Merge branch 'bpf: Remove recursion check for struct_ops prog'

Martin KaFai Lau says:

====================

From: Martin KaFai Lau <martin.lau@kernel.org>

The struct_ops is sharing the tracing-trampoline's enter/exit
function which tracks prog->active to avoid recursion.  It turns
out the struct_ops bpf prog will hit this prog->active and
unnecessarily skipped running the struct_ops prog.  eg.  The
'.ssthresh' may run in_task() and then interrupted by softirq
that runs the same '.ssthresh'.

The kernel does not call the tcp-cc's ops in a recursive way,
so this set is to remove the recursion check for struct_ops prog.

v3:
- Clear the bpf_chg_cc_inprogress from the newly cloned tcp_sock
  in tcp_create_openreq_child() because the listen sk can
  be cloned without lock being held. (Eric Dumazet)

v2:
- v1 [0] turned into a long discussion on a few cases and also
  whether it needs to follow the bpf_run_ctx chain if there is
  tracing bpf_run_ctx (kprobe/trace/trampoline) running in between.

  It is a good signal that it is not obvious enough to reason
  about it and needs a tradeoff for a more straight forward approach.

  This revision uses one bit out of an existing 1 byte hole
  in the tcp_sock.  It is in Patch 4.

  [0]: https://lore.kernel.org/bpf/20220922225616.3054840-1-kafai@fb.com/T/#md98d40ac5ec295fdadef476c227a3401b2b6b911
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

arch/x86/net/bpf_jit_comp.c

@@ -1836,6 +1836,9 @@ static int invoke_bpf_prog(const struct btf_func_model *m, u8 **pprog,
 	if (p->aux->sleepable) {
 		enter = __bpf_prog_enter_sleepable;
 		exit = __bpf_prog_exit_sleepable;
+	} else if (p->type == BPF_PROG_TYPE_STRUCT_OPS) {
+		enter = __bpf_prog_enter_struct_ops;
+		exit = __bpf_prog_exit_struct_ops;
 	} else if (p->expected_attach_type == BPF_LSM_CGROUP) {
 		enter = __bpf_prog_enter_lsm_cgroup;
 		exit = __bpf_prog_exit_lsm_cgroup;

include/linux/bpf.h

@@ -864,6 +864,10 @@ u64 notrace __bpf_prog_enter_lsm_cgroup(struct bpf_prog *prog,
 					struct bpf_tramp_run_ctx *run_ctx);
 void notrace __bpf_prog_exit_lsm_cgroup(struct bpf_prog *prog, u64 start,
 					struct bpf_tramp_run_ctx *run_ctx);
+u64 notrace __bpf_prog_enter_struct_ops(struct bpf_prog *prog,
+					struct bpf_tramp_run_ctx *run_ctx);
+void notrace __bpf_prog_exit_struct_ops(struct bpf_prog *prog, u64 start,
+					struct bpf_tramp_run_ctx *run_ctx);
 void notrace __bpf_tramp_enter(struct bpf_tramp_image *tr);
 void notrace __bpf_tramp_exit(struct bpf_tramp_image *tr);
 

include/linux/tcp.h

@@ -388,6 +388,12 @@ struct tcp_sock {
 	u8	bpf_sock_ops_cb_flags;  /* Control calling BPF programs
 					 * values defined in uapi/linux/tcp.h
 					 */
+	u8	bpf_chg_cc_inprogress:1; /* In the middle of
+					  * bpf_setsockopt(TCP_CONGESTION),
+					  * it is to avoid the bpf_tcp_cc->init()
+					  * to recur itself by calling
+					  * bpf_setsockopt(TCP_CONGESTION, "itself").
+					  */
 #define BPF_SOCK_OPS_TEST_FLAG(TP, ARG) (TP->bpf_sock_ops_cb_flags & ARG)
 #else
 #define BPF_SOCK_OPS_TEST_FLAG(TP, ARG) 0

kernel/bpf/trampoline.c

@@ -964,6 +964,29 @@ void notrace __bpf_prog_exit_sleepable(struct bpf_prog *prog, u64 start,
 	rcu_read_unlock_trace();
 }
 
