X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=tools%2Fperf%2Fbuiltin-record.c;h=a1b99eeac3c0e9887af395a969d781865aab8905;hb=84b13fd596522db47f9545d5124c30cc00dfdf5a;hp=718b8f7b6aac005e34e912fea880968d395d899d;hpb=1fe2c1066ce6a30bda7b27785ee3d9b8e62ffbbd;p=safe%2Fjmp%2Flinux-2.6

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 718b8f7..a1b99ee 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -5,59 +5,79 @@
  * (or a CPU, or a PID) into the perf.data output file - for
  * later analysis via perf report.
  */
+#define _FILE_OFFSET_BITS 64
+
 #include "builtin.h"
 
 #include "perf.h"
 
+#include "util/build-id.h"
 #include "util/util.h"
 #include "util/parse-options.h"
 #include "util/parse-events.h"
-#include "util/string.h"
 
 #include "util/header.h"
+#include "util/event.h"
+#include "util/debug.h"
+#include "util/session.h"
+#include "util/symbol.h"
+#include "util/cpumap.h"
 
 #include <unistd.h>
 #include <sched.h>
 
-#define ALIGN(x, a)		__ALIGN_MASK(x, (typeof(x))(a)-1)
-#define __ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
+enum write_mode_t {
+	WRITE_FORCE,
+	WRITE_APPEND
+};
 
-static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
+static int			*fd[MAX_NR_CPUS][MAX_COUNTERS];
 
-static long			default_interval		= 100000;
+static unsigned int		user_interval 			= UINT_MAX;
+static long			default_interval		=      0;
 
-static int			nr_cpus				= 0;
+static int			nr_cpus				=      0;
 static unsigned int		page_size;
-static unsigned int		mmap_pages			= 128;
-static int			freq				= 0;
+static unsigned int		mmap_pages			=    128;
+static unsigned int		user_freq 			= UINT_MAX;
+static int			freq				=   1000;
 static int			output;
+static int			pipe_output			=      0;
 static const char		*output_name			= "perf.data";
-static int			group				= 0;
-static unsigned int		realtime_prio			= 0;
-static int			system_wide			= 0;
-static pid_t			target_pid			= -1;
-static int			inherit				= 1;
-static int			force				= 0;
-static int			append_file			= 0;
-static int			call_graph			= 0;
-static int			inherit_stat			= 0;
-static int			no_samples			= 0;
-static int			sample_address			= 0;
-
-static long			samples;
+static int			group				=      0;
+static unsigned int		realtime_prio			=      0;
+static bool			raw_samples			=  false;
+static bool			system_wide			=  false;
+static int			profile_cpu			=     -1;
+static pid_t			target_pid			=     -1;
+static pid_t			target_tid			=     -1;
+static pid_t			*all_tids			=      NULL;
+static int			thread_num			=      0;
+static pid_t			child_pid			=     -1;
+static bool			inherit				=   true;
+static enum write_mode_t	write_mode			= WRITE_FORCE;
+static bool			call_graph			=  false;
+static bool			inherit_stat			=  false;
+static bool			no_samples			=  false;
+static bool			sample_address			=  false;
+static bool			multiplex			=  false;
+static int			multiplex_fd			=     -1;
+
+static long			samples				=      0;
 static struct timeval		last_read;
 static struct timeval		this_read;
 
-static u64			bytes_written;
+static u64			bytes_written			=      0;
 
-static struct pollfd		event_array[MAX_NR_CPUS * MAX_COUNTERS];
+static struct pollfd		*event_array;
 
-static int			nr_poll;
-static int			nr_cpu;
+static int			nr_poll				=      0;
+static int			nr_cpu				=      0;
 
-static int			file_new = 1;
+static int			file_new			=      1;
+static off_t			post_processing_offset;
 
-struct perf_header		*header;
+static struct perf_session	*session;
 
 struct mmap_data {
 	int			counter;
@@ -66,11 +86,11 @@ struct mmap_data {
 	unsigned int		prev;
 };
 
-static struct mmap_data		mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
+static struct mmap_data		*mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
 
 static unsigned long mmap_read_head(struct mmap_data *md)
 {
-	struct perf_counter_mmap_page *pc = md->base;
+	struct perf_event_mmap_page *pc = md->base;
 	long head;
 
 	head = pc->data_head;
@@ -81,7 +101,7 @@ static unsigned long mmap_read_head(struct mmap_data *md)
 
 static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
 {
-	struct perf_counter_mmap_page *pc = md->base;
+	struct perf_event_mmap_page *pc = md->base;
 
