perf/core improvements and fixes

Hi Ingo,

Please consider pulling,

- Arnaldo

The following changes since commit a95a49fa0cc5eec730d8703b1544fa7ea6a11dec:

Merge tag 'perf-core-for-mingo-2' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2015-11-26 09:13:50 +0100)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo

for you to fetch changes up to 43798bf37215fe242e592fd4605d804e2da0781b:

bpf tools: Add helper function for updating bpf maps elements (2015-11-26 17:21:24 -0300)

----------------------------------------------------------------
perf/core improvements and fixes:

User visible:

- Add 'vmlinux.debug' to the vmlinux seach path (Ekaterina Tumanova)

- Do not show sample_(type|period) in the perf_event_attr dump when using -v with 'perf stat' (Jiri Olsa)

- Display the WEIGHT sample bit, when set, in 'perf evlist -v' (Jiri Olsa)

- Honour --hide-unresolved in 'report', will honour it as well in 'top' when --hide-unresolved gets supported in that tool (Namhyung Kim)

- Fix freeze wit h--call-graph 'flat/folded' due to not properly reinitializing the callchain rb_tree (Namhyumg Kim)

- Set dso->long_name when a module name is passed as a parameter to tools like 'perf probe' but the 'struct dso' associated to that module still doesn't have the full path for the module, just the '[name]' one obtained from /proc/modules (Wang Nan)

- Fix anon_hugepage mmaps detection using scanf on /proc/PID/smaps (Yannick Brosseau)

Infrastructure:

- Add helper function for updating bpf maps elements (He Kuang)

- Fix traceevents plugins build race (Jiri Olsa)

- Add the $OUTPUT path prefix with 'fixdep' (Jiri Olsa)

----------------------------------------------------------------
Ekaterina Tumanova (2):
perf symbols: Refactor vmlinux_path__init() to ease path additions perf symbols: Add the path to vmlinux.debug

He Kuang (1):
bpf tools: Add helper function for updating bpf maps elements

Jiri Olsa (6):
perf script: Remove default_scripting_ops perf build: Fix traceevent plugins build race perf script: Pass perf_script into process_event tools build: Use fixdep with OUTPUT path prefix perf stat: Clear sample_(type|period) for counting perf evlist: Display WEIGHT sample type bit

Namhyung Kim (2):
perf callchain: Honor hide_unresolved perf top: Fix freeze on --call-graph flat/folded

Wang Nan (1):

      perf machine: Adjust dso->long_name for offline module

Yannick Brosseau (1):
perf tools: Correctly identify anon_hugepage when generating map (v2)

 tools/build/Makefile         |  2 +-
 tools/build/Makefile.include |  2 +-
 tools/lib/bpf/bpf.c          | 14 +++++++++
 tools/lib/bpf/bpf.h          |  2 ++
 tools/perf/Makefile.perf     |  2 +-
 tools/perf/builtin-report.c  |  7 ++---
 tools/perf/builtin-script.c  | 67 +++++++++++++-------------------------------
 tools/perf/builtin-stat.c    |  7 +++++
 tools/perf/util/callchain.c  |  1 +
 tools/perf/util/evsel.c      |  1 +
 tools/perf/util/machine.c    | 32 ++++++++++++++++++++-
 tools/perf/util/map.c        |  4 +--
 tools/perf/util/symbol.c     | 65 ++++++++++++++++++++++--------------------
 tools/perf/util/symbol.h     |  3 +-
 14 files changed, 120 insertions(+), 89 deletions(-)

From: Namhyung Kim <[email protected]>

If user requested to hide unresolved entries, skip unresolved callchains as well as hist entries.

---
 tools/perf/builtin-report.c | 7 +++----
 tools/perf/util/machine.c   | 5 +++++
 tools/perf/util/symbol.h    | 3 ++-
 3 files changed, 10 insertions(+), 5 deletions(-)

index 14428342b47b..8a9c6908f54e 100644
@@ -45,7 +45,6 @@ struct report {
 	struct perf_tool	tool;
 	struct perf_session	*session;
 	bool			use_tui, use_gtk, use_stdio;
-	bool			hide_unresolved;
 	bool			dont_use_callchains;
 	bool			show_full_info;
 	bool			show_threads;

@@ -146,7 +145,7 @@ static int process_sample_event(struct perf_tool *tool,
 	struct hist_entry_iter iter = {
 		.evsel 			= evsel,
 		.sample 		= sample,
-		.hide_unresolved 	= rep->hide_unresolved,
+		.hide_unresolved 	= symbol_conf.hide_unresolved,
 		.add_entry_cb 		= hist_iter__report_callback,
 	};
 	int ret = 0;

@@ -157,7 +156,7 @@ static int process_sample_event(struct perf_tool *tool,
 		return -1;
 	}
 
-	if (rep->hide_unresolved && al.sym == NULL)
+	if (symbol_conf.hide_unresolved && al.sym == NULL)
 		goto out_put;
 
 	if (rep->cpu_list && !test_bit(sample->cpu, rep->cpu_bitmap))

@@ -740,7 +739,7 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 	OPT_STRING_NOEMPTY('t', "field-separator", &symbol_conf.field_sep, "separator",
 		   "separator for columns, no spaces will be added between "
 		   "columns '.' is reserved."),
-	OPT_BOOLEAN('U', "hide-unresolved", &report.hide_unresolved,
+	OPT_BOOLEAN('U', "hide-unresolved", &symbol_conf.hide_unresolved,
 		    "Only display entries resolved to a symbol"),
 	OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory",
 		    "Look for files with symbols relative to this directory"),

index 7f5071a4d9aa..f0019b72db48 100644
@@ -1618,6 +1618,8 @@ static int add_callchain_ip(struct thread *thread,
 		}
 	}
 
+	if (symbol_conf.hide_unresolved && al.sym == NULL)
+		return 0;
 	return callchain_cursor_append(&callchain_cursor, al.addr, al.map, al.sym);
 }
 

@@ -1872,6 +1874,9 @@ check_calls:
 static int unwind_entry(struct unwind_entry *entry, void *arg)
 {
 	struct callchain_cursor *cursor = arg;
+
+	if (symbol_conf.hide_unresolved && entry->sym == NULL)
+		return 0;
 	return callchain_cursor_append(cursor, entry->ip,
 				       entry->map, entry->sym);
 }

index dcd786e364f2..857f707ac12b 100644
@@ -108,7 +108,8 @@ struct symbol_conf {
 			show_hist_headers,
 			branch_callstack,
 			has_filter,
-			show_ref_callgraph;
+			show_ref_callgraph,
+			hide_unresolved;
 	const char	*vmlinux_name,
 			*kallsyms_name,
 			*source_prefix,

2.1.0

From: Namhyung Kim <[email protected]>

The callchain rbtree is rebuilt periodically, so it needs to reinitialize the root everytime. Otherwise it can be stuck in the rbtree insertion with stale pointers.

---
 tools/perf/util/callchain.c | 1 +
 1 file changed, 1 insertion(+)

index fc3b1e0d09ee..564377d2bebf 100644
@@ -290,6 +290,7 @@ static void
 sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
 		u64 min_hit, struct callchain_param *param __maybe_unused)
 {
+	*rb_root = RB_ROOT;
 	__sort_chain_flat(rb_root, &root->node, min_hit);
 }
 

2.1.0

From: Jiri Olsa <[email protected]>

The default script handler (the one that displays samples on screen) is implemented scripting_ops instance with process_event callback.

This way we can't pass any script config into display function, because we don't want perl or python handlers to be depended on perf script internals.

Removing the default_scripting_ops and calling process event function directly. This way it's possible to pass perf_script struct and process configuration data in following commit.

---
 tools/perf/builtin-script.c | 43 +++++++------------------------------------
 1 file changed, 7 insertions(+), 36 deletions(-)

index 72b5deb4bd79..8e3f8048d2d0 100644
@@ -643,57 +643,24 @@ static void process_event(union perf_event *event, struct perf_sample *sample,
 	printf("\n");
 }
 
