1 files changed, 499 insertions, 153 deletions

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 135b7837e6bf..8cb58d68a006 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -10,9 +10,9 @@
 #include "util/util.h"
 
 #include "util/color.h"
-#include "util/list.h"
+#include <linux/list.h>
 #include "util/cache.h"
-#include "util/rbtree.h"
+#include <linux/rbtree.h>
 #include "util/symbol.h"
 #include "util/string.h"
 #include "util/callchain.h"
@@ -31,10 +31,12 @@
 static char             const *input_name = "perf.data";
 static char             *vmlinux = NULL;
 
-static char             default_sort_order[] = "comm,dso";
+static char             default_sort_order[] = "comm,dso,symbol";
 static char             *sort_order = default_sort_order;
-static char             *dso_list_str, *comm_list_str, *sym_list_str;
+static char             *dso_list_str, *comm_list_str, *sym_list_str,
+                        *col_width_list_str;
 static struct strlist   *dso_list, *comm_list, *sym_list;
+static char             *field_sep;
 
 static int              input;
 static int              show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
@@ -46,7 +48,10 @@ static int		dump_trace = 0;
 static int              verbose;
 #define eprintf(x...)   do { if (verbose) fprintf(stderr, x); } while (0)
 
+static int              modules;
+
 static int              full_paths;
+static int              show_nr_samples;
 
 static unsigned long    page_size;
 static unsigned long    mmap_window = 32;
@@ -56,8 +61,17 @@ static char		*parent_pattern = default_parent_pattern;
 static regex_t          parent_regex;
 
 static int              exclude_other = 1;
+
+static char             callchain_default_opt[] = "fractal,0.5";
+
 static int              callchain;
 
+static
+struct callchain_param  callchain_param = {
+        .mode   = CHAIN_GRAPH_ABS,
+        .min_percent = 0.5
+};
+
 static u64              sample_type;
 
 struct ip_event {
@@ -85,13 +99,7 @@ struct comm_event {
 struct fork_event {
         struct perf_event_header header;
         u32 pid, ppid;
-};
-
-struct period_event {
-        struct perf_event_header header;
-        u64 time;
-        u64 id;
-        u64 sample_period;
+        u32 tid, ptid;
 };
 
 struct lost_event {
@@ -113,14 +121,41 @@ typedef union event_union {
         struct mmap_event               mmap;
         struct comm_event               comm;
         struct fork_event               fork;
-        struct period_event             period;
         struct lost_event               lost;
         struct read_event               read;
 } event_t;
 
+static int repsep_fprintf(FILE *fp, const char *fmt, ...)
+{
+        int n;
+        va_list ap;
+
+        va_start(ap, fmt);
+        if (!field_sep)
+                n = vfprintf(fp, fmt, ap);
+        else {
+                char *bf = NULL;
+                n = vasprintf(&bf, fmt, ap);
+                if (n > 0) {
+                        char *sep = bf;
+                        while (1) {
+                                sep = strchr(sep, *field_sep);
+                                if (sep == NULL)
+                                        break;
+                                *sep = '.';
+                        }
+                }
+                fputs(bf, fp);
+                free(bf);
+        }
+        va_end(ap);
+        return n;
+}
+
 static LIST_HEAD(dsos);
 static struct dso *kernel_dso;
 static struct dso *vdso;
+static struct dso *hypervisor_dso;
 
 static void dsos__add(struct dso *dso)
 {
@@ -176,7 +211,7 @@ static void dsos__fprintf(FILE *fp)
 
 static struct symbol *vdso__find_symbol(struct dso *dso, u64 ip)
 {
-        return dso__find_symbol(kernel_dso, ip);
+        return dso__find_symbol(dso, ip);
 }
 
 static int load_kernel(void)
@@ -187,8 +222,8 @@ static int load_kernel(void)
         if (!kernel_dso)
                 return -1;
 
-        err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose);
-        if (err) {
+        err = dso__load_kernel(kernel_dso, vmlinux, NULL, verbose, modules);
+        if (err <= 0) {
                 dso__delete(kernel_dso);
                 kernel_dso = NULL;
         } else
@@ -202,6 +237,11 @@ static int load_kernel(void)
 
         dsos__add(vdso);
 
+        hypervisor_dso = dso__new("[hypervisor]", 0);
+        if (!hypervisor_dso)
+                return -1;
+        dsos__add(hypervisor_dso);
+
         return err;
 }
 
@@ -213,7 +253,7 @@ static int strcommon(const char *pathname)
 {
         int n = 0;
 
-        while (pathname[n] == cwd[n] && n < cwdlen)
+        while (n < cwdlen && pathname[n] == cwd[n])
                 ++n;
 
         return n;
@@ -233,7 +273,7 @@ static u64 map__map_ip(struct map *map, u64 ip)
         return ip - map->start + map->pgoff;
 }
 
