Merge tag 'perf-tools-fixes-for-v5.16-2021-11-19' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull perf tools fixes from Arnaldo Carvalho de Melo:

 - Fix the 'local_weight', 'weight' (memory access latency),
   'local_ins_lat', 'ins_lat' (instruction latency) and 'pstage_cyc'
   (pipeline stage cycles) sort key sample aggregation.

 - Fix 'perf test' entry for watchpoints on s/390.

 - Fix branch_stack entry endianness check in the 'perf test' sample
   parsing test.

 - Fix ARM SPE handling on 'perf inject'.

 - Fix memory leaks detected with ASan.

 - Fix build on arm64 related to reallocarray() availability.

 - Sync copies of kernel headers: cpufeatures, kvm, MIPS syscalltable
   (futex_waitv).

* tag 'perf-tools-fixes-for-v5.16-2021-11-19' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf evsel: Fix memory leaks relating to unit
  perf report: Fix memory leaks around perf_tip()
  perf hist: Fix memory leak of a perf_hpp_fmt
  tools headers UAPI: Sync MIPS syscall table file changed by new futex_waitv syscall
  tools build: Fix removal of feature-sync-compare-and-swap feature detection
  perf inject: Fix ARM SPE handling
  perf bench: Fix two memory leaks detected with ASan
  perf test sample-parsing: Fix branch_stack entry endianness check
  tools headers UAPI: Sync x86's asm/kvm.h with the kernel sources
  perf sort: Fix the 'p_stage_cyc' sort key behavior
  perf sort: Fix the 'ins_lat' sort key behavior
  perf sort: Fix the 'weight' sort key behavior
  perf tools: Set COMPAT_NEED_REALLOCARRAY for CONFIG_AUXTRACE=1
  perf tests wp: Remove unused functions on s390
  tools headers UAPI: Sync linux/kvm.h with the kernel sources
  tools headers cpufeatures: Sync with the kernel sources

22 files changed, 136 insertions, 109 deletions

diff --git a/tools/arch/x86/include/asm/cpufeatures.h b/tools/arch/x86/include/asm/cpufeatures.h
index d0ce5cfd3ac1..d5b5f2ab87a0 100644
--- a/tools/arch/x86/include/asm/cpufeatures.h
+++ b/tools/arch/x86/include/asm/cpufeatures.h
@@ -277,6 +277,7 @@
 #define X86_FEATURE_XSAVEC		(10*32+ 1) /* XSAVEC instruction */
 #define X86_FEATURE_XGETBV1		(10*32+ 2) /* XGETBV with ECX = 1 instruction */
 #define X86_FEATURE_XSAVES		(10*32+ 3) /* XSAVES/XRSTORS instructions */
+#define X86_FEATURE_XFD			(10*32+ 4) /* "" eXtended Feature Disabling */
 
 /*
  * Extended auxiliary flags: Linux defined - for features scattered in various
@@ -298,6 +299,7 @@
 /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */
 #define X86_FEATURE_AVX_VNNI		(12*32+ 4) /* AVX VNNI instructions */
 #define X86_FEATURE_AVX512_BF16		(12*32+ 5) /* AVX512 BFLOAT16 instructions */
+#define X86_FEATURE_AMX_TILE		(18*32+24) /* AMX tile Support */
 
 /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
 #define X86_FEATURE_CLZERO		(13*32+ 0) /* CLZERO instruction */
diff --git a/tools/arch/x86/include/uapi/asm/kvm.h b/tools/arch/x86/include/uapi/asm/kvm.h
index 2ef1f6513c68..5a776a08f78c 100644
--- a/tools/arch/x86/include/uapi/asm/kvm.h
+++ b/tools/arch/x86/include/uapi/asm/kvm.h
@@ -504,4 +504,8 @@ struct kvm_pmu_event_filter {
 #define KVM_PMU_EVENT_ALLOW 0
 #define KVM_PMU_EVENT_DENY 1
 
