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

Pull perf/core updates from Arnaldo Carvalho de Melo:

Changes in user visible interfaces:

  * Rename 'record's --no-delay option to --no-buffering, better reflecting its
    purpose and freeing up '--delay' to take the place of '--initial-delay', so that
    'record' and 'stat' are consistent.

Refactorings:

  * Get rid of die() and friends (good riddance!) in libtraceevent (Namhyung Kim)

Infrastructure enhancements:

  * Fix cross build problems related to pkgconfig and CROSS_COMPILE not being
    propagated to the feature tests, leading to features being tested in the
    host and then being enabled on the target. (Mark Rutland)

  * Fix pointer-integer size mismatch in some libtraceevent plugins (Mark Rutland)

  * Fix build error due to zfree() cast (Namhyung Kim)

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

tools/lib/traceevent/Makefile

@@ -136,7 +136,7 @@ export Q VERBOSE
 
 EVENT_PARSE_VERSION = $(EP_VERSION).$(EP_PATCHLEVEL).$(EP_EXTRAVERSION)
 
-INCLUDES = -I. $(CONFIG_INCLUDES)
+INCLUDES = -I. -I $(srctree)/../../include $(CONFIG_INCLUDES)
 
 # Set compile option CFLAGS if not set elsewhere
 CFLAGS ?= -g -Wall

tools/lib/traceevent/event-parse.h

@@ -58,6 +58,12 @@ struct pevent_record {
 #endif
 };
 
+enum trace_seq_fail {
+	TRACE_SEQ__GOOD,
+	TRACE_SEQ__BUFFER_POISONED,
+	TRACE_SEQ__MEM_ALLOC_FAILED,
+};
+
 /*
  * Trace sequences are used to allow a function to call several other functions
  * to create a string of data to use (up to a max of PAGE_SIZE).
@@ -68,6 +74,7 @@ struct trace_seq {
 	unsigned int		buffer_size;
 	unsigned int		len;
 	unsigned int		readpos;
+	enum trace_seq_fail	state;
 };
 
 void trace_seq_init(struct trace_seq *s);
@@ -98,7 +105,7 @@ typedef int (*pevent_event_handler_func)(struct trace_seq *s,
 					 void *context);
 
 typedef int (*pevent_plugin_load_func)(struct pevent *pevent);
-typedef int (*pevent_plugin_unload_func)(void);
+typedef int (*pevent_plugin_unload_func)(struct pevent *pevent);
 
 struct plugin_option {
 	struct plugin_option		*next;
@@ -123,7 +130,7 @@ struct plugin_option {
  * PEVENT_PLUGIN_UNLOADER:  (optional)
  *   The function called just before unloading
  *
- *   int PEVENT_PLUGIN_UNLOADER(void)
+ *   int PEVENT_PLUGIN_UNLOADER(struct pevent *pevent)
  *
  * PEVENT_PLUGIN_OPTIONS:  (optional)
  *   Plugin options that can be set before loading
@@ -404,7 +411,8 @@ enum pevent_errno {
 struct plugin_list;
 
 struct plugin_list *traceevent_load_plugins(struct pevent *pevent);
-void traceevent_unload_plugins(struct plugin_list *plugin_list);
+void traceevent_unload_plugins(struct plugin_list *plugin_list,
+			       struct pevent *pevent);
 
 struct cmdline;
 struct cmdline_list;

tools/lib/traceevent/event-plugin.c

@@ -197,7 +197,7 @@ traceevent_load_plugins(struct pevent *pevent)
 }
 
 void
-traceevent_unload_plugins(struct plugin_list *plugin_list)
+traceevent_unload_plugins(struct plugin_list *plugin_list, struct pevent *pevent)
 {
 	pevent_plugin_unload_func func;
 	struct plugin_list *list;
@@ -207,7 +207,7 @@ traceevent_unload_plugins(struct plugin_list *plugin_list)
 		plugin_list = list->next;
 		func = dlsym(list->handle, PEVENT_PLUGIN_UNLOADER_NAME);
 		if (func)
-			func();
+			func(pevent);
 		dlclose(list->handle);
 		free(list->name);
 		free(list);

tools/lib/traceevent/event-utils.h

@@ -23,18 +23,14 @@
 #include <ctype.h>
 
 /* Can be overridden */
-void die(const char *fmt, ...);
-void *malloc_or_die(unsigned int size);
 void warning(const char *fmt, ...);
 void pr_stat(const char *fmt, ...);
 void vpr_stat(const char *fmt, va_list ap);
 
