Merge tag 'perf-urgent-for-mingo-5.5-20191216' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

Pull perf/urgent fixes:

perf top:

 Arnaldo Carvalho de Melo:

 - Do not bail out when perf_env__read_cpuid() returns ENOSYS, which
   has been reported happening on aarch64.

perf metricgroup:

  Kajol Jain:

  - Fix printing event names of metric group with multiple events

vendor events:

x86:

  Ravi Bangoria:

  - Fix Kernel_Utilization metric.

s390:

  Ed Maste:

  - Fix counter long description for DTLB1_GPAGE_WRITES and L1D_RO_EXCL_WRITES.

perf header:

  Michael Petlan:

  - Fix false warning when there are no duplicate cache entries

libtraceevent:

  Sudip Mukherjee:

  - Allow custom libdir path

API headers:

  Arnaldo Carvalho de Melo:

  - Sync linux/kvm.h with the kernel sources.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

44 files changed, 3280 insertions, 128 deletions

diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h
index e6f17c8e2dba..f0a16b4adbbd 100644
--- a/tools/include/uapi/linux/kvm.h
+++ b/tools/include/uapi/linux/kvm.h
@@ -1348,6 +1348,7 @@ struct kvm_s390_ucas_mapping {
 #define KVM_PPC_GET_CPU_CHAR	  _IOR(KVMIO,  0xb1, struct kvm_ppc_cpu_char)
 /* Available with KVM_CAP_PMU_EVENT_FILTER */
 #define KVM_SET_PMU_EVENT_FILTER  _IOW(KVMIO,  0xb2, struct kvm_pmu_event_filter)
+#define KVM_PPC_SVM_OFF		  _IO(KVMIO,  0xb3)
 
 /* ioctl for vm fd */
 #define KVM_CREATE_DEVICE	  _IOWR(KVMIO,  0xe0, struct kvm_create_device)
diff --git a/tools/lib/bpf/.gitignore b/tools/lib/bpf/.gitignore
index 35bf013e368c..e97c2ebcf447 100644
--- a/tools/lib/bpf/.gitignore
+++ b/tools/lib/bpf/.gitignore
@@ -1,7 +1,6 @@
 libbpf_version.h
 libbpf.pc
 FEATURE-DUMP.libbpf
-test_libbpf
 libbpf.so.*
 TAGS
 tags
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index 99425d0be6ff..defae23a0169 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -147,12 +147,12 @@ TAGS_PROG := $(if $(shell which etags 2>/dev/null),etags,ctags)
 
 GLOBAL_SYM_COUNT = $(shell readelf -s --wide $(BPF_IN_SHARED) | \
 			   cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' | \
-			   awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$8}' | \
+			   awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}' | \
 			   sort -u | wc -l)
 VERSIONED_SYM_COUNT = $(shell readelf -s --wide $(OUTPUT)libbpf.so | \
 			      grep -Eo '[^ ]+@LIBBPF_' | cut -d@ -f1 | sort -u | wc -l)
 
-CMD_TARGETS = $(LIB_TARGET) $(PC_FILE) $(OUTPUT)test_libbpf
+CMD_TARGETS = $(LIB_TARGET) $(PC_FILE)
 
 all: fixdep
 	$(Q)$(MAKE) all_cmd
@@ -180,9 +180,9 @@ $(BPF_IN_SHARED): force elfdep bpfdep bpf_helper_defs.h
 $(BPF_IN_STATIC): force elfdep bpfdep bpf_helper_defs.h
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(STATIC_OBJDIR)
 
-bpf_helper_defs.h: $(srctree)/include/uapi/linux/bpf.h
+bpf_helper_defs.h: $(srctree)/tools/include/uapi/linux/bpf.h
 	$(Q)$(srctree)/scripts/bpf_helpers_doc.py --header 		\
-		--file $(srctree)/include/uapi/linux/bpf.h > bpf_helper_defs.h
+		--file $(srctree)/tools/include/uapi/linux/bpf.h > bpf_helper_defs.h
 
 $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION)
 
@@ -196,9 +196,6 @@ $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED)
 $(OUTPUT)libbpf.a: $(BPF_IN_STATIC)
 	$(QUIET_LINK)$(RM) $@; $(AR) rcs $@ $^
 