+/* for KVM_{GET,SET,HAS}_DEVICE_ATTR */
+#define KVM_VCPU_TSC_CTRL 0 /* control group for the timestamp counter (TSC) */
+#define   KVM_VCPU_TSC_OFFSET 0 /* attribute for the TSC offset */
+
 #endif /* _ASM_X86_KVM_H */
diff --git a/tools/build/feature/test-all.c b/tools/build/feature/test-all.c
index 920439527291..0b243ce842be 100644
--- a/tools/build/feature/test-all.c
+++ b/tools/build/feature/test-all.c
@@ -200,7 +200,6 @@ int main(int argc, char *argv[])
 	main_test_timerfd();
 	main_test_stackprotector_all();
 	main_test_libdw_dwarf_unwind();
-	main_test_sync_compare_and_swap(argc, argv);
 	main_test_zlib();
 	main_test_pthread_attr_setaffinity_np();
 	main_test_pthread_barrier();
diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h
index a067410ebea5..1daa45268de2 100644
--- a/tools/include/uapi/linux/kvm.h
+++ b/tools/include/uapi/linux/kvm.h
@@ -269,6 +269,7 @@ struct kvm_xen_exit {
 #define KVM_EXIT_AP_RESET_HOLD    32
 #define KVM_EXIT_X86_BUS_LOCK     33
 #define KVM_EXIT_XEN              34
+#define KVM_EXIT_RISCV_SBI        35
 
 /* For KVM_EXIT_INTERNAL_ERROR */
 /* Emulate instruction failed. */
@@ -397,13 +398,23 @@ struct kvm_run {
 		 * "ndata" is correct, that new fields are enumerated in "flags",
 		 * and that each flag enumerates fields that are 64-bit aligned
 		 * and sized (so that ndata+internal.data[] is valid/accurate).
+		 *
+		 * Space beyond the defined fields may be used to store arbitrary
+		 * debug information relating to the emulation failure. It is
+		 * accounted for in "ndata" but the format is unspecified and is
+		 * not represented in "flags". Any such information is *not* ABI!
 		 */
 		struct {
 			__u32 suberror;
 			__u32 ndata;
 			__u64 flags;
-			__u8  insn_size;
-			__u8  insn_bytes[15];
+			union {
+				struct {
+					__u8  insn_size;
+					__u8  insn_bytes[15];
+				};
+			};
+			/* Arbitrary debug data may follow. */
 		} emulation_failure;
 		/* KVM_EXIT_OSI */
 		struct {
@@ -469,6 +480,13 @@ struct kvm_run {
 		} msr;
 		/* KVM_EXIT_XEN */
 		struct kvm_xen_exit xen;
+		/* KVM_EXIT_RISCV_SBI */
+		struct {
+			unsigned long extension_id;
+			unsigned long function_id;
+			unsigned long args[6];
+			unsigned long ret[2];
+		} riscv_sbi;
 		/* Fix the size of the union. */
 		char padding[256];
 	};
@@ -1112,6 +1130,7 @@ struct kvm_ppc_resize_hpt {
 #define KVM_CAP_BINARY_STATS_FD 203
 #define KVM_CAP_EXIT_ON_EMULATION_FAILURE 204
 #define KVM_CAP_ARM_MTE 205
+#define KVM_CAP_VM_MOVE_ENC_CONTEXT_FROM 206
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
@@ -1223,11 +1242,16 @@ struct kvm_irqfd {
 
 /* Do not use 1, KVM_CHECK_EXTENSION returned it before we had flags.  */
 #define KVM_CLOCK_TSC_STABLE		2
+#define KVM_CLOCK_REALTIME		(1 << 2)
+#define KVM_CLOCK_HOST_TSC		(1 << 3)
 
 struct kvm_clock_data {
 	__u64 clock;
 	__u32 flags;
-	__u32 pad[9];
+	__u32 pad0;
+	__u64 realtime;
+	__u64 host_tsc;
+	__u32 pad[4];
 };
 