 /* Always available */
-void __die(const char *fmt, ...);
 void __warning(const char *fmt, ...);
 void __pr_stat(const char *fmt, ...);
 
-void __vdie(const char *fmt, ...);
 void __vwarning(const char *fmt, ...);
 void __vpr_stat(const char *fmt, ...);
 

tools/lib/traceevent/parse-utils.c

@@ -25,40 +25,6 @@
 
 #define __weak __attribute__((weak))
 
-void __vdie(const char *fmt, va_list ap)
-{
-	int ret = errno;
-
-	if (errno)
-		perror("trace-cmd");
-	else
-		ret = -1;
-
-	fprintf(stderr, "  ");
-	vfprintf(stderr, fmt, ap);
-
-	fprintf(stderr, "\n");
-	exit(ret);
-}
-
-void __die(const char *fmt, ...)
-{
-	va_list ap;
-
-	va_start(ap, fmt);
-	__vdie(fmt, ap);
-	va_end(ap);
-}
-
-void __weak die(const char *fmt, ...)
-{
-	va_list ap;
-
-	va_start(ap, fmt);
-	__vdie(fmt, ap);
-	va_end(ap);
-}
-
 void __vwarning(const char *fmt, va_list ap)
 {
 	if (errno)
@@ -117,13 +83,3 @@ void __weak pr_stat(const char *fmt, ...)
 	__vpr_stat(fmt, ap);
 	va_end(ap);
 }
-
-void __weak *malloc_or_die(unsigned int size)
-{
-	void *data;
-
-	data = malloc(size);
-	if (!data)
-		die("malloc");
-	return data;
-}

tools/lib/traceevent/plugin_cfg80211.c

@@ -8,7 +8,7 @@ static unsigned long long
 process___le16_to_cpup(struct trace_seq *s,
 		       unsigned long long *args)
 {
-	uint16_t *val = (uint16_t *) args[0];
+	uint16_t *val = (uint16_t *) (unsigned long) args[0];
 	return val ? (long long) le16toh(*val) : 0;
 }
 

tools/lib/traceevent/plugin_function.c

@@ -144,7 +144,7 @@ int PEVENT_PLUGIN_LOADER(struct pevent *pevent)
 	return 0;
 }
 
-void PEVENT_PLUGIN_UNLOADER(void)
+void PEVENT_PLUGIN_UNLOADER(struct pevent *pevent)
 {
 	int i, x;
 

tools/lib/traceevent/plugin_scsi.c

@@ -405,7 +405,7 @@ scsi_trace_parse_cdb(struct trace_seq *p, unsigned char *cdb, int len)
 unsigned long long process_scsi_trace_parse_cdb(struct trace_seq *s,
 						unsigned long long *args)
 {
-	scsi_trace_parse_cdb(s, (unsigned char *) args[1], args[2]);
+	scsi_trace_parse_cdb(s, (unsigned char *) (unsigned long) args[1], args[2]);
 	return 0;
 }
 

tools/lib/traceevent/trace-seq.c

@@ -22,6 +22,7 @@
 #include <string.h>
 #include <stdarg.h>
 
+#include <asm/bug.h>
 #include "event-parse.h"
 #include "event-utils.h"
 
@@ -32,10 +33,21 @@
 #define TRACE_SEQ_POISON	((void *)0xdeadbeef)
 #define TRACE_SEQ_CHECK(s)						\
 do {									\
-	if ((s)->buffer == TRACE_SEQ_POISON)			\
-		die("Usage of trace_seq after it was destroyed");	\
+	if (WARN_ONCE((s)->buffer == TRACE_SEQ_POISON,			\
+		      "Usage of trace_seq after it was destroyed"))	\
+		(s)->state = TRACE_SEQ__BUFFER_POISONED;		\
 } while (0)
 
