtering an address_location
  perf sched: Fixup header alignment in 'latency' output
  perf timechart: Fix off-by-one error in 'record' argv handling
  perf machine: Factor machine__find_thread to take tid argument
  perf tools: Speed up thread map generation
  perf kvm: introduce --list-cmds for use by scripts
  perf ui hists: Pass evsel to hpp->header/width functions explicitly
  perf symbols: Introduce thread__find_cpumode_addr_location
  perf session: Change header.misc dump from decimal to hex
  perf ui/tui: Reuse generic __hpp__fmt() code
  ...

6 files changed, 573 insertions, 102 deletions

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 79f9f84..ae407f7 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -892,7 +892,6 @@ static void x86_pmu_enable(struct pmu *pmu)
 		 * hw_perf_group_sched_in() or x86_pmu_enable()
 		 *
 		 * step1: save events moving to new counters
-		 * step2: reprogram moved events into new counters
 		 */
 		for (i = 0; i < n_running; i++) {
 			event = cpuc->event_list[i];
@@ -918,6 +917,9 @@ static void x86_pmu_enable(struct pmu *pmu)
 			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;
@@ -1043,7 +1045,7 @@ static int x86_pmu_add(struct perf_event *event, int flags)
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be performed
-	 * at commit time (->commit_txn) as a whole
+	 * at commit time (->commit_txn) as a whole.
 	 */
 	if (cpuc->group_flag & PERF_EVENT_TXN)
 		goto done_collect;
@@ -1058,6 +1060,10 @@ static int x86_pmu_add(struct perf_event *event, int flags)
 	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;
@@ -1183,28 +1189,38 @@ static void x86_pmu_del(struct perf_event *event, int flags)
 	 * If we're called during a txn, we don't need to do anything.
 	 * 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->group_flag & PERF_EVENT_TXN)
 		return;
 
+	/*
+	 * 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]) {
+		if (event == cpuc->event_list[i])
+			break;
+	}
 
-			if (i >= cpuc->n_events - cpuc->n_added)
-				--cpuc->n_added;
+	if (WARN_ON_ONCE(i == cpuc->n_events)) /* called ->del() without ->add() ? */
+		return;
 
-			if (x86_pmu.put_event_constraints)
-				x86_pmu.put_event_constraints(cpuc, event);
+	/* If we have a newly added event; make sure to decrease n_added. */
+	if (i >= cpuc->n_events - cpuc->n_added)
+		--cpuc->n_added;
 
-			while (++i < cpuc->n_events)
-				cpuc->event_list[i-1] = cpuc->event_list[i];
+	if (x86_pmu.put_event_constraints)
+		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->n_events;
 
-			--cpuc->n_events;
-			break;
-		}
-	}
 	perf_event_update_userpage(event);
 }
 
@@ -1598,7 +1614,8 @@ static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
 	__this_cpu_and(cpu_hw_events.group_flag, ~PERF_EVENT_TXN);
 	/*
-	 * Truncate the collected events.
+	 * 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));
@@ -1609,6 +1626,8 @@ static void x86_pmu_cancel_txn(struct pmu *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)
 {
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 4972c24..3b2f9bd 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -130,9 +130,11 @@ struct cpu_hw_events {
 	unsigned long		running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
 	int			enabled;
 
