perf hist: Only allocate callchain_node if processing callchains

The struct callchain_node size is 120 bytes, that are never used when
there are no callchains or '-g none' is specified, so conditionally
allocate it, reducing sizeof(struct hist_entry) from 210 bytes to only
96, greatly speeding the non-callchain processing.

LKML-Reference: <new-submission>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/builtin-report.c

@@ -110,8 +110,8 @@ static int perf_session__add_hist_entry(struct perf_session *self,
 
 	if (symbol_conf.use_callchain) {
 		if (!hit)
-			callchain_init(&he->callchain);
-		err = append_chain(&he->callchain, data->callchain, syms);
+			callchain_init(he->callchain);
+		err = append_chain(he->callchain, data->callchain, syms);
 		free(syms);
 
 		if (err)

tools/perf/util/hist.c

@@ -50,7 +50,8 @@ struct hist_entry *__perf_session__add_hist_entry(struct rb_root *hists,
 			p = &(*p)->rb_right;
 	}
 
-	he = malloc(sizeof(*he));
+	he = malloc(sizeof(*he) + (symbol_conf.use_callchain ?
+				    sizeof(struct callchain_node) : 0));
 	if (!he)
 		return NULL;
 	*he = entry;
@@ -168,7 +169,7 @@ static void perf_session__insert_output_hist_entry(struct rb_root *root,
 	struct hist_entry *iter;
 
 	if (symbol_conf.use_callchain)
-		callchain_param.sort(&he->sorted_chain, &he->callchain,
+		callchain_param.sort(&he->sorted_chain, he->callchain,
 				      min_callchain_hits, &callchain_param);
 
 	while (*p != NULL) {

tools/perf/util/sort.h

@@ -49,12 +49,12 @@ struct hist_entry {
 	u64			ip;
 	char			level;
 	struct symbol	  *parent;
-	struct callchain_node	callchain;
 	union {
 		unsigned long	  position;
 		struct hist_entry *pair;
 		struct rb_root	  sorted_chain;
 	};
+	struct callchain_node	callchain[0];
 };
 
 enum sort_type {