Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) Support ipv6 checksum offload in sunvnet driver, from Shannon
    Nelson.

 2) Move to RB-tree instead of custom AVL code in inetpeer, from Eric
    Dumazet.

 3) Allow generic XDP to work on virtual devices, from John Fastabend.

 4) Add bpf device maps and XDP_REDIRECT, which can be used to build
    arbitrary switching frameworks using XDP. From John Fastabend.

 5) Remove UFO offloads from the tree, gave us little other than bugs.

 6) Remove the IPSEC flow cache, from Florian Westphal.

 7) Support ipv6 route offload in mlxsw driver.

 8) Support VF representors in bnxt_en, from Sathya Perla.

 9) Add support for forward error correction modes to ethtool, from
    Vidya Sagar Ravipati.

10) Add time filter for packet scheduler action dumping, from Jamal Hadi
    Salim.

11) Extend the zerocopy sendmsg() used by virtio and tap to regular
    sockets via MSG_ZEROCOPY. From Willem de Bruijn.

12) Significantly rework value tracking in the BPF verifier, from Edward
    Cree.

13) Add new jump instructions to eBPF, from Daniel Borkmann.

14) Rework rtnetlink plumbing so that operations can be run without
    taking the RTNL semaphore. From Florian Westphal.

15) Support XDP in tap driver, from Jason Wang.

16) Add 32-bit eBPF JIT for ARM, from Shubham Bansal.

17) Add Huawei hinic ethernet driver.

18) Allow to report MD5 keys in TCP inet_diag dumps, from Ivan
    Delalande.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1780 commits)
  i40e: point wb_desc at the nvm_wb_desc during i40e_read_nvm_aq
  i40e: avoid NVM acquire deadlock during NVM update
  drivers: net: xgene: Remove return statement from void function
  drivers: net: xgene: Configure tx/rx delay for ACPI
  drivers: net: xgene: Read tx/rx delay for ACPI
  rocker: fix kcalloc parameter order
  rds: Fix non-atomic operation on shared flag variable
  net: sched: don't use GFP_KERNEL under spin lock
  vhost_net: correctly check tx avail during rx busy polling
  net: mdio-mux: add mdio_mux parameter to mdio_mux_init()
  rxrpc: Make service connection lookup always check for retry
  net: stmmac: Delete dead code for MDIO registration
  gianfar: Fix Tx flow control deactivation
  cxgb4: Ignore MPS_TX_INT_CAUSE[Bubble] for T6
  cxgb4: Fix pause frame count in t4_get_port_stats
  cxgb4: fix memory leak
  tun: rename generic_xdp to skb_xdp
  tun: reserve extra headroom only when XDP is set
  net: dsa: bcm_sf2: Configure IMP port TC2QOS mapping
  net: dsa: bcm_sf2: Advertise number of egress queues
  ...

1 files changed, 83 insertions, 6 deletions

diff --git a/samples/bpf/map_perf_test_user.c b/samples/bpf/map_perf_test_user.c
index 1a8894b5ac51..f388254896f6 100644
--- a/samples/bpf/map_perf_test_user.c
+++ b/samples/bpf/map_perf_test_user.c
@@ -46,6 +46,7 @@ enum test_type {
 	HASH_LOOKUP,
 	ARRAY_LOOKUP,
 	INNER_LRU_HASH_PREALLOC,
+	LRU_HASH_LOOKUP,
 	NR_TESTS,
 };
 
@@ -60,6 +61,7 @@ const char *test_map_names[NR_TESTS] = {
 	[HASH_LOOKUP] = "hash_map",
 	[ARRAY_LOOKUP] = "array_map",
 	[INNER_LRU_HASH_PREALLOC] = "inner_lru_hash_map",
+	[LRU_HASH_LOOKUP] = "lru_hash_lookup_map",
 };
 
 static int test_flags = ~0;
@@ -67,6 +69,8 @@ static uint32_t num_map_entries;
 static uint32_t inner_lru_hash_size;
 static int inner_lru_hash_idx = -1;
 static int array_of_lru_hashs_idx = -1;
+static int lru_hash_lookup_idx = -1;
+static int lru_hash_lookup_test_entries = 32;
 static uint32_t max_cnt = 1000000;
 
 static int check_test_flags(enum test_type t)
@@ -86,6 +90,32 @@ static void test_hash_prealloc(int cpu)
 	       cpu, max_cnt * 1000000000ll / (time_get_ns() - start_time));
 }
 
+static int pre_test_lru_hash_lookup(int tasks)
+{
+	int fd = map_fd[lru_hash_lookup_idx];
+	uint32_t key;
+	long val = 1;
+	int ret;
+
+	if (num_map_entries > lru_hash_lookup_test_entries)
+		lru_hash_lookup_test_entries = num_map_entries;
+
+	/* Populate the lru_hash_map for LRU_HASH_LOOKUP perf test.
+	 *
+	 * It is fine that the user requests for a map with
+	 * num_map_entries < 32 and some of the later lru hash lookup
+	 * may return not found.  For LRU map, we are not interested
+	 * in such small map performance.
+	 */
+	for (key = 0; key < lru_hash_lookup_test_entries; key++) {
+		ret = bpf_map_update_elem(fd, &key, &val, BPF_NOEXIST);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static void do_test_lru(enum test_type test, int cpu)
 {
 	static int inner_lru_map_fds[MAX_NR_CPUS];
@@ -97,14 +127,20 @@ static void do_test_lru(enum test_type test, int cpu)
 
