perf tools: Preserve eBPF maps when loading kcore

We need to preserve eBPF maps even if they are covered by kcore, because
we need to access eBPF dso for source data.

Add the map_groups__merge_in function to do that.  It merges a map into
map_groups by splitting the new map within the existing map regions.

Suggested-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stanislav Fomichev <sdf@google.com>
Link: http://lkml.kernel.org/r/20190508132010.14512-9-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

1 files changed, 93 insertions, 4 deletions

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 5cbad55cd99d..29780fcd049c 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1166,6 +1166,85 @@ static int kcore_mapfn(u64 start, u64 len, u64 pgoff, void *data)
 	return 0;
 }
 
+/*
+ * Merges map into map_groups by splitting the new map
+ * within the existing map regions.
+ */
+static int map_groups__merge_in(struct map_groups *kmaps, struct map *new_map)
+{
+	struct map *old_map;
+	LIST_HEAD(merged);
+
+	for (old_map = map_groups__first(kmaps); old_map;
+	     old_map = map_groups__next(old_map)) {
+
+		/* no overload with this one */
+		if (new_map->end < old_map->start ||
+		    new_map->start >= old_map->end)
+			continue;
+
+		if (new_map->start < old_map->start) {
+			/*
+			 * |new......
+			 *       |old....
+			 */
+			if (new_map->end < old_map->end) {
+				/*
+				 * |new......|     -> |new..|
+				 *       |old....| ->       |old....|
+				 */
+				new_map->end = old_map->start;
+			} else {
+				/*
+				 * |new.............| -> |new..|       |new..|
+				 *       |old....|    ->       |old....|
+				 */
+				struct map *m = map__clone(new_map);
+
+				if (!m)
+					return -ENOMEM;
+
+				m->end = old_map->start;
+				list_add_tail(&m->node, &merged);
+				new_map->start = old_map->end;
+			}
+		} else {
+			/*
+			 *      |new......
+			 * |old....
+			 */
+			if (new_map->end < old_map->end) {
+				/*
+				 *      |new..|   -> x
+				 * |old.........| -> |old.........|
+				 */
+				map__put(new_map);
+				new_map = NULL;
+				break;
+			} else {
+				/*
+				 *      |new......| ->         |new...|
+				 * |old....|        -> |old....|
+				 */
+				new_map->start = old_map->end;
+			}
+		}
+	}
+
+	while (!list_empty(&merged)) {
+		old_map = list_entry(merged.next, struct map, node);
+		list_del_init(&old_map->node);
+		map_groups__insert(kmaps, old_map);
+		map__put(old_map);
+	}
+
+	if (new_map) {
+		map_groups__insert(kmaps, new_map);
+		map__put(new_map);
+	}
+	return 0;
+}
+
 static int dso__load_kcore(struct dso *dso, struct map *map,
 			   const char *kallsyms_filename)
 {
@@ -1222,7 +1301,12 @@ static int dso__load_kcore(struct dso *dso, struct map *map,
 	while (old_map) {
 		struct map *next = map_groups__next(old_map);
 
-		if (old_map != map)
+		/*
+		 * We need to preserve eBPF maps even if they are
+		 * covered by kcore, because we need to access
+		 * eBPF dso for source data.
+		 */
+		if (old_map != map && !__map__is_bpf_prog(old_map))
 			map_groups__remove(kmaps, old_map);
 		old_map = next;
 	}
@@ -1256,11 +1340,16 @@ static int dso__load_kcore(struct dso *dso, struct map *map,
 			map_groups__remove(kmaps, map);
 			map_groups__insert(kmaps, map);
 			map__put(map);
+			map__put(new_map);
 		} else {
-			map_groups__insert(kmaps, new_map);
+			/*
+			 * Merge kcore map into existing maps,
+			 * and ensure that current maps (eBPF)
+			 * stay intact.
+			 */
+			if (map_groups__merge_in(kmaps, new_map))
+				goto out_err;
 		}
-
-		map__put(new_map);
 	}
 
 	if (machine__is(machine, "x86_64")) {