8 files changed, 124 insertions, 71 deletions

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index cb9033d3f72f..68218cfd38b3 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -386,22 +386,29 @@ endif
 ifdef NO_DEMANGLE
         BASIC_CFLAGS += -DNO_DEMANGLE
 else
-
         has_bfd := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd > /dev/null 2>&1 && echo y")
 
-        has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y")
-
-        has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y")
-
         ifeq ($(has_bfd),y)
                 EXTLIBS += -lbfd
-        else ifeq ($(has_bfd_iberty),y)
-                EXTLIBS += -lbfd -liberty
-        else ifeq ($(has_bfd_iberty_z),y)
-                EXTLIBS += -lbfd -liberty -lz
         else
-                msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling)
-                BASIC_CFLAGS += -DNO_DEMANGLE
+                has_bfd_iberty := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y")
+                ifeq ($(has_bfd_iberty),y)
+                        EXTLIBS += -lbfd -liberty
+                else
+                        has_bfd_iberty_z := $(shell sh -c "(echo '\#include <bfd.h>'; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y")
+                        ifeq ($(has_bfd_iberty_z),y)
+                                EXTLIBS += -lbfd -liberty -lz
+                        else
+                                has_cplus_demangle := $(shell sh -c "(echo 'extern char *cplus_demangle(const char *, int);'; echo 'int main(void) { cplus_demangle(0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -liberty > /dev/null 2>&1 && echo y")
+                                ifeq ($(has_cplus_demangle),y)
+                                        EXTLIBS += -liberty
+                                        BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE
+                                else
+                                        msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling)
+                                        BASIC_CFLAGS += -DNO_DEMANGLE
+                                endif
+                        endif
+                endif
         endif
 endif
 
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
index f990fa8a35c9..d88c6961274c 100644
--- a/tools/perf/builtin-list.c
+++ b/tools/perf/builtin-list.c
@@ -10,11 +10,12 @@
 
 #include "perf.h"
 
-#include "util/parse-options.h"
 #include "util/parse-events.h"
+#include "util/cache.h"
 
 int cmd_list(int argc __used, const char **argv __used, const char *prefix __used)
 {
+        setup_pager();
         print_events();
         return 0;
 }
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 106c6abd1c38..65b4115e417d 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -35,7 +35,9 @@ static int			output;
 static const char               *output_name                    = "perf.data";
 static int                      group                           = 0;
 static unsigned int             realtime_prio                   = 0;
+static int                      raw_samples                     = 0;
 static int                      system_wide                     = 0;
+static int                      profile_cpu                     = -1;
 static pid_t                    target_pid                      = -1;
 static int                      inherit                         = 1;
 static int                      force                           = 0;
@@ -185,46 +187,48 @@ static void sig_atexit(void)
         kill(getpid(), signr);
 }
 
-static void pid_synthesize_comm_event(pid_t pid, int full)
+static pid_t 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;
+        FILE *fp;
+        size_t size = 0;
         DIR *tasks;
         struct dirent dirent, *next;
+        pid_t tgid = 0;
 
-        snprintf(filename, sizeof(filename), "/proc/%d/stat", pid);
+        snprintf(filename, sizeof(filename), "/proc/%d/status", pid);
 
-        fd = open(filename, O_RDONLY);
-        if (fd < 0) {
+        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;
+                return 0;
         }
-        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;
+        while (!comm_ev.comm[0] || !comm_ev.pid) {
+                if (fgets(bf, sizeof(bf), fp) == NULL)
+                        goto out_failure;
+
+                if (memcmp(bf, "Name:", 5) == 0) {
+                        char *name = bf + 5;
+                        while (*name && isspace(*name))
+                                ++name;
+                        size = strlen(name) - 1;
+                        memcpy(comm_ev.comm, name, size++);
+                } else if (memcmp(bf, "Tgid:", 5) == 0) {
+                        char *tgids = bf + 5;
+                        while (*tgids && isspace(*tgids))
+                                ++tgids;
+                        tgid = comm_ev.pid = atoi(tgids);
+                }
+        }
+
         comm_ev.header.type = PERF_EVENT_COMM;
         size = ALIGN(size, sizeof(u64));
         comm_ev.header.size = sizeof(comm_ev) - (sizeof(comm_ev.comm) - size);
@@ -233,7 +237,7 @@ static void pid_synthesize_comm_event(pid_t pid, int full)
                 comm_ev.tid = pid;
 