+#define TRACE_SEQ_CHECK_RET_N(s, n)		\
+do {						\
+	TRACE_SEQ_CHECK(s);			\
+	if ((s)->state != TRACE_SEQ__GOOD)	\
+		return n; 			\
+} while (0)
+
+#define TRACE_SEQ_CHECK_RET(s)   TRACE_SEQ_CHECK_RET_N(s, )
+#define TRACE_SEQ_CHECK_RET0(s)  TRACE_SEQ_CHECK_RET_N(s, 0)
+
 /**
  * trace_seq_init - initialize the trace_seq structure
  * @s: a pointer to the trace_seq structure to initialize
@@ -45,7 +57,11 @@ void trace_seq_init(struct trace_seq *s)
 	s->len = 0;
 	s->readpos = 0;
 	s->buffer_size = TRACE_SEQ_BUF_SIZE;
-	s->buffer = malloc_or_die(s->buffer_size);
+	s->buffer = malloc(s->buffer_size);
+	if (s->buffer != NULL)
+		s->state = TRACE_SEQ__GOOD;
+	else
+		s->state = TRACE_SEQ__MEM_ALLOC_FAILED;
 }
 
 /**
@@ -71,17 +87,23 @@ void trace_seq_destroy(struct trace_seq *s)
 {
 	if (!s)
 		return;
-	TRACE_SEQ_CHECK(s);
+	TRACE_SEQ_CHECK_RET(s);
 	free(s->buffer);
 	s->buffer = TRACE_SEQ_POISON;
 }
 
 static void expand_buffer(struct trace_seq *s)
 {
+	char *buf;
+
+	buf = realloc(s->buffer, s->buffer_size + TRACE_SEQ_BUF_SIZE);
+	if (WARN_ONCE(!buf, "Can't allocate trace_seq buffer memory")) {
+		s->state = TRACE_SEQ__MEM_ALLOC_FAILED;
+		return;
+	}
+
+	s->buffer = buf;
 	s->buffer_size += TRACE_SEQ_BUF_SIZE;
-	s->buffer = realloc(s->buffer, s->buffer_size);
-	if (!s->buffer)
-		die("Can't allocate trace_seq buffer memory");
 }
 
 /**
@@ -105,9 +127,9 @@ trace_seq_printf(struct trace_seq *s, const char *fmt, ...)
 	int len;
 	int ret;
 
-	TRACE_SEQ_CHECK(s);
-
  try_again:
+	TRACE_SEQ_CHECK_RET0(s);
+
 	len = (s->buffer_size - 1) - s->len;
 
 	va_start(ap, fmt);
@@ -141,9 +163,9 @@ trace_seq_vprintf(struct trace_seq *s, const char *fmt, va_list args)
 	int len;
 	int ret;
 
-	TRACE_SEQ_CHECK(s);
-
  try_again:
+	TRACE_SEQ_CHECK_RET0(s);
+
 	len = (s->buffer_size - 1) - s->len;
 
 	ret = vsnprintf(s->buffer + s->len, len, fmt, args);
@@ -172,13 +194,15 @@ int trace_seq_puts(struct trace_seq *s, const char *str)
 {
 	int len;
 
-	TRACE_SEQ_CHECK(s);
+	TRACE_SEQ_CHECK_RET0(s);
 
 	len = strlen(str);
 
 	while (len > ((s->buffer_size - 1) - s->len))
 		expand_buffer(s);
 
+	TRACE_SEQ_CHECK_RET0(s);
+
 	memcpy(s->buffer + s->len, str, len);
 	s->len += len;
 
@@ -187,11 +211,13 @@ int trace_seq_puts(struct trace_seq *s, const char *str)
 
 int trace_seq_putc(struct trace_seq *s, unsigned char c)
 {
-	TRACE_SEQ_CHECK(s);
+	TRACE_SEQ_CHECK_RET0(s);
 
 	while (s->len >= (s->buffer_size - 1))
 		expand_buffer(s);
 
+	TRACE_SEQ_CHECK_RET0(s);
+
 	s->buffer[s->len++] = c;
 
 	return 1;
@@ -199,7 +225,7 @@ int trace_seq_putc(struct trace_seq *s, unsigned char c)
 
 void trace_seq_terminate(struct trace_seq *s)
 {
-	TRACE_SEQ_CHECK(s);
+	TRACE_SEQ_CHECK_RET(s);
 
