Merge branch 'perfcounters/tracing' into perfcounters/core

Merge reason: this topic is ready now to merge into the main
              development branch for .32, with functional
              perf trace output.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

tools/perf/Makefile

@@ -347,6 +347,9 @@ LIB_OBJS += util/values.o
 LIB_OBJS += util/debug.o
 LIB_OBJS += util/map.o
 LIB_OBJS += util/thread.o
+LIB_OBJS += util/trace-event-parse.o
+LIB_OBJS += util/trace-event-read.o
+LIB_OBJS += util/trace-event-info.o
 
 BUILTIN_OBJS += builtin-annotate.o
 BUILTIN_OBJS += builtin-help.o
@@ -355,6 +358,7 @@ BUILTIN_OBJS += builtin-record.o
 BUILTIN_OBJS += builtin-report.o
 BUILTIN_OBJS += builtin-stat.o
 BUILTIN_OBJS += builtin-top.o
+BUILTIN_OBJS += builtin-trace.o
 
 PERFLIBS = $(LIB_FILE)
 

tools/perf/builtin-record.c

@@ -17,6 +17,7 @@
 #include "util/header.h"
 #include "util/event.h"
 #include "util/debug.h"
+#include "util/trace-event.h"
 
 #include <unistd.h>
 #include <sched.h>
@@ -546,6 +547,17 @@ static int __cmd_record(int argc, const char **argv)
 	else
 		header = perf_header__new();
 
+
+	if (raw_samples) {
+		read_tracing_data(attrs, nr_counters);
+	} else {
+		for (i = 0; i < nr_counters; i++) {
+			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
+				read_tracing_data(attrs, nr_counters);
+				break;
+			}
+		}
+	}
 	atexit(atexit_header);
 
 	if (!system_wide) {

tools/perf/builtin-report.c

@@ -1428,7 +1428,7 @@ static int __cmd_report(void)
 			exit(-1);
 		}
 		if (callchain) {
-			fprintf(stderr, "selected -c but no callchain data."
+			fprintf(stderr, "selected -g but no callchain data."
 					" Did you call perf record without"
 					" -g?\n");
 			exit(-1);