-	int			n_events;
-	int			n_added;
-	int			n_txn;
+	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			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 */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
index 047f540..bd2253d 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
@@ -66,6 +66,47 @@ 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 uncore_pmu_start_hrtimer(struct intel_uncore_box *box);
+static void uncore_pmu_cancel_hrtimer(struct intel_uncore_box *box);
+static void uncore_perf_event_update(struct intel_uncore_box *box, struct perf_event *event);
+static void uncore_pmu_event_read(struct perf_event *event);
+
+static struct intel_uncore_pmu *uncore_event_to_pmu(struct perf_event *event)
+{
+	return container_of(event->pmu, struct intel_uncore_pmu, pmu);
+}
+
+static struct intel_uncore_box *
+uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu)
+{
+	struct intel_uncore_box *box;
+
+	box = *per_cpu_ptr(pmu->box, cpu);
+	if (box)
+		return box;
+
+	raw_spin_lock(&uncore_box_lock);
+	list_for_each_entry(box, &pmu->box_list, list) {
+		if (box->phys_id == topology_physical_package_id(cpu)) {
+			atomic_inc(&box->refcnt);
+			*per_cpu_ptr(pmu->box, cpu) = box;
+			break;
+		}
+	}
+	raw_spin_unlock(&uncore_box_lock);
+
+	return *per_cpu_ptr(pmu->box, cpu);
+}
+
+static struct intel_uncore_box *uncore_event_to_box(struct perf_event *event)
+{
+	/*
+	 * perf core schedules event on the basis of cpu, uncore events are
+	 * collected by one of the cpus inside a physical package.
+	 */
+	return uncore_pmu_to_box(uncore_event_to_pmu(event), smp_processor_id());
+}
+
 static u64 uncore_msr_read_counter(struct intel_uncore_box *box, struct perf_event *event)
 {
 	u64 count;
@@ -1639,6 +1680,349 @@ static struct intel_uncore_type *snb_msr_uncores[] = {
 	&snb_uncore_cbox,
 	NULL,
 };
+
+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"),
+
+	{ /* 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
+
+static struct attribute *snb_uncore_imc_formats_attr[] = {
+	&format_attr_event.attr,
+	NULL,
+};
+
+static 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 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, SNB_UNCORE_PCI_IMC_MAP_SIZE);
+	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)
+{}
+
+static u64 snb_uncore_imc_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	return (u64)*(unsigned int *)(box->io_addr + hwc->event_base);
+}
+
+/*
+ * custom event_init() function because we define our own fixed, free
+ * running counters, so we do not want to conflict with generic uncore
+ * logic. Also simplifies processing
+ */
+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.exclude_user   ||
+	    event->attr.exclude_kernel ||
+	    event->attr.exclude_hv     ||
+	    event->attr.exclude_idle   ||
+	    event->attr.exclude_host   ||
+	    event->attr.exclude_guest  ||
+	    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->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_FIXED;
+		break;
+	case SNB_UNCORE_PCI_IMC_DATA_WRITES:
+		base = SNB_UNCORE_PCI_IMC_DATA_WRITES_BASE;
+		idx = UNCORE_PMC_IDX_FIXED + 1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	/* must be done before validate_group */
+	event->hw.event_base = base;
+	event->hw.config = cfg;
+	event->hw.idx = idx;
+
+	/* 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;
+}
+
+static void snb_uncore_imc_event_start(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	u64 count;
+
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
+
+	event->hw.state = 0;
+	box->n_active++;
+
+	list_add_tail(&event->active_entry, &box->active_list);
+
+	count = snb_uncore_imc_read_counter(box, event);
+	local64_set(&event->hw.prev_count, count);
+
+	if (box->n_active == 1)
+		uncore_pmu_start_hrtimer(box);
+}
+
+static void snb_uncore_imc_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;
+
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		box->n_active--;
+
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+
+		list_del(&event->active_entry);
+
+		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;
+	}
+}
+
+static int snb_uncore_imc_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;
+
+	if (!box)
+		return -ENODEV;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
+
+	snb_uncore_imc_event_start(event, 0);
+
+	box->n_events++;
+
+	return 0;
+}
+
+static void snb_uncore_imc_event_del(struct perf_event *event, int flags)
+{
+	struct intel_uncore_box *box = uncore_event_to_box(event);
+	int i;
+
+	snb_uncore_imc_event_stop(event, PERF_EF_UPDATE);
+
+	for (i = 0; i < box->n_events; i++) {
+		if (event == box->event_list[i]) {
+			--box->n_events;
+			break;
+		}
+	}
+}
+
+static int snb_pci2phy_map_init(int devid)
+{
+	struct pci_dev *dev = NULL;
+	int bus;
+
+	dev = pci_get_device(PCI_VENDOR_ID_INTEL, devid, dev);
+	if (!dev)
+		return -ENOTTY;
+
+	bus = dev->bus->number;
+
+	pcibus_to_physid[bus] = 0;
+
+	pci_dev_put(dev);
+
+	return 0;
+}
+
+static struct pmu snb_uncore_imc_pmu = {
+	.task_ctx_nr	= perf_invalid_context,
+	.event_init	= snb_uncore_imc_event_init,
+	.add		= snb_uncore_imc_event_add,
+	.del		= snb_uncore_imc_event_del,
+	.start		= snb_uncore_imc_event_start,
+	.stop		= snb_uncore_imc_event_stop,
+	.read		= uncore_pmu_event_read,
+};
+
+static struct intel_uncore_ops snb_uncore_imc_ops = {
+	.init_box	= snb_uncore_imc_init_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   = 2,
+	.num_boxes	= 1,
+	.fixed_ctr_bits	= 32,
+	.fixed_ctr	= SNB_UNCORE_PCI_IMC_CTR_BASE,
+	.event_descs	= snb_uncore_imc_events,
+	.format_group	= &snb_uncore_imc_format_group,
+	.perf_ctr	= SNB_UNCORE_PCI_IMC_DATA_READS_BASE,
+	.event_mask	= SNB_UNCORE_PCI_IMC_EVENT_MASK,
+	.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 DEFINE_PCI_DEVICE_TABLE(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 DEFINE_PCI_DEVICE_TABLE(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),
+	},
+	{ /* end: all zeroes */ },
+};
+
+static DEFINE_PCI_DEVICE_TABLE(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),
+	},
+	{ /* 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,
+};
+
 /* end of Sandy Bridge uncore support */
 