 	/* There's always one character left on the buffer */
 	s->buffer[s->len] = 0;
@@ -208,5 +234,16 @@ void trace_seq_terminate(struct trace_seq *s)
 int trace_seq_do_printf(struct trace_seq *s)
 {
 	TRACE_SEQ_CHECK(s);
-	return printf("%.*s", s->len, s->buffer);
+
+	switch (s->state) {
+	case TRACE_SEQ__GOOD:
+		return printf("%.*s", s->len, s->buffer);
+	case TRACE_SEQ__BUFFER_POISONED:
+		puts("Usage of trace_seq after it was destroyed");
+		break;
+	case TRACE_SEQ__MEM_ALLOC_FAILED:
+		puts("Can't allocate trace_seq buffer memory");
+		break;
+	}
+	return -1;
 }

tools/perf/Documentation/perf-record.txt

@@ -68,8 +68,7 @@ OPTIONS
 --realtime=::
 	Collect data with this RT SCHED_FIFO priority.
 
--D::
---no-delay::
+--no-buffering::
 	Collect data without buffering.
 
 -c::
@@ -209,7 +208,8 @@ overrides that and uses per-thread mmaps.  A side-effect of that is that
 inheritance is automatically disabled.  --per-thread is ignored with a warning
 if combined with -a or -C options.
 
---initial-delay msecs::
+-D::
+--delay=::
 After starting the program, wait msecs before measuring. This is useful to
 filter out the startup phase of the program, which is often very different.
 

tools/perf/Makefile.perf

@@ -76,6 +76,7 @@ $(OUTPUT)PERF-VERSION-FILE: ../../.git/HEAD
 
 CC = $(CROSS_COMPILE)gcc
 AR = $(CROSS_COMPILE)ar
+PKG_CONFIG = $(CROSS_COMPILE)pkg-config
 
 RM      = rm -f
 LN      = ln -f

tools/perf/builtin-record.c

@@ -838,7 +838,7 @@ const struct option record_options[] = {
 		    "record events on existing thread id"),
 	OPT_INTEGER('r', "realtime", &record.realtime_prio,
 		    "collect data with this RT SCHED_FIFO priority"),
-	OPT_BOOLEAN('D', "no-delay", &record.opts.no_delay,
+	OPT_BOOLEAN(0, "no-buffering", &record.opts.no_buffering,
 		    "collect data without buffering"),
 	OPT_BOOLEAN('R', "raw-samples", &record.opts.raw_samples,
 		    "collect raw sample records from all opened counters"),
@@ -882,7 +882,7 @@ const struct option record_options[] = {
 	OPT_CALLBACK('G', "cgroup", &record.evlist, "name",
 		     "monitor event in cgroup name only",
 		     parse_cgroups),
-	OPT_UINTEGER(0, "initial-delay", &record.opts.initial_delay,
+	OPT_UINTEGER('D', "delay", &record.opts.initial_delay,
 		  "ms to wait before starting measurement after program start"),
 	OPT_STRING('u', "uid", &record.opts.target.uid_str, "user",
 		   "user to profile"),

tools/perf/builtin-report.c

@@ -75,24 +75,6 @@ static int report__config(const char *var, const char *value, void *cb)
 	return perf_default_config(var, value, cb);
 }
 
-static int report__resolve_callchain(struct report *rep, struct symbol **parent,
-				     struct perf_evsel *evsel, struct addr_location *al,
-				     struct perf_sample *sample)
-{
-	if ((sort__has_parent || symbol_conf.use_callchain) && sample->callchain) {
-		return machine__resolve_callchain(al->machine, evsel, al->thread, sample,
-						  parent, al, rep->max_stack);
-	}
-	return 0;
-}
-
-static int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample)
-{
-	if (!symbol_conf.use_callchain)
-		return 0;
-	return callchain_append(he->callchain, &callchain_cursor, sample->period);
-}
-
 static int report__add_mem_hist_entry(struct perf_tool *tool, struct addr_location *al,
 				      struct perf_sample *sample, struct perf_evsel *evsel,
 				      union perf_event *event)
@@ -103,7 +85,7 @@ static int report__add_mem_hist_entry(struct perf_tool *tool, struct addr_locati
 	struct hist_entry *he;
 	struct mem_info *mi, *mx;
 	uint64_t cost;
-	int err = report__resolve_callchain(rep, &parent, evsel, al, sample);
+	int err = sample__resolve_callchain(sample, &parent, evsel, al, rep->max_stack);
 