-static int default_start_script(const char *script __maybe_unused,
-				int argc __maybe_unused,
-				const char **argv __maybe_unused)
-{
-	return 0;
-}
-
-static int default_flush_script(void)
-{
-	return 0;
-}
-
-static int default_stop_script(void)
-{
-	return 0;
-}
-
-static int default_generate_script(struct pevent *pevent __maybe_unused,
-				   const char *outfile __maybe_unused)
-{
-	return 0;
-}
-
-static struct scripting_ops default_scripting_ops = {
-	.start_script		= default_start_script,
-	.flush_script		= default_flush_script,
-	.stop_script		= default_stop_script,
-	.process_event		= process_event,
-	.generate_script	= default_generate_script,
-};
-
 static struct scripting_ops	*scripting_ops;
 
 static void setup_scripting(void)
 {
 	setup_perl_scripting();
 	setup_python_scripting();
-
-	scripting_ops = &default_scripting_ops;
 }
 
 static int flush_scripting(void)
 {
-	return scripting_ops->flush_script();
+	return scripting_ops ? scripting_ops->flush_script() : 0;
 }
 
 static int cleanup_scripting(void)
 {
 	pr_debug("\nperf script stopped\n");
 
-	return scripting_ops->stop_script();
+	return scripting_ops ? scripting_ops->stop_script() : 0;
 }
 
 static int process_sample_event(struct perf_tool *tool __maybe_unused,

@@ -727,7 +694,11 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 	if (cpu_list && !test_bit(sample->cpu, cpu_bitmap))
 		goto out_put;
 
-	scripting_ops->process_event(event, sample, evsel, &al);
+	if (scripting_ops)
+		scripting_ops->process_event(event, sample, evsel, &al);
+	else
+		process_event(event, sample, evsel, &al);
+
 out_put:
 	addr_location__put(&al);
 	return 0;

2.1.0

From: Jiri Olsa <[email protected]>

Ingo reported following build failure:

  $ make clean install
  ...
    CC       plugin_kmem.o
  fixdep: error opening depfile: ./.plugin_hrtimer.o.d: No such file or directory
  /home/mingo/tip/tools/build/Makefile.build:77: recipe for target
  'plugin_hrtimer.o' failed
  make[3]: *** [plugin_hrtimer.o] Error 2
  Makefile:189: recipe for target 'plugin_hrtimer-in.o' failed
  make[2]: *** [plugin_hrtimer-in.o] Error 2
  Makefile.perf:414: recipe for target 'libtraceevent_plugins' failed
  make[1]: *** [libtraceevent_plugins] Error 2
  make[1]: *** Waiting for unfinished jobs....

Currently we have the install-traceevent-plugins target being dependent on $(LIBTRACEEVENT), which will actualy not build any plugin. So the install-traceevent-plugins target itself will try to build plugins, but..

Plugins built is also triggered by perf build itself via libtraceevent_plugins target.

This might cause a race having one make thread removing temp files from another and result in above error. Fixing this by having proper plugins build dependency before installing plugins.

---
 tools/perf/Makefile.perf | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 0d19d5447d6c..929a32ba15f5 100644
@@ -420,7 +420,7 @@ $(LIBTRACEEVENT)-clean:
 	$(call QUIET_CLEAN, libtraceevent)
 	$(Q)$(MAKE) -C $(TRACE_EVENT_DIR) O=$(OUTPUT) clean >/dev/null
 
-install-traceevent-plugins: $(LIBTRACEEVENT)
+install-traceevent-plugins: libtraceevent_plugins
 	$(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) install_plugins
 
 $(LIBAPI): fixdep FORCE

2.1.0

From: Wang Nan <[email protected]>

Something unexpected may happen if copy statically linked perf to a production environment:

  # ./perf probe -m ./mymodule.ko my_func
  [mymodule] with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events.
  # ./perf buildid-cache -a ./mymodule.ko
  # ./perf probe -m ./mymodule.ko my_func
  Added new event:
    probe:my_func        (on my_func in /home/wangnan/kmodule/mymodule.ko)

You can now use it in all perf tools, such as:

perf record -e probe:my_func -aR sleep 1

Where:

  # ldd ./perf
 	not a dynamic executable
  # strace -e open ./perf probe -m ./mymodule.ko my_func
  ...
  open("/home/wangnan/kmodule/mymodule.ko", O_RDONLY) = 3
  open("/home/wangnan/kmodule/../lib64/elfutils/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  ...
  open("/lib64/tls/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/lib64/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/usr/lib64/tls/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("/usr/lib64/libebl_x86_64.so", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
  open("[mymodule]", O_RDONLY)            = -1 ENOENT (No such file or directory)
  open("/home/wangnan/.debug/.build-id/32/6ab42550ef3d24944f53c817533728367effeb", O_RDONLY) = -1 ENOENT (No such file or directory)
  open("[mymodule]", O_RDONLY)            = -1 ENOENT (No such file or directory)

In the above example, probe fails before we put the module into buildid-cache. However, user would expect it success in both case because perf is able to find probe points actually.