 	/*
 	 * ensure all reads are done before we write the tail out.
@@ -90,6 +110,11 @@ static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
 	pc->data_tail = tail;
 }
 
+static void advance_output(size_t size)
+{
+	bytes_written += size;
+}
+
 static void write_output(void *buf, size_t size)
 {
 	while (size) {
@@ -105,6 +130,13 @@ static void write_output(void *buf, size_t size)
 	}
 }
 
+static int process_synthesized_event(event_t *event,
+				     struct perf_session *self __used)
+{
+	write_output(event, event->header.size);
+	return 0;
+}
+
 static void mmap_read(struct mmap_data *md)
 {
 	unsigned int head = mmap_read_head(md);
@@ -177,6 +209,9 @@ static void sig_handler(int sig)
 
 static void sig_atexit(void)
 {
+	if (child_pid != -1)
+		kill(child_pid, SIGTERM);
+
 	if (signr == -1)
 		return;
 
@@ -184,184 +219,34 @@ static void sig_atexit(void)
 	kill(getpid(), signr);
 }
 
-static void pid_synthesize_comm_event(pid_t pid, int full)
-{
-	struct comm_event comm_ev;
-	char filename[PATH_MAX];
-	char bf[BUFSIZ];
-	int fd;
-	size_t size;
-	char *field, *sep;
-	DIR *tasks;
-	struct dirent dirent, *next;
-
-	snprintf(filename, sizeof(filename), "/proc/%d/stat", pid);
-
-	fd = open(filename, O_RDONLY);
-	if (fd < 0) {
-		/*
-		 * We raced with a task exiting - just return:
-		 */
-		if (verbose)
-			fprintf(stderr, "couldn't open %s\n", filename);
-		return;
-	}
-	if (read(fd, bf, sizeof(bf)) < 0) {
-		fprintf(stderr, "couldn't read %s\n", filename);
-		exit(EXIT_FAILURE);
-	}
-	close(fd);
-
-	/* 9027 (cat) R 6747 9027 6747 34816 9027 ... */
-	memset(&comm_ev, 0, sizeof(comm_ev));
-	field = strchr(bf, '(');
-	if (field == NULL)
-		goto out_failure;
-	sep = strchr(++field, ')');
-	if (sep == NULL)
-		goto out_failure;
-	size = sep - field;
-	memcpy(comm_ev.comm, field, size++);
-
-	comm_ev.pid = pid;
-	comm_ev.header.type = PERF_EVENT_COMM;
-	size = ALIGN(size, sizeof(u64));
-	comm_ev.header.size = sizeof(comm_ev) - (sizeof(comm_ev.comm) - size);
-
-	if (!full) {
-		comm_ev.tid = pid;
-
-		write_output(&comm_ev, comm_ev.header.size);
-		return;
-	}
-
-	snprintf(filename, sizeof(filename), "/proc/%d/task", pid);
-
-	tasks = opendir(filename);
-	while (!readdir_r(tasks, &dirent, &next) && next) {
-		char *end;
-		pid = strtol(dirent.d_name, &end, 10);
-		if (*end)
-			continue;
-
-		comm_ev.tid = pid;
-
-		write_output(&comm_ev, comm_ev.header.size);
-	}
-	closedir(tasks);
-	return;
-
-out_failure:
-	fprintf(stderr, "couldn't get COMM and pgid, malformed %s\n",
-		filename);
-	exit(EXIT_FAILURE);
-}
-
-static void pid_synthesize_mmap_samples(pid_t pid)
-{
-	char filename[PATH_MAX];
-	FILE *fp;
-
-	snprintf(filename, sizeof(filename), "/proc/%d/maps", pid);
-
-	fp = fopen(filename, "r");
-	if (fp == NULL) {
-		/*
-		 * We raced with a task exiting - just return:
-		 */
-		if (verbose)
-			fprintf(stderr, "couldn't open %s\n", filename);
-		return;
-	}
-	while (1) {
-		char bf[BUFSIZ], *pbf = bf;
-		struct mmap_event mmap_ev = {
-			.header = { .type = PERF_EVENT_MMAP },
-		};
-		int n;
-		size_t size;
-		if (fgets(bf, sizeof(bf), fp) == NULL)
-			break;
-
-		/* 00400000-0040c000 r-xp 00000000 fd:01 41038  /bin/cat */
-		n = hex2u64(pbf, &mmap_ev.start);
-		if (n < 0)
-			continue;
-		pbf += n + 1;
-		n = hex2u64(pbf, &mmap_ev.len);
-		if (n < 0)
-			continue;
-		pbf += n + 3;
-		if (*pbf == 'x') { /* vm_exec */
-			char *execname = strchr(bf, '/');
-
-			/* Catch VDSO */
-			if (execname == NULL)
-				execname = strstr(bf, "[vdso]");
-
-			if (execname == NULL)
-				continue;
-
-			size = strlen(execname);
-			execname[size - 1] = '\0'; /* Remove \n */
-			memcpy(mmap_ev.filename, execname, size);
-			size = ALIGN(size, sizeof(u64));
-			mmap_ev.len -= mmap_ev.start;
-			mmap_ev.header.size = (sizeof(mmap_ev) -
-					       (sizeof(mmap_ev.filename) - size));
-			mmap_ev.pid = pid;
-			mmap_ev.tid = pid;
-
-			write_output(&mmap_ev, mmap_ev.header.size);
-		}
-	}
-
-	fclose(fp);
-}
-
-static void synthesize_all(void)
-{
-	DIR *proc;
-	struct dirent dirent, *next;
-
-	proc = opendir("/proc");
-
-	while (!readdir_r(proc, &dirent, &next) && next) {
-		char *end;
-		pid_t pid;
-
-		pid = strtol(dirent.d_name, &end, 10);
-		if (*end) /* only interested in proper numerical dirents */
-			continue;
-
-		pid_synthesize_comm_event(pid, 1);
-		pid_synthesize_mmap_samples(pid);
-	}
-
-	closedir(proc);
-}
-
 static int group_fd;
 
-static struct perf_header_attr *get_header_attr(struct perf_counter_attr *a, int nr)
+static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int nr)
 {
 	struct perf_header_attr *h_attr;
 
-	if (nr < header->attrs) {
-		h_attr = header->attr[nr];
+	if (nr < session->header.attrs) {
+		h_attr = session->header.attr[nr];
 	} else {
 		h_attr = perf_header_attr__new(a);
-		perf_header__add_attr(header, h_attr);
+		if (h_attr != NULL)
+			if (perf_header__add_attr(&session->header, h_attr) < 0) {
+				perf_header_attr__delete(h_attr);
+				h_attr = NULL;
+			}
 	}
 
 	return h_attr;
 }
 
-static void create_counter(int counter, int cpu, pid_t pid)
+static void create_counter(int counter, int cpu)
 {
-	struct perf_counter_attr *attr = attrs + counter;
+	char *filter = filters[counter];
+	struct perf_event_attr *attr = attrs + counter;
 	struct perf_header_attr *h_attr;
 	int track = !counter; /* only the first counter needs these */
+	int thread_index;
+	int ret;
 	struct {
 		u64 count;
 		u64 time_enabled;
@@ -373,12 +258,24 @@ static void create_counter(int counter, int cpu, pid_t pid)
 				  PERF_FORMAT_TOTAL_TIME_RUNNING |
 				  PERF_FORMAT_ID;
 
-	attr->sample_type	= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
+	attr->sample_type	|= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
+
+	if (nr_counters > 1)
+		attr->sample_type |= PERF_SAMPLE_ID;
 
-	if (freq) {
-		attr->sample_type	|= PERF_SAMPLE_PERIOD;
-		attr->freq		= 1;
-		attr->sample_freq	= freq;
+	/*
+	 * We default some events to a 1 default interval. But keep
+	 * it a weak assumption overridable by the user.
+	 */
+	if (!attr->sample_period || (user_freq != UINT_MAX &&
+				     user_interval != UINT_MAX)) {
+		if (freq) {
+			attr->sample_type	|= PERF_SAMPLE_PERIOD;
+			attr->freq		= 1;
+			attr->sample_freq	= freq;
+		} else {
+			attr->sample_period = default_interval;
+		}
 	}
 
 	if (no_samples)
@@ -393,100 +290,170 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	if (call_graph)
 		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;
 