 /* Nehalem uncore support */
@@ -2789,6 +3173,7 @@ again:
 static enum hrtimer_restart uncore_pmu_hrtimer(struct hrtimer *hrtimer)
 {
 	struct intel_uncore_box *box;
+	struct perf_event *event;
 	unsigned long flags;
 	int bit;
 
@@ -2801,19 +3186,27 @@ static enum hrtimer_restart uncore_pmu_hrtimer(struct hrtimer *hrtimer)
 	 */
 	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(UNCORE_PMU_HRTIMER_INTERVAL));
+	hrtimer_forward_now(hrtimer, ns_to_ktime(box->hrtimer_duration));
 	return HRTIMER_RESTART;
 }
 
 static void uncore_pmu_start_hrtimer(struct intel_uncore_box *box)
 {
 	__hrtimer_start_range_ns(&box->hrtimer,
-			ns_to_ktime(UNCORE_PMU_HRTIMER_INTERVAL), 0,
+			ns_to_ktime(box->hrtimer_duration), 0,
 			HRTIMER_MODE_REL_PINNED, 0);
 }
 
@@ -2847,43 +3240,12 @@ static struct intel_uncore_box *uncore_alloc_box(struct intel_uncore_type *type,
 	box->cpu = -1;
 	box->phys_id = -1;
 
-	return box;
-}
+	/* set default hrtimer timeout */
+	box->hrtimer_duration = UNCORE_PMU_HRTIMER_INTERVAL;
 
-static struct intel_uncore_box *
-uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu)
-{
-	struct intel_uncore_box *box;
+	INIT_LIST_HEAD(&box->active_list);
 
-	box = *per_cpu_ptr(pmu->box, cpu);
-	if (box)
-		return box;
-
-	raw_spin_lock(&uncore_box_lock);
-	list_for_each_entry(box, &pmu->box_list, list) {
-		if (box->phys_id == topology_physical_package_id(cpu)) {
-			atomic_inc(&box->refcnt);
-			*per_cpu_ptr(pmu->box, cpu) = box;
-			break;
-		}
-	}
-	raw_spin_unlock(&uncore_box_lock);
-
-	return *per_cpu_ptr(pmu->box, cpu);
-}
-
-static struct intel_uncore_pmu *uncore_event_to_pmu(struct perf_event *event)
-{
-	return container_of(event->pmu, struct intel_uncore_pmu, pmu);
-}
-
-static struct intel_uncore_box *uncore_event_to_box(struct perf_event *event)
-{
-	/*
-	 * perf core schedules event on the basis of cpu, uncore events are
-	 * collected by one of the cpus inside a physical package.
-	 */
-	return uncore_pmu_to_box(uncore_event_to_pmu(event), smp_processor_id());
+	return box;
 }
 
 static int
@@ -3279,16 +3641,21 @@ static int __init uncore_pmu_register(struct intel_uncore_pmu *pmu)
 {
 	int ret;
 