-static u64 vdso__map_ip(struct map *map, u64 ip)
+static u64 vdso__map_ip(struct map *map __used, u64 ip)
 {
         return ip;
 }
@@ -343,12 +383,28 @@ static struct thread *thread__new(pid_t pid)
         return self;
 }
 
+static unsigned int dsos__col_width,
+                    comms__col_width,
+                    threads__col_width;
+
 static int thread__set_comm(struct thread *self, const char *comm)
 {
         if (self->comm)
                 free(self->comm);
         self->comm = strdup(comm);
-        return self->comm ? 0 : -ENOMEM;
+        if (!self->comm)
+                return -ENOMEM;
+
+        if (!col_width_list_str && !field_sep &&
+            (!comm_list || strlist__has_entry(comm_list, comm))) {
+                unsigned int slen = strlen(comm);
+                if (slen > comms__col_width) {
+                        comms__col_width = slen;
+                        threads__col_width = slen + 6;
+                }
+        }
+
+        return 0;
 }
 
 static size_t thread__fprintf(struct thread *self, FILE *fp)
@@ -519,7 +575,9 @@ struct sort_entry {
 
         int64_t (*cmp)(struct hist_entry *, struct hist_entry *);
         int64_t (*collapse)(struct hist_entry *, struct hist_entry *);
-        size_t  (*print)(FILE *fp, struct hist_entry *);
+        size_t  (*print)(FILE *fp, struct hist_entry *, unsigned int width);
+        unsigned int *width;
+        bool    elide;
 };
 
 static int64_t cmp_null(void *l, void *r)
@@ -541,15 +599,17 @@ sort__thread_cmp(struct hist_entry *left, struct hist_entry *right)
 }
 
 static size_t
-sort__thread_print(FILE *fp, struct hist_entry *self)
+sort__thread_print(FILE *fp, struct hist_entry *self, unsigned int width)
 {
-        return fprintf(fp, "%16s:%5d", self->thread->comm ?: "", self->thread->pid);
+        return repsep_fprintf(fp, "%*s:%5d", width - 6,
+                              self->thread->comm ?: "", self->thread->pid);
 }
 
 static struct sort_entry sort_thread = {
-        .header = "         Command:  Pid",
+        .header = "Command:  Pid",
         .cmp    = sort__thread_cmp,
         .print  = sort__thread_print,
+        .width  = &threads__col_width,
 };
 
 /* --sort comm */
@@ -573,16 +633,17 @@ sort__comm_collapse(struct hist_entry *left, struct hist_entry *right)
 }
 
 static size_t
-sort__comm_print(FILE *fp, struct hist_entry *self)
+sort__comm_print(FILE *fp, struct hist_entry *self, unsigned int width)
 {
-        return fprintf(fp, "%16s", self->thread->comm);
+        return repsep_fprintf(fp, "%*s", width, self->thread->comm);
 }
 
 static struct sort_entry sort_comm = {
-        .header         = "         Command",
+        .header         = "Command",
         .cmp            = sort__comm_cmp,
         .collapse       = sort__comm_collapse,
         .print          = sort__comm_print,
+        .width          = &comms__col_width,
 };
 
 /* --sort dso */
@@ -600,18 +661,19 @@ sort__dso_cmp(struct hist_entry *left, struct hist_entry *right)
 }
 
 static size_t
-sort__dso_print(FILE *fp, struct hist_entry *self)
+sort__dso_print(FILE *fp, struct hist_entry *self, unsigned int width)
 {
         if (self->dso)
-                return fprintf(fp, "%-25s", self->dso->name);
+                return repsep_fprintf(fp, "%-*s", width, self->dso->name);
 
-        return fprintf(fp, "%016llx         ", (u64)self->ip);
+        return repsep_fprintf(fp, "%*llx", width, (u64)self->ip);
 }
 
 static struct sort_entry sort_dso = {
-        .header = "Shared Object            ",
+        .header = "Shared Object",
         .cmp    = sort__dso_cmp,
         .print  = sort__dso_print,
+        .width  = &dsos__col_width,
 };
 
 /* --sort symbol */
@@ -631,18 +693,22 @@ sort__sym_cmp(struct hist_entry *left, struct hist_entry *right)
 }
 
 static size_t
-sort__sym_print(FILE *fp, struct hist_entry *self)
+sort__sym_print(FILE *fp, struct hist_entry *self, unsigned int width __used)
 {
         size_t ret = 0;
 
         if (verbose)
-                ret += fprintf(fp, "%#018llx  ", (u64)self->ip);
+                ret += repsep_fprintf(fp, "%#018llx  ", (u64)self->ip);
 
+        ret += repsep_fprintf(fp, "[%c] ", self->level);
         if (self->sym) {
-                ret += fprintf(fp, "[%c] %s",
-                        self->dso == kernel_dso ? 'k' : '.', self->sym->name);
+                ret += repsep_fprintf(fp, "%s", self->sym->name);
+
+                if (self->sym->module)
+                        ret += repsep_fprintf(fp, "\t[%s]",
+                                             self->sym->module->name);
         } else {
-                ret += fprintf(fp, "%#016llx", (u64)self->ip);
+                ret += repsep_fprintf(fp, "%#016llx", (u64)self->ip);
         }
 
         return ret;
@@ -669,19 +735,19 @@ sort__parent_cmp(struct hist_entry *left, struct hist_entry *right)
 }
 
 static size_t
-sort__parent_print(FILE *fp, struct hist_entry *self)
+sort__parent_print(FILE *fp, struct hist_entry *self, unsigned int width)
 {
-        size_t ret = 0;
-
-        ret += fprintf(fp, "%-20s", self->parent ? self->parent->name : "[other]");
-
-        return ret;
+        return repsep_fprintf(fp, "%-*s", width,
+                              self->parent ? self->parent->name : "[other]");
 }
 
+static unsigned int parent_symbol__col_width;
+
 static struct sort_entry sort_parent = {
-        .header = "Parent symbol       ",
+        .header = "Parent symbol",
         .cmp    = sort__parent_cmp,
         .print  = sort__parent_print,
+        .width  = &parent_symbol__col_width,
 };
 
 static int sort__need_collapse = 0;
@@ -705,7 +771,7 @@ static LIST_HEAD(hist_entry__sort_list);
 
 static int sort_dimension__add(char *tok)
 {
-        int i;
+        unsigned int i;
 
         for (i = 0; i < ARRAY_SIZE(sort_dimensions); i++) {
                 struct sort_dimension *sd = &sort_dimensions[i];
@@ -775,8 +841,109 @@ hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
         return cmp;
 }
 
+static size_t ipchain__fprintf_graph_line(FILE *fp, int depth, int depth_mask)
+{
+        int i;
+        size_t ret = 0;
+
+        ret += fprintf(fp, "%s", "                ");
+
+        for (i = 0; i < depth; i++)
+                if (depth_mask & (1 << i))
+                        ret += fprintf(fp, "|          ");
+                else
+                        ret += fprintf(fp, "           ");
+
+        ret += fprintf(fp, "\n");
+
+        return ret;
+}
 static size_t
-callchain__fprintf(FILE *fp, struct callchain_node *self, u64 total_samples)
+ipchain__fprintf_graph(FILE *fp, struct callchain_list *chain, int depth,
+                       int depth_mask, int count, u64 total_samples,
+                       int hits)
+{
+        int i;
+        size_t ret = 0;
+
+        ret += fprintf(fp, "%s", "                ");
+        for (i = 0; i < depth; i++) {
+                if (depth_mask & (1 << i))
+                        ret += fprintf(fp, "|");
+                else
+                        ret += fprintf(fp, " ");
+                if (!count && i == depth - 1) {
+                        double percent;
+
+                        percent = hits * 100.0 / total_samples;
+                        ret += percent_color_fprintf(fp, "--%2.2f%%-- ", percent);
+                } else
+                        ret += fprintf(fp, "%s", "          ");
+        }
+        if (chain->sym)
+                ret += fprintf(fp, "%s\n", chain->sym->name);
+        else
+                ret += fprintf(fp, "%p\n", (void *)(long)chain->ip);
+
+        return ret;
+}
+
+static size_t
+callchain__fprintf_graph(FILE *fp, struct callchain_node *self,
+                        u64 total_samples, int depth, int depth_mask)
+{
+        struct rb_node *node, *next;
+        struct callchain_node *child;
+        struct callchain_list *chain;
+        int new_depth_mask = depth_mask;
+        u64 new_total;
+        size_t ret = 0;
+        int i;
+
+        if (callchain_param.mode == CHAIN_GRAPH_REL)
+                new_total = self->cumul_hit;
+        else
+                new_total = total_samples;
+
+        node = rb_first(&self->rb_root);
+        while (node) {
+                child = rb_entry(node, struct callchain_node, rb_node);
+
+                /*
+                 * The depth mask manages the output of pipes that show
+                 * the depth. We don't want to keep the pipes of the current
+                 * level for the last child of this depth
+                 */
+                next = rb_next(node);
+                if (!next)
+                        new_depth_mask &= ~(1 << (depth - 1));
+
+                /*
+                 * But we keep the older depth mask for the line seperator
+                 * to keep the level link until we reach the last child
+                 */
+                ret += ipchain__fprintf_graph_line(fp, depth, depth_mask);
+                i = 0;
+                list_for_each_entry(chain, &child->val, list) {
+                        if (chain->ip >= PERF_CONTEXT_MAX)
+                                continue;
+                        ret += ipchain__fprintf_graph(fp, chain, depth,
+                                                      new_depth_mask, i++,
+                                                      new_total,
+                                                      child->cumul_hit);
+                }
+                ret += callchain__fprintf_graph(fp, child, new_total,
+                                                depth + 1,
+                                                new_depth_mask | (1 << depth));
+                node = next;
+        }
+
+        return ret;
+}
+
+static size_t
+callchain__fprintf_flat(FILE *fp, struct callchain_node *self,
+                        u64 total_samples)
 {
         struct callchain_list *chain;
         size_t ret = 0;
@@ -784,11 +951,18 @@ callchain__fprintf(FILE *fp, struct callchain_node *self, u64 total_samples)
         if (!self)
                 return 0;
 