tools/perf/builtin-trace.c

@@ -0,0 +1,282 @@
+#include "builtin.h"
+
+#include "util/util.h"
+#include "util/cache.h"
+#include "util/symbol.h"
+#include "util/thread.h"
+#include "util/header.h"
+
+#include "util/parse-options.h"
+
+#include "perf.h"
+#include "util/debug.h"
+
+#include "util/trace-event.h"
+
+static char		const *input_name = "perf.data";
+static int		input;
+static unsigned long	page_size;
+static unsigned long	mmap_window = 32;
+
+static unsigned long	total = 0;
+static unsigned long	total_comm = 0;
+
+static struct rb_root	threads;
+static struct thread	*last_match;
+
+static struct perf_header *header;
+static u64		sample_type;
+
+
+static int
+process_comm_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	struct thread *thread;
+
+	thread = threads__findnew(event->comm.pid, &threads, &last_match);
+
+	dump_printf("%p [%p]: PERF_EVENT_COMM: %s:%d\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->comm.comm, event->comm.pid);
+
+	if (thread == NULL ||
+	    thread__set_comm(thread, event->comm.comm)) {
+		dump_printf("problem processing PERF_EVENT_COMM, skipping event.\n");
+		return -1;
+	}
+	total_comm++;
+
+	return 0;
+}
+
+static int
+process_sample_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	char level;
+	int show = 0;
+	struct dso *dso = NULL;
+	struct thread *thread;
+	u64 ip = event->ip.ip;
+	u64 period = 1;
+	void *more_data = event->ip.__more_data;
+	int cpumode;
+
+	thread = threads__findnew(event->ip.pid, &threads, &last_match);
+
+	if (sample_type & PERF_SAMPLE_PERIOD) {
+		period = *(u64 *)more_data;
+		more_data += sizeof(u64);
+	}
+
+	dump_printf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n",
+		(void *)(offset + head),
+		(void *)(long)(event->header.size),
+		event->header.misc,
+		event->ip.pid, event->ip.tid,
+		(void *)(long)ip,
+		(long long)period);
+
+	dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid);
+
+	if (thread == NULL) {
+		eprintf("problem processing %d event, skipping it.\n",
+			event->header.type);
+		return -1;
+	}
+
+	cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
+
+	if (cpumode == PERF_EVENT_MISC_KERNEL) {
+		show = SHOW_KERNEL;
+		level = 'k';
+
+		dso = kernel_dso;
+
+		dump_printf(" ...... dso: %s\n", dso->name);
+
+	} else if (cpumode == PERF_EVENT_MISC_USER) {
+
+		show = SHOW_USER;
+		level = '.';
+
+	} else {
+		show = SHOW_HV;
+		level = 'H';
+
+		dso = hypervisor_dso;
+
+		dump_printf(" ...... dso: [hypervisor]\n");
+	}
+
+	if (sample_type & PERF_SAMPLE_RAW) {
+		struct {
+			u32 size;
+			char data[0];
+		} *raw = more_data;
+
+		/*
+		 * FIXME: better resolve from pid from the struct trace_entry
+		 * field, although it should be the same than this perf
+		 * event pid
+		 */
+		print_event(0, raw->data, raw->size, 0, thread->comm);
+	}
+	total += period;
+
+	return 0;
+}
+
+static int
+process_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	trace_event(event);
+
+	switch (event->header.type) {
+	case PERF_EVENT_MMAP ... PERF_EVENT_LOST:
+		return 0;
+
+	case PERF_EVENT_COMM:
+		return process_comm_event(event, offset, head);
+
+	case PERF_EVENT_EXIT ... PERF_EVENT_READ:
+		return 0;
+
+	case PERF_EVENT_SAMPLE:
+		return process_sample_event(event, offset, head);
+
+	case PERF_EVENT_MAX:
+	default:
+		return -1;
+	}
+
+	return 0;
+}
+
+static int __cmd_trace(void)
+{
+	int ret, rc = EXIT_FAILURE;
+	unsigned long offset = 0;
+	unsigned long head = 0;
+	struct stat perf_stat;
+	event_t *event;
+	uint32_t size;
+	char *buf;
+
+	trace_report();
+
+	input = open(input_name, O_RDONLY);
+	if (input < 0) {
+		perror("failed to open file");
+		exit(-1);
+	}
+
+	ret = fstat(input, &perf_stat);
+	if (ret < 0) {
+		perror("failed to stat file");
+		exit(-1);
+	}
+
+	if (!perf_stat.st_size) {
+		fprintf(stderr, "zero-sized file, nothing to do!\n");
+		exit(0);
+	}
+	header = perf_header__read(input);
+	sample_type = perf_header__sample_type(header);
+
+	if (!(sample_type & PERF_SAMPLE_RAW))
+		die("No trace sample to read. Did you call perf record "
+		    "without -R?");
+
+	if (load_kernel() < 0) {
+		perror("failed to load kernel symbols");
+		return EXIT_FAILURE;
+	}
+
+remap:
+	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
+			   MAP_SHARED, input, offset);
+	if (buf == MAP_FAILED) {
+		perror("failed to mmap file");
+		exit(-1);
+	}
+
+more:
+	event = (event_t *)(buf + head);
+
+	size = event->header.size;
+	if (!size)
+		size = 8;
+
+	if (head + event->header.size >= page_size * mmap_window) {
+		unsigned long shift = page_size * (head / page_size);
+		int res;
+
+		res = munmap(buf, page_size * mmap_window);
+		assert(res == 0);
+
+		offset += shift;
+		head -= shift;
+		goto remap;
+	}
+
+	size = event->header.size;
+
+
+	if (!size || process_event(event, offset, head) < 0) {
+
+		/*
+		 * assume we lost track of the stream, check alignment, and
+		 * increment a single u64 in the hope to catch on again 'soon'.
+		 */
+
+		if (unlikely(head & 7))
+			head &= ~7ULL;
+
+		size = 8;
+	}
+
+	head += size;
+
+	if (offset + head < (unsigned long)perf_stat.st_size)
+		goto more;
+
+	rc = EXIT_SUCCESS;
+	close(input);
+
+	return rc;
+}
+
+static const char * const annotate_usage[] = {
+	"perf trace [<options>] <command>",
+	NULL
+};
+
+static const struct option options[] = {
+	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
+		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN('v', "verbose", &verbose,
+		    "be more verbose (show symbol address, etc)"),
+	OPT_END()
+};
+
+int cmd_trace(int argc, const char **argv, const char *prefix __used)
+{
+	symbol__init();
+	page_size = getpagesize();
+
+	argc = parse_options(argc, argv, options, annotate_usage, 0);
+	if (argc) {
+		/*
+		 * Special case: if there's an argument left then assume tha
+		 * it's a symbol filter:
+		 */
+		if (argc > 1)
+			usage_with_options(annotate_usage, options);
+	}
+
+
+	setup_pager();
+
+	return __cmd_trace();
+}

