perf callchain: Per-event type selection support

This patchkit adds the ability to set callgraph mode (fp, dwarf, lbr) per
event. This in term can reduce sampling overhead and the size of the
perf.data.

Here is an example.

  perf record -e 'cpu/cpu-cycles,period=1000,call-graph=fp,time=1/,cpu/instructions,call-graph=lbr/' sleep 1

 perf evlist -v
 cpu/cpu-cycles,period=1000,call-graph=fp,time=1/: type: 4, size: 112,
 config: 0x3c, { sample_period, sample_freq }: 1000, sample_type:
 IP|TID|TIME|CALLCHAIN|PERIOD|IDENTIFIER, read_format: ID, disabled: 1,
 inherit: 1, mmap: 1, comm: 1, enable_on_exec: 1, task: 1, sample_id_all:
 1, exclude_guest: 1, mmap2: 1, comm_exec: 1
 cpu/instructions,call-graph=lbr/: type: 4, size: 112, config: 0xc0, {
 sample_period, sample_freq }: 4000, sample_type:
 IP|TID|TIME|CALLCHAIN|PERIOD|BRANCH_STACK|IDENTIFIER, read_format: ID,
 disabled: 1, inherit: 1, freq: 1, enable_on_exec: 1, sample_id_all: 1,
 exclude_guest: 1

Signed-off-by: Kan Liang <kan.liang@intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1439289050-40510-1-git-send-email-kan.liang@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

7 files changed, 86 insertions, 3 deletions

diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
index afbe45ef7e3e..7f82dec2b541 100644
--- a/tools/perf/Documentation/perf-record.txt
+++ b/tools/perf/Documentation/perf-record.txt
@@ -53,6 +53,9 @@ OPTIONS
           - 'time': Disable/enable time stamping. Acceptable values are 1 for
                     enabling time stamping. 0 for disabling time stamping.
                     The default is 1.
+          - 'call-graph': Disable/enable callgraph. Acceptable str are "fp" for
+                         FP mode, "dwarf" for DWARF mode, "lbr" for LBR mode.
+          - 'stack-size': user stack size for dwarf mode
           Note: If user explicitly sets options which conflict with the params,
           the value set by the params will be overridden.
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 04fddddc6b6f..6647925d5f28 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -588,11 +588,36 @@ perf_evsel__config_callgraph(struct perf_evsel *evsel,
         }
 }
 
-static void apply_config_terms(struct perf_evsel *evsel)
+static void
+perf_evsel__reset_callgraph(struct perf_evsel *evsel,
+                            struct callchain_param *param)
+{
+        struct perf_event_attr *attr = &evsel->attr;
+
+        perf_evsel__reset_sample_bit(evsel, CALLCHAIN);
+        if (param->record_mode == CALLCHAIN_LBR) {
+                perf_evsel__reset_sample_bit(evsel, BRANCH_STACK);
+                attr->branch_sample_type &= ~(PERF_SAMPLE_BRANCH_USER |
+                                              PERF_SAMPLE_BRANCH_CALL_STACK);
+        }
+        if (param->record_mode == CALLCHAIN_DWARF) {
+                perf_evsel__reset_sample_bit(evsel, REGS_USER);
+                perf_evsel__reset_sample_bit(evsel, STACK_USER);
+        }
+}
+
+static void apply_config_terms(struct perf_evsel *evsel,
+                               struct record_opts *opts)
 {
         struct perf_evsel_config_term *term;
         struct list_head *config_terms = &evsel->config_terms;
         struct perf_event_attr *attr = &evsel->attr;
+        struct callchain_param param;
+        u32 dump_size = 0;
+        char *callgraph_buf = NULL;
+
+        /* callgraph default */
+        param.record_mode = callchain_param.record_mode;
 
         list_for_each_entry(term, config_terms, list) {
                 switch (term->type) {
@@ -610,10 +635,43 @@ static void apply_config_terms(struct perf_evsel *evsel)
                         else
                                 perf_evsel__reset_sample_bit(evsel, TIME);
                         break;
+                case PERF_EVSEL__CONFIG_TERM_CALLGRAPH:
+                        callgraph_buf = term->val.callgraph;
+                        break;
+                case PERF_EVSEL__CONFIG_TERM_STACK_USER:
+                        dump_size = term->val.stack_user;
+                        break;
                 default:
                         break;
                 }
         }
+
+        /* User explicitly set per-event callgraph, clear the old setting and reset. */
+        if ((callgraph_buf != NULL) || (dump_size > 0)) {
+
+                /* parse callgraph parameters */
+                if (callgraph_buf != NULL) {
+                        param.enabled = true;
+                        if (parse_callchain_record(callgraph_buf, &param)) {
+                                pr_err("per-event callgraph setting for %s failed. "
+                                       "Apply callgraph global setting for it\n",
+                                       evsel->name);
+                                return;
+                        }
+                }
+                if (dump_size > 0) {
+                        dump_size = round_up(dump_size, sizeof(u64));
+                        param.dump_size = dump_size;
+                }
+
+                /* If global callgraph set, clear it */
+                if (callchain_param.enabled)
+                        perf_evsel__reset_callgraph(evsel, &callchain_param);
+
+                /* set perf-event callgraph */
+                if (param.enabled)
+                        perf_evsel__config_callgraph(evsel, opts, &param);
+        }
 }
 
