Merge tag 'perf-core-for-mingo' of...

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

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

 * Don't install scripting files files when perl/python support is disabled.

 * Support ! in -e expressions in 'perf trace', to filter a list of syscalls.

 * Add --verbose and -o/--output options to 'perf trace'.

 * Introduce better formatting of syscall arguments in 'perf trace',
   including so far beautifiers for mmap, madvise, syscall return
   values.

 * Fixup jobserver setup in libtraceevent makefile.

 * Debug improvements from Adrian Hunter.

 * Try to increase the file descriptor limits on EMFILE, from Andi Kleen.

 * Remove unused force option in 'perf kvm', from David Ahern.

 * Make 'perf trace' command line arguments consistent with 'perf record',
   from David Ahern.

 * Fix correlation of samples coming after PERF_RECORD_EXIT event, from
   David Ahern.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

tools/lib/traceevent/Makefile

@@ -60,7 +60,7 @@ ifeq ($(BUILD_SRC),)
 ifneq ($(BUILD_OUTPUT),)
 
 define build_output
-	$(if $(VERBOSE:1=),@)$(MAKE) -C $(BUILD_OUTPUT) 	\
+	$(if $(VERBOSE:1=),@)+$(MAKE) -C $(BUILD_OUTPUT) 	\
 	BUILD_SRC=$(CURDIR) -f $(CURDIR)/Makefile $1
 endef
 

tools/perf/Documentation/perf-trace.txt

@@ -23,29 +23,45 @@ analysis phases.
 OPTIONS
 -------
 
+-a::
 --all-cpus::
         System-wide collection from all CPUs.
 
 -e::
 --expr::
 	List of events to show, currently only syscall names.
+	Prefixing with ! shows all syscalls but the ones specified.  You may
+	need to escape it.
+
+-o::
+--output=::
+	Output file name.
 
 -p::
 --pid=::
 	Record events on existing process ID (comma separated list).
 
+-t::
 --tid=::
         Record events on existing thread ID (comma separated list).
 
+-u::
 --uid=::
         Record events in threads owned by uid. Name or number.
 
+-v::
+--verbose=::
+        Verbosity level.
+
+-i::
 --no-inherit::
 	Child tasks do not inherit counters.
 
+-m::
 --mmap-pages=::
 	Number of mmap data pages. Must be a power of two.
 
+-C::
 --cpu::
 Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a
 comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2.

tools/perf/Makefile

@@ -766,17 +766,21 @@ check: $(OUTPUT)common-cmds.h
 install-bin: all
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)'
 	$(INSTALL) $(OUTPUT)perf '$(DESTDIR_SQ)$(bindir_SQ)'
+	$(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
+ifndef NO_LIBPERL
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
-	$(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)'
 	$(INSTALL) scripts/perl/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) scripts/perl/*.pl -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl'
 	$(INSTALL) scripts/perl/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/bin'
+endif
+ifndef NO_LIBPYTHON
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
 	$(INSTALL) scripts/python/Perf-Trace-Util/lib/Perf/Trace/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/Perf-Trace-Util/lib/Perf/Trace'
 	$(INSTALL) scripts/python/*.py -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python'
 	$(INSTALL) scripts/python/bin/* -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/python/bin'
+endif
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d'
 	$(INSTALL) bash_completion '$(DESTDIR_SQ)$(sysconfdir_SQ)/bash_completion.d/perf'
 	$(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/tests'

tools/perf/builtin-kvm.c

@@ -1305,7 +1305,6 @@ kvm_events_record(struct perf_kvm_stat *kvm, int argc, const char **argv)
 	const char * const record_args[] = {
 		"record",
 		"-R",
-		"-f",
 		"-m", "1024",
 		"-c", "1",
 	};

tools/perf/util/evlist.c

@@ -14,6 +14,7 @@
 #include "target.h"
 #include "evlist.h"
 #include "evsel.h"
+#include "debug.h"
 #include <unistd.h>
 
 #include "parse-events.h"
@@ -486,6 +487,7 @@ static int perf_evlist__mmap_per_cpu(struct perf_evlist *evlist, int prot, int m
 	int nr_cpus = cpu_map__nr(evlist->cpus);
 	int nr_threads = thread_map__nr(evlist->threads);
 
