Merge branch 'vsock-sockmap-support'

Bobby Eshleman says:

====================
Add support for sockmap to vsock.

We're testing usage of vsock as a way to redirect guest-local UDS
requests to the host and this patch series greatly improves the
performance of such a setup.

Compared to copying packets via userspace, this improves throughput by
121% in basic testing.

Tested as follows.

Setup: guest unix dgram sender -> guest vsock redirector -> host vsock
       server
Threads: 1
Payload: 64k
No sockmap:
- 76.3 MB/s
- The guest vsock redirector was
  "socat VSOCK-CONNECT:2:1234 UNIX-RECV:/path/to/sock"
Using sockmap (this patch):
- 168.8 MB/s (+121%)
- The guest redirector was a simple sockmap echo server,
  redirecting unix ingress to vsock 2:1234 egress.
- Same sender and server programs

*Note: these numbers are from RFC v1

Only the virtio transport has been tested. The loopback transport was
used in writing bpf/selftests, but not thoroughly tested otherwise.

This series requires the skb patch.

Changes in v4:
- af_vsock: fix parameter alignment in vsock_dgram_recvmsg()
- af_vsock: add TCP_ESTABLISHED comment in vsock_dgram_connect()
- vsock/bpf: change ret type to bool

Changes in v3:
- vsock/bpf: Refactor wait logic in vsock_bpf_recvmsg() to avoid
  backwards goto
- vsock/bpf: Check psock before acquiring slock
- vsock/bpf: Return bool instead of int of 0 or 1
- vsock/bpf: Wrap macro args __sk/__psock in parens
- vsock/bpf: Place comment trailer */ on separate line

Changes in v2:
- vsock/bpf: rename vsock_dgram_* -> vsock_*
- vsock/bpf: change sk_psock_{get,put} and {lock,release}_sock() order
  to minimize slock hold time
- vsock/bpf: use "new style" wait
- vsock/bpf: fix bug in wait log
- vsock/bpf: add check that recvmsg sk_type is one dgram, seqpacket, or
  stream.  Return error if not one of the three.
- virtio/vsock: comment __skb_recv_datagram() usage
- virtio/vsock: do not init copied in read_skb()
- vsock/bpf: add ifdef guard around struct proto in dgram_recvmsg()
- selftests/bpf: add vsock loopback config for aarch64
- selftests/bpf: add vsock loopback config for s390x
- selftests/bpf: remove vsock device from vmtest.sh qemu machine
- selftests/bpf: remove CONFIG_VIRTIO_VSOCKETS=y from config.x86_64
- vsock/bpf: move transport-related (e.g., if (!vsk->transport)) checks
  out of fast path
====================

Signed-off-by: Bobby Eshleman <bobby.eshleman@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

4 files changed, 171 insertions, 0 deletions

diff --git a/tools/testing/selftests/bpf/config.aarch64 b/tools/testing/selftests/bpf/config.aarch64
index 1f0437644186..253821494884 100644
--- a/tools/testing/selftests/bpf/config.aarch64
+++ b/tools/testing/selftests/bpf/config.aarch64
@@ -176,6 +176,8 @@ CONFIG_VIRTIO_MMIO_CMDLINE_DEVICES=y
 CONFIG_VIRTIO_MMIO=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
 CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_XFRM_USER=y
diff --git a/tools/testing/selftests/bpf/config.s390x b/tools/testing/selftests/bpf/config.s390x
index d49f6170e7bd..2ba92167be35 100644
--- a/tools/testing/selftests/bpf/config.s390x
+++ b/tools/testing/selftests/bpf/config.s390x
@@ -140,5 +140,8 @@ CONFIG_VIRTIO_BALLOON=y
 CONFIG_VIRTIO_BLK=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
+CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_XFRM_USER=y
diff --git a/tools/testing/selftests/bpf/config.x86_64 b/tools/testing/selftests/bpf/config.x86_64
index dd97d61d325c..b650b2e617b8 100644
--- a/tools/testing/selftests/bpf/config.x86_64
+++ b/tools/testing/selftests/bpf/config.x86_64
@@ -234,7 +234,10 @@ CONFIG_VIRTIO_BLK=y
 CONFIG_VIRTIO_CONSOLE=y
 CONFIG_VIRTIO_NET=y
 CONFIG_VIRTIO_PCI=y
+CONFIG_VIRTIO_VSOCKETS_COMMON=y
 CONFIG_VLAN_8021Q=y
+CONFIG_VSOCKETS=y
+CONFIG_VSOCKETS_LOOPBACK=y
 CONFIG_X86_ACPI_CPUFREQ=y
 CONFIG_X86_CPUID=y
 CONFIG_X86_MSR=y
diff --git a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
index 567e07c19ecc..8f09e1ea3ba7 100644
--- a/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
+++ b/tools/testing/selftests/bpf/prog_tests/sockmap_listen.c
@@ -18,6 +18,7 @@
 #include <string.h>
 #include <sys/select.h>
 #include <unistd.h>
+#include <linux/vm_sockets.h>
 