 	if (err)
 		return err;
@@ -155,7 +137,7 @@ static int report__add_branch_hist_entry(struct perf_tool *tool, struct addr_loc
 	unsigned i;
 	struct hist_entry *he;
 	struct branch_info *bi, *bx;
-	int err = report__resolve_callchain(rep, &parent, evsel, al, sample);
+	int err = sample__resolve_callchain(sample, &parent, evsel, al, rep->max_stack);
 
 	if (err)
 		return err;
@@ -208,7 +190,7 @@ static int report__add_hist_entry(struct perf_tool *tool, struct perf_evsel *evs
 	struct report *rep = container_of(tool, struct report, tool);
 	struct symbol *parent = NULL;
 	struct hist_entry *he;
-	int err = report__resolve_callchain(rep, &parent, evsel, al, sample);
+	int err = sample__resolve_callchain(sample, &parent, evsel, al, rep->max_stack);
 
 	if (err)
 		return err;

tools/perf/builtin-top.c

@@ -743,15 +743,10 @@ static void perf_event__process_sample(struct perf_tool *tool,
 	if (al.sym == NULL || !al.sym->ignore) {
 		struct hist_entry *he;
 
-		if ((sort__has_parent || symbol_conf.use_callchain) &&
-		    sample->callchain) {
-			err = machine__resolve_callchain(machine, evsel,
-							 al.thread, sample,
-							 &parent, &al,
-							 top->max_stack);
-			if (err)
-				return;
-		}
+		err = sample__resolve_callchain(sample, &parent, evsel, &al,
+						top->max_stack);
+		if (err)
+			return;
 
 		he = perf_evsel__add_hist_entry(evsel, &al, sample);
 		if (he == NULL) {
@@ -759,12 +754,9 @@ static void perf_event__process_sample(struct perf_tool *tool,
 			return;
 		}
 
-		if (symbol_conf.use_callchain) {
-			err = callchain_append(he->callchain, &callchain_cursor,
-					       sample->period);
-			if (err)
-				return;
-		}
+		err = hist_entry__append_callchain(he, sample);
+		if (err)
+			return;
 
 		if (sort__has_sym)
 			perf_top__record_precise_ip(top, he, evsel->idx, ip);

tools/perf/builtin-trace.c

@@ -2258,7 +2258,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
 			},
 			.user_freq     = UINT_MAX,
 			.user_interval = ULLONG_MAX,
-			.no_delay      = true,
+			.no_buffering  = true,
 			.mmap_pages    = 1024,
 		},
 		.output = stdout,

tools/perf/config/Makefile

@@ -372,7 +372,7 @@ ifndef NO_SLANG
 endif
 
 ifndef NO_GTK2
-  FLAGS_GTK2=$(CFLAGS) $(LDFLAGS) $(EXTLIBS) $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null)
+  FLAGS_GTK2=$(CFLAGS) $(LDFLAGS) $(EXTLIBS) $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null)
   ifneq ($(feature-gtk2), 1)
     msg := $(warning GTK2 not found, disables GTK2 support. Please install gtk2-devel or libgtk2.0-dev);
     NO_GTK2 := 1
@@ -381,8 +381,8 @@ ifndef NO_GTK2
       GTK_CFLAGS := -DHAVE_GTK_INFO_BAR_SUPPORT
     endif
     CFLAGS += -DHAVE_GTK2_SUPPORT
-    GTK_CFLAGS += $(shell pkg-config --cflags gtk+-2.0 2>/dev/null)
-    GTK_LIBS := $(shell pkg-config --libs gtk+-2.0 2>/dev/null)
+    GTK_CFLAGS += $(shell $(PKG_CONFIG) --cflags gtk+-2.0 2>/dev/null)
+    GTK_LIBS := $(shell $(PKG_CONFIG) --libs gtk+-2.0 2>/dev/null)
     EXTLIBS += -ldl
   endif
 endif

tools/perf/config/feature-checks/Makefile

@@ -28,7 +28,8 @@ FILES=					\
 	test-stackprotector-all.bin	\
 	test-timerfd.bin
 
-CC := $(CC) -MD
+CC := $(CROSS_COMPILE)gcc -MD
+PKG_CONFIG := $(CROSS_COMPILE)pkg-config
 
 all: $(FILES)
 
