From: Eric Dumazet <edumazet@google.com>
Date: Fri, 18 Feb 2022 18:18:01 +0000 (-0800)
Subject: bpf: Call maybe_wait_bpf_programs() only once from generic_map_delete_batch()
X-Git-Url: http://git.maquefel.me/?a=commitdiff_plain;h=9087c6ff8dfe0a070e4e05a434399080603c29de;p=linux.git

bpf: Call maybe_wait_bpf_programs() only once from generic_map_delete_batch()

As stated in the comment found in maybe_wait_bpf_programs(),
the synchronize_rcu() barrier is only needed before returning
to userspace, not after each deletion in the batch.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/bpf/20220218181801.2971275-1-eric.dumazet@gmail.com
---

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index a72f63d5a7dae..9c7a72b65eee0 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1352,7 +1352,6 @@ int generic_map_delete_batch(struct bpf_map *map,
 		err = map->ops->map_delete_elem(map, key);
 		rcu_read_unlock();
 		bpf_enable_instrumentation();
-		maybe_wait_bpf_programs(map);
 		if (err)
 			break;
 		cond_resched();
@@ -1361,6 +1360,8 @@ int generic_map_delete_batch(struct bpf_map *map,
 		err = -EFAULT;
 
 	kvfree(key);
+
+	maybe_wait_bpf_programs(map);
 	return err;
 }