+	if (raw_samples) {
+		attr->sample_type	|= PERF_SAMPLE_TIME;
+		attr->sample_type	|= PERF_SAMPLE_RAW;
+		attr->sample_type	|= PERF_SAMPLE_CPU;
+	}
 
 	attr->mmap		= track;
 	attr->comm		= track;
-	attr->inherit		= (cpu < 0) && inherit;
-	attr->disabled		= 1;
+	attr->inherit		= inherit;
+	if (target_pid == -1 && !system_wide) {
+		attr->disabled = 1;
+		attr->enable_on_exec = 1;
+	}
 
+	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
-	fd[nr_cpu][counter] = sys_perf_counter_open(attr, pid, cpu, group_fd, 0);
+		fd[nr_cpu][counter][thread_index] = sys_perf_event_open(attr,
+				all_tids[thread_index], cpu, group_fd, 0);
+
+		if (fd[nr_cpu][counter][thread_index] < 0) {
+			int err = errno;
+
+			if (err == EPERM || err == EACCES)
+				die("Permission error - are you root?\n"
+					"\t Consider tweaking"
+					" /proc/sys/kernel/perf_event_paranoid.\n");
+			else if (err ==  ENODEV && profile_cpu != -1) {
+				die("No such device - did you specify"
+					" an out-of-range profile CPU?\n");
+			}
 
-	if (fd[nr_cpu][counter] < 0) {
-		int err = errno;
+			/*
+			 * If it's cycles then fall back to hrtimer
+			 * based cpu-clock-tick sw counter, which
+			 * is always available even if no PMU support:
+			 */
+			if (attr->type == PERF_TYPE_HARDWARE
+					&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
+
+				if (verbose)
+					warning(" ... trying to fall back to cpu-clock-ticks\n");
+				attr->type = PERF_TYPE_SOFTWARE;
+				attr->config = PERF_COUNT_SW_CPU_CLOCK;
+				goto try_again;
+			}
+			printf("\n");
+			error("perfcounter syscall returned with %d (%s)\n",
+					fd[nr_cpu][counter][thread_index], strerror(err));
+
+#if defined(__i386__) || defined(__x86_64__)
+			if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP)
+				die("No hardware sampling interrupt available."
+				    " No APIC? If so then you can boot the kernel"
+				    " with the \"lapic\" boot parameter to"
+				    " force-enable it.\n");
+#endif
+
+			die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
+			exit(-1);
+		}
 
-		if (err == EPERM)
-			die("Permission error - are you root?\n");
+		h_attr = get_header_attr(attr, counter);
+		if (h_attr == NULL)
+			die("nomem\n");
 
-		/*
-		 * If it's cycles then fall back to hrtimer
-		 * based cpu-clock-tick sw counter, which
-		 * is always available even if no PMU support:
-		 */
-		if (attr->type == PERF_TYPE_HARDWARE
-			&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
-
-			if (verbose)
-				warning(" ... trying to fall back to cpu-clock-ticks\n");
-			attr->type = PERF_TYPE_SOFTWARE;
-			attr->config = PERF_COUNT_SW_CPU_CLOCK;
-			goto try_again;
+		if (!file_new) {
+			if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
+				fprintf(stderr, "incompatible append\n");
+				exit(-1);
+			}
 		}
-		printf("\n");
-		error("perfcounter syscall returned with %d (%s)\n",
-			fd[nr_cpu][counter], strerror(err));
-		die("No CONFIG_PERF_COUNTERS=y kernel support configured?\n");
-		exit(-1);
-	}
 
-	h_attr = get_header_attr(attr, counter);
-
-	if (!file_new) {
-		if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
-			fprintf(stderr, "incompatible append\n");
+		if (read(fd[nr_cpu][counter][thread_index], &read_data, sizeof(read_data)) == -1) {
+			perror("Unable to read perf file descriptor\n");
 			exit(-1);
 		}
-	}
 
-	if (read(fd[nr_cpu][counter], &read_data, sizeof(read_data)) == -1) {
-		perror("Unable to read perf file descriptor\n");
-		exit(-1);
-	}
+		if (perf_header_attr__add_id(h_attr, read_data.id) < 0) {
+			pr_warning("Not enough memory to add id\n");
+			exit(-1);
+		}
 
-	perf_header_attr__add_id(h_attr, read_data.id);
+		assert(fd[nr_cpu][counter][thread_index] >= 0);
+		fcntl(fd[nr_cpu][counter][thread_index], F_SETFL, O_NONBLOCK);
 