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
@@ -251,6 +252,16 @@ static void init_addr_loopback6(struct sockaddr_storage *ss, socklen_t *len)
 	*len = sizeof(*addr6);
 }
 
+static void init_addr_loopback_vsock(struct sockaddr_storage *ss, socklen_t *len)
+{
+	struct sockaddr_vm *addr = memset(ss, 0, sizeof(*ss));
+
+	addr->svm_family = AF_VSOCK;
+	addr->svm_port = VMADDR_PORT_ANY;
+	addr->svm_cid = VMADDR_CID_LOCAL;
+	*len = sizeof(*addr);
+}
+
 static void init_addr_loopback(int family, struct sockaddr_storage *ss,
 			       socklen_t *len)
 {
@@ -261,6 +272,9 @@ static void init_addr_loopback(int family, struct sockaddr_storage *ss,
 	case AF_INET6:
 		init_addr_loopback6(ss, len);
 		return;
+	case AF_VSOCK:
+		init_addr_loopback_vsock(ss, len);
+		return;
 	default:
 		FAIL("unsupported address family %d", family);
 	}
@@ -1478,6 +1492,8 @@ static const char *family_str(sa_family_t family)
 		return "IPv6";
 	case AF_UNIX:
 		return "Unix";
+	case AF_VSOCK:
+		return "VSOCK";
 	default:
 		return "unknown";
 	}
@@ -1689,6 +1705,151 @@ static void test_unix_redir(struct test_sockmap_listen *skel, struct bpf_map *ma
 	unix_skb_redir_to_connected(skel, map, sotype);
 }
 