 	if (test == INNER_LRU_HASH_PREALLOC) {
 		int outer_fd = map_fd[array_of_lru_hashs_idx];
+		unsigned int mycpu, mynode;
 
 		assert(cpu < MAX_NR_CPUS);
 
 		if (cpu) {
+			ret = syscall(__NR_getcpu, &mycpu, &mynode, NULL);
+			assert(!ret);
+
 			inner_lru_map_fds[cpu] =
-				bpf_create_map(BPF_MAP_TYPE_LRU_HASH,
-					       sizeof(uint32_t), sizeof(long),
-					       inner_lru_hash_size, 0);
+				bpf_create_map_node(BPF_MAP_TYPE_LRU_HASH,
+						    sizeof(uint32_t),
+						    sizeof(long),
+						    inner_lru_hash_size, 0,
+						    mynode);
 			if (inner_lru_map_fds[cpu] == -1) {
 				printf("cannot create BPF_MAP_TYPE_LRU_HASH %s(%d)\n",
 				       strerror(errno), errno);
@@ -129,13 +165,17 @@ static void do_test_lru(enum test_type test, int cpu)
 
 	if (test == LRU_HASH_PREALLOC) {
 		test_name = "lru_hash_map_perf";
-		in6.sin6_addr.s6_addr16[7] = 0;
+		in6.sin6_addr.s6_addr16[2] = 0;
 	} else if (test == NOCOMMON_LRU_HASH_PREALLOC) {
 		test_name = "nocommon_lru_hash_map_perf";
-		in6.sin6_addr.s6_addr16[7] = 1;
+		in6.sin6_addr.s6_addr16[2] = 1;
 	} else if (test == INNER_LRU_HASH_PREALLOC) {
 		test_name = "inner_lru_hash_map_perf";
-		in6.sin6_addr.s6_addr16[7] = 2;
+		in6.sin6_addr.s6_addr16[2] = 2;
+	} else if (test == LRU_HASH_LOOKUP) {
+		test_name = "lru_hash_lookup_perf";
+		in6.sin6_addr.s6_addr16[2] = 3;
+		in6.sin6_addr.s6_addr32[3] = 0;
 	} else {
 		assert(0);
 	}
@@ -144,6 +184,11 @@ static void do_test_lru(enum test_type test, int cpu)
 	for (i = 0; i < max_cnt; i++) {
 		ret = connect(-1, (const struct sockaddr *)&in6, sizeof(in6));
 		assert(ret == -1 && errno == EBADF);
+		if (in6.sin6_addr.s6_addr32[3] <
+		    lru_hash_lookup_test_entries - 32)
+			in6.sin6_addr.s6_addr32[3] += 32;
+		else
+			in6.sin6_addr.s6_addr32[3] = 0;
 	}
 	printf("%d:%s pre-alloc %lld events per sec\n",
 	       cpu, test_name,
@@ -165,6 +210,11 @@ static void test_inner_lru_hash_prealloc(int cpu)
 	do_test_lru(INNER_LRU_HASH_PREALLOC, cpu);
 }
 
+static void test_lru_hash_lookup(int cpu)
+{
+	do_test_lru(LRU_HASH_LOOKUP, cpu);
+}
+
 static void test_percpu_hash_prealloc(int cpu)
 {
 	__u64 start_time;
@@ -237,6 +287,11 @@ static void test_array_lookup(int cpu)
 	       cpu, max_cnt * 1000000000ll * 64 / (time_get_ns() - start_time));
 }
 
+typedef int (*pre_test_func)(int tasks);
+const pre_test_func pre_test_funcs[] = {
+	[LRU_HASH_LOOKUP] = pre_test_lru_hash_lookup,
+};
+
 typedef void (*test_func)(int cpu);
 const test_func test_funcs[] = {
 	[HASH_PREALLOC] = test_hash_prealloc,
@@ -249,8 +304,25 @@ const test_func test_funcs[] = {
 	[HASH_LOOKUP] = test_hash_lookup,
 	[ARRAY_LOOKUP] = test_array_lookup,
 	[INNER_LRU_HASH_PREALLOC] = test_inner_lru_hash_prealloc,
+	[LRU_HASH_LOOKUP] = test_lru_hash_lookup,
 };
 
+static int pre_test(int tasks)
+{
+	int i;
+
+	for (i = 0; i < NR_TESTS; i++) {
+		if (pre_test_funcs[i] && check_test_flags(i)) {
+			int ret = pre_test_funcs[i](tasks);
+
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
 static void loop(int cpu)
 {
 	cpu_set_t cpuset;
@@ -271,6 +343,8 @@ static void run_perf_test(int tasks)
 	pid_t pid[tasks];
 	int i;
 
+	assert(!pre_test(tasks));
+
 	for (i = 0; i < tasks; i++) {
 		pid[i] = fork();
 		if (pid[i] == 0) {
@@ -338,6 +412,9 @@ static void fixup_map(struct bpf_map_data *map, int idx)
 		array_of_lru_hashs_idx = idx;
 	}
 
+	if (!strcmp("lru_hash_lookup_map", map->name))
+		lru_hash_lookup_idx = idx;
+
 	if (num_map_entries <= 0)
 		return;