 /* For KVM_CAP_SW_TLB */
diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
index 07e65a061fd3..afd144725a0b 100644
--- a/tools/perf/Makefile.config
+++ b/tools/perf/Makefile.config
@@ -1010,6 +1010,9 @@ ifndef NO_AUXTRACE
   ifndef NO_AUXTRACE
     $(call detected,CONFIG_AUXTRACE)
     CFLAGS += -DHAVE_AUXTRACE_SUPPORT
+    ifeq ($(feature-reallocarray), 0)
+      CFLAGS += -DCOMPAT_NEED_REALLOCARRAY
+    endif
   endif
 endif
 
diff --git a/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl b/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
index 1ca7bc337932..e2c481fcede6 100644
--- a/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
+++ b/tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl
@@ -363,3 +363,4 @@
 446	n64	landlock_restrict_self		sys_landlock_restrict_self
 # 447 reserved for memfd_secret
 448	n64	process_mrelease		sys_process_mrelease
+449	n64	futex_waitv			sys_futex_waitv
diff --git a/tools/perf/bench/sched-messaging.c b/tools/perf/bench/sched-messaging.c
index 488f6e6ba1a5..fa0ff4ce2b74 100644
--- a/tools/perf/bench/sched-messaging.c
+++ b/tools/perf/bench/sched-messaging.c
@@ -223,6 +223,8 @@ static unsigned int group(pthread_t *pth,
 		snd_ctx->out_fds[i] = fds[1];
 		if (!thread_mode)
 			close(fds[0]);
+
+		free(ctx);
 	}
 
 	/* Now we have all the fds, fork the senders */
@@ -239,6 +241,8 @@ static unsigned int group(pthread_t *pth,
 		for (i = 0; i < num_fds; i++)
 			close(snd_ctx->out_fds[i]);
 
+	free(snd_ctx);
+
 	/* Return number of children to reap */
 	return num_fds * 2;
 }
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 8167ebfe776a..8ae400429870 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -619,14 +619,17 @@ static int report__browse_hists(struct report *rep)
 	int ret;
 	struct perf_session *session = rep->session;
 	struct evlist *evlist = session->evlist;
-	const char *help = perf_tip(system_path(TIPDIR));
+	char *help = NULL, *path = NULL;
 
-	if (help == NULL) {
+	path = system_path(TIPDIR);
+	if (perf_tip(&help, path) || help == NULL) {
 		/* fallback for people who don't install perf ;-) */
-		help = perf_tip(DOCDIR);
-		if (help == NULL)
-			help = "Cannot load tips.txt file, please install perf!";
+		free(path);
+		path = system_path(DOCDIR);
+		if (perf_tip(&help, path) || help == NULL)
+			help = strdup("Cannot load tips.txt file, please install perf!");
 	}
+	free(path);
 
 	switch (use_browser) {
 	case 1:
@@ -651,7 +654,7 @@ static int report__browse_hists(struct report *rep)
 		ret = evlist__tty_browse_hists(evlist, rep, help);
 		break;
 	}
-
+	free(help);
 	return ret;
 }
 
diff --git a/tools/perf/tests/event_update.c b/tools/perf/tests/event_update.c
index fbb68deba59f..d01532d40acb 100644
--- a/tools/perf/tests/event_update.c
+++ b/tools/perf/tests/event_update.c
@@ -88,7 +88,6 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtes
 	struct evsel *evsel;
 	struct event_name tmp;
 	struct evlist *evlist = evlist__new_default();
-	char *unit = strdup("KRAVA");
 
 	TEST_ASSERT_VAL("failed to get evlist", evlist);
 
@@ -99,7 +98,8 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtes
 
 	perf_evlist__id_add(&evlist->core, &evsel->core, 0, 0, 123);
 
-	evsel->unit = unit;
+	free((char *)evsel->unit);
+	evsel->unit = strdup("KRAVA");
 
 	TEST_ASSERT_VAL("failed to synthesize attr update unit",
 			!perf_event__synthesize_event_update_unit(NULL, evsel, process_event_unit));
@@ -119,7 +119,6 @@ static int test__event_update(struct test_suite *test __maybe_unused, int subtes
 	TEST_ASSERT_VAL("failed to synthesize attr update cpus",
 			!perf_event__synthesize_event_update_cpus(&tmp.tool, evsel, process_event_cpus));
 
