From 9a3efb6b661f71d5675369ace9257833f0e78ef3 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Tue, 13 Feb 2018 19:00:21 -0800 Subject: bpf: fix memory leak in lpm_trie map_free callback function There is a memory leak happening in lpm_trie map_free callback function trie_free. The trie structure itself does not get freed. Also, trie_free function did not do synchronize_rcu before freeing various data structures. This is incorrect as some rcu_read_lock region(s) for lookup, update, delete or get_next_key may not complete yet. The fix is to add synchronize_rcu in the beginning of trie_free. The useless spin_lock is removed from this function as well. Fixes: b95a5c4db09b ("bpf: add a longest prefix match trie map implementation") Reported-by: Mathieu Malaterre Reported-by: Alexei Starovoitov Tested-by: Mathieu Malaterre Signed-off-by: Yonghong Song Signed-off-by: Alexei Starovoitov --- kernel/bpf/lpm_trie.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'kernel/bpf/lpm_trie.c') diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c index 7b469d10d0e9..a75e02c961b5 100644 --- a/kernel/bpf/lpm_trie.c +++ b/kernel/bpf/lpm_trie.c @@ -555,7 +555,10 @@ static void trie_free(struct bpf_map *map) struct lpm_trie_node __rcu **slot; struct lpm_trie_node *node; - raw_spin_lock(&trie->lock); + /* Wait for outstanding programs to complete + * update/lookup/delete/get_next_key and free the trie. + */ + synchronize_rcu(); /* Always start at the root and walk down to a node that has no * children. Then free that node, nullify its reference in the parent @@ -569,7 +572,7 @@ static void trie_free(struct bpf_map *map) node = rcu_dereference_protected(*slot, lockdep_is_held(&trie->lock)); if (!node) - goto unlock; + goto out; if (rcu_access_pointer(node->child[0])) { slot = &node->child[0]; @@ -587,8 +590,8 @@ static void trie_free(struct bpf_map *map) } } -unlock: - raw_spin_unlock(&trie->lock); +out: + kfree(trie); } static int trie_get_next_key(struct bpf_map *map, void *_key, void *_next_key) -- cgit From 6c5f61023c5b0edb0c8a64c902fe97c6453b1852 Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Thu, 22 Feb 2018 10:10:35 -0800 Subject: bpf: fix rcu lockdep warning for lpm_trie map_free callback Commit 9a3efb6b661f ("bpf: fix memory leak in lpm_trie map_free callback function") fixed a memory leak and removed unnecessary locks in map_free callback function. Unfortrunately, it introduced a lockdep warning. When lockdep checking is turned on, running tools/testing/selftests/bpf/test_lpm_map will have: [ 98.294321] ============================= [ 98.294807] WARNING: suspicious RCU usage [ 98.295359] 4.16.0-rc2+ #193 Not tainted [ 98.295907] ----------------------------- [ 98.296486] /home/yhs/work/bpf/kernel/bpf/lpm_trie.c:572 suspicious rcu_dereference_check() usage! [ 98.297657] [ 98.297657] other info that might help us debug this: [ 98.297657] [ 98.298663] [ 98.298663] rcu_scheduler_active = 2, debug_locks = 1 [ 98.299536] 2 locks held by kworker/2:1/54: [ 98.300152] #0: ((wq_completion)"events"){+.+.}, at: [<00000000196bc1f0>] process_one_work+0x157/0x5c0 [ 98.301381] #1: ((work_completion)(&map->work)){+.+.}, at: [<00000000196bc1f0>] process_one_work+0x157/0x5c0 Since actual trie tree removal happens only after no other accesses to the tree are possible, replacing rcu_dereference_protected(*slot, lockdep_is_held(&trie->lock)) with rcu_dereference_protected(*slot, 1) fixed the issue. Fixes: 9a3efb6b661f ("bpf: fix memory leak in lpm_trie map_free callback function") Reported-by: Eric Dumazet Suggested-by: Eric Dumazet Signed-off-by: Yonghong Song Reviewed-by: Eric Dumazet Acked-by: David S. Miller Signed-off-by: Daniel Borkmann --- kernel/bpf/lpm_trie.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel/bpf/lpm_trie.c') diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c index a75e02c961b5..b4b5b81e7251 100644 --- a/kernel/bpf/lpm_trie.c +++ b/kernel/bpf/lpm_trie.c @@ -569,8 +569,7 @@ static void trie_free(struct bpf_map *map) slot = &trie->root; for (;;) { - node = rcu_dereference_protected(*slot, - lockdep_is_held(&trie->lock)); + node = rcu_dereference_protected(*slot, 1); if (!node) goto out; -- cgit