-	assert(fd[nr_cpu][counter] >= 0);
-	fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);
+		/*
+		 * First counter acts as the group leader:
+		 */
+		if (group && group_fd == -1)
+			group_fd = fd[nr_cpu][counter][thread_index];
+		if (multiplex && multiplex_fd == -1)
+			multiplex_fd = fd[nr_cpu][counter][thread_index];
+
+		if (multiplex && fd[nr_cpu][counter][thread_index] != multiplex_fd) {
+
+			ret = ioctl(fd[nr_cpu][counter][thread_index], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd);
+			assert(ret != -1);
+		} else {
+			event_array[nr_poll].fd = fd[nr_cpu][counter][thread_index];
+			event_array[nr_poll].events = POLLIN;
+			nr_poll++;
+
+			mmap_array[nr_cpu][counter][thread_index].counter = counter;
+			mmap_array[nr_cpu][counter][thread_index].prev = 0;
+			mmap_array[nr_cpu][counter][thread_index].mask = mmap_pages*page_size - 1;
+			mmap_array[nr_cpu][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size,
+				PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter][thread_index], 0);
+			if (mmap_array[nr_cpu][counter][thread_index].base == MAP_FAILED) {
+				error("failed to mmap with %d (%s)\n", errno, strerror(errno));
+				exit(-1);
+			}
+		}
 
-	/*
-	 * First counter acts as the group leader:
-	 */
-	if (group && group_fd == -1)
-		group_fd = fd[nr_cpu][counter];
-
-	event_array[nr_poll].fd = fd[nr_cpu][counter];
-	event_array[nr_poll].events = POLLIN;
-	nr_poll++;
-
-	mmap_array[nr_cpu][counter].counter = counter;
-	mmap_array[nr_cpu][counter].prev = 0;
-	mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
-	mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
-			PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
-	if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
-		error("failed to mmap with %d (%s)\n", errno, strerror(errno));
-		exit(-1);
+		if (filter != NULL) {
+			ret = ioctl(fd[nr_cpu][counter][thread_index],
+					PERF_EVENT_IOC_SET_FILTER, filter);
+			if (ret) {
+				error("failed to set filter with %d (%s)\n", errno,
+						strerror(errno));
+				exit(-1);
+			}
+		}
 	}
-
-	ioctl(fd[nr_cpu][counter], PERF_COUNTER_IOC_ENABLE);
 }
 
-static void open_counters(int cpu, pid_t pid)
+static void open_counters(int cpu)
 {
 	int counter;
 
 	group_fd = -1;
 	for (counter = 0; counter < nr_counters; counter++)
-		create_counter(counter, cpu, pid);
+		create_counter(counter, cpu);
 
 	nr_cpu++;
 }
 
+static int process_buildids(void)
+{
+	u64 size = lseek(output, 0, SEEK_CUR);
+
+	if (size == 0)
+		return 0;
+
+	session->fd = output;
+	return __perf_session__process_events(session, post_processing_offset,
+					      size - post_processing_offset,
+					      size, &build_id__mark_dso_hit_ops);
+}
+
 static void atexit_header(void)
 {
-	header->data_size += bytes_written;
+	if (!pipe_output) {
+		session->header.data_size += bytes_written;
 
-	perf_header__write(header, output);
+		process_buildids();
+		perf_header__write(&session->header, output, true);
+	} else {
+		int err;
+
+		err = event__synthesize_build_ids(process_synthesized_event,
+						  session);
+		if (err < 0)
+			pr_err("Couldn't synthesize build ids.\n");
+	}
 }
 
 static int __cmd_record(int argc, const char **argv)
@@ -495,102 +462,256 @@ static int __cmd_record(int argc, const char **argv)
 	struct stat st;
 	pid_t pid = 0;
 	int flags;
-	int ret;
+	int err;
+	unsigned long waking = 0;
+	int child_ready_pipe[2], go_pipe[2];
+	const bool forks = argc > 0;
+	char buf;
 
 	page_size = sysconf(_SC_PAGE_SIZE);
-	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-	assert(nr_cpus <= MAX_NR_CPUS);
-	assert(nr_cpus >= 0);
 
 	atexit(sig_atexit);
 	signal(SIGCHLD, sig_handler);
 	signal(SIGINT, sig_handler);
 