                 write_output(&comm_ev, comm_ev.header.size);
-                return;
+                goto out_fclose;
         }
 
         snprintf(filename, sizeof(filename), "/proc/%d/task", pid);
@@ -250,7 +254,10 @@ static void pid_synthesize_comm_event(pid_t pid, int full)
                 write_output(&comm_ev, comm_ev.header.size);
         }
         closedir(tasks);
-        return;
+
+out_fclose:
+        fclose(fp);
+        return tgid;
 
 out_failure:
         fprintf(stderr, "couldn't get COMM and pgid, malformed %s\n",
@@ -258,7 +265,7 @@ out_failure:
         exit(EXIT_FAILURE);
 }
 
-static void pid_synthesize_mmap_samples(pid_t pid)
+static void pid_synthesize_mmap_samples(pid_t pid, pid_t tgid)
 {
         char filename[PATH_MAX];
         FILE *fp;
@@ -310,7 +317,7 @@ static void pid_synthesize_mmap_samples(pid_t pid)
                         mmap_ev.len -= mmap_ev.start;
                         mmap_ev.header.size = (sizeof(mmap_ev) -
                                                (sizeof(mmap_ev.filename) - size));
-                        mmap_ev.pid = pid;
+                        mmap_ev.pid = tgid;
                         mmap_ev.tid = pid;
 
                         write_output(&mmap_ev, mmap_ev.header.size);
@@ -329,14 +336,14 @@ static void synthesize_all(void)
 
         while (!readdir_r(proc, &dirent, &next) && next) {
                 char *end;
-                pid_t pid;
+                pid_t pid, tgid;
 
                 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);
+                tgid = pid_synthesize_comm_event(pid, 1);
+                pid_synthesize_mmap_samples(pid, tgid);
         }
 
         closedir(proc);
@@ -374,7 +381,7 @@ 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 (freq) {
                 attr->sample_type       |= PERF_SAMPLE_PERIOD;
@@ -394,6 +401,8 @@ 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_RAW;
 
         attr->mmap              = track;
         attr->comm              = track;
@@ -408,6 +417,8 @@ try_again:
 
                 if (err == EPERM)
                         die("Permission error - are you root?\n");
+                else if (err ==  ENODEV && profile_cpu != -1)
+                        die("No such device - did you specify an out-of-range profile CPU?\n");
 
                 /*
                  * If it's cycles then fall back to hrtimer
@@ -541,16 +552,22 @@ static int __cmd_record(int argc, const char **argv)
                 if (pid == -1)
                         pid = getpid();
 
-                open_counters(-1, pid);
-        } else for (i = 0; i < nr_cpus; i++)
-                open_counters(i, target_pid);
+                open_counters(profile_cpu, pid);
+        } else {
+                if (profile_cpu != -1) {
+                        open_counters(profile_cpu, target_pid);
+                } else {
+                        for (i = 0; i < nr_cpus; i++)
+                                open_counters(i, target_pid);
+                }
+        }
 
         if (file_new)
                 perf_header__write(header, output);
 
         if (!system_wide) {
-                pid_synthesize_comm_event(pid, 0);
-                pid_synthesize_mmap_samples(pid);
+                pid_t tgid = pid_synthesize_comm_event(pid, 0);
+                pid_synthesize_mmap_samples(pid, tgid);
         } else
                 synthesize_all();
 
@@ -618,10 +635,14 @@ static const struct option options[] = {
                     "record events on existing pid"),
         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,
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 93945ecdac86..6321951fe1bf 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -1276,11 +1276,11 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
                 more_data += sizeof(u64);
         }
 
-        dprintf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d: %p period: %Ld\n",
+        dprintf("%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.pid, event->ip.tid,
                 (void *)(long)ip,
                 (long long)period);
 
@@ -1340,10 +1340,11 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
         if (show & show_mask) {
                 struct symbol *sym = resolve_symbol(thread, &map, &dso, &ip);
 
-                if (dso_list && dso && dso->name && !strlist__has_entry(dso_list, dso->name))
+                if (dso_list && (!dso || !dso->name ||
+                                 !strlist__has_entry(dso_list, dso->name)))
                         return 0;
 
-                if (sym_list && sym && !strlist__has_entry(sym_list, sym->name))
+                if (sym_list && (!sym || !strlist__has_entry(sym_list, sym->name)))
                         return 0;
 
                 if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) {
@@ -1362,10 +1363,11 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head)
         struct thread *thread = threads__findnew(event->mmap.pid);
         struct map *map = map__new(&event->mmap, cwd, cwdlen);
 