-        ret += callchain__fprintf(fp, self->parent, total_samples);
+        ret += callchain__fprintf_flat(fp, self->parent, total_samples);
 
 
-        list_for_each_entry(chain, &self->val, list)
-                ret += fprintf(fp, "                %p\n", (void *)chain->ip);
+        list_for_each_entry(chain, &self->val, list) {
+                if (chain->ip >= PERF_CONTEXT_MAX)
+                        continue;
+                if (chain->sym)
+                        ret += fprintf(fp, "                %s\n", chain->sym->name);
+                else
+                        ret += fprintf(fp, "                %p\n",
+                                        (void *)(long)chain->ip);
+        }
 
         return ret;
 }
@@ -807,8 +981,19 @@ hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
 
                 chain = rb_entry(rb_node, struct callchain_node, rb_node);
                 percent = chain->hit * 100.0 / total_samples;
-                ret += fprintf(fp, "           %6.2f%%\n", percent);
-                ret += callchain__fprintf(fp, chain, total_samples);
+                switch (callchain_param.mode) {
+                case CHAIN_FLAT:
+                        ret += percent_color_fprintf(fp, "           %6.2f%%\n",
+                                                     percent);
+                        ret += callchain__fprintf_flat(fp, chain, total_samples);
+                        break;
+                case CHAIN_GRAPH_ABS: /* Falldown */
+                case CHAIN_GRAPH_REL:
+                        ret += callchain__fprintf_graph(fp, chain,
+                                                        total_samples, 1, 1);
+                default:
+                        break;
+                }
                 ret += fprintf(fp, "\n");
                 rb_node = rb_next(rb_node);
         }
@@ -826,33 +1011,26 @@ hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
         if (exclude_other && !self->parent)
                 return 0;
 
-        if (total_samples) {
-                double percent = self->count * 100.0 / total_samples;
-                char *color = PERF_COLOR_NORMAL;
-
-                /*
-                 * We color high-overhead entries in red, mid-overhead
-                 * entries in green - and keep the low overhead places
-                 * normal:
-                 */
-                if (percent >= 5.0) {
-                        color = PERF_COLOR_RED;
-                } else {
-                        if (percent >= 0.5)
-                                color = PERF_COLOR_GREEN;
-                }
+        if (total_samples)
+                ret = percent_color_fprintf(fp,
+                                            field_sep ? "%.2f" : "   %6.2f%%",
+                                        (self->count * 100.0) / total_samples);
+        else
+                ret = fprintf(fp, field_sep ? "%lld" : "%12lld ", self->count);
 
-                ret = color_fprintf(fp, color, "   %6.2f%%",
-                                (self->count * 100.0) / total_samples);
-        } else
-                ret = fprintf(fp, "%12Ld ", self->count);
+        if (show_nr_samples) {
+                if (field_sep)
+                        fprintf(fp, "%c%lld", *field_sep, self->count);
+                else
+                        fprintf(fp, "%11lld", self->count);
+        }
 
         list_for_each_entry(se, &hist_entry__sort_list, list) {
-                if (exclude_other && (se == &sort_parent))
+                if (se->elide)
                         continue;
 
-                fprintf(fp, "  ");
-                ret += se->print(fp, self);
+                fprintf(fp, "%s", field_sep ?: "  ");
+                ret += se->print(fp, self, se->width ? *se->width : 0);
         }
 
         ret += fprintf(fp, "\n");
@@ -867,6 +1045,18 @@ hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
  *
  */
 
+static void dso__calc_col_width(struct dso *self)
+{
+        if (!col_width_list_str && !field_sep &&
+            (!dso_list || strlist__has_entry(dso_list, self->name))) {
+                unsigned int slen = strlen(self->name);
+                if (slen > dsos__col_width)
+                        dsos__col_width = slen;
+        }
+
+        self->slen_calculated = 1;
+}
+
 static struct symbol *
 resolve_symbol(struct thread *thread, struct map **mapp,
                struct dso **dsop, u64 *ipp)
@@ -886,6 +1076,14 @@ resolve_symbol(struct thread *thread, struct map **mapp,
 
         map = thread__find_map(thread, ip);
         if (map != NULL) {
+                /*
+                 * We have to do this here as we may have a dso
+                 * with no symbol hit that has a name longer than
+                 * the ones with symbols sampled.
+                 */
+                if (!sort_dso.elide && !map->dso->slen_calculated)
+                        dso__calc_col_width(map->dso);
+
                 if (mapp)
                         *mapp = map;
 got_map:
@@ -923,6 +1121,58 @@ static int call__match(struct symbol *sym)
         return 0;
 }
 