-	free(unit);
 	evlist__delete(evlist);
 	return 0;
 }
diff --git a/tools/perf/tests/sample-parsing.c b/tools/perf/tests/sample-parsing.c
index b669d22f2b13..07f2411b0ad4 100644
--- a/tools/perf/tests/sample-parsing.c
+++ b/tools/perf/tests/sample-parsing.c
@@ -36,7 +36,7 @@
  * These are based on the input value (213) specified
  * in branch_stack variable.
  */
-#define BS_EXPECTED_BE	0xa00d000000000000
+#define BS_EXPECTED_BE	0xa000d00000000000
 #define BS_EXPECTED_LE	0xd5000000
 #define FLAG(s)	s->branch_stack->entries[i].flags
 
diff --git a/tools/perf/tests/wp.c b/tools/perf/tests/wp.c
index 820d942b30c3..9d4c45184e71 100644
--- a/tools/perf/tests/wp.c
+++ b/tools/perf/tests/wp.c
@@ -21,6 +21,7 @@ do {                                            \
 volatile u64 data1;
 volatile u8 data2[3];
 
+#ifndef __s390x__
 static int wp_read(int fd, long long *count, int size)
 {
 	int ret = read(fd, count, size);
@@ -48,7 +49,6 @@ static void get__perf_event_attr(struct perf_event_attr *attr, int wp_type,
 	attr->exclude_hv     = 1;
 }
 
-#ifndef __s390x__
 static int __event(int wp_type, void *wp_addr, unsigned long wp_len)
 {
 	int fd;
diff --git a/tools/perf/ui/hist.c b/tools/perf/ui/hist.c
index c1f24d004852..5075ecead5f3 100644
--- a/tools/perf/ui/hist.c
+++ b/tools/perf/ui/hist.c
@@ -535,6 +535,18 @@ struct perf_hpp_list perf_hpp_list = {
 #undef __HPP_SORT_ACC_FN
 #undef __HPP_SORT_RAW_FN
 
+static void fmt_free(struct perf_hpp_fmt *fmt)
+{
+	/*
+	 * At this point fmt should be completely
+	 * unhooked, if not it's a bug.
+	 */
+	BUG_ON(!list_empty(&fmt->list));
+	BUG_ON(!list_empty(&fmt->sort_list));
+
+	if (fmt->free)
+		fmt->free(fmt);
+}
 
 void perf_hpp__init(void)
 {
@@ -598,9 +610,10 @@ void perf_hpp_list__prepend_sort_field(struct perf_hpp_list *list,
 	list_add(&format->sort_list, &list->sorts);
 }
 
-void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
+static void perf_hpp__column_unregister(struct perf_hpp_fmt *format)
 {
 	list_del_init(&format->list);
+	fmt_free(format);
 }
 
 void perf_hpp__cancel_cumulate(void)
@@ -672,19 +685,6 @@ next:
 }
 
 
-static void fmt_free(struct perf_hpp_fmt *fmt)
-{
-	/*
-	 * At this point fmt should be completely
-	 * unhooked, if not it's a bug.
-	 */
-	BUG_ON(!list_empty(&fmt->list));
-	BUG_ON(!list_empty(&fmt->sort_list));
-
-	if (fmt->free)
-		fmt->free(fmt);
-}
-
 void perf_hpp__reset_output_field(struct perf_hpp_list *list)
 {
 	struct perf_hpp_fmt *fmt, *tmp;
diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c
index 4748bcfe61de..fccac06b573a 100644
--- a/tools/perf/util/arm-spe.c
+++ b/tools/perf/util/arm-spe.c
@@ -51,6 +51,7 @@ struct arm_spe {
 	u8				timeless_decoding;
 	u8				data_queued;
 
+	u64				sample_type;
 	u8				sample_flc;
 	u8				sample_llc;
 	u8				sample_tlb;
@@ -287,6 +288,12 @@ static void arm_spe_prep_sample(struct arm_spe *spe,
 	event->sample.header.size = sizeof(struct perf_event_header);
 }
 
+static int arm_spe__inject_event(union perf_event *event, struct perf_sample *sample, u64 type)
+{
+	event->header.size = perf_event__sample_event_size(sample, type, 0);
+	return perf_event__synthesize_sample(event, type, 0, sample);
+}
+
 static inline int
 arm_spe_deliver_synth_event(struct arm_spe *spe,
 			    struct arm_spe_queue *speq __maybe_unused,
@@ -295,6 +302,12 @@ arm_spe_deliver_synth_event(struct arm_spe *spe,
 {
 	int ret;
 