-	pmu->pmu = (struct pmu) {
-		.attr_groups	= pmu->type->attr_groups,
-		.task_ctx_nr	= perf_invalid_context,
-		.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,
-	};
+	if (!pmu->type->pmu) {
+		pmu->pmu = (struct pmu) {
+			.attr_groups	= pmu->type->attr_groups,
+			.task_ctx_nr	= perf_invalid_context,
+			.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,
+		};
+	} else {
+		pmu->pmu = *pmu->type->pmu;
+		pmu->pmu.attr_groups = pmu->type->attr_groups;
+	}
 
 	if (pmu->type->num_boxes == 1) {
 		if (strlen(pmu->type->name) > 0)
@@ -3502,6 +3869,28 @@ static int __init uncore_pci_init(void)
 		pci_uncores = ivt_pci_uncores;
 		uncore_pci_driver = &ivt_uncore_pci_driver;
 		break;
+	case 42: /* Sandy Bridge */
+		ret = snb_pci2phy_map_init(PCI_DEVICE_ID_INTEL_SNB_IMC);
+		if (ret)
+			return ret;
+		pci_uncores = snb_pci_uncores;
+		uncore_pci_driver = &snb_uncore_pci_driver;
+		break;
+	case 58: /* Ivy Bridge */
+		ret = snb_pci2phy_map_init(PCI_DEVICE_ID_INTEL_IVB_IMC);
+		if (ret)
+			return ret;
+		pci_uncores = snb_pci_uncores;
+		uncore_pci_driver = &ivb_uncore_pci_driver;
+		break;
+	case 60: /* Haswell */
+	case 69: /* Haswell Celeron */
+		ret = snb_pci2phy_map_init(PCI_DEVICE_ID_INTEL_HSW_IMC);
+		if (ret)
+			return ret;
+		pci_uncores = snb_pci_uncores;
+		uncore_pci_driver = &hsw_uncore_pci_driver;
+		break;
 	default:
 		return 0;
 	}
@@ -3773,7 +4162,7 @@ static void __init uncore_cpu_setup(void *dummy)
 
 static int __init uncore_cpu_init(void)
 {
-	int ret, cpu, max_cores;
+	int ret, max_cores;
 
 	max_cores = boot_cpu_data.x86_max_cores;
 	switch (boot_cpu_data.x86_model) {
@@ -3817,29 +4206,6 @@ static int __init uncore_cpu_init(void)
 	if (ret)
 		return ret;
 
-	get_online_cpus();
-
-	for_each_online_cpu(cpu) {
-		int i, phys_id = topology_physical_package_id(cpu);
-
-		for_each_cpu(i, &uncore_cpu_mask) {
-			if (phys_id == topology_physical_package_id(i)) {
-				phys_id = -1;
-				break;
-			}
-		}
-		if (phys_id < 0)
-			continue;
-
-		uncore_cpu_prepare(cpu, phys_id);
-		uncore_event_init_cpu(cpu);
-	}
-	on_each_cpu(uncore_cpu_setup, NULL, 1);
-
-	register_cpu_notifier(&uncore_cpu_nb);
-
-	put_online_cpus();
-
 	return 0;
 }
 
@@ -3868,6 +4234,41 @@ static int __init uncore_pmus_register(void)
 	return 0;
 }
 
+static void __init uncore_cpumask_init(void)
+{
+	int cpu;
+
+	/*
+	 * ony invoke once from msr or pci init code
+	 */
+	if (!cpumask_empty(&uncore_cpu_mask))
+		return;
+
+	get_online_cpus();
+
+	for_each_online_cpu(cpu) {
+		int i, phys_id = topology_physical_package_id(cpu);
+
+		for_each_cpu(i, &uncore_cpu_mask) {
+			if (phys_id == topology_physical_package_id(i)) {
+				phys_id = -1;
+				break;
+			}
+		}
+		if (phys_id < 0)
+			continue;
+
+		uncore_cpu_prepare(cpu, phys_id);
+		uncore_event_init_cpu(cpu);
+	}
+	on_each_cpu(uncore_cpu_setup, NULL, 1);
+
+	register_cpu_notifier(&uncore_cpu_nb);
+
+	put_online_cpus();
+}
+
+
 static int __init intel_uncore_init(void)
 {
 	int ret;
@@ -3886,6 +4287,7 @@ static int __init intel_uncore_init(void)
 		uncore_pci_exit();
 		goto fail;
 	}
+	uncore_cpumask_init();
 
 	uncore_pmus_register();
 	return 0;
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.h b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
index a80ab71..90236f0 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.h
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
@@ -6,6 +6,7 @@
 