+u64 notrace __bpf_prog_enter_struct_ops(struct bpf_prog *prog,
+					struct bpf_tramp_run_ctx *run_ctx)
+	__acquires(RCU)
+{
+	rcu_read_lock();
+	migrate_disable();
+
+	run_ctx->saved_run_ctx = bpf_set_run_ctx(&run_ctx->run_ctx);
+
+	return bpf_prog_start_time();
+}
+
+void notrace __bpf_prog_exit_struct_ops(struct bpf_prog *prog, u64 start,
+					struct bpf_tramp_run_ctx *run_ctx)
+	__releases(RCU)
+{
+	bpf_reset_run_ctx(run_ctx->saved_run_ctx);
+
+	update_prog_stats(prog, start);
+	migrate_enable();
+	rcu_read_unlock();
+}
+
 void notrace __bpf_tramp_enter(struct bpf_tramp_image *tr)
 {
 	percpu_ref_get(&tr->pcref);

net/core/filter.c

@@ -5102,6 +5102,59 @@ static int bpf_sol_tcp_setsockopt(struct sock *sk, int optname,
 	return 0;
 }
 
+static int sol_tcp_sockopt_congestion(struct sock *sk, char *optval,
+				      int *optlen, bool getopt)
+{
+	struct tcp_sock *tp;
+	int ret;
+
+	if (*optlen < 2)
+		return -EINVAL;
+
+	if (getopt) {
+		if (!inet_csk(sk)->icsk_ca_ops)
+			return -EINVAL;
+		/* BPF expects NULL-terminated tcp-cc string */
+		optval[--(*optlen)] = '\0';
+		return do_tcp_getsockopt(sk, SOL_TCP, TCP_CONGESTION,
+					 KERNEL_SOCKPTR(optval),
+					 KERNEL_SOCKPTR(optlen));
+	}
+
+	/* "cdg" is the only cc that alloc a ptr
+	 * in inet_csk_ca area.  The bpf-tcp-cc may
+	 * overwrite this ptr after switching to cdg.
+	 */
+	if (*optlen >= sizeof("cdg") - 1 && !strncmp("cdg", optval, *optlen))
+		return -ENOTSUPP;
+
+	/* It stops this looping
+	 *
+	 * .init => bpf_setsockopt(tcp_cc) => .init =>
+	 * bpf_setsockopt(tcp_cc)" => .init => ....
+	 *
+	 * The second bpf_setsockopt(tcp_cc) is not allowed
+	 * in order to break the loop when both .init
+	 * are the same bpf prog.
+	 *
+	 * This applies even the second bpf_setsockopt(tcp_cc)
+	 * does not cause a loop.  This limits only the first
+	 * '.init' can call bpf_setsockopt(TCP_CONGESTION) to
+	 * pick a fallback cc (eg. peer does not support ECN)
+	 * and the second '.init' cannot fallback to
+	 * another.
+	 */
+	tp = tcp_sk(sk);
+	if (tp->bpf_chg_cc_inprogress)
+		return -EBUSY;
+
+	tp->bpf_chg_cc_inprogress = 1;
+	ret = do_tcp_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
+				KERNEL_SOCKPTR(optval), *optlen);
+	tp->bpf_chg_cc_inprogress = 0;
+	return ret;
+}
+
 static int sol_tcp_sockopt(struct sock *sk, int optname,
 			   char *optval, int *optlen,
 			   bool getopt)
@@ -5125,9 +5178,7 @@ static int sol_tcp_sockopt(struct sock *sk, int optname,
 			return -EINVAL;
 		break;
 	case TCP_CONGESTION:
-		if (*optlen < 2)
-			return -EINVAL;
-		break;
+		return sol_tcp_sockopt_congestion(sk, optval, optlen, getopt);
 	case TCP_SAVED_SYN:
 		if (*optlen < 1)
 			return -EINVAL;
@@ -5152,13 +5203,6 @@ static int sol_tcp_sockopt(struct sock *sk, int optname,
 			return 0;
 		}
 
