Merge branch 'bpf-fixes'

Daniel Borkmann says:

====================
Various BPF fixes

Follow-up to fix incorrect pruning when alignment tracking is
in use and to properly clear regs after call to not leave stale
data behind, also a fix that adds bpf_clone_redirect to the
bpf_helper_changes_pkt_data helper and exposes correct map_flags
for lpm map into fdinfo. For details, please see individual
patches.

v1 -> v2:
  - Reworked first patch so that env->strict_alignment is the
    final indicator on whether we have to deal with strict
    alignment rather than having CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
    checks on various locations, so only checking env->strict_alignment
    is sufficient after that. Thanks for spotting, Dave!
  - Added patch 3 and 4.
  - Rest as is.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/filter.h

@@ -272,6 +272,16 @@ struct bpf_prog_aux;
 		.off   = OFF,					\
 		.imm   = IMM })
 
+/* Unconditional jumps, goto pc + off16 */
+
+#define BPF_JMP_A(OFF)						\
+	((struct bpf_insn) {					\
+		.code  = BPF_JMP | BPF_JA,			\
+		.dst_reg = 0,					\
+		.src_reg = 0,					\
+		.off   = OFF,					\
+		.imm   = 0 })
+
 /* Function call */
 
 #define BPF_EMIT_CALL(FUNC)					\

kernel/bpf/arraymap.c

@@ -86,6 +86,7 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr)
 	array->map.key_size = attr->key_size;
 	array->map.value_size = attr->value_size;
 	array->map.max_entries = attr->max_entries;
+	array->map.map_flags = attr->map_flags;
 	array->elem_size = elem_size;
 
 	if (!percpu)

kernel/bpf/lpm_trie.c

@@ -432,6 +432,7 @@ static struct bpf_map *trie_alloc(union bpf_attr *attr)
 	trie->map.key_size = attr->key_size;
 	trie->map.value_size = attr->value_size;
 	trie->map.max_entries = attr->max_entries;
+	trie->map.map_flags = attr->map_flags;
 	trie->data_size = attr->key_size -
 			  offsetof(struct bpf_lpm_trie_key, data);
 	trie->max_prefixlen = trie->data_size * 8;

kernel/bpf/stackmap.c

@@ -88,6 +88,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr)
 	smap->map.key_size = attr->key_size;
 	smap->map.value_size = value_size;
 	smap->map.max_entries = attr->max_entries;
+	smap->map.map_flags = attr->map_flags;
 	smap->n_buckets = n_buckets;
 	smap->map.pages = round_up(cost, PAGE_SIZE) >> PAGE_SHIFT;
 

kernel/bpf/verifier.c

@@ -463,19 +463,22 @@ static const int caller_saved[CALLER_SAVED_REGS] = {
 	BPF_REG_0, BPF_REG_1, BPF_REG_2, BPF_REG_3, BPF_REG_4, BPF_REG_5
 };
 
+static void mark_reg_not_init(struct bpf_reg_state *regs, u32 regno)
+{
+	BUG_ON(regno >= MAX_BPF_REG);
+
+	memset(&regs[regno], 0, sizeof(regs[regno]));
+	regs[regno].type = NOT_INIT;
+	regs[regno].min_value = BPF_REGISTER_MIN_RANGE;
+	regs[regno].max_value = BPF_REGISTER_MAX_RANGE;
+}
+
 static void init_reg_state(struct bpf_reg_state *regs)
 {
 	int i;
 
-	for (i = 0; i < MAX_BPF_REG; i++) {
-		regs[i].type = NOT_INIT;
-		regs[i].imm = 0;
-		regs[i].min_value = BPF_REGISTER_MIN_RANGE;
-		regs[i].max_value = BPF_REGISTER_MAX_RANGE;
-		regs[i].min_align = 0;
-		regs[i].aux_off = 0;
-		regs[i].aux_off_align = 0;
-	}
+	for (i = 0; i < MAX_BPF_REG; i++)
+		mark_reg_not_init(regs, i);
 
 	/* frame pointer */
 	regs[BPF_REG_FP].type = FRAME_PTR;
@@ -843,9 +846,6 @@ static int check_ptr_alignment(struct bpf_verifier_env *env,
 {
 	bool strict = env->strict_alignment;
 
-	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS))
-		strict = true;
-
 	switch (reg->type) {
 	case PTR_TO_PACKET:
 		return check_pkt_ptr_alignment(reg, off, size, strict);
@@ -1349,7 +1349,6 @@ static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx)
 	struct bpf_verifier_state *state = &env->cur_state;
 	const struct bpf_func_proto *fn = NULL;
 	struct bpf_reg_state *regs = state->regs;
-	struct bpf_reg_state *reg;
 	struct bpf_call_arg_meta meta;
 	bool changes_data;
 	int i, err;
@@ -1416,11 +1415,8 @@ static int check_call(struct bpf_verifier_env *env, int func_id, int insn_idx)
 	}
 
 	/* reset caller saved regs */
-	for (i = 0; i < CALLER_SAVED_REGS; i++) {
-		reg = regs + caller_saved[i];
-		reg->type = NOT_INIT;
-		reg->imm = 0;
-	}
+	for (i = 0; i < CALLER_SAVED_REGS; i++)
+		mark_reg_not_init(regs, caller_saved[i]);
 