+	pr_debug2("perf event ring buffer mmapped per cpu\n");
 	for (cpu = 0; cpu < nr_cpus; cpu++) {
 		int output = -1;
 
@@ -524,6 +526,7 @@ static int perf_evlist__mmap_per_thread(struct perf_evlist *evlist, int prot, in
 	int thread;
 	int nr_threads = thread_map__nr(evlist->threads);
 
+	pr_debug2("perf event ring buffer mmapped per thread\n");
 	for (thread = 0; thread < nr_threads; thread++) {
 		int output = -1;
 

tools/perf/util/evsel.c

@@ -13,6 +13,7 @@
 #include <traceevent/event-parse.h>
 #include <linux/hw_breakpoint.h>
 #include <linux/perf_event.h>
+#include <sys/resource.h>
 #include "asm/bug.h"
 #include "evsel.h"
 #include "evlist.h"
@@ -21,6 +22,7 @@
 #include "thread_map.h"
 #include "target.h"
 #include "perf_regs.h"
+#include "debug.h"
 
 static struct {
 	bool sample_id_all;
@@ -861,12 +863,72 @@ static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
 	return fd;
 }
 
+#define __PRINT_ATTR(fmt, cast, field)  \
+	fprintf(fp, "  %-19s "fmt"\n", #field, cast attr->field)
+
+#define PRINT_ATTR_U32(field)  __PRINT_ATTR("%u" , , field)
+#define PRINT_ATTR_X32(field)  __PRINT_ATTR("%#x", , field)
+#define PRINT_ATTR_U64(field)  __PRINT_ATTR("%" PRIu64, (uint64_t), field)
+#define PRINT_ATTR_X64(field)  __PRINT_ATTR("%#"PRIx64, (uint64_t), field)
+
+#define PRINT_ATTR2N(name1, field1, name2, field2)	\
+	fprintf(fp, "  %-19s %u    %-19s %u\n",		\
+	name1, attr->field1, name2, attr->field2)
+
+#define PRINT_ATTR2(field1, field2) \
+	PRINT_ATTR2N(#field1, field1, #field2, field2)
+
+static size_t perf_event_attr__fprintf(struct perf_event_attr *attr, FILE *fp)
+{
+	size_t ret = 0;
+
+	ret += fprintf(fp, "%.60s\n", graph_dotted_line);
+	ret += fprintf(fp, "perf_event_attr:\n");
+
+	ret += PRINT_ATTR_U32(type);
+	ret += PRINT_ATTR_U32(size);
+	ret += PRINT_ATTR_X64(config);
+	ret += PRINT_ATTR_U64(sample_period);
+	ret += PRINT_ATTR_U64(sample_freq);
+	ret += PRINT_ATTR_X64(sample_type);
+	ret += PRINT_ATTR_X64(read_format);
+
+	ret += PRINT_ATTR2(disabled, inherit);
+	ret += PRINT_ATTR2(pinned, exclusive);
+	ret += PRINT_ATTR2(exclude_user, exclude_kernel);
+	ret += PRINT_ATTR2(exclude_hv, exclude_idle);
+	ret += PRINT_ATTR2(mmap, comm);
+	ret += PRINT_ATTR2(freq, inherit_stat);
+	ret += PRINT_ATTR2(enable_on_exec, task);
+	ret += PRINT_ATTR2(watermark, precise_ip);
+	ret += PRINT_ATTR2(mmap_data, sample_id_all);
+	ret += PRINT_ATTR2(exclude_host, exclude_guest);
+	ret += PRINT_ATTR2N("excl.callchain_kern", exclude_callchain_kernel,
+			    "excl.callchain_user", exclude_callchain_user);
+
+	ret += PRINT_ATTR_U32(wakeup_events);
+	ret += PRINT_ATTR_U32(wakeup_watermark);
+	ret += PRINT_ATTR_X32(bp_type);
+	ret += PRINT_ATTR_X64(bp_addr);
+	ret += PRINT_ATTR_X64(config1);
+	ret += PRINT_ATTR_U64(bp_len);
+	ret += PRINT_ATTR_X64(config2);
+	ret += PRINT_ATTR_X64(branch_sample_type);
+	ret += PRINT_ATTR_X64(sample_regs_user);
+	ret += PRINT_ATTR_U32(sample_stack_user);
+
+	ret += fprintf(fp, "%.60s\n", graph_dotted_line);
+
+	return ret;
+}
+
 static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 			      struct thread_map *threads)
 {
 	int cpu, thread;
 	unsigned long flags = 0;
 	int pid = -1, err;
+	enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit = NO_CHANGE;
 
 	if (evsel->fd == NULL &&
 	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
@@ -884,6 +946,9 @@ static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 	if (perf_missing_features.sample_id_all)
 		evsel->attr.sample_id_all = 0;
 
+	if (verbose >= 2)
+		perf_event_attr__fprintf(&evsel->attr, stderr);
+
 	for (cpu = 0; cpu < cpus->nr; cpu++) {
 
 		for (thread = 0; thread < threads->nr; thread++) {
@@ -893,6 +958,9 @@ static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 				pid = threads->map[thread];
 
 			group_fd = get_group_fd(evsel, cpu, thread);
+retry_open:
+			pr_debug2("perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx\n",
+				  pid, cpus->map[cpu], group_fd, flags);
 
 			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
 								     pid,
@@ -902,12 +970,37 @@ static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
 				err = -errno;
 				goto try_fallback;
 			}
+			set_rlimit = NO_CHANGE;
 		}
 	}
 
 	return 0;
 
 try_fallback:
+	/*
+	 * perf stat needs between 5 and 22 fds per CPU. When we run out
+	 * of them try to increase the limits.
+	 */
+	if (err == -EMFILE && set_rlimit < INCREASED_MAX) {
+		struct rlimit l;
+		int old_errno = errno;
+
+		if (getrlimit(RLIMIT_NOFILE, &l) == 0) {
+			if (set_rlimit == NO_CHANGE)
+				l.rlim_cur = l.rlim_max;
+			else {
+				l.rlim_cur = l.rlim_max + 1000;
+				l.rlim_max = l.rlim_cur;
+			}
+			if (setrlimit(RLIMIT_NOFILE, &l) == 0) {
+				set_rlimit++;
+				errno = old_errno;
+				goto retry_open;
+			}
+		}
+		errno = old_errno;
+	}
+
 	if (err != -EINVAL || cpu > 0 || thread > 0)
 		goto out_close;
 

tools/perf/util/machine.c

@@ -253,7 +253,8 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size)
 	return;
 }
 
-static struct thread *__machine__findnew_thread(struct machine *machine, pid_t tid,
+static struct thread *__machine__findnew_thread(struct machine *machine,
+						pid_t pid, pid_t tid,
 						bool create)
 {
 	struct rb_node **p = &machine->threads.rb_node;
@@ -265,8 +266,11 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t
 	 * so most of the time we dont have to look up
 	 * the full rbtree:
 	 */
-	if (machine->last_match && machine->last_match->tid == tid)
+	if (machine->last_match && machine->last_match->tid == tid) {
+		if (pid && pid != machine->last_match->pid_)
+			machine->last_match->pid_ = pid;
 		return machine->last_match;
+	}
 
 	while (*p != NULL) {
 		parent = *p;
@@ -274,6 +278,8 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t
 
 		if (th->tid == tid) {
 			machine->last_match = th;
+			if (pid && pid != th->pid_)
+				th->pid_ = pid;
 			return th;
 		}
 
@@ -286,7 +292,7 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t
 	if (!create)
 		return NULL;
 
-	th = thread__new(tid);
+	th = thread__new(pid, tid);
 	if (th != NULL) {
 		rb_link_node(&th->rb_node, parent, p);
 		rb_insert_color(&th->rb_node, &machine->threads);
@@ -298,12 +304,12 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t
 
 struct thread *machine__findnew_thread(struct machine *machine, pid_t tid)
 {
-	return __machine__findnew_thread(machine, tid, true);
+	return __machine__findnew_thread(machine, 0, tid, true);
 }
 
 struct thread *machine__find_thread(struct machine *machine, pid_t tid)
 {
-	return __machine__findnew_thread(machine, tid, false);
+	return __machine__findnew_thread(machine, 0, tid, false);
 }
 
 int machine__process_comm_event(struct machine *machine, union perf_event *event)
@@ -1031,11 +1037,27 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event
 	return 0;
 }
 