 /*
@@ -812,7 +870,7 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
          * Apply event specific term settings,
          * it overloads any global configuration.
          */
-        apply_config_terms(evsel);
+        apply_config_terms(evsel, opts);
 }
 
 static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index fdf2674ab339..93ac6b128149 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -41,6 +41,8 @@ enum {
         PERF_EVSEL__CONFIG_TERM_PERIOD,
         PERF_EVSEL__CONFIG_TERM_FREQ,
         PERF_EVSEL__CONFIG_TERM_TIME,
+        PERF_EVSEL__CONFIG_TERM_CALLGRAPH,
+        PERF_EVSEL__CONFIG_TERM_STACK_USER,
         PERF_EVSEL__CONFIG_TERM_MAX,
 };
 
@@ -51,6 +53,8 @@ struct perf_evsel_config_term {
                 u64     period;
                 u64     freq;
                 bool    time;
+                char    *callgraph;
+                u64     stack_user;
         } val;
 };
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index dbf315df4220..d826e6f515db 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -614,6 +614,12 @@ do {									   \
                         return -EINVAL;
                 }
                 break;
+        case PARSE_EVENTS__TERM_TYPE_CALLGRAPH:
+                CHECK_TYPE_VAL(STR);
+                break;
+        case PARSE_EVENTS__TERM_TYPE_STACKSIZE:
+                CHECK_TYPE_VAL(NUM);
+                break;
         case PARSE_EVENTS__TERM_TYPE_NAME:
                 CHECK_TYPE_VAL(STR);
                 break;
@@ -668,6 +674,12 @@ do {								\
                 case PARSE_EVENTS__TERM_TYPE_TIME:
                         ADD_CONFIG_TERM(TIME, time, term->val.num);
                         break;
+                case PARSE_EVENTS__TERM_TYPE_CALLGRAPH:
+                        ADD_CONFIG_TERM(CALLGRAPH, callgraph, term->val.str);
+                        break;
+                case PARSE_EVENTS__TERM_TYPE_STACKSIZE:
+                        ADD_CONFIG_TERM(STACK_USER, stack_user, term->val.num);
+                        break;
                 default:
                         break;
                 }
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
index ce2d13a16226..a09b0e210997 100644
--- a/tools/perf/util/parse-events.h
+++ b/tools/perf/util/parse-events.h
@@ -65,6 +65,8 @@ enum {
         PARSE_EVENTS__TERM_TYPE_SAMPLE_FREQ,
         PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE,
         PARSE_EVENTS__TERM_TYPE_TIME,
+        PARSE_EVENTS__TERM_TYPE_CALLGRAPH,
+        PARSE_EVENTS__TERM_TYPE_STACKSIZE,
 };
 
 struct parse_events_term {
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index 4306f5ad75c7..936d566f48d8 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -185,6 +185,8 @@ period			{ return term(yyscanner, PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD); }
 freq                    { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_SAMPLE_FREQ); }
 branch_type             { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_BRANCH_SAMPLE_TYPE); }
 time                    { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_TIME); }
+call-graph              { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_CALLGRAPH); }
+stack-size              { return term(yyscanner, PARSE_EVENTS__TERM_TYPE_STACKSIZE); }
 ,                       { return ','; }
 "/"                     { BEGIN(INITIAL); return '/'; }
 {name_minus}            { return str(yyscanner, PE_NAME); }
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index d85f11b8cacf..84cad054d6f7 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -634,7 +634,9 @@ static char *formats_error_string(struct list_head *formats)
 {
         struct perf_pmu_format *format;
         char *err, *str;
-        static const char *static_terms = "config,config1,config2,name,period,freq,branch_type,time\n";
+        static const char *static_terms = "config,config1,config2,name,"
+                                          "period,freq,branch_type,time,"
+                                          "call-graph,stack-size\n";
         unsigned i = 0;
 
         if (!asprintf(&str, "valid terms:"))