+	if (spe->synth_opts.inject) {
+		ret = arm_spe__inject_event(event, sample, spe->sample_type);
+		if (ret)
+			return ret;
+	}
+
 	ret = perf_session__deliver_synth_event(spe->session, event, sample);
 	if (ret)
 		pr_err("ARM SPE: failed to deliver event, error %d\n", ret);
@@ -986,6 +999,8 @@ arm_spe_synth_events(struct arm_spe *spe, struct perf_session *session)
 	else
 		attr.sample_type |= PERF_SAMPLE_TIME;
 
+	spe->sample_type = attr.sample_type;
+
 	attr.exclude_user = evsel->core.attr.exclude_user;
 	attr.exclude_kernel = evsel->core.attr.exclude_kernel;
 	attr.exclude_hv = evsel->core.attr.exclude_hv;
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index a59fb2ecb84e..ac0127be0459 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -241,7 +241,7 @@ void evsel__init(struct evsel *evsel,
 {
 	perf_evsel__init(&evsel->core, attr, idx);
 	evsel->tracking	   = !idx;
-	evsel->unit	   = "";
+	evsel->unit	   = strdup("");
 	evsel->scale	   = 1.0;
 	evsel->max_events  = ULONG_MAX;
 	evsel->evlist	   = NULL;
@@ -276,13 +276,8 @@ struct evsel *evsel__new_idx(struct perf_event_attr *attr, int idx)
 	}
 
 	if (evsel__is_clock(evsel)) {
-		/*
-		 * The evsel->unit points to static alias->unit
-		 * so it's ok to use static string in here.
-		 */
-		static const char *unit = "msec";
-
-		evsel->unit = unit;
+		free((char *)evsel->unit);
+		evsel->unit = strdup("msec");
 		evsel->scale = 1e-6;
 	}
 
@@ -420,7 +415,11 @@ struct evsel *evsel__clone(struct evsel *orig)
 
 	evsel->max_events = orig->max_events;
 	evsel->tool_event = orig->tool_event;
-	evsel->unit = orig->unit;
+	free((char *)evsel->unit);
+	evsel->unit = strdup(orig->unit);
+	if (evsel->unit == NULL)
+		goto out_err;
+
 	evsel->scale = orig->scale;
 	evsel->snapshot = orig->snapshot;
 	evsel->per_pkg = orig->per_pkg;
@@ -1441,6 +1440,7 @@ void evsel__exit(struct evsel *evsel)
 	zfree(&evsel->group_name);
 	zfree(&evsel->name);
 	zfree(&evsel->pmu_name);
+	zfree(&evsel->unit);
 	zfree(&evsel->metric_id);
 	evsel__zero_per_pkg(evsel);
 	hashmap__free(evsel->per_pkg_mask);
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index fda8d14c891f..79cce216727e 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -4257,9 +4257,11 @@ int perf_event__process_event_update(struct perf_tool *tool __maybe_unused,
 
 	switch (ev->type) {
 	case PERF_EVENT_UPDATE__UNIT:
+		free((char *)evsel->unit);
 		evsel->unit = strdup(ev->data);
 		break;
 	case PERF_EVENT_UPDATE__NAME:
+		free(evsel->name);
 		evsel->name = strdup(ev->data);
 		break;
 	case PERF_EVENT_UPDATE__SCALE:
@@ -4268,11 +4270,11 @@ int perf_event__process_event_update(struct perf_tool *tool __maybe_unused,
 		break;
 	case PERF_EVENT_UPDATE__CPUS:
 		ev_cpus = (struct perf_record_event_update_cpus *)ev->data;
-
 		map = cpu_map__new_data(&ev_cpus->cpus);
-		if (map)
+		if (map) {
+			perf_cpu_map__put(evsel->core.own_cpus);
 			evsel->core.own_cpus = map;
-		else
+		} else
 			pr_err("failed to get event_update cpus\n");
 	default:
 		break;
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index 65fe65ba03c2..b776465e04ef 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -289,15 +289,10 @@ static long hist_time(unsigned long htime)
 	return htime;
 }
 