+static void machine__remove_thread(struct machine *machine, struct thread *th)
+{
+	machine->last_match = NULL;
+	rb_erase(&th->rb_node, &machine->threads);
+	/*
+	 * We may have references to this thread, for instance in some hist_entry
+	 * instances, so just move them to a separate list.
+	 */
+	list_add_tail(&th->node, &machine->dead_threads);
+}
+
 int machine__process_fork_event(struct machine *machine, union perf_event *event)
 {
-	struct thread *thread = machine__findnew_thread(machine, event->fork.tid);
+	struct thread *thread = machine__find_thread(machine, event->fork.tid);
 	struct thread *parent = machine__findnew_thread(machine, event->fork.ptid);
 
+	/* if a thread currently exists for the thread id remove it */
+	if (thread != NULL)
+		machine__remove_thread(machine, thread);
+
+	thread = machine__findnew_thread(machine, event->fork.tid);
 	if (dump_trace)
 		perf_event__fprintf_task(event, stdout);
 
@@ -1048,18 +1070,8 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event
 	return 0;
 }
 
-static void machine__remove_thread(struct machine *machine, struct thread *th)
-{
-	machine->last_match = NULL;
-	rb_erase(&th->rb_node, &machine->threads);
-	/*
-	 * We may have references to this thread, for instance in some hist_entry
-	 * instances, so just move them to a separate list.
-	 */
-	list_add_tail(&th->node, &machine->dead_threads);
-}
-
-int machine__process_exit_event(struct machine *machine, union perf_event *event)
+int machine__process_exit_event(struct machine *machine __maybe_unused,
+				union perf_event *event)
 {
 	struct thread *thread = machine__find_thread(machine, event->fork.tid);
 
@@ -1067,7 +1079,7 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event
 		perf_event__fprintf_task(event, stdout);
 
 	if (thread != NULL)
-		machine__remove_thread(machine, thread);
+		thread__exited(thread);
 
 	return 0;
 }

tools/perf/util/python.c

@@ -8,6 +8,26 @@
 #include "cpumap.h"
 #include "thread_map.h"
 
+/*
+ * Support debug printing even though util/debug.c is not linked.  That means
+ * implementing 'verbose' and 'eprintf'.
+ */
+int verbose;
+
+int eprintf(int level, const char *fmt, ...)
+{
+	va_list args;
+	int ret = 0;
+
+	if (verbose >= level) {
+		va_start(args, fmt);
+		ret = vfprintf(stderr, fmt, args);
+		va_end(args);
+	}
+
+	return ret;
+}
+
 /* Define PyVarObject_HEAD_INIT for python 2.5 */
 #ifndef PyVarObject_HEAD_INIT
 # define PyVarObject_HEAD_INIT(type, size) PyObject_HEAD_INIT(type) size,

tools/perf/util/thread.c

@@ -7,12 +7,13 @@
 #include "util.h"
 #include "debug.h"
 
-struct thread *thread__new(pid_t tid)
+struct thread *thread__new(pid_t pid, pid_t tid)
 {
 	struct thread *self = zalloc(sizeof(*self));
 
 	if (self != NULL) {
 		map_groups__init(&self->mg);
+		self->pid_ = pid;
 		self->tid = tid;
 		self->ppid = -1;
 		self->comm = malloc(32);

tools/perf/util/thread.h

@@ -12,10 +12,12 @@ struct thread {
 		struct list_head node;
 	};
 	struct map_groups	mg;
+	pid_t			pid_; /* Not all tools update this */
 	pid_t			tid;
 	pid_t			ppid;
 	char			shortname[3];
 	bool			comm_set;
+	bool			dead; /* if set thread has exited */
 	char			*comm;
 	int			comm_len;
 
@@ -24,8 +26,12 @@ struct thread {
 
 struct machine;
 
-struct thread *thread__new(pid_t tid);
+struct thread *thread__new(pid_t pid, pid_t tid);
 void thread__delete(struct thread *self);
+static inline void thread__exited(struct thread *thread)
+{
+	thread->dead = true;
+}
 
 int thread__set_comm(struct thread *self, const char *comm);
 int thread__comm_len(struct thread *self);