+static struct symbol **
+resolve_callchain(struct thread *thread, struct map *map __used,
+                    struct ip_callchain *chain, struct hist_entry *entry)
+{
+        u64 context = PERF_CONTEXT_MAX;
+        struct symbol **syms = NULL;
+        unsigned int i;
+
+        if (callchain) {
+                syms = calloc(chain->nr, sizeof(*syms));
+                if (!syms) {
+                        fprintf(stderr, "Can't allocate memory for symbols\n");
+                        exit(-1);
+                }
+        }
+
+        for (i = 0; i < chain->nr; i++) {
+                u64 ip = chain->ips[i];
+                struct dso *dso = NULL;
+                struct symbol *sym;
+
+                if (ip >= PERF_CONTEXT_MAX) {
+                        context = ip;
+                        continue;
+                }
+
+                switch (context) {
+                case PERF_CONTEXT_HV:
+                        dso = hypervisor_dso;
+                        break;
+                case PERF_CONTEXT_KERNEL:
+                        dso = kernel_dso;
+                        break;
+                default:
+                        break;
+                }
+
+                sym = resolve_symbol(thread, NULL, &dso, &ip);
+
+                if (sym) {
+                        if (sort__has_parent && call__match(sym) &&
+                            !entry->parent)
+                                entry->parent = sym;
+                        if (!callchain)
+                                break;
+                        syms[i] = sym;
+                }
+        }
+
+        return syms;
+}
+
 /*
  * collect histogram counts
  */
@@ -935,6 +1185,7 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
         struct rb_node **p = &hist.rb_node;
         struct rb_node *parent = NULL;
         struct hist_entry *he;
+        struct symbol **syms = NULL;
         struct hist_entry entry = {
                 .thread = thread,
                 .map    = map,
@@ -948,36 +1199,8 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
         };
         int cmp;
 
-        if (sort__has_parent && chain) {
-                u64 context = PERF_CONTEXT_MAX;
-                int i;
-
-                for (i = 0; i < chain->nr; i++) {
-                        u64 ip = chain->ips[i];
-                        struct dso *dso = NULL;
-                        struct symbol *sym;
-
-                        if (ip >= PERF_CONTEXT_MAX) {
-                                context = ip;
-                                continue;
-                        }
-
-                        switch (context) {
-                        case PERF_CONTEXT_KERNEL:
-                                dso = kernel_dso;
-                                break;
-                        default:
-                                break;
-                        }
-
-                        sym = resolve_symbol(thread, NULL, &dso, &ip);
-
-                        if (sym && call__match(sym)) {
-                                entry.parent = sym;
-                                break;
-                        }
-                }
-        }
+        if ((sort__has_parent || callchain) && chain)
+                syms = resolve_callchain(thread, map, chain, &entry);
 
         while (*p != NULL) {
                 parent = *p;
@@ -987,8 +1210,10 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 
                 if (!cmp) {
                         he->count += count;
-                        if (callchain)
-                                append_chain(&he->callchain, chain);
+                        if (callchain) {
+                                append_chain(&he->callchain, chain, syms);
+                                free(syms);
+                        }
                         return 0;
                 }
 
@@ -1004,7 +1229,8 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
         *he = entry;
         if (callchain) {
                 callchain_init(&he->callchain);
-                append_chain(&he->callchain, chain);
+                append_chain(&he->callchain, chain, syms);
+                free(syms);
         }
         rb_link_node(&he->rb_node, parent, p);
         rb_insert_color(&he->rb_node, &hist);
@@ -1076,14 +1302,15 @@ static void collapse__resort(void)
 
 static struct rb_root output_hists;
 
-static void output__insert_entry(struct hist_entry *he)
+static void output__insert_entry(struct hist_entry *he, u64 min_callchain_hits)
 {
         struct rb_node **p = &output_hists.rb_node;
         struct rb_node *parent = NULL;
         struct hist_entry *iter;
 
         if (callchain)
-                sort_chain_to_rbtree(&he->sorted_chain, &he->callchain);
+                callchain_param.sort(&he->sorted_chain, &he->callchain,
+                                      min_callchain_hits, &callchain_param);
 
         while (*p != NULL) {
                 parent = *p;
@@ -1099,11 +1326,14 @@ static void output__insert_entry(struct hist_entry *he)
         rb_insert_color(&he->rb_node, &output_hists);
 }
 