-static void he_stat__add_period(struct he_stat *he_stat, u64 period,
-				u64 weight, u64 ins_lat, u64 p_stage_cyc)
+static void he_stat__add_period(struct he_stat *he_stat, u64 period)
 {
-
 	he_stat->period		+= period;
-	he_stat->weight		+= weight;
 	he_stat->nr_events	+= 1;
-	he_stat->ins_lat	+= ins_lat;
-	he_stat->p_stage_cyc	+= p_stage_cyc;
 }
 
 static void he_stat__add_stat(struct he_stat *dest, struct he_stat *src)
@@ -308,9 +303,6 @@ static void he_stat__add_stat(struct he_stat *dest, struct he_stat *src)
 	dest->period_guest_sys	+= src->period_guest_sys;
 	dest->period_guest_us	+= src->period_guest_us;
 	dest->nr_events		+= src->nr_events;
-	dest->weight		+= src->weight;
-	dest->ins_lat		+= src->ins_lat;
-	dest->p_stage_cyc		+= src->p_stage_cyc;
 }
 
 static void he_stat__decay(struct he_stat *he_stat)
@@ -598,9 +590,6 @@ static struct hist_entry *hists__findnew_entry(struct hists *hists,
 	struct hist_entry *he;
 	int64_t cmp;
 	u64 period = entry->stat.period;
-	u64 weight = entry->stat.weight;
-	u64 ins_lat = entry->stat.ins_lat;
-	u64 p_stage_cyc = entry->stat.p_stage_cyc;
 	bool leftmost = true;
 
 	p = &hists->entries_in->rb_root.rb_node;
@@ -619,11 +608,11 @@ static struct hist_entry *hists__findnew_entry(struct hists *hists,
 
 		if (!cmp) {
 			if (sample_self) {
-				he_stat__add_period(&he->stat, period, weight, ins_lat, p_stage_cyc);
+				he_stat__add_period(&he->stat, period);
 				hist_entry__add_callchain_period(he, period);
 			}
 			if (symbol_conf.cumulate_callchain)
-				he_stat__add_period(he->stat_acc, period, weight, ins_lat, p_stage_cyc);
+				he_stat__add_period(he->stat_acc, period);
 
 			/*
 			 * This mem info was allocated from sample__resolve_mem
@@ -733,9 +722,6 @@ __hists__add_entry(struct hists *hists,
 		.stat = {
 			.nr_events = 1,
 			.period	= sample->period,
-			.weight = sample->weight,
-			.ins_lat = sample->ins_lat,
-			.p_stage_cyc = sample->p_stage_cyc,
 		},
 		.parent = sym_parent,
 		.filtered = symbol__parent_filter(sym_parent) | al->filtered,
@@ -748,6 +734,9 @@ __hists__add_entry(struct hists *hists,
 		.raw_size = sample->raw_size,
 		.ops = ops,
 		.time = hist_time(sample->time),
+		.weight = sample->weight,
+		.ins_lat = sample->ins_lat,
+		.p_stage_cyc = sample->p_stage_cyc,
 	}, *he = hists__findnew_entry(hists, &entry, al, sample_self);
 
 	if (!hists->has_callchains && he && he->callchain_size != 0)
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index 5343b62476e6..621f35ae1efa 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -369,7 +369,6 @@ enum {
 };
 
 void perf_hpp__init(void);
-void perf_hpp__column_unregister(struct perf_hpp_fmt *format);
 void perf_hpp__cancel_cumulate(void);
 void perf_hpp__setup_output_field(struct perf_hpp_list *list);
 void perf_hpp__reset_output_field(struct perf_hpp_list *list);
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 5bfb6f892489..ba74fdf74af9 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -402,8 +402,10 @@ static int add_event_tool(struct list_head *list, int *idx,
 	if (!evsel)
 		return -ENOMEM;
 	evsel->tool_event = tool_event;
-	if (tool_event == PERF_TOOL_DURATION_TIME)
-		evsel->unit = "ns";
+	if (tool_event == PERF_TOOL_DURATION_TIME) {
+		free((char *)evsel->unit);
+		evsel->unit = strdup("ns");
+	}
 	return 0;
 }
 
