net: Allow opt-out from global protocol memory accounting.

Some protocols (e.g., TCP, UDP) implement memory accounting for socket
buffers and charge memory to per-protocol global counters pointed to by
sk->sk_proto->memory_allocated.

Sometimes, system processes do not want that limitation. For a similar
purpose, there is SO_RESERVE_MEM for sockets under memcg.

Also, by opting out of the per-protocol accounting, sockets under memcg
can avoid paying costs for two orthogonal memory accounting mechanisms.
A microbenchmark result is in the subsequent bpf patch.

Let's allow opt-out from the per-protocol memory accounting if
sk->sk_bypass_prot_mem is true.

sk->sk_bypass_prot_mem and sk->sk_prot are placed in the same cache
line, and sk_has_account() always fetches sk->sk_prot before accessing
sk->sk_bypass_prot_mem, so there is no extra cache miss for this patch.

The following patches will set sk->sk_bypass_prot_mem to true, and
then, the per-protocol memory accounting will be skipped.

Note that this does NOT disable memcg, but rather the per-protocol one.

Another option not to use the hole in struct sock_common is create
sk_prot variants like tcp_prot_bypass, but this would complicate
SOCKMAP logic, tcp_bpf_prots etc.

Signed-off-by: Kuniyuki Iwashima <[e-mail]>
Signed-off-by: Martin KaFai Lau <[e-mail]>
Reviewed-by: Shakeel Butt <[e-mail]>
Reviewed-by: Eric Dumazet <[e-mail]>
Acked-by: Roman Gushchin <[e-mail]>
Link: https://patch.msgid.link/[e-mail]

include/net/proto_memory.h

···
 	    mem_cgroup_sk_under_memory_pressure(sk))
 		return true;
 
+	if (sk->sk_bypass_prot_mem)
+		return false;
+
 	return !!READ_ONCE(*sk->sk_prot->memory_pressure);
 }
 

include/net/sock.h

···
  *	@skc_reuseport: %SO_REUSEPORT setting
  *	@skc_ipv6only: socket is IPV6 only
  *	@skc_net_refcnt: socket is using net ref counting
+ *	@skc_bypass_prot_mem: bypass the per-protocol memory accounting for skb
  *	@skc_bound_dev_if: bound device index if != 0
  *	@skc_bind_node: bind hash linkage for various protocol lookup tables
  *	@skc_portaddr_node: second hash linkage for UDP/UDP-Lite protocol
···
 	unsigned char		skc_reuseport:1;
 	unsigned char		skc_ipv6only:1;
 	unsigned char		skc_net_refcnt:1;
+	unsigned char		skc_bypass_prot_mem:1;
 	int			skc_bound_dev_if;
 	union {
 		struct hlist_node	skc_bind_node;
···
 #define sk_reuseport		__sk_common.skc_reuseport
 #define sk_ipv6only		__sk_common.skc_ipv6only
 #define sk_net_refcnt		__sk_common.skc_net_refcnt
+#define sk_bypass_prot_mem	__sk_common.skc_bypass_prot_mem
 #define sk_bound_dev_if		__sk_common.skc_bound_dev_if
 #define sk_bind_node		__sk_common.skc_bind_node
 #define sk_prot			__sk_common.skc_prot

include/net/tcp.h

···
 	    mem_cgroup_sk_under_memory_pressure(sk))
 		return true;
 
+	if (sk->sk_bypass_prot_mem)
+		return false;
+
 	return READ_ONCE(tcp_memory_pressure);
 }
 /*

net/core/sock.c

···
 	if (!charged)
 		return -ENOMEM;
 
+	if (sk->sk_bypass_prot_mem)
+		goto success;
+
 	/* pre-charge to forward_alloc */
 	sk_memory_allocated_add(sk, pages);
 	allocated = sk_memory_allocated(sk);
+
 	/* If the system goes into memory pressure with this
 	 * precharge, give up and return error.
 	 */
···
 		mem_cgroup_sk_uncharge(sk, pages);
 		return -ENOMEM;
 	}
+
+success:
 	sk_forward_alloc_add(sk, pages << PAGE_SHIFT);
 