tools/perf/perf.c

@@ -292,6 +292,7 @@ static void handle_internal_command(int argc, const char **argv)
 		{ "top", cmd_top, 0 },
 		{ "annotate", cmd_annotate, 0 },
 		{ "version", cmd_version, 0 },
+		{ "trace", cmd_trace, 0 },
 	};
 	unsigned int i;
 	static const char ext[] = STRIP_EXTENSION;

tools/perf/util/parse-events.c

@@ -158,9 +158,9 @@ int valid_debugfs_mount(const char *debugfs)
 	return 0;
 }
 
-static const char *tracepoint_id_to_name(u64 config)
+struct tracepoint_path *tracepoint_id_to_path(u64 config)
 {
-	static char tracepoint_name[2 * MAX_EVENT_LENGTH];
+	struct tracepoint_path *path = NULL;
 	DIR *sys_dir, *evt_dir;
 	struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent;
 	struct stat st;
@@ -170,7 +170,7 @@ static const char *tracepoint_id_to_name(u64 config)
 	char evt_path[MAXPATHLEN];
 
 	if (valid_debugfs_mount(debugfs_path))
-		return "unkown";
+		return NULL;
 
 	sys_dir = opendir(debugfs_path);
 	if (!sys_dir)
@@ -197,10 +197,23 @@ static const char *tracepoint_id_to_name(u64 config)
 			if (id == config) {
 				closedir(evt_dir);
 				closedir(sys_dir);
-				snprintf(tracepoint_name, 2 * MAX_EVENT_LENGTH,
-					"%s:%s", sys_dirent.d_name,
-					evt_dirent.d_name);
-				return tracepoint_name;
+				path = calloc(1, sizeof(path));
+				path->system = malloc(MAX_EVENT_LENGTH);
+				if (!path->system) {
+					free(path);
+					return NULL;
+				}
+				path->name = malloc(MAX_EVENT_LENGTH);
+				if (!path->name) {
+					free(path->system);
+					free(path);
+					return NULL;
+				}
+				strncpy(path->system, sys_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				strncpy(path->name, evt_dirent.d_name,
+					MAX_EVENT_LENGTH);
+				return path;
 			}
 		}
 		closedir(evt_dir);
@@ -208,7 +221,25 @@ static const char *tracepoint_id_to_name(u64 config)
 