The reason is because perf won't utilize module's full path if it failed to open debuginfo. In:

     convert_to_probe_trace_events ->
        find_probe_trace_events_from_map ->
            get_target_map ->
                kernel_get_module_map ->
                    machine__findnew_module_map ->
                        map_groups__find_by_name

map_groups__find_by_name() is able to find the map of that module, but this information is found from /proc/module before it knows the real path of the offline module. Therefore, the map->dso->long_name is set to something like '[mymodule]', which prevent dso__load() find the real path of the module file.

In another aspect, if dso__load() can get the offline module through buildid cache, it can read symble table from that ko. Even if debuginfo is not available, 'perf probe' can success if the '.symtab' can be found.

This patch improves machine__findnew_module_map(): when dso->long_name is leading with '[' (doesn't find path of module when parsing /proc/modules), fixes it by dso__set_long_name(), so following dso__load() is possible to find the symbol table.

This patch won't interfere with buildid matching. Here is the test result:

  # ./perf probe -m ./mymodule.ko my_func
  Added new event:
    probe:my_func        (on my_func in /home/wangnan/kmodule/mymodule.ko)

You can now use it in all perf tools, such as:

perf record -e probe:my_func -aR sleep 1

  # ./perf probe -d '*'
  Removed event: probe:my_func
  # mv ./mymodule.{ko,.bak}
  # mv ./moduleb.ko mymodule.ko
  # ./perf probe -m ./mymodule.ko my_func
  /home/wangnan/kmodule/mymodule.ko with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events.

  # ./perf probe -v -m ./mymodule.ko my_func
  probe-definition(0): my_func
  symbol:my_func file:(null) line:0 offset:0 return:0 lazy:(null)
  0 arguments
  Could not open debuginfo. Try to use symbols.
  symsrc__init: build id mismatch for /home/wangnan/kmodule/mymodule.ko.
  /home/wangnan/kmodule/mymodule.ko with build id 326ab42550ef3d24944f53c817533728367effeb not found, continuing without symbols
  Failed to find symbol my_func in /home/wangnan/kmodule/mymodule.ko
    Error: Failed to add events. Reason: No such file or directory (Code: -2)

---
 tools/perf/util/machine.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

index f0019b72db48..95a7f6087346 100644
@@ -561,6 +561,24 @@ int machine__process_switch_event(struct machine *machine __maybe_unused,
 	return 0;
 }
 
+static void dso__adjust_kmod_long_name(struct dso *dso, const char *filename)
+{
+	const char *dup_filename;
+
+	if (!filename || !dso || !dso->long_name)
+		return;
+	if (dso->long_name[0] != '[')
+		return;
+	if (!strchr(filename, '/'))
+		return;
+
+	dup_filename = strdup(filename);
+	if (!dup_filename)
+		return;
+
+	dso__set_long_name(dso, filename, true);
+}
+
 struct map *machine__findnew_module_map(struct machine *machine, u64 start,
 					const char *filename)
 {

@@ -573,8 +591,15 @@ struct map *machine__findnew_module_map(struct machine *machine, u64 start,
 
 	map = map_groups__find_by_name(&machine->kmaps, MAP__FUNCTION,
 				       m.name);
-	if (map)
+	if (map) {
+		/*
+		 * If the map's dso is an offline module, give dso__load()
+		 * a chance to find the file path of that module by fixing
+		 * long_name.
+		 */
+		dso__adjust_kmod_long_name(map->dso, filename);
 		goto out;
+	}
 
 	dso = machine__findnew_module_dso(machine, &m, filename);
 	if (dso == NULL)

2.1.0

From: Yannick Brosseau <[email protected]>

When parsing /proc/xxx/maps, the sscanf in perf_event__synthesize_mmap_events truncate the map name at the space in "/anon_hugepage (deleted)".

is_anon_memory() then only receives the string "/anon_hugepage" and does not detect it. We change is_anon_memory() to only compare the first part of the string, effectively ignoring if " (deleted)" is there.