-static void output__resort(void)
+static void output__resort(u64 total_samples)
 {
         struct rb_node *next;
         struct hist_entry *n;
         struct rb_root *tree = &hist;
+        u64 min_callchain_hits;
+
+        min_callchain_hits = total_samples * (callchain_param.min_percent / 100);
 
         if (sort__need_collapse)
                 tree = &collapse_hists;
@@ -1115,7 +1345,7 @@ static void output__resort(void)
                 next = rb_next(&n->rb_node);
 
                 rb_erase(&n->rb_node, tree);
-                output__insert_entry(n);
+                output__insert_entry(n, min_callchain_hits);
         }
 }
 
@@ -1125,35 +1355,67 @@ static size_t output__fprintf(FILE *fp, u64 total_samples)
         struct sort_entry *se;
         struct rb_node *nd;
         size_t ret = 0;
+        unsigned int width;
+        char *col_width = col_width_list_str;
 
-        fprintf(fp, "\n");
-        fprintf(fp, "#\n");
-        fprintf(fp, "# (%Ld samples)\n", (u64)total_samples);
+        fprintf(fp, "# Samples: %Ld\n", (u64)total_samples);
         fprintf(fp, "#\n");
 
         fprintf(fp, "# Overhead");
+        if (show_nr_samples) {
+                if (field_sep)
+                        fprintf(fp, "%cSamples", *field_sep);
+                else
+                        fputs("  Samples  ", fp);
+        }
         list_for_each_entry(se, &hist_entry__sort_list, list) {
-                if (exclude_other && (se == &sort_parent))
+                if (se->elide)
+                        continue;
+                if (field_sep) {
+                        fprintf(fp, "%c%s", *field_sep, se->header);
                         continue;
-                fprintf(fp, "  %s", se->header);
+                }
+                width = strlen(se->header);
+                if (se->width) {
+                        if (col_width_list_str) {
+                                if (col_width) {
+                                        *se->width = atoi(col_width);
+                                        col_width = strchr(col_width, ',');
+                                        if (col_width)
+                                                ++col_width;
+                                }
+                        }
+                        width = *se->width = max(*se->width, width);
+                }
+                fprintf(fp, "  %*s", width, se->header);
         }
         fprintf(fp, "\n");
 
+        if (field_sep)
+                goto print_entries;
+
         fprintf(fp, "# ........");
+        if (show_nr_samples)
+                fprintf(fp, " ..........");
         list_for_each_entry(se, &hist_entry__sort_list, list) {
-                int i;
+                unsigned int i;
 
-                if (exclude_other && (se == &sort_parent))
+                if (se->elide)
                         continue;
 
                 fprintf(fp, "  ");
-                for (i = 0; i < strlen(se->header); i++)
+                if (se->width)
+                        width = *se->width;
+                else
+                        width = strlen(se->header);
+                for (i = 0; i < width; i++)
                         fprintf(fp, ".");
         }
         fprintf(fp, "\n");
 
         fprintf(fp, "#\n");
 
+print_entries:
         for (nd = rb_first(&output_hists); nd; nd = rb_next(nd)) {
                 pos = rb_entry(nd, struct hist_entry, rb_node);
                 ret += hist_entry__fprintf(fp, pos, total_samples);
@@ -1162,7 +1424,7 @@ static size_t output__fprintf(FILE *fp, u64 total_samples)
         if (sort_order == default_sort_order &&
                         parent_pattern == default_parent_pattern) {
                 fprintf(fp, "#\n");
-                fprintf(fp, "# (For more details, try: perf report --sort comm,dso,symbol)\n");
+                fprintf(fp, "# (For a higher level overview, try: perf report --sort comm,dso)\n");
                 fprintf(fp, "#\n");
         }
         fprintf(fp, "\n");
@@ -1213,6 +1475,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
         struct map *map = NULL;
         void *more_data = event->ip.__more_data;
         struct ip_callchain *chain = NULL;
+        int cpumode;
 
         if (sample_type & PERF_SAMPLE_PERIOD) {
                 period = *(u64 *)more_data;
@@ -1228,7 +1491,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
                 (long long)period);
 
         if (sample_type & PERF_SAMPLE_CALLCHAIN) {
-                int i;
+                unsigned int i;
 
                 chain = (void *)more_data;
 
@@ -1256,7 +1519,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
         if (comm_list && !strlist__has_entry(comm_list, thread->comm))
                 return 0;
 
-        if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
+        cpumode = event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK;
+
+        if (cpumode == PERF_EVENT_MISC_KERNEL) {
                 show = SHOW_KERNEL;
                 level = 'k';
 
@@ -1264,7 +1529,7 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 
                 dprintf(" ...... dso: %s\n", dso->name);
 