@@ -1630,7 +1632,8 @@ int parse_events_add_pmu(struct parse_events_state *parse_state,
 	if (parse_state->fake_pmu)
 		return 0;
 
-	evsel->unit = info.unit;
+	free((char *)evsel->unit);
+	evsel->unit = strdup(info.unit);
 	evsel->scale = info.scale;
 	evsel->per_pkg = info.per_pkg;
 	evsel->snapshot = info.snapshot;
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index 568a88c001c6..a111065b484e 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -1325,88 +1325,68 @@ struct sort_entry sort_mispredict = {
 	.se_width_idx	= HISTC_MISPREDICT,
 };
 
-static u64 he_weight(struct hist_entry *he)
-{
-	return he->stat.nr_events ? he->stat.weight / he->stat.nr_events : 0;
-}
-
 static int64_t
-sort__local_weight_cmp(struct hist_entry *left, struct hist_entry *right)
+sort__weight_cmp(struct hist_entry *left, struct hist_entry *right)
 {
-	return he_weight(left) - he_weight(right);
+	return left->weight - right->weight;
 }
 
 static int hist_entry__local_weight_snprintf(struct hist_entry *he, char *bf,
 				    size_t size, unsigned int width)
 {
-	return repsep_snprintf(bf, size, "%-*llu", width, he_weight(he));
+	return repsep_snprintf(bf, size, "%-*llu", width, he->weight);
 }
 
 struct sort_entry sort_local_weight = {
 	.se_header	= "Local Weight",
-	.se_cmp		= sort__local_weight_cmp,
+	.se_cmp		= sort__weight_cmp,
 	.se_snprintf	= hist_entry__local_weight_snprintf,
 	.se_width_idx	= HISTC_LOCAL_WEIGHT,
 };
 
-static int64_t
-sort__global_weight_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-	return left->stat.weight - right->stat.weight;
-}
-
 static int hist_entry__global_weight_snprintf(struct hist_entry *he, char *bf,
 					      size_t size, unsigned int width)
 {
-	return repsep_snprintf(bf, size, "%-*llu", width, he->stat.weight);
+	return repsep_snprintf(bf, size, "%-*llu", width,
+			       he->weight * he->stat.nr_events);
 }
 
 struct sort_entry sort_global_weight = {
 	.se_header	= "Weight",
-	.se_cmp		= sort__global_weight_cmp,
+	.se_cmp		= sort__weight_cmp,
 	.se_snprintf	= hist_entry__global_weight_snprintf,
 	.se_width_idx	= HISTC_GLOBAL_WEIGHT,
 };
 
-static u64 he_ins_lat(struct hist_entry *he)
-{
-		return he->stat.nr_events ? he->stat.ins_lat / he->stat.nr_events : 0;
-}
-
 static int64_t
-sort__local_ins_lat_cmp(struct hist_entry *left, struct hist_entry *right)
+sort__ins_lat_cmp(struct hist_entry *left, struct hist_entry *right)
 {
-		return he_ins_lat(left) - he_ins_lat(right);
+	return left->ins_lat - right->ins_lat;
 }
 
 static int hist_entry__local_ins_lat_snprintf(struct hist_entry *he, char *bf,
 					      size_t size, unsigned int width)
 {
-		return repsep_snprintf(bf, size, "%-*u", width, he_ins_lat(he));
+	return repsep_snprintf(bf, size, "%-*u", width, he->ins_lat);
 }
 
 struct sort_entry sort_local_ins_lat = {
 	.se_header	= "Local INSTR Latency",
-	.se_cmp		= sort__local_ins_lat_cmp,
+	.se_cmp		= sort__ins_lat_cmp,
 	.se_snprintf	= hist_entry__local_ins_lat_snprintf,
 	.se_width_idx	= HISTC_LOCAL_INS_LAT,
 };
 