---
 tools/perf/util/map.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index afc6b56cf749..93d9f1ce3baa 100644
@@ -26,8 +26,8 @@ const char *map_type__name[MAP__NR_TYPES] = {
 static inline int is_anon_memory(const char *filename)
 {
 	return !strcmp(filename, "//anon") ||
-	       !strcmp(filename, "/dev/zero (deleted)") ||
-	       !strcmp(filename, "/anon_hugepage (deleted)");
+	       !strncmp(filename, "/dev/zero", sizeof("/dev/zero") - 1) ||
+	       !strncmp(filename, "/anon_hugepage", sizeof("/anon_hugepage") - 1);
 }
 
 static inline int is_no_dso_memory(const char *filename)

2.1.0

From: Jiri Olsa <[email protected]>

Passing perf_script struct into process_event function, so we could process configuration data for event printing.

It will be used in following patch to get event name string width.

---
 tools/perf/builtin-script.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

index 8e3f8048d2d0..3c3f8d0e3064 100644
@@ -588,8 +588,17 @@ static void print_sample_flags(u32 flags)
 	printf("  %-4s ", str);
 }
 
-static void process_event(union perf_event *event, struct perf_sample *sample,
-			  struct perf_evsel *evsel, struct addr_location *al)
+struct perf_script {
+	struct perf_tool	tool;
+	struct perf_session	*session;
+	bool			show_task_events;
+	bool			show_mmap_events;
+	bool			show_switch_events;
+};
+
+static void process_event(struct perf_script *script __maybe_unused, union perf_event *event,
+			  struct perf_sample *sample, struct perf_evsel *evsel,
+			  struct addr_location *al)
 {
 	struct thread *thread = al->thread;
 	struct perf_event_attr *attr = &evsel->attr;

@@ -663,12 +672,13 @@ static int cleanup_scripting(void)
 	return scripting_ops ? scripting_ops->stop_script() : 0;
 }
 
-static int process_sample_event(struct perf_tool *tool __maybe_unused,
+static int process_sample_event(struct perf_tool *tool,
 				union perf_event *event,
 				struct perf_sample *sample,
 				struct perf_evsel *evsel,
 				struct machine *machine)
 {
+	struct perf_script *scr = container_of(tool, struct perf_script, tool);
 	struct addr_location al;
 
 	if (debug_mode) {

@@ -697,21 +707,13 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused,
 	if (scripting_ops)
 		scripting_ops->process_event(event, sample, evsel, &al);
 	else
-		process_event(event, sample, evsel, &al);
+		process_event(scr, event, sample, evsel, &al);
 
 out_put:
 	addr_location__put(&al);
 	return 0;
 }
 
-struct perf_script {
-	struct perf_tool	tool;
-	struct perf_session	*session;
-	bool			show_task_events;
-	bool			show_mmap_events;
-	bool			show_switch_events;
-};
-
 static int process_attr(struct perf_tool *tool, union perf_event *event,
 			struct perf_evlist **pevlist)
 {

2.1.0

From: Jiri Olsa <[email protected]>

Adding OUTPUT path prefix for fixdep target so we use it properly in out of tree builds.

If the fixdep already existed in the tree, the out of tree build would see it already exist and did not build the out of tree version, as reported by Arnaldo:

[acme@zoo linux]$ make O=/tmp/build/perf -C tools/perf
make: Entering directory '/home/git/linux/tools/perf'
BUILD: Doing 'make -j4' parallel build
make[2]: Nothing to be done for 'fixdep'.
make: Leaving directory '/home/git/linux/tools/perf'

---
 tools/build/Makefile         | 2 +-
 tools/build/Makefile.include | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

index a93036272d43..0d5a0e3a8fa9 100644
@@ -25,7 +25,7 @@ export Q srctree CC LD
 MAKEFLAGS := --no-print-directory
 build     := -f $(srctree)/tools/build/Makefile.build dir=. obj
 
-all: fixdep
+all: $(OUTPUT)fixdep
 
 clean:
 	$(call QUIET_CLEAN, fixdep)

index 6254760290c9..be630bed66d2 100644
@@ -4,7 +4,7 @@ ifdef CROSS_COMPILE
 fixdep:
 else
 fixdep:
-	$(Q)$(MAKE) -C $(srctree)/tools/build CFLAGS= LDFLAGS= fixdep
+	$(Q)$(MAKE) -C $(srctree)/tools/build CFLAGS= LDFLAGS= $(OUTPUT)fixdep
 endif
 
 .PHONY: fixdep

2.1.0

From: Ekaterina Tumanova <[email protected]>

Refactor vmlinux_path__init() to ease subsequent additions of new vmlinux locations.

