bpf: net: Avoid copying sk_user_data of reuseport_array during sk_clone

It makes little sense for copying sk_user_data of reuseport_array during sk_clone_lock(). This patch reuses the SK_USER_DATA_NOCOPY bit introduced in commit f1ff5ce2cd5e ("net, sk_msg: Clear sk_user_data pointer on clone if tagged"). It is used to mark the sk_user_data is not supposed to be copied to its clone. Although the cloned sk's sk_user_data will not be used/freed in bpf_sk_reuseport_detach(), this change can still allow the cloned sk's sk_user_data to be used by some other means. Freeing the reuseport_array's sk_user_data does not require a rcu grace period. Thus, the existing rcu_assign_sk_user_data_nocopy() is not used.

Martin KaFai Lau committed 6 years ago

Commit f3dda7a1 parent 365f9ae1 file · +9 −4Browse files

Unified · Split

kernel/bpf/reuseport_array.c+9−4

@@ -20,11 +20,14 @@ static struct reuseport_array reuseport_array(struct bpf_map map) · 19 unmodified lines
20	20	/* The caller must hold the reuseport_lock */
21	21	void bpf_sk_reuseport_detach(struct sock *sk)
22	22	{
23		− struct sock __rcu **socks;
	23	+ uintptr_t sk_user_data;
24	24
25	25	write_lock_bh(&sk->sk_callback_lock);
26		− socks = sk->sk_user_data;
27		− if (socks) {
	26	+ sk_user_data = (uintptr_t)sk->sk_user_data;
	27	+ if (sk_user_data) {
	28	+ struct sock __rcu **socks;
	29	+
	30	+ socks = (void *)(sk_user_data & SK_USER_DATA_PTRMASK);
28	31	WRITE_ONCE(sk->sk_user_data, NULL);
29	32	/*
30	33	* Do not move this NULL assignment outside of
@@ -252,6 +255,7 @@ int bpf_fd_reuseport_array_update_elem(struct bpf_map map, void key, · 221 unmodified lines
252	255	struct sock free_osk = NULL, osk, *nsk;
253	256	struct sock_reuseport *reuse;
254	257	u32 index = (u32 )key;
	258	+ uintptr_t sk_user_data;
255	259	struct socket *socket;
256	260	int err, fd;
257	261
@@ -305,7 +309,8 @@ int bpf_fd_reuseport_array_update_elem(struct bpf_map map, void key, · 47 unmodified lines
305	309	if (err)
306	310	goto put_file_unlock;
307	311
308		− WRITE_ONCE(nsk->sk_user_data, &array->ptrs[index]);
	312	+ sk_user_data = (uintptr_t)&array->ptrs[index] \| SK_USER_DATA_NOCOPY;
	313	+ WRITE_ONCE(nsk->sk_user_data, (void *)sk_user_data);
309	314	rcu_assign_pointer(array->ptrs[index], nsk);
310	315	free_osk = osk;
311	316	err = 0;