-static int64_t
-sort__global_ins_lat_cmp(struct hist_entry *left, struct hist_entry *right)
-{
-		return left->stat.ins_lat - right->stat.ins_lat;
-}
-
 static int hist_entry__global_ins_lat_snprintf(struct hist_entry *he, char *bf,
 					       size_t size, unsigned int width)
 {
-		return repsep_snprintf(bf, size, "%-*u", width, he->stat.ins_lat);
+	return repsep_snprintf(bf, size, "%-*u", width,
+			       he->ins_lat * he->stat.nr_events);
 }
 
 struct sort_entry sort_global_ins_lat = {
 	.se_header	= "INSTR Latency",
-	.se_cmp		= sort__global_ins_lat_cmp,
+	.se_cmp		= sort__ins_lat_cmp,
 	.se_snprintf	= hist_entry__global_ins_lat_snprintf,
 	.se_width_idx	= HISTC_GLOBAL_INS_LAT,
 };
@@ -1414,13 +1394,13 @@ struct sort_entry sort_global_ins_lat = {
 static int64_t
 sort__global_p_stage_cyc_cmp(struct hist_entry *left, struct hist_entry *right)
 {
-	return left->stat.p_stage_cyc - right->stat.p_stage_cyc;
+	return left->p_stage_cyc - right->p_stage_cyc;
 }
 
 static int hist_entry__p_stage_cyc_snprintf(struct hist_entry *he, char *bf,
 					size_t size, unsigned int width)
 {
-	return repsep_snprintf(bf, size, "%-*u", width, he->stat.p_stage_cyc);
+	return repsep_snprintf(bf, size, "%-*u", width, he->p_stage_cyc);
 }
 
 struct sort_entry sort_p_stage_cyc = {
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index b67c469aba79..7b7145501933 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -49,9 +49,6 @@ struct he_stat {
 	u64			period_us;
 	u64			period_guest_sys;
 	u64			period_guest_us;
-	u64			weight;
-	u64			ins_lat;
-	u64			p_stage_cyc;
 	u32			nr_events;
 };
 
@@ -109,6 +106,9 @@ struct hist_entry {
 	s32			socket;
 	s32			cpu;
 	u64			code_page_size;
+	u64			weight;
+	u64			ins_lat;
+	u64			p_stage_cyc;
 	u8			cpumode;
 	u8			depth;
 
diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c
index 37a9492edb3e..df3c4671be72 100644
--- a/tools/perf/util/util.c
+++ b/tools/perf/util/util.c
@@ -379,32 +379,32 @@ fetch_kernel_version(unsigned int *puint, char *str,
 	return 0;
 }
 
-const char *perf_tip(const char *dirpath)
+int perf_tip(char **strp, const char *dirpath)
 {
 	struct strlist *tips;
 	struct str_node *node;
-	char *tip = NULL;
 	struct strlist_config conf = {
 		.dirname = dirpath,
 		.file_only = true,
 	};
+	int ret = 0;
 
+	*strp = NULL;
 	tips = strlist__new("tips.txt", &conf);
 	if (tips == NULL)
-		return errno == ENOENT ? NULL :
-			"Tip: check path of tips.txt or get more memory! ;-p";
+		return -errno;
 
 	if (strlist__nr_entries(tips) == 0)
 		goto out;
 
 	node = strlist__entry(tips, random() % strlist__nr_entries(tips));
-	if (asprintf(&tip, "Tip: %s", node->s) < 0)
-		tip = (char *)"Tip: get more memory! ;-)";
+	if (asprintf(strp, "Tip: %s", node->s) < 0)
+		ret = -ENOMEM;
 
 out:
 	strlist__delete(tips);
 
-	return tip;
+	return ret;
 }
 
 char *perf_exe(char *buf, int len)
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index ad737052e597..9f0d36ba77f2 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -39,7 +39,7 @@ int fetch_kernel_version(unsigned int *puint,
 #define KVER_FMT	"%d.%d.%d"
 #define KVER_PARAM(x)	KVER_VERSION(x), KVER_PATCHLEVEL(x), KVER_SUBLEVEL(x)
 
-const char *perf_tip(const char *dirpath);
+int perf_tip(char **strp, const char *dirpath);
 
 #ifndef HAVE_SCHED_GETCPU_SUPPORT
 int sched_getcpu(void);