 #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
@@ -440,6 +441,7 @@ struct intel_uncore_type {
 	struct intel_uncore_ops *ops;
 	struct uncore_event_desc *event_descs;
 	const struct attribute_group *attr_groups[4];
+	struct pmu *pmu; /* for custom pmu ops */
 };
 
 #define pmu_group attr_groups[0]
@@ -488,8 +490,11 @@ struct intel_uncore_box {
 	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 *io_addr;
 	struct intel_uncore_extra_reg shared_regs[0];
 };
 
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index 3486e66..5d466b7 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -1257,7 +1257,24 @@ again:
 			pass++;
 			goto again;
 		}
-
+		/*
+		 * Perf does test runs to see if a whole group can be assigned
+		 * together succesfully.  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;
@@ -1322,6 +1339,7 @@ static __initconst const struct x86_pmu p4_pmu = {
 __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);
@@ -1340,5 +1358,19 @@ __init int p4_pmu_init(void)
 
 	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/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 6fcb49c..b4872b9 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -87,6 +87,7 @@ __setup("unknown_nmi_panic", setup_unknown_nmi_panic);
 #define nmi_to_desc(type) (&nmi_desc[type])
 
 static u64 nmi_longest_ns = 1 * NSEC_PER_MSEC;
+
 static int __init nmi_warning_debugfs(void)
 {
 	debugfs_create_u64("nmi_longest_ns", 0644,
@@ -95,6 +96,20 @@ static int __init nmi_warning_debugfs(void)
 }
 fs_initcall(nmi_warning_debugfs);
 
+static void nmi_max_handler(struct irq_work *w)
+{
+	struct nmiaction *a = container_of(w, struct nmiaction, irq_work);
+	int remainder_ns, decimal_msecs;
+	u64 whole_msecs = ACCESS_ONCE(a->max_duration);
+
+	remainder_ns = do_div(whole_msecs, (1000 * 1000));
+	decimal_msecs = remainder_ns / 1000;
+
+	printk_ratelimited(KERN_INFO
+		"INFO: NMI handler (%ps) took too long to run: %lld.%03d msecs\n",
+		a->handler, whole_msecs, decimal_msecs);
+}
+
 static int __kprobes nmi_handle(unsigned int type, struct pt_regs *regs, bool b2b)
 {
 	struct nmi_desc *desc = nmi_to_desc(type);
@@ -110,26 +125,20 @@ static int __kprobes nmi_handle(unsigned int type, struct pt_regs *regs, bool b2
 	 * to handle those situations.
 	 */
 	list_for_each_entry_rcu(a, &desc->head, list) {
-		u64 before, delta, whole_msecs;
-		int remainder_ns, decimal_msecs, thishandled;
+		int thishandled;
+		u64 delta;
 
-		before = sched_clock();
+		delta = sched_clock();
 		thishandled = a->handler(type, regs);
 		handled += thishandled;
-		delta = sched_clock() - before;
+		delta = sched_clock() - delta;
 		trace_nmi_handler(a->handler, (int)delta, thishandled);
 
-		if (delta < nmi_longest_ns)
+		if (delta < nmi_longest_ns || delta < a->max_duration)
 			continue;
 
-		nmi_longest_ns = delta;
-		whole_msecs = delta;
-		remainder_ns = do_div(whole_msecs, (1000 * 1000));
-		decimal_msecs = remainder_ns / 1000;
-		printk_ratelimited(KERN_INFO
-			"INFO: NMI handler (%ps) took too long to run: "
-			"%lld.%03d msecs\n", a->handler, whole_msecs,
-			decimal_msecs);
+		a->max_duration = delta;
+		irq_work_queue(&a->irq_work);
 	}
 
 	rcu_read_unlock();
@@ -146,6 +155,8 @@ int __register_nmi_handler(unsigned int type, struct nmiaction *action)
 	if (!action->handler)
 		return -EINVAL;
 
+	init_irq_work(&action->irq_work, nmi_max_handler);
+
 	spin_lock_irqsave(&desc->lock, flags);
 
 	/*