Merge branch 'BPF-inline-improvements'

Daniel Borkmann says:

====================
BPF inline improvements

First one makes htab inlining more robust wrt future jits and
second one inlines map in map lookups through map_gen_lookup()
callback.

v1 -> v2:
  - BITS_PER_LONG guard in patch 1
  - BPF_EMIT_CALL is on __htab_map_lookup_elem
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

kernel/bpf/arraymap.c

@@ -606,6 +606,31 @@ static void *array_of_map_lookup_elem(struct bpf_map *map, void *key)
 	return READ_ONCE(*inner_map);
 }
 
+static u32 array_of_map_gen_lookup(struct bpf_map *map,
+				   struct bpf_insn *insn_buf)
+{
+	u32 elem_size = round_up(map->value_size, 8);
+	struct bpf_insn *insn = insn_buf;
+	const int ret = BPF_REG_0;
+	const int map_ptr = BPF_REG_1;
+	const int index = BPF_REG_2;
+
+	*insn++ = BPF_ALU64_IMM(BPF_ADD, map_ptr, offsetof(struct bpf_array, value));
+	*insn++ = BPF_LDX_MEM(BPF_W, ret, index, 0);
+	*insn++ = BPF_JMP_IMM(BPF_JGE, ret, map->max_entries, 5);
+	if (is_power_of_2(elem_size))
+		*insn++ = BPF_ALU64_IMM(BPF_LSH, ret, ilog2(elem_size));
+	else
+		*insn++ = BPF_ALU64_IMM(BPF_MUL, ret, elem_size);
+	*insn++ = BPF_ALU64_REG(BPF_ADD, ret, map_ptr);
+	*insn++ = BPF_LDX_MEM(BPF_DW, ret, ret, 0);
+	*insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 1);
+	*insn++ = BPF_JMP_IMM(BPF_JA, 0, 0, 1);
+	*insn++ = BPF_MOV64_IMM(ret, 0);
+
+	return insn - insn_buf;
+}
+
 const struct bpf_map_ops array_of_maps_map_ops = {
 	.map_alloc = array_of_map_alloc,
 	.map_free = array_of_map_free,
@@ -615,4 +640,5 @@ const struct bpf_map_ops array_of_maps_map_ops = {
 	.map_fd_get_ptr = bpf_map_fd_get_ptr,
 	.map_fd_put_ptr = bpf_map_fd_put_ptr,
 	.map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem,
+	.map_gen_lookup = array_of_map_gen_lookup,
 };

kernel/bpf/hashtab.c

@@ -1322,6 +1322,22 @@ static void *htab_of_map_lookup_elem(struct bpf_map *map, void *key)
 	return READ_ONCE(*inner_map);
 }
 
+static u32 htab_of_map_gen_lookup(struct bpf_map *map,
+				  struct bpf_insn *insn_buf)
+{
+	struct bpf_insn *insn = insn_buf;
+	const int ret = BPF_REG_0;
+
+	*insn++ = BPF_EMIT_CALL((u64 (*)(u64, u64, u64, u64, u64))__htab_map_lookup_elem);
+	*insn++ = BPF_JMP_IMM(BPF_JEQ, ret, 0, 2);
+	*insn++ = BPF_ALU64_IMM(BPF_ADD, ret,
+				offsetof(struct htab_elem, key) +
+				round_up(map->key_size, 8));
+	*insn++ = BPF_LDX_MEM(BPF_DW, ret, ret, 0);
+
+	return insn - insn_buf;
+}
+
 static void htab_of_map_free(struct bpf_map *map)
 {
 	bpf_map_meta_free(map->inner_map_meta);
@@ -1337,4 +1353,5 @@ const struct bpf_map_ops htab_of_maps_map_ops = {
 	.map_fd_get_ptr = bpf_map_fd_get_ptr,
 	.map_fd_put_ptr = bpf_map_fd_put_ptr,
 	.map_fd_sys_lookup_elem = bpf_map_fd_sys_lookup_elem,
+	.map_gen_lookup = htab_of_map_gen_lookup,
 };

kernel/bpf/verifier.c

@@ -4160,7 +4160,11 @@ static int fixup_bpf_calls(struct bpf_verifier_env *env)
 			continue;
 		}
 
-		if (ebpf_jit_enabled() && insn->imm == BPF_FUNC_map_lookup_elem) {
+		/* BPF_EMIT_CALL() assumptions in some of the map_gen_lookup
+		 * handlers are currently limited to 64 bit only.
+		 */
+		if (ebpf_jit_enabled() && BITS_PER_LONG == 64 &&
+		    insn->imm == BPF_FUNC_map_lookup_elem) {
 			map_ptr = env->insn_aux_data[i + delta].map_ptr;
 			if (map_ptr == BPF_MAP_PTR_POISON ||
 			    !map_ptr->ops->map_gen_lookup)