-        dprintf("%p [%p]: PERF_EVENT_MMAP %d: [%p(%p) @ %p]: %s\n",
+        dprintf("%p [%p]: PERF_EVENT_MMAP %d/%d: [%p(%p) @ %p]: %s\n",
                 (void *)(offset + head),
                 (void *)(long)(event->header.size),
                 event->mmap.pid,
+                event->mmap.tid,
                 (void *)(long)event->mmap.start,
                 (void *)(long)event->mmap.len,
                 (void *)(long)event->mmap.pgoff,
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index a8e67aa9ef49..011473411642 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -13,6 +13,7 @@
 #include <stdio.h>
 #include <stdbool.h>
 #include <errno.h>
+#include <math.h>
 
 #include "callchain.h"
 
@@ -112,7 +113,7 @@ static void __sort_chain_graph_rel(struct callchain_node *node,
         u64 min_hit;
 
         node->rb_root = RB_ROOT;
-        min_hit = node->children_hit * min_percent / 100.0;
+        min_hit = ceil(node->children_hit * min_percent);
 
         chain_for_each_child(child, node) {
                 __sort_chain_graph_rel(child, min_percent);
@@ -126,7 +127,7 @@ static void
 sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root,
                      u64 min_hit __used, struct callchain_param *param)
 {
-        __sort_chain_graph_rel(chain_root, param->min_percent);
+        __sort_chain_graph_rel(chain_root, param->min_percent / 100.0);
         rb_root->rb_node = chain_root->rb_root.rb_node;
 }
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 4858d83b3b67..044178408783 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -379,6 +379,7 @@ static int parse_tracepoint_event(const char **strp,
                                     struct perf_counter_attr *attr)
 {
         const char *evt_name;
+        char *flags;
         char sys_name[MAX_EVENT_LENGTH];
         char id_buf[4];
         int fd;
@@ -400,6 +401,15 @@ static int parse_tracepoint_event(const char **strp,
         strncpy(sys_name, *strp, sys_length);
         sys_name[sys_length] = '\0';
         evt_name = evt_name + 1;
+
+        flags = strchr(evt_name, ':');
+        if (flags) {
+                *flags = '\0';
+                flags++;
+                if (!strncmp(flags, "record", strlen(flags)))
+                        attr->sample_type |= PERF_SAMPLE_RAW;
+        }
+
         evt_length = strlen(evt_name);
         if (evt_length >= MAX_EVENT_LENGTH)
                 return 0;
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index e9b13b414955..0b9862351260 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -7,23 +7,8 @@
 #include <gelf.h>
 #include <elf.h>
 
-#ifndef NO_DEMANGLE
-#include <bfd.h>
-#else
-static inline
-char *bfd_demangle(void __used *v, const char __used *c, int __used i)
-{
-        return NULL;
-}
-#endif
-
 const char *sym_hist_filter;
 
-#ifndef DMGL_PARAMS
-#define DMGL_PARAMS      (1 << 0)       /* Include function args */
-#define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
-#endif
-
 enum dso_origin {
         DSO__ORIG_KERNEL = 0,
         DSO__ORIG_JAVA_JIT,
@@ -816,6 +801,8 @@ more:
         }
 out:
         free(name);
+        if (ret < 0 && strstr(self->name, " (deleted)") != NULL)
+                return 0;
         return ret;
 }
 
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 50f723571241..48b8e5759af9 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -8,6 +8,30 @@
 #include "module.h"
 #include "event.h"
 
+#ifdef HAVE_CPLUS_DEMANGLE
+extern char *cplus_demangle(const char *, int);
+
+static inline char *bfd_demangle(void __used *v, const char *c, int i)
+{
+        return cplus_demangle(c, i);
+}
+#else
+#ifdef NO_DEMANGLE
+static inline char *bfd_demangle(void __used *v, const char __used *c,
+                                 int __used i)
+{
+        return NULL;
+}
+#else
+#include <bfd.h>
+#endif
+#endif
+
+#ifndef DMGL_PARAMS
+#define DMGL_PARAMS      (1 << 0)       /* Include function args */
+#define DMGL_ANSI        (1 << 1)       /* Include const, volatile, etc */
+#endif
+
 struct symbol {
         struct rb_node  rb_node;
         u64             start;