 	WRITE_ONCE(sk->sk_reserved_mem,
···
 	if (likely(skb_page_frag_refill(32U, pfrag, sk->sk_allocation)))
 		return true;
 
-	sk_enter_memory_pressure(sk);
+	if (!sk->sk_bypass_prot_mem)
+		sk_enter_memory_pressure(sk);
+
 	sk_stream_moderate_sndbuf(sk);
+
 	return false;
 }
 EXPORT_SYMBOL(sk_page_frag_refill);
···
 {
 	bool memcg_enabled = false, charged = false;
 	struct proto *prot = sk->sk_prot;
-	long allocated;
+	long allocated = 0;
 
-	sk_memory_allocated_add(sk, amt);
-	allocated = sk_memory_allocated(sk);
+	if (!sk->sk_bypass_prot_mem) {
+		sk_memory_allocated_add(sk, amt);
+		allocated = sk_memory_allocated(sk);
+	}
 
 	if (mem_cgroup_sk_enabled(sk)) {
 		memcg_enabled = true;
···
 		if (!charged)
 			goto suppress_allocation;
 	}
+
+	if (!allocated)
+		return 1;
 
 	/* Under limit. */
 	if (allocated <= sk_prot_mem_limits(sk, 0)) {
···
 
 	trace_sock_exceed_buf_limit(sk, prot, allocated, kind);
 
-	sk_memory_allocated_sub(sk, amt);
+	if (allocated)
+		sk_memory_allocated_sub(sk, amt);
 
 	if (charged)
 		mem_cgroup_sk_uncharge(sk, amt);
···
  */
 void __sk_mem_reduce_allocated(struct sock *sk, int amount)
 {
-	sk_memory_allocated_sub(sk, amount);
-
 	if (mem_cgroup_sk_enabled(sk))
 		mem_cgroup_sk_uncharge(sk, amount);
+
+	if (sk->sk_bypass_prot_mem)
+		return;
+
+	sk_memory_allocated_sub(sk, amount);
 
 	if (sk_under_global_memory_pressure(sk) &&
 	    (sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)))

net/ipv4/tcp.c

···
 		}
 		__kfree_skb(skb);
 	} else {
-		sk->sk_prot->enter_memory_pressure(sk);
+		if (!sk->sk_bypass_prot_mem)
+			tcp_enter_memory_pressure(sk);
 		sk_stream_moderate_sndbuf(sk);
 	}
 	return NULL;

net/ipv4/tcp_output.c

···
 	delta = size - sk->sk_forward_alloc;
 	if (delta <= 0)
 		return;
+
 	amt = sk_mem_pages(delta);
 	sk_forward_alloc_add(sk, amt << PAGE_SHIFT);
-	sk_memory_allocated_add(sk, amt);
 
 	if (mem_cgroup_sk_enabled(sk))
 		mem_cgroup_sk_charge(sk, amt, gfp_memcg_charge() | __GFP_NOFAIL);
+
+	if (sk->sk_bypass_prot_mem)
+		return;
+
+	sk_memory_allocated_add(sk, amt);
 }
 
 /* Send a FIN. The caller locks the socket for us.

net/mptcp/protocol.c

···
 	mptcp_for_each_subflow(msk, subflow) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 
-		if (first)
+		if (first && !ssk->sk_bypass_prot_mem) {
 			tcp_enter_memory_pressure(ssk);
-		sk_stream_moderate_sndbuf(ssk);
+			first = false;
+		}
 
-		first = false;
+		sk_stream_moderate_sndbuf(ssk);
 	}
 	__mptcp_sync_sndbuf(sk);
 }

net/tls/tls_device.c

···
 	if (!offload_ctx->open_record) {
 		if (unlikely(!skb_page_frag_refill(prepend_size, pfrag,
 						   sk->sk_allocation))) {
-			READ_ONCE(sk->sk_prot)->enter_memory_pressure(sk);
+			if (!sk->sk_bypass_prot_mem)
+				READ_ONCE(sk->sk_prot)->enter_memory_pressure(sk);
 			sk_stream_moderate_sndbuf(sk);
 			return -ENOMEM;
 		}