---
 tools/perf/util/symbol.c | 64 +++++++++++++++++++++++++-----------------------
 1 file changed, 33 insertions(+), 31 deletions(-)

index cd08027a6d2c..e2ac6b6676e4 100644
@@ -1860,24 +1860,43 @@ static void vmlinux_path__exit(void)
 	zfree(&vmlinux_path);
 }
 
+static const char * const vmlinux_paths[] = {
+	"vmlinux",
+	"/boot/vmlinux"
+};
+
+static const char * const vmlinux_paths_upd[] = {
+	"/boot/vmlinux-%s",
+	"/usr/lib/debug/boot/vmlinux-%s",
+	"/lib/modules/%s/build/vmlinux",
+	"/usr/lib/debug/lib/modules/%s/vmlinux"
+};
+
+static int vmlinux_path__add(const char *new_entry)
+{
+	vmlinux_path[vmlinux_path__nr_entries] = strdup(new_entry);
+	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
+		return -1;
+	++vmlinux_path__nr_entries;
+
+	return 0;
+}
+
 static int vmlinux_path__init(struct perf_env *env)
 {
 	struct utsname uts;
 	char bf[PATH_MAX];
 	char *kernel_version;
+	unsigned int i;
 
-	vmlinux_path = malloc(sizeof(char *) * 6);
+	vmlinux_path = malloc(sizeof(char *) * (ARRAY_SIZE(vmlinux_paths) +
+			      ARRAY_SIZE(vmlinux_paths_upd)));
 	if (vmlinux_path == NULL)
 		return -1;
 
-	vmlinux_path[vmlinux_path__nr_entries] = strdup("vmlinux");
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-	++vmlinux_path__nr_entries;
-	vmlinux_path[vmlinux_path__nr_entries] = strdup("/boot/vmlinux");
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-	++vmlinux_path__nr_entries;
+	for (i = 0; i < ARRAY_SIZE(vmlinux_paths); i++)
+		if (vmlinux_path__add(vmlinux_paths[i]) < 0)
+			goto out_fail;
 
 	/* only try kernel version if no symfs was given */
 	if (symbol_conf.symfs[0] != 0)

@@ -1892,28 +1911,11 @@ static int vmlinux_path__init(struct perf_env *env)
 		kernel_version = uts.release;
 	}
 
-	snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", kernel_version);
-	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-	++vmlinux_path__nr_entries;
-	snprintf(bf, sizeof(bf), "/usr/lib/debug/boot/vmlinux-%s",
-		 kernel_version);
-	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-        ++vmlinux_path__nr_entries;
-	snprintf(bf, sizeof(bf), "/lib/modules/%s/build/vmlinux", kernel_version);
-	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-	++vmlinux_path__nr_entries;
-	snprintf(bf, sizeof(bf), "/usr/lib/debug/lib/modules/%s/vmlinux",
-		 kernel_version);
-	vmlinux_path[vmlinux_path__nr_entries] = strdup(bf);
-	if (vmlinux_path[vmlinux_path__nr_entries] == NULL)
-		goto out_fail;
-	++vmlinux_path__nr_entries;
+	for (i = 0; i < ARRAY_SIZE(vmlinux_paths_upd); i++) {
+		snprintf(bf, sizeof(bf), vmlinux_paths_upd[i], kernel_version);
+		if (vmlinux_path__add(bf) < 0)
+			goto out_fail;
+	}
 
 	return 0;
 

2.1.0

From: Ekaterina Tumanova <[email protected]>

Currently when debuginfo is separated to vmlinux.debug, it's contents get ignored. Let's change that and add it to the vmlinux_path list.