-$(OUTPUT)test_libbpf: test_libbpf.c $(OUTPUT)libbpf.a
-	$(QUIET_LINK)$(CC) $(CFLAGS) $(LDFLAGS) $(INCLUDES) $^ -lelf -o $@
-
 $(OUTPUT)libbpf.pc:
 	$(QUIET_GEN)sed -e "s|@PREFIX@|$(prefix)|" \
 		-e "s|@LIBDIR@|$(libdir_SQ)|" \
@@ -214,9 +211,9 @@ check_abi: $(OUTPUT)libbpf.so
 		     "versioned symbols in $^ ($(VERSIONED_SYM_COUNT))." \
 		     "Please make sure all LIBBPF_API symbols are"	 \
 		     "versioned in $(VERSION_SCRIPT)." >&2;		 \
-		readelf -s --wide $(OUTPUT)libbpf-in.o |		 \
+		readelf -s --wide $(BPF_IN_SHARED) |			 \
 		    cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' |	 \
-		    awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$8}'|   \
+		    awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}'|  \
 		    sort -u > $(OUTPUT)libbpf_global_syms.tmp;		 \
 		readelf -s --wide $(OUTPUT)libbpf.so |			 \
 		    grep -Eo '[^ ]+@LIBBPF_' | cut -d@ -f1 |		 \
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index b20f82e58989..3f09772192f1 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -171,10 +171,8 @@ struct bpf_program {
 			RELO_DATA,
 		} type;
 		int insn_idx;
-		union {
-			int map_idx;
-			int text_off;
-		};
+		int map_idx;
+		int sym_off;
 	} *reloc_desc;
 	int nr_reloc;
 	int log_level;
@@ -1819,12 +1817,12 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		if (sym->st_value % 8) {
-			pr_warn("bad call relo offset: %lu\n", sym->st_value);
+			pr_warn("bad call relo offset: %llu\n", (__u64)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_CALL;
 		reloc_desc->insn_idx = insn_idx;
-		reloc_desc->text_off = sym->st_value / 8;
+		reloc_desc->sym_off = sym->st_value;
 		obj->has_pseudo_calls = true;
 		return 0;
 	}
@@ -1868,6 +1866,7 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 		reloc_desc->type = RELO_LD64;
 		reloc_desc->insn_idx = insn_idx;
 		reloc_desc->map_idx = map_idx;
+		reloc_desc->sym_off = 0; /* sym->st_value determines map_idx */
 		return 0;
 	}
 
@@ -1899,6 +1898,7 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 	reloc_desc->type = RELO_DATA;
 	reloc_desc->insn_idx = insn_idx;
 	reloc_desc->map_idx = map_idx;
+	reloc_desc->sym_off = sym->st_value;
 	return 0;
 }
 
@@ -3563,8 +3563,8 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 		return -LIBBPF_ERRNO__RELOC;
 
 	if (prog->idx == obj->efile.text_shndx) {
-		pr_warn("relo in .text insn %d into off %d\n",
-			relo->insn_idx, relo->text_off);
+		pr_warn("relo in .text insn %d into off %d (insn #%d)\n",
+			relo->insn_idx, relo->sym_off, relo->sym_off / 8);
 		return -LIBBPF_ERRNO__RELOC;
 	}
 
@@ -3599,7 +3599,7 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 			 prog->section_name);
 	}
 	insn = &prog->insns[relo->insn_idx];
-	insn->imm += relo->text_off + prog->main_prog_cnt - relo->insn_idx;
+	insn->imm += relo->sym_off / 8 + prog->main_prog_cnt - relo->insn_idx;
 	return 0;
 }
 
@@ -3622,31 +3622,26 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
 		return 0;
 
 	for (i = 0; i < prog->nr_reloc; i++) {
-		if (prog->reloc_desc[i].type == RELO_LD64 ||
-		    prog->reloc_desc[i].type == RELO_DATA) {
-			bool relo_data = prog->reloc_desc[i].type == RELO_DATA;
-			struct bpf_insn *insns = prog->insns;
-			int insn_idx, map_idx;
+		struct reloc_desc *relo = &prog->reloc_desc[i];
 
-			insn_idx = prog->reloc_desc[i].insn_idx;
-			map_idx = prog->reloc_desc[i].map_idx;
+		if (relo->type == RELO_LD64 || relo->type == RELO_DATA) {
+			struct bpf_insn *insn = &prog->insns[relo->insn_idx];
 
-			if (insn_idx + 1 >= (int)prog->insns_cnt) {
+			if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
 				pr_warn("relocation out of range: '%s'\n",
 					prog->section_name);
 				return -LIBBPF_ERRNO__RELOC;
 			}
 
-			if (!relo_data) {
-				insns[insn_idx].src_reg = BPF_PSEUDO_MAP_FD;
+			if (relo->type != RELO_DATA) {
+				insn[0].src_reg = BPF_PSEUDO_MAP_FD;
 			} else {
-				insns[insn_idx].src_reg = BPF_PSEUDO_MAP_VALUE;
-				insns[insn_idx + 1].imm = insns[insn_idx].imm;
+				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+				insn[1].imm = insn[0].imm + relo->sym_off;
 			}
-			insns[insn_idx].imm = obj->maps[map_idx].fd;
-		} else if (prog->reloc_desc[i].type == RELO_CALL) {
-			err = bpf_program__reloc_text(prog, obj,
-						      &prog->reloc_desc[i]);
+			insn[0].imm = obj->maps[relo->map_idx].fd;
+		} else if (relo->type == RELO_CALL) {
+			err = bpf_program__reloc_text(prog, obj, relo);
 			if (err)
 				return err;
 		}
diff --git a/tools/lib/traceevent/Makefile b/tools/lib/traceevent/Makefile
index c5a03356a999..c874c017c636 100644
--- a/tools/lib/traceevent/Makefile
+++ b/tools/lib/traceevent/Makefile
@@ -39,11 +39,12 @@ DESTDIR_SQ = '$(subst ','\'',$(DESTDIR))'
 
 LP64 := $(shell echo __LP64__ | ${CC} ${CFLAGS} -E -x c - | tail -n 1)
 ifeq ($(LP64), 1)
-  libdir_relative = lib64
+  libdir_relative_temp = lib64
 else
-  libdir_relative = lib
+  libdir_relative_temp = lib
 endif
 
+libdir_relative ?= $(libdir_relative_temp)
 prefix ?= /usr/local
 libdir = $(prefix)/$(libdir_relative)
 man_dir = $(prefix)/share/man
diff --git a/tools/lib/traceevent/plugins/Makefile b/tools/lib/traceevent/plugins/Makefile
index f440989fa55e..349bb81482ab 100644
--- a/tools/lib/traceevent/plugins/Makefile
+++ b/tools/lib/traceevent/plugins/Makefile
@@ -32,11 +32,12 @@ DESTDIR_SQ = '$(subst ','\'',$(DESTDIR))'
 
 LP64 := $(shell echo __LP64__ | ${CC} ${CFLAGS} -E -x c - | tail -n 1)
 ifeq ($(LP64), 1)
-  libdir_relative = lib64
+  libdir_relative_tmp = lib64
 else
-  libdir_relative = lib
+  libdir_relative_tmp = lib
 endif
 
+libdir_relative ?= $(libdir_relative_tmp)
 prefix ?= /usr/local
 libdir = $(prefix)/$(libdir_relative)
 
diff --git a/tools/pci/pcitest.c b/tools/pci/pcitest.c
index cb1e51fcc84e..32b7c6f9043d 100644
--- a/tools/pci/pcitest.c
+++ b/tools/pci/pcitest.c
@@ -129,6 +129,7 @@ static int run_test(struct pci_test *test)
 	}
 
 	fflush(stdout);
+	close(fd);
 	return (ret < 0) ? ret : 1 - ret; /* return 0 if test succeeded */
 }
 
diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST
index 70f1ff4e2eb4..4934edb5adfd 100644
--- a/tools/perf/MANIFEST
+++ b/tools/perf/MANIFEST
@@ -19,3 +19,4 @@ tools/lib/bitmap.c
 tools/lib/str_error_r.c
 tools/lib/vsprintf.c
 tools/lib/zalloc.c
+scripts/bpf_helpers_doc.py
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index dc80044bc46f..795e353de095 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1568,9 +1568,13 @@ int cmd_top(int argc, const char **argv)
 	 */
 	status = perf_env__read_cpuid(&perf_env);
 	if (status) {
-		pr_err("Couldn't read the cpuid for this machine: %s\n",
-		       str_error_r(errno, errbuf, sizeof(errbuf)));
-		goto out_delete_evlist;
+		/*
+		 * Some arches do not provide a get_cpuid(), so just use pr_debug, otherwise
+		 * warn the user explicitely.
+		 */
+		eprintf(status == ENOSYS ? 1 : 0, verbose,
+			"Couldn't read the cpuid for this machine: %s\n",
+			str_error_r(errno, errbuf, sizeof(errbuf)));
 	}
 	top.evlist->env = &perf_env;
 