@@ -37,7 +38,7 @@ BUILD = $(CC) $(CFLAGS) -o $(OUTPUT)$@ $(patsubst %.bin,%.c,$@) $(LDFLAGS)
 ###############################
 
 test-all.bin:
-	$(BUILD) -Werror -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma -lelf -laudit -I/usr/include/slang -lslang $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl
+	$(BUILD) -Werror -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma -lelf -laudit -I/usr/include/slang -lslang $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl
 
 test-hello.bin:
 	$(BUILD)
@@ -82,10 +83,10 @@ test-libslang.bin:
 	$(BUILD) -I/usr/include/slang -lslang
 
 test-gtk2.bin:
-	$(BUILD) $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null)
+	$(BUILD) $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null)
 
 test-gtk2-infobar.bin:
-	$(BUILD) $(shell pkg-config --libs --cflags gtk+-2.0 2>/dev/null)
+	$(BUILD) $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null)
 
 grep-libs  = $(filter -l%,$(1))
 strip-libs = $(filter-out -l%,$(1))

tools/perf/perf.h

@@ -252,7 +252,7 @@ struct record_opts {
 	int	     call_graph;
 	bool	     group;
 	bool	     inherit_stat;
-	bool	     no_delay;
+	bool	     no_buffering;
 	bool	     no_inherit;
 	bool	     no_inherit_set;
 	bool	     no_samples;

tools/perf/tests/open-syscall-tp-fields.c

@@ -11,10 +11,10 @@ int test__syscall_open_tp_fields(void)
 			.uid = UINT_MAX,
 			.uses_mmap = true,
 		},
-		.no_delay   = true,
-		.freq	    = 1,
-		.mmap_pages = 256,
-		.raw_samples = true,
+		.no_buffering = true,
+		.freq	      = 1,
+		.mmap_pages   = 256,
+		.raw_samples  = true,
 	};
 	const char *filename = "/etc/passwd";
 	int flags = O_RDONLY | O_DIRECTORY;

tools/perf/tests/perf-record.c

@@ -39,9 +39,9 @@ int test__PERF_RECORD(void)
 			.uid = UINT_MAX,
 			.uses_mmap = true,
 		},
-		.no_delay   = true,
-		.freq	    = 10,
-		.mmap_pages = 256,
+		.no_buffering = true,
+		.freq	      = 10,
+		.mmap_pages   = 256,
 	};
 	cpu_set_t cpu_mask;
 	size_t cpu_mask_size = sizeof(cpu_mask);

tools/perf/util/callchain.c

@@ -17,6 +17,8 @@
 
 #include "hist.h"
 #include "util.h"
+#include "sort.h"
+#include "machine.h"
 #include "callchain.h"
 
 __thread struct callchain_cursor callchain_cursor;
@@ -531,3 +533,24 @@ int callchain_cursor_append(struct callchain_cursor *cursor,
 
 	return 0;
 }
+
+int sample__resolve_callchain(struct perf_sample *sample, struct symbol **parent,
+			      struct perf_evsel *evsel, struct addr_location *al,
+			      int max_stack)
+{
+	if (sample->callchain == NULL)
+		return 0;
+
+	if (symbol_conf.use_callchain || sort__has_parent) {
+		return machine__resolve_callchain(al->machine, evsel, al->thread,
+						  sample, parent, al, max_stack);
+	}
+	return 0;
+}
+
+int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample)
+{
+	if (!symbol_conf.use_callchain)
+		return 0;
+	return callchain_append(he->callchain, &callchain_cursor, sample->period);
+}

tools/perf/util/callchain.h

@@ -145,10 +145,16 @@ static inline void callchain_cursor_advance(struct callchain_cursor *cursor)
 }
 
 struct option;
+struct hist_entry;
 
 int record_parse_callchain(const char *arg, struct record_opts *opts);
 int record_parse_callchain_opt(const struct option *opt, const char *arg, int unset);
 int record_callchain_opt(const struct option *opt, const char *arg, int unset);
 
+int sample__resolve_callchain(struct perf_sample *sample, struct symbol **parent,
+			      struct perf_evsel *evsel, struct addr_location *al,
+			      int max_stack);
+int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample);
+
 extern const char record_callchain_help[];
 #endif	/* __PERF_CALLCHAIN_H */

tools/perf/util/evsel.c

@@ -627,7 +627,7 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
 	if (opts->sample_address)
 		perf_evsel__set_sample_bit(evsel, DATA_SRC);
 