-	if (!stat(output_name, &st) && st.st_size) {
-		if (!force && !append_file) {
-			fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
-					output_name);
-			exit(-1);
+	if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) {
+		perror("failed to create pipes");
+		exit(-1);
+	}
+
+	if (!strcmp(output_name, "-"))
+		pipe_output = 1;
+	else if (!stat(output_name, &st) && st.st_size) {
+		if (write_mode == WRITE_FORCE) {
+			char oldname[PATH_MAX];
+			snprintf(oldname, sizeof(oldname), "%s.old",
+				 output_name);
+			unlink(oldname);
+			rename(output_name, oldname);
 		}
-	} else {
-		append_file = 0;
+	} else if (write_mode == WRITE_APPEND) {
+		write_mode = WRITE_FORCE;
 	}
 
 	flags = O_CREAT|O_RDWR;
-	if (append_file)
+	if (write_mode == WRITE_APPEND)
 		file_new = 0;
 	else
 		flags |= O_TRUNC;
 
-	output = open(output_name, flags, S_IRUSR|S_IWUSR);
+	if (pipe_output)
+		output = STDOUT_FILENO;
+	else
+		output = open(output_name, flags, S_IRUSR | S_IWUSR);
 	if (output < 0) {
 		perror("failed to create output file");
 		exit(-1);
 	}
 
-	if (!file_new)
-		header = perf_header__read(output);
-	else
-		header = perf_header__new();
+	session = perf_session__new(output_name, O_WRONLY,
+				    write_mode == WRITE_FORCE);
+	if (session == NULL) {
+		pr_err("Not enough memory for reading perf file header\n");
+		return -1;
+	}
+
+	if (!file_new) {
+		err = perf_header__read(session, output);
+		if (err < 0)
+			return err;
+	}
+
+	if (raw_samples) {
+		perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
+	} else {
+		for (i = 0; i < nr_counters; i++) {
+			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
+				perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
+				break;
+			}
+		}
+	}
 
 	atexit(atexit_header);
 
-	if (!system_wide) {
-		pid = target_pid;
-		if (pid == -1)
-			pid = getpid();
+	if (forks) {
+		child_pid = fork();
+		if (pid < 0) {
+			perror("failed to fork");
+			exit(-1);
+		}
 
-		open_counters(-1, pid);
-	} else for (i = 0; i < nr_cpus; i++)
-		open_counters(i, target_pid);
+		if (!child_pid) {
+			if (pipe_output)
+				dup2(2, 1);
+			close(child_ready_pipe[0]);
+			close(go_pipe[1]);
+			fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+			/*
+			 * Do a dummy execvp to get the PLT entry resolved,
+			 * so we avoid the resolver overhead on the real
+			 * execvp call.
+			 */
+			execvp("", (char **)argv);
+
+			/*
+			 * Tell the parent we're ready to go
+			 */
+			close(child_ready_pipe[1]);
+
+			/*
+			 * Wait until the parent tells us to go.
+			 */
+			if (read(go_pipe[0], &buf, 1) == -1)
+				perror("unable to read pipe");
+
+			execvp(argv[0], (char **)argv);
+
+			perror(argv[0]);
+			exit(-1);
+		}
 
-	if (file_new)
-		perf_header__write(header, output);
+		if (!system_wide && target_tid == -1 && target_pid == -1)
+			all_tids[0] = child_pid;
 
-	if (!system_wide) {
-		pid_synthesize_comm_event(pid, 0);
-		pid_synthesize_mmap_samples(pid);
-	} else
-		synthesize_all();
+		close(child_ready_pipe[1]);
+		close(go_pipe[0]);
+		/*
+		 * wait for child to settle
+		 */
+		if (read(child_ready_pipe[0], &buf, 1) == -1) {
+			perror("unable to read pipe");
+			exit(-1);
+		}
+		close(child_ready_pipe[0]);
+	}
 
-	if (target_pid == -1 && argc) {
-		pid = fork();
-		if (pid < 0)
-			perror("failed to fork");
+	if ((!system_wide && !inherit) || profile_cpu != -1) {
+		open_counters(profile_cpu);
+	} else {
+		nr_cpus = read_cpu_map();
+		for (i = 0; i < nr_cpus; i++)
+			open_counters(cpumap[i]);
+	}
 