+/* Returns two connected loopback vsock sockets */
+static int vsock_socketpair_connectible(int sotype, int *v0, int *v1)
+{
+	struct sockaddr_storage addr;
+	socklen_t len = sizeof(addr);
+	int s, p, c;
+
+	s = socket_loopback(AF_VSOCK, sotype);
+	if (s < 0)
+		return -1;
+
+	c = xsocket(AF_VSOCK, sotype | SOCK_NONBLOCK, 0);
+	if (c == -1)
+		goto close_srv;
+
+	if (getsockname(s, sockaddr(&addr), &len) < 0)
+		goto close_cli;
+
+	if (connect(c, sockaddr(&addr), len) < 0 && errno != EINPROGRESS) {
+		FAIL_ERRNO("connect");
+		goto close_cli;
+	}
+
+	len = sizeof(addr);
+	p = accept_timeout(s, sockaddr(&addr), &len, IO_TIMEOUT_SEC);
+	if (p < 0)
+		goto close_cli;
+
+	*v0 = p;
+	*v1 = c;
+
+	return 0;
+
+close_cli:
+	close(c);
+close_srv:
+	close(s);
+
+	return -1;
+}
+
+static void vsock_unix_redir_connectible(int sock_mapfd, int verd_mapfd,
+					 enum redir_mode mode, int sotype)
+{
+	const char *log_prefix = redir_mode_str(mode);
+	char a = 'a', b = 'b';
+	int u0, u1, v0, v1;
+	int sfd[2];
+	unsigned int pass;
+	int err, n;
+	u32 key;
+
+	zero_verdict_count(verd_mapfd);
+
+	if (socketpair(AF_UNIX, SOCK_STREAM | SOCK_NONBLOCK, 0, sfd))
+		return;
+
+	u0 = sfd[0];
+	u1 = sfd[1];
+
+	err = vsock_socketpair_connectible(sotype, &v0, &v1);
+	if (err) {
+		FAIL("vsock_socketpair_connectible() failed");
+		goto close_uds;
+	}
+
+	err = add_to_sockmap(sock_mapfd, u0, v0);
+	if (err) {
+		FAIL("add_to_sockmap failed");
+		goto close_vsock;
+	}
+
+	n = write(v1, &a, sizeof(a));
+	if (n < 0)
+		FAIL_ERRNO("%s: write", log_prefix);
+	if (n == 0)
+		FAIL("%s: incomplete write", log_prefix);
+	if (n < 1)
+		goto out;
+
+	n = recv(mode == REDIR_INGRESS ? u0 : u1, &b, sizeof(b), MSG_DONTWAIT);
+	if (n < 0)
+		FAIL("%s: recv() err, errno=%d", log_prefix, errno);
+	if (n == 0)
+		FAIL("%s: incomplete recv", log_prefix);
+	if (b != a)
+		FAIL("%s: vsock socket map failed, %c != %c", log_prefix, a, b);
+
+	key = SK_PASS;
+	err = xbpf_map_lookup_elem(verd_mapfd, &key, &pass);
+	if (err)
+		goto out;
+	if (pass != 1)
+		FAIL("%s: want pass count 1, have %d", log_prefix, pass);
+out:
+	key = 0;
+	bpf_map_delete_elem(sock_mapfd, &key);
+	key = 1;
+	bpf_map_delete_elem(sock_mapfd, &key);
+
+close_vsock:
+	close(v0);
+	close(v1);
+
+close_uds:
+	close(u0);
+	close(u1);
+}
+
+static void vsock_unix_skb_redir_connectible(struct test_sockmap_listen *skel,
+					     struct bpf_map *inner_map,
+					     int sotype)
+{
+	int verdict = bpf_program__fd(skel->progs.prog_skb_verdict);
+	int verdict_map = bpf_map__fd(skel->maps.verdict_map);
+	int sock_map = bpf_map__fd(inner_map);
+	int err;
+
+	err = xbpf_prog_attach(verdict, sock_map, BPF_SK_SKB_VERDICT, 0);
+	if (err)
+		return;
+
+	skel->bss->test_ingress = false;
+	vsock_unix_redir_connectible(sock_map, verdict_map, REDIR_EGRESS, sotype);
+	skel->bss->test_ingress = true;
+	vsock_unix_redir_connectible(sock_map, verdict_map, REDIR_INGRESS, sotype);
+
+	xbpf_prog_detach2(verdict, sock_map, BPF_SK_SKB_VERDICT);
+}
+
+static void test_vsock_redir(struct test_sockmap_listen *skel, struct bpf_map *map)
+{
+	const char *family_name, *map_name;
+	char s[MAX_TEST_NAME];
+
+	family_name = family_str(AF_VSOCK);
+	map_name = map_type_str(map);
+	snprintf(s, sizeof(s), "%s %s %s", map_name, family_name, __func__);
+	if (!test__start_subtest(s))
+		return;
+
+	vsock_unix_skb_redir_connectible(skel, map, SOCK_STREAM);
+	vsock_unix_skb_redir_connectible(skel, map, SOCK_SEQPACKET);
+}
+
 static void test_reuseport(struct test_sockmap_listen *skel,
 			   struct bpf_map *map, int family, int sotype)
 {
@@ -2060,12 +2221,14 @@ void serial_test_sockmap_listen(void)
 	run_tests(skel, skel->maps.sock_map, AF_INET6);
 	test_unix_redir(skel, skel->maps.sock_map, SOCK_DGRAM);
 	test_unix_redir(skel, skel->maps.sock_map, SOCK_STREAM);
+	test_vsock_redir(skel, skel->maps.sock_map);
 
 	skel->bss->test_sockmap = false;
 	run_tests(skel, skel->maps.sock_hash, AF_INET);
 	run_tests(skel, skel->maps.sock_hash, AF_INET6);
 	test_unix_redir(skel, skel->maps.sock_hash, SOCK_DGRAM);
 	test_unix_redir(skel, skel->maps.sock_hash, SOCK_STREAM);
+	test_vsock_redir(skel, skel->maps.sock_hash);
 
 	test_sockmap_listen__destroy(skel);
 }