 cleanup:
 	closedir(sys_dir);
-	return "unkown";
+	return NULL;
+}
+
+#define TP_PATH_LEN (MAX_EVENT_LENGTH * 2 + 1)
+static const char *tracepoint_id_to_name(u64 config)
+{
+	static char buf[TP_PATH_LEN];
+	struct tracepoint_path *path;
+
+	path = tracepoint_id_to_path(config);
+	if (path) {
+		snprintf(buf, TP_PATH_LEN, "%s:%s", path->system, path->name);
+		free(path->name);
+		free(path->system);
+		free(path);
+	} else
+		snprintf(buf, TP_PATH_LEN, "%s:%s", "unknown", "unknown");
+
+	return buf;
 }
 
 static int is_cache_op_valid(u8 cache_type, u8 cache_op)

tools/perf/util/parse-events.h

@@ -1,10 +1,19 @@
-
+#ifndef _PARSE_EVENTS_H
+#define _PARSE_EVENTS_H
 /*
  * Parse symbolic events/counts passed in as options:
  */
 
 struct option;
 
+struct tracepoint_path {
+	char *system;
+	char *name;
+	struct tracepoint_path *next;
+};
+
+extern struct tracepoint_path *tracepoint_id_to_path(u64 config);
+
 extern int			nr_counters;
 
 extern struct perf_counter_attr attrs[MAX_COUNTERS];
@@ -21,3 +30,5 @@ extern void print_events(void);
 extern char debugfs_path[];
 extern int valid_debugfs_mount(const char *debugfs);
 