-		if (!pid) {
-			if (execvp(argv[0], (char **)argv)) {
-				perror(argv[0]);
-				exit(-1);
-			}
+	if (pipe_output) {
+		err = perf_header__write_pipe(output);
+		if (err < 0)
+			return err;
+	} else if (file_new) {
+		err = perf_header__write(&session->header, output, false);
+		if (err < 0)
+			return err;
+	}
+
+	post_processing_offset = lseek(output, 0, SEEK_CUR);
+
+	if (pipe_output) {
+		err = event__synthesize_attrs(&session->header,
+					      process_synthesized_event,
+					      session);
+		if (err < 0) {
+			pr_err("Couldn't synthesize attrs.\n");
+			return err;
 		}
+
+		err = event__synthesize_event_types(process_synthesized_event,
+						    session);
+		if (err < 0) {
+			pr_err("Couldn't synthesize event_types.\n");
+			return err;
+		}
+
+		err = event__synthesize_tracing_data(output, attrs,
+						     nr_counters,
+						     process_synthesized_event,
+						     session);
+		if (err <= 0) {
+			pr_err("Couldn't record tracing data.\n");
+			return err;
+		}
+
+		advance_output(err);
 	}
 
+	err = event__synthesize_kernel_mmap(process_synthesized_event,
+					    session, "_text");
+	if (err < 0)
+		err = event__synthesize_kernel_mmap(process_synthesized_event,
+						    session, "_stext");
+	if (err < 0) {
+		pr_err("Couldn't record kernel reference relocation symbol.\n");
+		return err;
+	}
+
+	err = event__synthesize_modules(process_synthesized_event, session);
+	if (err < 0) {
+		pr_err("Couldn't record kernel reference relocation symbol.\n");
+		return err;
+	}
+
+	if (!system_wide && profile_cpu == -1)
+		event__synthesize_thread(target_tid, process_synthesized_event,
+					 session);
+	else
+		event__synthesize_threads(process_synthesized_event, session);
+
 	if (realtime_prio) {
 		struct sched_param param;
 
 		param.sched_priority = realtime_prio;
 		if (sched_setscheduler(0, SCHED_FIFO, &param)) {
-			printf("Could not set realtime priority.\n");
+			pr_err("Could not set realtime priority.\n");
 			exit(-1);
 		}
 	}
 
+	/*
+	 * Let the child rip
+	 */
+	if (forks)
+		close(go_pipe[1]);
+
 	for (;;) {
 		int hits = samples;
+		int thread;
 
 		for (i = 0; i < nr_cpu; i++) {
-			for (counter = 0; counter < nr_counters; counter++)
-				mmap_read(&mmap_array[i][counter]);
+			for (counter = 0; counter < nr_counters; counter++) {
+				for (thread = 0;
+					thread < thread_num; thread++) {
+					if (mmap_array[i][counter][thread].base)
+						mmap_read(&mmap_array[i][counter][thread]);
+				}
+
+			}
 		}
 
 		if (hits == samples) {
 			if (done)
 				break;
-			ret = poll(event_array, nr_poll, 100);
+			err = poll(event_array, nr_poll, -1);
+			waking++;
+		}
+
+		if (done) {
+			for (i = 0; i < nr_cpu; i++) {
+				for (counter = 0;
+					counter < nr_counters;
+					counter++) {
+					for (thread = 0;
+						thread < thread_num;
+						thread++)
+						ioctl(fd[i][counter][thread],
+							PERF_EVENT_IOC_DISABLE);
+				}
+			}
 		}
 	}
 
+	fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
+
 	/*
 	 * Approximate RIP event size: 24 bytes.
 	 */
@@ -609,33 +730,43 @@ static const char * const record_usage[] = {
 	NULL
 };
 