---
 tools/perf/util/symbol.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index e2ac6b6676e4..d51abd2e7865 100644
@@ -1869,7 +1869,8 @@ static const char * const vmlinux_paths_upd[] = {
 	"/boot/vmlinux-%s",
 	"/usr/lib/debug/boot/vmlinux-%s",
 	"/lib/modules/%s/build/vmlinux",
-	"/usr/lib/debug/lib/modules/%s/vmlinux"
+	"/usr/lib/debug/lib/modules/%s/vmlinux",
+	"/usr/lib/debug/boot/vmlinux-%s.debug"
 };
 
 static int vmlinux_path__add(const char *new_entry)

2.1.0

From: Jiri Olsa <[email protected]>

Clear sample_(type|period) for counting, as it only confuses debug output with unwanted sampling details:

Before:

  $ sudo perf stat -e 'raw_syscalls:sys_enter' -vv ls
  ------------------------------------------------------------
  perf_event_attr:
    type                             2
    size                             112
    config                           0x11
    { sample_period, sample_freq }   1
    sample_type                      TIME|CPU|PERIOD|RAW
    read_format                      TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING
    disabled                         1
    inherit                          1
    enable_on_exec                   1
    exclude_guest                    1
  ...

After:

  $ sudo perf stat -e 'raw_syscalls:sys_enter' -vv ls
  ------------------------------------------------------------
  perf_event_attr:
    type                             2
    size                             112
    config                           0x11
    read_format                      TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING
    disabled                         1
    inherit                          1
    enable_on_exec                   1
    exclude_guest                    1
  ...

---
 tools/perf/builtin-stat.c | 7 +++++++
 1 file changed, 7 insertions(+)

index e77880b5094d..df2fbf046ee2 100644
@@ -161,6 +161,13 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
 
 	attr->inherit = !no_inherit;
 
+	/*
+	 * Some events get initialized with sample_(period/type) set,
+	 * like tracepoints. Clear it up for counting.
+	 */
+	attr->sample_period = 0;
+	attr->sample_type   = 0;
+
 	if (target__has_cpu(&target))
 		return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
 

2.1.0

From: Jiri Olsa <[email protected]>

Adding WIEGHT bit_name call to display sample_type properly.

  $ perf evlist -v
  cpu/mem-loads/pp: ...SNIP... sample_type: IP|TID|TIME|ADDR|ID|CPU|DATA_SRC|WEIGHT ...

---
 tools/perf/util/evsel.c | 1 +
 1 file changed, 1 insertion(+)

index 397fb4ed3c97..0a1f4d9e52fc 100644
@@ -1192,6 +1192,7 @@ static void __p_sample_type(char *buf, size_t size, u64 value)
 		bit_name(PERIOD), bit_name(STREAM_ID), bit_name(RAW),
 		bit_name(BRANCH_STACK), bit_name(REGS_USER), bit_name(STACK_USER),
 		bit_name(IDENTIFIER), bit_name(REGS_INTR), bit_name(DATA_SRC),
+		bit_name(WEIGHT),
 		{ .name = NULL, }
 	};
 #undef bit_name

2.1.0

From: He Kuang <[email protected]>

Add bpf_map_update_elem() helper function which calls the sys_bpf syscall to update elements in bpf maps. Upcoming patches will use it to adjust data in map through the perf command line.

---
 tools/lib/bpf/bpf.c | 14 ++++++++++++++
 tools/lib/bpf/bpf.h |  2 ++
 2 files changed, 16 insertions(+)

index a6331050ab79..5bdc6eab6852 100644
@@ -83,3 +83,17 @@ int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
 	log_buf[0] = 0;
 	return sys_bpf(BPF_PROG_LOAD, &attr, sizeof(attr));
 }
+
+int bpf_map_update_elem(int fd, void *key, void *value,
+			u64 flags)
+{
+	union bpf_attr attr;
+
+	bzero(&attr, sizeof(attr));
+	attr.map_fd = fd;
+	attr.key = ptr_to_u64(key);
+	attr.value = ptr_to_u64(value);
+	attr.flags = flags;
+
+	return sys_bpf(BPF_MAP_UPDATE_ELEM, &attr, sizeof(attr));
+}

index 854b7361b784..a76465541292 100644
@@ -20,4 +20,6 @@ int bpf_load_program(enum bpf_prog_type type, struct bpf_insn *insns,
 		     u32 kern_version, char *log_buf,
 		     size_t log_buf_sz);
 
+int bpf_map_update_elem(int fd, void *key, void *value,
+			u64 flags);
 #endif

2.1.0

Pulled, thanks a lot Arnaldo!

Ingo