diff --git a/tools/perf/pmu-events/arch/s390/cf_z13/extended.json b/tools/perf/pmu-events/arch/s390/cf_z13/extended.json
index 436ce33f1182..5da8296b667e 100644
--- a/tools/perf/pmu-events/arch/s390/cf_z13/extended.json
+++ b/tools/perf/pmu-events/arch/s390/cf_z13/extended.json
@@ -32,7 +32,7 @@
 		"EventCode": "132",
 		"EventName": "DTLB1_GPAGE_WRITES",
 		"BriefDescription": "DTLB1 Two-Gigabyte Page Writes",
-		"PublicDescription": "Counter:132	Name:DTLB1_GPAGE_WRITES A translation entry has been written to the Level-1 Data Translation Lookaside Buffer for a two-gigabyte page."
+		"PublicDescription": "A translation entry has been written to the Level-1 Data Translation Lookaside Buffer for a two-gigabyte page."
 	},
 	{
 		"Unit": "CPU-M-CF",
diff --git a/tools/perf/pmu-events/arch/s390/cf_z14/extended.json b/tools/perf/pmu-events/arch/s390/cf_z14/extended.json
index 68618152ea2c..89e070727e1b 100644
--- a/tools/perf/pmu-events/arch/s390/cf_z14/extended.json
+++ b/tools/perf/pmu-events/arch/s390/cf_z14/extended.json
@@ -4,7 +4,7 @@
 		"EventCode": "128",
 		"EventName": "L1D_RO_EXCL_WRITES",
 		"BriefDescription": "L1D Read-only Exclusive Writes",
-		"PublicDescription": "L1D_RO_EXCL_WRITES A directory write to the Level-1 Data cache where the line was originally in a Read-Only state in the cache but has been updated to be in the Exclusive state that allows stores to the cache line"
+		"PublicDescription": "A directory write to the Level-1 Data cache where the line was originally in a Read-Only state in the cache but has been updated to be in the Exclusive state that allows stores to the cache line"
 	},
 	{
 		"Unit": "CPU-M-CF",
diff --git a/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json b/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json
index bc7151d639d7..45a34ce4fe89 100644
--- a/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json
@@ -297,7 +297,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json b/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json
index 49c5f123d811..961fe4395758 100644
--- a/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json
@@ -115,7 +115,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json b/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json
index 113d19e92678..746734ce09be 100644
--- a/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json
@@ -297,7 +297,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json b/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json
index 2ba32af9bc36..f94653229dd4 100644
--- a/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json
@@ -315,7 +315,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json b/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json
index c80f16fde6d0..5402cd3120f9 100644
--- a/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json
@@ -267,7 +267,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json b/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json
index e501729c3dd1..832f3cb40b34 100644
--- a/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json
@@ -267,7 +267,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json b/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json
index e2446966b651..d69b2a8fc0bc 100644
--- a/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json
@@ -285,7 +285,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json b/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json
index 9294769dec64..5f465fd81315 100644
--- a/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json
@@ -285,7 +285,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
index 603ff9c2e9a1..3e909b306003 100644
--- a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json
@@ -171,7 +171,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
index c6b485b3a2cb..50c053235752 100644
--- a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json
@@ -171,7 +171,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json b/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json
index 0ca539bb60f6..e7feb60f9fa9 100644
--- a/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json
@@ -303,7 +303,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
index 047d7e11aa6f..21d7a0c2c2e8 100644
--- a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
+++ b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json
@@ -315,7 +315,7 @@
     },
     {
         "BriefDescription": "Fraction of cycles spent in Kernel mode",
-        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:u / CPU_CLK_UNHALTED.REF_TSC",
+        "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC:k / CPU_CLK_UNHALTED.REF_TSC",
         "MetricGroup": "Summary",
         "MetricName": "Kernel_Utilization"
     },
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
index becc2d109423..93ad27830e2b 100644
--- a/tools/perf/util/header.c
+++ b/tools/perf/util/header.c
@@ -850,7 +850,7 @@ int __weak strcmp_cpuid_str(const char *mapcpuid, const char *cpuid)
  */
 int __weak get_cpuid(char *buffer __maybe_unused, size_t sz __maybe_unused)
 {
-	return -1;
+	return ENOSYS; /* Not implemented */
 }
 
 static int write_cpuid(struct feat_fd *ff,
@@ -1089,21 +1089,18 @@ static void cpu_cache_level__fprintf(FILE *out, struct cpu_cache_level *c)
 	fprintf(out, "L%d %-15s %8s [%s]\n", c->level, c->type, c->size, c->map);
 }
 
-static int build_caches(struct cpu_cache_level caches[], u32 size, u32 *cntp)
+#define MAX_CACHE_LVL 4
+
+static int build_caches(struct cpu_cache_level caches[], u32 *cntp)
 {
 	u32 i, cnt = 0;
-	long ncpus;
 	u32 nr, cpu;
 	u16 level;
 
-	ncpus = sysconf(_SC_NPROCESSORS_CONF);
-	if (ncpus < 0)
-		return -1;
-
-	nr = (u32)(ncpus & UINT_MAX);
+	nr = cpu__max_cpu();
 
 	for (cpu = 0; cpu < nr; cpu++) {
-		for (level = 0; level < 10; level++) {
+		for (level = 0; level < MAX_CACHE_LVL; level++) {
 			struct cpu_cache_level c;
 			int err;
 
@@ -1123,18 +1120,12 @@ static int build_caches(struct cpu_cache_level caches[], u32 size, u32 *cntp)
 				caches[cnt++] = c;
 			else
 				cpu_cache_level__free(&c);
-
-			if (WARN_ONCE(cnt == size, "way too many cpu caches.."))
-				goto out;
 		}
 	}
- out:
 	*cntp = cnt;
 	return 0;
 }
 
-#define MAX_CACHE_LVL 4
-
 static int write_cache(struct feat_fd *ff,