It's dangerous to free the original metric when it's called from
resolve_metric() as it's already in the metric_list and might have other
resources too. Instead, it'd better let them bail out and be released
properly at the later stage.
So add a check when it's called from metricgroup__add_metric() and
release it. Also make sure that mp is set properly.
Fixes: aec16d025ee01 ("perf metric: Collect referenced metrics in struct metric_ref_node")
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lore.kernel.org/lkml/20200915031819.386559-10-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
m->has_constraint = metric_no_group || metricgroup__has_constraint(pe);
INIT_LIST_HEAD(&m->metric_refs);
m->metric_refs_cnt = 0;
- *mp = m;
parent = expr_ids__alloc(ids);
if (!parent) {
free(m);
return -ENOMEM;
}
+ *mp = m;
} else {
/*
* We got here for the referenced metric, via the
* all the metric's IDs and add it to the parent context.
*/
if (expr__find_other(pe->metric_expr, NULL, &m->pctx, runtime) < 0) {
- expr__ctx_clear(&m->pctx);
- free(m);
+ if (m->metric_refs_cnt == 0) {
+ expr__ctx_clear(&m->pctx);
+ free(m);
+ *mp = NULL;
+ }
return -EINVAL;
}