-	if (opts->no_delay) {
+	if (opts->no_buffering) {
 		attr->watermark = 0;
 		attr->wakeup_events = 1;
 	}

tools/perf/util/hist.c

@@ -181,21 +181,21 @@ void hists__output_recalc_col_len(struct hists *hists, int max_rows)
 	}
 }
 
-static void hist_entry__add_cpumode_period(struct hist_entry *he,
-					   unsigned int cpumode, u64 period)
+static void he_stat__add_cpumode_period(struct he_stat *he_stat,
+					unsigned int cpumode, u64 period)
 {
 	switch (cpumode) {
 	case PERF_RECORD_MISC_KERNEL:
-		he->stat.period_sys += period;
+		he_stat->period_sys += period;
 		break;
 	case PERF_RECORD_MISC_USER:
-		he->stat.period_us += period;
+		he_stat->period_us += period;
 		break;
 	case PERF_RECORD_MISC_GUEST_KERNEL:
-		he->stat.period_guest_sys += period;
+		he_stat->period_guest_sys += period;
 		break;
 	case PERF_RECORD_MISC_GUEST_USER:
-		he->stat.period_guest_us += period;
+		he_stat->period_guest_us += period;
 		break;
 	default:
 		break;
@@ -222,10 +222,10 @@ static void he_stat__add_stat(struct he_stat *dest, struct he_stat *src)
 	dest->weight		+= src->weight;
 }
 
-static void hist_entry__decay(struct hist_entry *he)
+static void he_stat__decay(struct he_stat *he_stat)
 {
-	he->stat.period = (he->stat.period * 7) / 8;
-	he->stat.nr_events = (he->stat.nr_events * 7) / 8;
+	he_stat->period = (he_stat->period * 7) / 8;
+	he_stat->nr_events = (he_stat->nr_events * 7) / 8;
 	/* XXX need decay for weight too? */
 }
 
@@ -236,7 +236,7 @@ static bool hists__decay_entry(struct hists *hists, struct hist_entry *he)
 	if (prev_period == 0)
 		return true;
 
-	hist_entry__decay(he);
+	he_stat__decay(&he->stat);
 
 	if (!he->filtered)
 		hists->stats.total_period -= prev_period - he->stat.period;
@@ -402,7 +402,7 @@ static struct hist_entry *add_hist_entry(struct hists *hists,
 	rb_link_node(&he->rb_node_in, parent, p);
 	rb_insert_color(&he->rb_node_in, hists->entries_in);
 out:
-	hist_entry__add_cpumode_period(he, al->cpumode, period);
+	he_stat__add_cpumode_period(&he->stat, al->cpumode, period);
 	return he;
 }
 

tools/perf/util/machine.c

@@ -1314,8 +1314,6 @@ static int machine__resolve_callchain_sample(struct machine *machine,
 				*root_al = al;
 				callchain_cursor_reset(&callchain_cursor);
 			}
-			if (!symbol_conf.use_callchain)
-				break;
 		}
 
 		err = callchain_cursor_append(&callchain_cursor,

tools/perf/util/srcline.c

@@ -129,7 +129,7 @@ static struct a2l_data *addr2line_init(const char *path)
 
 out:
 	if (a2l) {
-		zfree((void **)&a2l->input);
+		zfree((char **)&a2l->input);
 		free(a2l);
 	}
 	bfd_close(abfd);
@@ -140,7 +140,7 @@ static void addr2line_cleanup(struct a2l_data *a2l)
 {
 	if (a2l->abfd)
 		bfd_close(a2l->abfd);
-	zfree((void **)&a2l->input);
+	zfree((char **)&a2l->input);
 	zfree(&a2l->syms);
 	free(a2l);
 }

tools/perf/util/strlist.c

@@ -39,7 +39,7 @@ out_delete:
 static void str_node__delete(struct str_node *snode, bool dupstr)
 {
 	if (dupstr)
-		zfree((void **)&snode->s);
+		zfree((char **)&snode->s);
 	free(snode);
 }
 

tools/perf/util/trace-event.c

@@ -34,8 +34,8 @@ int trace_event__init(struct trace_event *t)
 
 void trace_event__cleanup(struct trace_event *t)
 {
+	traceevent_unload_plugins(t->plugin_list, t->pevent);
 	pevent_free(t->pevent);
-	traceevent_unload_plugins(t->plugin_list);
 }
 
 static struct event_format*