Commit 3d906ef1 authored by Peter Zijlstra's avatar Peter Zijlstra Committed by Ingo Molnar

perf_counter tools: Handle overlapping MMAP events

Martin Schwidefsky reported "perf report" symbol resolution
problems on S390.

Since we only report MMAP, not MUNMAP, we have to deal with
overlapping maps.

We used to simply throw out the old map on the assumption whole
maps got unmapped. This obviously doesn't deal with partial
unmaps. However it appears some dynamic linkers do fancy
partial unmaps (s390), so do something more elaborate and
truncate the old maps, only removing them when they've been
fully covered.

This resolves (part of) the S390 symbol resolution problems.
Reported-by: default avatarMartin Schwidefsky <schwidefsky@de.ibm.com>
Tested-by: default avatarMartin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: default avatarPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
parent dee41206
...@@ -400,11 +400,29 @@ static void thread__insert_map(struct thread *self, struct map *map) ...@@ -400,11 +400,29 @@ static void thread__insert_map(struct thread *self, struct map *map)
list_for_each_entry_safe(pos, tmp, &self->maps, node) { list_for_each_entry_safe(pos, tmp, &self->maps, node) {
if (map__overlap(pos, map)) { if (map__overlap(pos, map)) {
if (verbose >= 2) {
printf("overlapping maps:\n");
map__fprintf(map, stdout);
map__fprintf(pos, stdout);
}
if (map->start <= pos->start && map->end > pos->start)
pos->start = map->end;
if (map->end >= pos->end && map->start < pos->end)
pos->end = map->start;
if (verbose >= 2) {
printf("after collision:\n");
map__fprintf(pos, stdout);
}
if (pos->start >= pos->end) {
list_del_init(&pos->node); list_del_init(&pos->node);
/* XXX leaks dsos */
free(pos); free(pos);
} }
} }
}
list_add_tail(&map->node, &self->maps); list_add_tail(&map->node, &self->maps);
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment