[GIT PULL] perf fixes

[Ingo Molnar <mingo@kernel.org>]

Linus,

Please pull the latest perf-urgent-for-linus git tree from:

   git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git perf-urgent-for-linus

   # HEAD: 2c81a6477081966fe80b8c6daa68459bca896774 perf/core: Fix pmu::filter_match for SW-led groups

Various fixes:

 - 32-bit callgraph bug fix
 - suboptimal event group scheduling bug fix
 - event constraint fixes for Broadwell/Skylake
 - RAPL module name collision fix

 Thanks,

	Ingo

------------------>
Josh Poimboeuf (1):
      perf/x86: Fix 32-bit perf user callgraph collection

Mark Rutland (1):
      perf/core: Fix pmu::filter_match for SW-led groups

Stephane Eranian (1):
      perf/x86/intel: Update event constraints when HT is off

Ville Syrjälä (1):
      x86/perf/intel/rapl: Fix module name collision with powercap intel-rapl


 arch/x86/events/core.c         | 11 ++++++-----
 arch/x86/events/intel/Makefile |  4 ++--
 arch/x86/events/intel/core.c   | 29 +++++++++++++++++++++++++++++
 kernel/events/core.c           | 23 ++++++++++++++++++++++-
 4 files changed, 59 insertions(+), 8 deletions(-)

diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 33787ee817f0..26ced536005a 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2319,7 +2319,7 @@ void
 perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs)
 {
 	struct stack_frame frame;
-	const void __user *fp;
+	const unsigned long __user *fp;
 
 	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
 		/* TODO: We don't support guest os callchain now */
@@ -2332,7 +2332,7 @@ perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs
 	if (regs->flags & (X86_VM_MASK | PERF_EFLAGS_VM))
 		return;
 
-	fp = (void __user *)regs->bp;
+	fp = (unsigned long __user *)regs->bp;
 
 	perf_callchain_store(entry, regs->ip);
 
@@ -2345,16 +2345,17 @@ perf_callchain_user(struct perf_callchain_entry_ctx *entry, struct pt_regs *regs
 	pagefault_disable();
 	while (entry->nr < entry->max_stack) {
 		unsigned long bytes;
+
 		frame.next_frame	     = NULL;
 		frame.return_address = 0;
 
-		if (!access_ok(VERIFY_READ, fp, 16))
+		if (!access_ok(VERIFY_READ, fp, sizeof(*fp) * 2))
 			break;
 
-		bytes = __copy_from_user_nmi(&frame.next_frame, fp, 8);
+		bytes = __copy_from_user_nmi(&frame.next_frame, fp, sizeof(*fp));
 		if (bytes != 0)
 			break;
-		bytes = __copy_from_user_nmi(&frame.return_address, fp+8, 8);
+		bytes = __copy_from_user_nmi(&frame.return_address, fp + 1, sizeof(*fp));
 		if (bytes != 0)
 			break;
 
diff --git a/arch/x86/events/intel/Makefile b/arch/x86/events/intel/Makefile
index 3660b2cf245a..06c2baa51814 100644
--- a/arch/x86/events/intel/Makefile
+++ b/arch/x86/events/intel/Makefile
@@ -1,8 +1,8 @@
 obj-$(CONFIG_CPU_SUP_INTEL)		+= core.o bts.o cqm.o
 obj-$(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_RAPL)	+= intel-rapl.o
-intel-rapl-objs				:= rapl.o
+obj-$(CONFIG_PERF_EVENTS_INTEL_RAPL)	+= intel-rapl-perf.o
+intel-rapl-perf-objs			:= rapl.o
 obj-$(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
diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 7c666958a625..9b4f9d3ce465 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -115,6 +115,10 @@ static struct event_constraint intel_snb_event_constraints[] __read_mostly =
 	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.* */
@@ -139,6 +143,10 @@ static struct event_constraint intel_ivb_event_constraints[] __read_mostly =
 	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.* */
@@ -182,6 +190,16 @@ struct event_constraint intel_skl_event_constraints[] = {
 	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
 };
 
@@ -250,6 +268,10 @@ static struct event_constraint intel_hsw_event_constraints[] = {
 	/* 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.* */
@@ -264,6 +286,13 @@ struct event_constraint intel_bdw_event_constraints[] = {
 	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
 };
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 85cd41878a74..43d43a2d5811 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1678,12 +1678,33 @@ static bool is_orphaned_event(struct perf_event *event)
 	return event->state == PERF_EVENT_STATE_DEAD;
 }
 
-static inline int pmu_filter_match(struct perf_event *event)
+static inline int __pmu_filter_match(struct perf_event *event)
 {
 	struct pmu *pmu = event->pmu;
 	return pmu->filter_match ? pmu->filter_match(event) : 1;
 }
 
+/*
+ * Check whether we should attempt to schedule an event group based on
+ * PMU-specific filtering. An event group can consist of HW and SW events,
+ * potentially with a SW leader, so we must check all the filters, to
+ * determine whether a group is schedulable:
+ */
+static inline int pmu_filter_match(struct perf_event *event)
+{
+	struct perf_event *child;
+
+	if (!__pmu_filter_match(event))
+		return 0;
+
+	list_for_each_entry(child, &event->sibling_list, group_entry) {
+		if (!__pmu_filter_match(child))
+			return 0;
+	}
+
+	return 1;
+}
+
 static inline int
 event_filter_match(struct perf_event *event)
 {