+static bool force, append_file;
+
 static const struct option options[] = {
 	OPT_CALLBACK('e', "event", NULL, "event",
 		     "event selector. use 'perf list' to list available events",
 		     parse_events),
+	OPT_CALLBACK(0, "filter", NULL, "filter",
+		     "event filter", parse_filter),
 	OPT_INTEGER('p', "pid", &target_pid,
-		    "record events on existing pid"),
+		    "record events on existing process id"),
+	OPT_INTEGER('t', "tid", &target_tid,
+		    "record events on existing thread id"),
 	OPT_INTEGER('r', "realtime", &realtime_prio,
 		    "collect data with this RT SCHED_FIFO priority"),
+	OPT_BOOLEAN('R', "raw-samples", &raw_samples,
+		    "collect raw sample records from all opened counters"),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
 			    "system-wide collection from all CPUs"),
 	OPT_BOOLEAN('A', "append", &append_file,
 			    "append to the output file to do incremental profiling"),
+	OPT_INTEGER('C', "profile_cpu", &profile_cpu,
+			    "CPU to profile on"),
 	OPT_BOOLEAN('f', "force", &force,
-			"overwrite existing data file"),
-	OPT_LONG('c', "count", &default_interval,
+			"overwrite existing data file (deprecated)"),
+	OPT_LONG('c', "count", &user_interval,
 		    "event period to sample"),
 	OPT_STRING('o', "output", &output_name, "file",
 		    "output file name"),
 	OPT_BOOLEAN('i', "inherit", &inherit,
 		    "child tasks inherit counters"),
-	OPT_INTEGER('F', "freq", &freq,
+	OPT_INTEGER('F', "freq", &user_freq,
 		    "profile at this frequency"),
 	OPT_INTEGER('m', "mmap-pages", &mmap_pages,
 		    "number of mmap data pages"),
 	OPT_BOOLEAN('g', "call-graph", &call_graph,
 		    "do call-graph (stack chain/backtrace) recording"),
-	OPT_BOOLEAN('v', "verbose", &verbose,
+	OPT_INCR('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_BOOLEAN('s', "stat", &inherit_stat,
 		    "per thread counts"),
@@ -643,29 +774,85 @@ static const struct option options[] = {
 		    "Sample addresses"),
 	OPT_BOOLEAN('n', "no-samples", &no_samples,
 		    "don't sample"),
+	OPT_BOOLEAN('M', "multiplex", &multiplex,
+		    "multiplex counter output in a single channel"),
 	OPT_END()
 };
 
 int cmd_record(int argc, const char **argv, const char *prefix __used)
 {
-	int counter;
+	int i,j;
 
 	argc = parse_options(argc, argv, options, record_usage,
-		PARSE_OPT_STOP_AT_NON_OPTION);
-	if (!argc && target_pid == -1 && !system_wide)
+			    PARSE_OPT_STOP_AT_NON_OPTION);
+	if (!argc && target_pid == -1 && target_tid == -1 &&
+		!system_wide && profile_cpu == -1)
 		usage_with_options(record_usage, options);
 
+	if (force && append_file) {
+		fprintf(stderr, "Can't overwrite and append at the same time."
+				" You need to choose between -f and -A");
+		usage_with_options(record_usage, options);
+	} else if (append_file) {
+		write_mode = WRITE_APPEND;
+	} else {
+		write_mode = WRITE_FORCE;
+	}
+
+	symbol__init();
+
 	if (!nr_counters) {
 		nr_counters	= 1;
 		attrs[0].type	= PERF_TYPE_HARDWARE;
 		attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
 	}
 
-	for (counter = 0; counter < nr_counters; counter++) {
-		if (attrs[counter].sample_period)
-			continue;
+	if (target_pid != -1) {
+		target_tid = target_pid;
+		thread_num = find_all_tid(target_pid, &all_tids);
+		if (thread_num <= 0) {
+			fprintf(stderr, "Can't find all threads of pid %d\n",
+					target_pid);
+			usage_with_options(record_usage, options);
+		}
+	} else {
+		all_tids=malloc(sizeof(pid_t));
+		if (!all_tids)
+			return -ENOMEM;
 
-		attrs[counter].sample_period = default_interval;
+		all_tids[0] = target_tid;
+		thread_num = 1;
+	}
+
+	for (i = 0; i < MAX_NR_CPUS; i++) {
+		for (j = 0; j < MAX_COUNTERS; j++) {
+			fd[i][j] = malloc(sizeof(int)*thread_num);
+			mmap_array[i][j] = zalloc(
+				sizeof(struct mmap_data)*thread_num);
+			if (!fd[i][j] || !mmap_array[i][j])
+				return -ENOMEM;
+		}
+	}
+	event_array = malloc(
+		sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num);
+	if (!event_array)
+		return -ENOMEM;
+
+	if (user_interval != UINT_MAX)
+		default_interval = user_interval;
+	if (user_freq != UINT_MAX)
+		freq = user_freq;
+
+	/*
+	 * User specified count overrides default frequency.
+	 */
+	if (default_interval)
+		freq = 0;
+	else if (freq) {
+		default_interval = freq;
+	} else {
+		fprintf(stderr, "frequency and count are zero, aborting\n");
+		exit(EXIT_FAILURE);
 	}
 
 	return __cmd_record(argc, argv);