-		if (optname == TCP_CONGESTION) {
-			if (!inet_csk(sk)->icsk_ca_ops)
-				return -EINVAL;
-			/* BPF expects NULL-terminated tcp-cc string */
-			optval[--(*optlen)] = '\0';
-		}
-
 		return do_tcp_getsockopt(sk, SOL_TCP, optname,
 					 KERNEL_SOCKPTR(optval),
 					 KERNEL_SOCKPTR(optlen));
@@ -5285,12 +5329,6 @@ static int _bpf_getsockopt(struct sock *sk, int level, int optname,
 BPF_CALL_5(bpf_sk_setsockopt, struct sock *, sk, int, level,
 	   int, optname, char *, optval, int, optlen)
 {
-	if (level == SOL_TCP && optname == TCP_CONGESTION) {
-		if (optlen >= sizeof("cdg") - 1 &&
-		    !strncmp("cdg", optval, optlen))
-			return -ENOTSUPP;
-	}
-
 	return _bpf_setsockopt(sk, level, optname, optval, optlen);
 }
 

net/ipv4/tcp_minisocks.c

@@ -541,6 +541,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 	newtp->fastopen_req = NULL;
 	RCU_INIT_POINTER(newtp->fastopen_rsk, NULL);
 
+	newtp->bpf_chg_cc_inprogress = 0;
 	tcp_bpf_clone(sk, newsk);
 
 	__TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS);

tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c

@@ -290,6 +290,10 @@ static void test_dctcp_fallback(void)
 		goto done;
 	ASSERT_STREQ(dctcp_skel->bss->cc_res, "cubic", "cc_res");
 	ASSERT_EQ(dctcp_skel->bss->tcp_cdg_res, -ENOTSUPP, "tcp_cdg_res");
+	/* All setsockopt(TCP_CONGESTION) in the recurred
+	 * bpf_dctcp->init() should fail with -EBUSY.
+	 */
+	ASSERT_EQ(dctcp_skel->bss->ebusy_cnt, 3, "ebusy_cnt");
 
 	err = getsockopt(srv_fd, SOL_TCP, TCP_CONGESTION, srv_cc, &cc_len);
 	if (!ASSERT_OK(err, "getsockopt(srv_fd, TCP_CONGESTION)"))

tools/testing/selftests/bpf/progs/bpf_dctcp.c

@@ -11,6 +11,7 @@
 #include <linux/types.h>
 #include <linux/stddef.h>
 #include <linux/tcp.h>
+#include <errno.h>
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 #include "bpf_tcp_helpers.h"
@@ -23,6 +24,7 @@ const char tcp_cdg[] = "cdg";
 char cc_res[TCP_CA_NAME_MAX];
 int tcp_cdg_res = 0;
 int stg_result = 0;
+int ebusy_cnt = 0;
 
 struct {
 	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
@@ -64,16 +66,23 @@ void BPF_PROG(dctcp_init, struct sock *sk)
 
 	if (!(tp->ecn_flags & TCP_ECN_OK) && fallback[0]) {
 		/* Switch to fallback */
-		bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
-			       (void *)fallback, sizeof(fallback));
-		/* Switch back to myself which the bpf trampoline
-		 * stopped calling dctcp_init recursively.
+		if (bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
+				   (void *)fallback, sizeof(fallback)) == -EBUSY)
+			ebusy_cnt++;
+
+		/* Switch back to myself and the recurred dctcp_init()
+		 * will get -EBUSY for all bpf_setsockopt(TCP_CONGESTION),
+		 * except the last "cdg" one.
 		 */
-		bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
-			       (void *)bpf_dctcp, sizeof(bpf_dctcp));
+		if (bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
+				   (void *)bpf_dctcp, sizeof(bpf_dctcp)) == -EBUSY)
+			ebusy_cnt++;
+
 		/* Switch back to fallback */
-		bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
-			       (void *)fallback, sizeof(fallback));
+		if (bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
+				   (void *)fallback, sizeof(fallback)) == -EBUSY)
+			ebusy_cnt++;
+
 		/* Expecting -ENOTSUPP for tcp_cdg_res */
 		tcp_cdg_res = bpf_setsockopt(sk, SOL_TCP, TCP_CONGESTION,
 					     (void *)tcp_cdg, sizeof(tcp_cdg));