-        } else if (event->header.misc & PERF_EVENT_MISC_USER) {
+        } else if (cpumode == PERF_EVENT_MISC_USER) {
 
                 show = SHOW_USER;
                 level = '.';
@@ -1272,6 +1537,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
         } else {
                 show = SHOW_HV;
                 level = 'H';
+
+                dso = hypervisor_dso;
+
                 dprintf(" ...... dso: [hypervisor]\n");
         }
 
@@ -1341,15 +1609,27 @@ process_comm_event(event_t *event, unsigned long offset, unsigned long head)
 }
 
 static int
-process_fork_event(event_t *event, unsigned long offset, unsigned long head)
+process_task_event(event_t *event, unsigned long offset, unsigned long head)
 {
         struct thread *thread = threads__findnew(event->fork.pid);
         struct thread *parent = threads__findnew(event->fork.ppid);
 
-        dprintf("%p [%p]: PERF_EVENT_FORK: %d:%d\n",
+        dprintf("%p [%p]: PERF_EVENT_%s: (%d:%d):(%d:%d)\n",
                 (void *)(offset + head),
                 (void *)(long)(event->header.size),
-                event->fork.pid, event->fork.ppid);
+                event->header.type == PERF_EVENT_FORK ? "FORK" : "EXIT",
+                event->fork.pid, event->fork.tid,
+                event->fork.ppid, event->fork.ptid);
+
+        /*
+         * A thread clone will have the same PID for both
+         * parent and child.
+         */
+        if (thread == parent)
+                return 0;
+
+        if (event->header.type == PERF_EVENT_EXIT)
+                return 0;
 
         if (!thread || !parent || thread__fork(thread, parent)) {
                 dprintf("problem processing PERF_EVENT_FORK, skipping event.\n");
@@ -1361,19 +1641,6 @@ process_fork_event(event_t *event, unsigned long offset, unsigned long head)
 }
 
 static int
-process_period_event(event_t *event, unsigned long offset, unsigned long head)
-{
-        dprintf("%p [%p]: PERF_EVENT_PERIOD: time:%Ld, id:%Ld: period:%Ld\n",
-                (void *)(offset + head),
-                (void *)(long)(event->header.size),
-                event->period.time,
-                event->period.id,
-                event->period.sample_period);
-
-        return 0;
-}
-
-static int
 process_lost_event(event_t *event, unsigned long offset, unsigned long head)
 {
         dprintf("%p [%p]: PERF_EVENT_LOST: id:%Ld: lost:%Ld\n",
@@ -1452,10 +1719,8 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
                 return process_comm_event(event, offset, head);
 
         case PERF_EVENT_FORK:
-                return process_fork_event(event, offset, head);
-
-        case PERF_EVENT_PERIOD:
-                return process_period_event(event, offset, head);
+        case PERF_EVENT_EXIT:
+                return process_task_event(event, offset, head);
 
         case PERF_EVENT_LOST:
                 return process_lost_event(event, offset, head);
@@ -1534,9 +1799,19 @@ static int __cmd_report(void)
 
         sample_type = perf_header__sample_type();
 
-        if (sort__has_parent && !(sample_type & PERF_SAMPLE_CALLCHAIN)) {
-                fprintf(stderr, "selected --sort parent, but no callchain data\n");
-                exit(-1);
+        if (!(sample_type & PERF_SAMPLE_CALLCHAIN)) {
+                if (sort__has_parent) {
+                        fprintf(stderr, "selected --sort parent, but no"
+                                        " callchain data. Did you call"
+                                        " perf record without -g?\n");
+                        exit(-1);
+                }
+                if (callchain) {
+                        fprintf(stderr, "selected -c but no callchain data."
+                                        " Did you call perf record without"
+                                        " -g?\n");
+                        exit(-1);
+                }
         }
 
         if (load_kernel() < 0) {
@@ -1619,7 +1894,7 @@ more:
         if (offset + head >= header->data_offset + header->data_size)
                 goto done;
 
-        if (offset + head < stat.st_size)
+        if (offset + head < (unsigned long)stat.st_size)
                 goto more;
 
 done:
@@ -1643,12 +1918,58 @@ done:
                 dsos__fprintf(stdout);
 
         collapse__resort();
-        output__resort();
+        output__resort(total);
         output__fprintf(stdout, total);
 