+
+#endif /* _PARSE_EVENTS_H */

tools/perf/util/trace-event.h

@@ -0,0 +1,239 @@
+#ifndef _TRACE_EVENTS_H
+#define _TRACE_EVENTS_H
+
+#include "parse-events.h"
+
+#define __unused __attribute__((unused))
+
+
+#ifndef PAGE_MASK
+#define PAGE_MASK (page_size - 1)
+#endif
+
+enum {
+	RINGBUF_TYPE_PADDING		= 29,
+	RINGBUF_TYPE_TIME_EXTEND	= 30,
+	RINGBUF_TYPE_TIME_STAMP		= 31,
+};
+
+#ifndef TS_SHIFT
+#define TS_SHIFT		27
+#endif
+
+#define NSECS_PER_SEC		1000000000ULL
+#define NSECS_PER_USEC		1000ULL
+
+enum format_flags {
+	FIELD_IS_ARRAY		= 1,
+	FIELD_IS_POINTER	= 2,
+};
+
+struct format_field {
+	struct format_field	*next;
+	char			*type;
+	char			*name;
+	int			offset;
+	int			size;
+	unsigned long		flags;
+};
+
+struct format {
+	int			nr_common;
+	int			nr_fields;
+	struct format_field	*common_fields;
+	struct format_field	*fields;
+};
+
+struct print_arg_atom {
+	char			*atom;
+};
+
+struct print_arg_string {
+	char			*string;
+};
+
+struct print_arg_field {
+	char			*name;
+	struct format_field	*field;
+};
+
+struct print_flag_sym {
+	struct print_flag_sym	*next;
+	char			*value;
+	char			*str;
+};
+
+struct print_arg_typecast {
+	char 			*type;
+	struct print_arg	*item;
+};
+
+struct print_arg_flags {
+	struct print_arg	*field;
+	char			*delim;
+	struct print_flag_sym	*flags;
+};
+
+struct print_arg_symbol {
+	struct print_arg	*field;
+	struct print_flag_sym	*symbols;
+};
+
+struct print_arg;
+
+struct print_arg_op {
+	char			*op;
+	int			prio;
+	struct print_arg	*left;
+	struct print_arg	*right;
+};
+
+struct print_arg_func {
+	char			*name;
+	struct print_arg	*args;
+};
+
+enum print_arg_type {
+	PRINT_NULL,
+	PRINT_ATOM,
+	PRINT_FIELD,
+	PRINT_FLAGS,
+	PRINT_SYMBOL,
+	PRINT_TYPE,
+	PRINT_STRING,
+	PRINT_OP,
+};
+
+struct print_arg {
+	struct print_arg		*next;
+	enum print_arg_type		type;
+	union {
+		struct print_arg_atom		atom;
+		struct print_arg_field		field;
+		struct print_arg_typecast	typecast;
+		struct print_arg_flags		flags;
+		struct print_arg_symbol		symbol;
+		struct print_arg_func		func;
+		struct print_arg_string		string;
+		struct print_arg_op		op;
+	};
+};
+
+struct print_fmt {
+	char			*format;
+	struct print_arg	*args;
+};
+
+struct event {
+	struct event		*next;
+	char			*name;
+	int			id;
+	int			flags;
+	struct format		format;
+	struct print_fmt	print_fmt;
+};
+
+enum {
+	EVENT_FL_ISFTRACE	= 1,
+	EVENT_FL_ISPRINT	= 2,
+	EVENT_FL_ISBPRINT	= 4,
+	EVENT_FL_ISFUNC		= 8,
+	EVENT_FL_ISFUNCENT	= 16,
+	EVENT_FL_ISFUNCRET	= 32,
+};
+
+struct record {
+	unsigned long long ts;
+	int size;
+	void *data;
+};
+
+struct record *trace_peek_data(int cpu);
+struct record *trace_read_data(int cpu);
+
+void parse_set_info(int nr_cpus, int long_sz);
+
+void trace_report(void);
+
+void *malloc_or_die(unsigned int size);
+
+void parse_cmdlines(char *file, int size);
+void parse_proc_kallsyms(char *file, unsigned int size);
+void parse_ftrace_printk(char *file, unsigned int size);
+
+void print_funcs(void);
+void print_printk(void);
+
+int parse_ftrace_file(char *buf, unsigned long size);
+int parse_event_file(char *buf, unsigned long size, char *system);
+void print_event(int cpu, void *data, int size, unsigned long long nsecs,
+		  char *comm);
+
+extern int file_bigendian;
+extern int host_bigendian;
+
+int bigendian(void);
+
+static inline unsigned short __data2host2(unsigned short data)
+{
+	unsigned short swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 8) |
+		((data & (0xffULL << 8)) >> 8);
+
+	return swap;
+}
+
+static inline unsigned int __data2host4(unsigned int data)
+{
+	unsigned int swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 24) |
+		((data & (0xffULL << 8)) << 8) |
+		((data & (0xffULL << 16)) >> 8) |
+		((data & (0xffULL << 24)) >> 24);
+
+	return swap;
+}
+
+static inline unsigned long long __data2host8(unsigned long long data)
+{
+	unsigned long long swap;
+
+	if (host_bigendian == file_bigendian)
+		return data;
+
+	swap = ((data & 0xffULL) << 56) |
+		((data & (0xffULL << 8)) << 40) |
+		((data & (0xffULL << 16)) << 24) |
+		((data & (0xffULL << 24)) << 8) |
+		((data & (0xffULL << 32)) >> 8) |
+		((data & (0xffULL << 40)) >> 24) |
+		((data & (0xffULL << 48)) >> 40) |
+		((data & (0xffULL << 56)) >> 56);
+
+	return swap;
+}
+
+#define data2host2(ptr)		__data2host2(*(unsigned short *)ptr)
+#define data2host4(ptr)		__data2host4(*(unsigned int *)ptr)
+#define data2host8(ptr)		__data2host8(*(unsigned long long *)ptr)
+
+extern int header_page_ts_offset;
+extern int header_page_ts_size;
+extern int header_page_size_offset;
+extern int header_page_size_size;
+extern int header_page_data_offset;
+extern int header_page_data_size;
+
+int parse_header_page(char *buf, unsigned long size);
+
+void read_tracing_data(struct perf_counter_attr *pattrs, int nb_counters);
+
+#endif /* _TRACE_EVENTS_H */

tools/perf/util/util.h

@@ -307,6 +307,7 @@ static inline int has_extension(const char *filename, const char *ext)
 #undef isspace
 #undef isdigit
 #undef isalpha
+#undef isprint
 #undef isalnum
 #undef tolower
 #undef toupper