         return rc;
 }
 
+static int
+parse_callchain_opt(const struct option *opt __used, const char *arg,
+                    int unset __used)
+{
+        char *tok;
+        char *endptr;
+
+        callchain = 1;
+
+        if (!arg)
+                return 0;
+
+        tok = strtok((char *)arg, ",");
+        if (!tok)
+                return -1;
+
+        /* get the output mode */
+        if (!strncmp(tok, "graph", strlen(arg)))
+                callchain_param.mode = CHAIN_GRAPH_ABS;
+
+        else if (!strncmp(tok, "flat", strlen(arg)))
+                callchain_param.mode = CHAIN_FLAT;
+
+        else if (!strncmp(tok, "fractal", strlen(arg)))
+                callchain_param.mode = CHAIN_GRAPH_REL;
+
+        else
+                return -1;
+
+        /* get the min percentage */
+        tok = strtok(NULL, ",");
+        if (!tok)
+                goto setup;
+
+        callchain_param.min_percent = strtod(tok, &endptr);
+        if (tok == endptr)
+                return -1;
+
+setup:
+        if (register_callchain_param(&callchain_param) < 0) {
+                fprintf(stderr, "Can't register callchain params\n");
+                return -1;
+        }
+        return 0;
+}
+
 static const char * const report_usage[] = {
         "perf report [<options>] <command>",
         NULL
@@ -1662,6 +1983,10 @@ static const struct option options[] = {
         OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
                     "dump raw trace in ASCII"),
         OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"),
+        OPT_BOOLEAN('m', "modules", &modules,
+                    "load module symbols - WARNING: use only with -k and LIVE kernel"),
+        OPT_BOOLEAN('n', "show-nr-samples", &show_nr_samples,
+                    "Show a column with the number of samples"),
         OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
                    "sort by key(s): pid, comm, dso, symbol, parent"),
         OPT_BOOLEAN('P', "full-paths", &full_paths,
@@ -1670,13 +1995,21 @@ static const struct option options[] = {
                    "regex filter to identify parent, see: '--sort parent'"),
         OPT_BOOLEAN('x', "exclude-other", &exclude_other,
                     "Only display entries with parent-match"),
-        OPT_BOOLEAN('c', "callchain", &callchain, "Display callchains"),
+        OPT_CALLBACK_DEFAULT('g', "call-graph", NULL, "output_type,min_percent",
+                     "Display callchains using output_type and min percent threshold. "
+                     "Default: fractal,0.5", &parse_callchain_opt, callchain_default_opt),
         OPT_STRING('d', "dsos", &dso_list_str, "dso[,dso...]",
                    "only consider symbols in these dsos"),
         OPT_STRING('C', "comms", &comm_list_str, "comm[,comm...]",
                    "only consider symbols in these comms"),
         OPT_STRING('S', "symbols", &sym_list_str, "symbol[,symbol...]",
                    "only consider these symbols"),
+        OPT_STRING('w', "column-widths", &col_width_list_str,
+                   "width[,width...]",
+                   "don't try to adjust column width, use these fixed values"),
+        OPT_STRING('t', "field-separator", &field_sep, "separator",
+                   "separator for columns, no spaces will be added between "
+                   "columns '.' is reserved."),
         OPT_END()
 };
 
@@ -1696,7 +2029,8 @@ static void setup_sorting(void)
 }
 
 static void setup_list(struct strlist **list, const char *list_str,
-                       const char *list_name)
+                       struct sort_entry *se, const char *list_name,
+                       FILE *fp)
 {
         if (list_str) {
                 *list = strlist__new(true, list_str);
@@ -1705,10 +2039,15 @@ static void setup_list(struct strlist **list, const char *list_str,
                                 list_name);
                         exit(129);
                 }
+                if (strlist__nr_entries(*list) == 1) {
+                        fprintf(fp, "# %s: %s\n", list_name,
+                                strlist__entry(*list, 0)->s);
+                        se->elide = true;
+                }
         }
 }
 
-int cmd_report(int argc, const char **argv, const char *prefix)
+int cmd_report(int argc, const char **argv, const char *prefix __used)
 {
         symbol__init();
 
@@ -1718,9 +2057,10 @@ int cmd_report(int argc, const char **argv, const char *prefix)
 
         setup_sorting();
 
-        if (parent_pattern != default_parent_pattern)
+        if (parent_pattern != default_parent_pattern) {
                 sort_dimension__add("parent");
-        else
+                sort_parent.elide = 1;
+        } else
                 exclude_other = 0;
 
         /*
@@ -1729,11 +2069,17 @@ int cmd_report(int argc, const char **argv, const char *prefix)
         if (argc)
                 usage_with_options(report_usage, options);
 
-        setup_list(&dso_list, dso_list_str, "dso");
-        setup_list(&comm_list, comm_list_str, "comm");
-        setup_list(&sym_list, sym_list_str, "symbol");
-
         setup_pager();
 
+        setup_list(&dso_list, dso_list_str, &sort_dso, "dso", stdout);
+        setup_list(&comm_list, comm_list_str, &sort_comm, "comm", stdout);
+        setup_list(&sym_list, sym_list_str, &sort_sym, "symbol", stdout);
+
+        if (field_sep && *field_sep == '.') {
+                fputs("'.' is the only non valid --field-separator argument\n",
+                      stderr);
+                exit(129);
+        }
+
         return __cmd_report();
 }