 	/* update return register */
 	if (fn->ret_type == RET_INTEGER) {
@@ -2448,7 +2444,6 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn)
 {
 	struct bpf_reg_state *regs = env->cur_state.regs;
 	u8 mode = BPF_MODE(insn->code);
-	struct bpf_reg_state *reg;
 	int i, err;
 
 	if (!may_access_skb(env->prog->type)) {
@@ -2481,11 +2476,8 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn)
 	}
 
 	/* reset caller saved regs to unreadable */
-	for (i = 0; i < CALLER_SAVED_REGS; i++) {
-		reg = regs + caller_saved[i];
-		reg->type = NOT_INIT;
-		reg->imm = 0;
-	}
+	for (i = 0; i < CALLER_SAVED_REGS; i++)
+		mark_reg_not_init(regs, caller_saved[i]);
 
 	/* mark destination R0 register as readable, since it contains
 	 * the value fetched from the packet
@@ -2696,7 +2688,8 @@ err_free:
 /* the following conditions reduce the number of explored insns
  * from ~140k to ~80k for ultra large programs that use a lot of ptr_to_packet
  */
-static bool compare_ptrs_to_packet(struct bpf_reg_state *old,
+static bool compare_ptrs_to_packet(struct bpf_verifier_env *env,
+				   struct bpf_reg_state *old,
 				   struct bpf_reg_state *cur)
 {
 	if (old->id != cur->id)
@@ -2739,7 +2732,7 @@ static bool compare_ptrs_to_packet(struct bpf_reg_state *old,
 	 * 'if (R4 > data_end)' and all further insn were already good with r=20,
 	 * so they will be good with r=30 and we can prune the search.
 	 */
-	if (old->off <= cur->off &&
+	if (!env->strict_alignment && old->off <= cur->off &&
 	    old->off >= old->range && cur->off >= cur->range)
 		return true;
 
@@ -2810,7 +2803,7 @@ static bool states_equal(struct bpf_verifier_env *env,
 			continue;
 
 		if (rold->type == PTR_TO_PACKET && rcur->type == PTR_TO_PACKET &&
-		    compare_ptrs_to_packet(rold, rcur))
+		    compare_ptrs_to_packet(env, rold, rcur))
 			continue;
 
 		return false;
@@ -3588,10 +3581,10 @@ int bpf_check(struct bpf_prog **prog, union bpf_attr *attr)
 	} else {
 		log_level = 0;
 	}
-	if (attr->prog_flags & BPF_F_STRICT_ALIGNMENT)
+
+	env->strict_alignment = !!(attr->prog_flags & BPF_F_STRICT_ALIGNMENT);
+	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS))
 		env->strict_alignment = true;
-	else
-		env->strict_alignment = false;
 
 	ret = replace_map_fd_with_map_ptr(env);
 	if (ret < 0)
@@ -3697,7 +3690,10 @@ int bpf_analyzer(struct bpf_prog *prog, const struct bpf_ext_analyzer_ops *ops,
 	mutex_lock(&bpf_verifier_lock);
 
 	log_level = 0;
+
 	env->strict_alignment = false;
+	if (!IS_ENABLED(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS))
+		env->strict_alignment = true;
 
 	env->explored_states = kcalloc(env->prog->len,
 				       sizeof(struct bpf_verifier_state_list *),

net/core/filter.c

@@ -2281,6 +2281,7 @@ bool bpf_helper_changes_pkt_data(void *func)
 	    func == bpf_skb_change_head ||
 	    func == bpf_skb_change_tail ||
 	    func == bpf_skb_pull_data ||
+	    func == bpf_clone_redirect ||
 	    func == bpf_l3_csum_replace ||
 	    func == bpf_l4_csum_replace ||
 	    func == bpf_xdp_adjust_head)

tools/include/linux/filter.h

@@ -208,6 +208,16 @@
 		.off   = OFF,					\
 		.imm   = IMM })
 
+/* Unconditional jumps, goto pc + off16 */
+
+#define BPF_JMP_A(OFF)						\
+	((struct bpf_insn) {					\
+		.code  = BPF_JMP | BPF_JA,			\
+		.dst_reg = 0,					\
+		.src_reg = 0,					\
+		.off   = OFF,					\
+		.imm   = 0 })
+
 /* Function call */
 
 #define BPF_EMIT_CALL(FUNC)					\

tools/testing/selftests/bpf/test_verifier.c

@@ -49,6 +49,7 @@
 #define MAX_NR_MAPS	4
 
 #define F_NEEDS_EFFICIENT_UNALIGNED_ACCESS	(1 << 0)
+#define F_LOAD_WITH_STRICT_ALIGNMENT		(1 << 1)
 
 struct bpf_test {
 	const char *descr;
@@ -2614,6 +2615,30 @@ static struct bpf_test tests[] = {
 		.result = REJECT,
 		.prog_type = BPF_PROG_TYPE_SCHED_CLS,
 	},
+	{
+		"direct packet access: test17 (pruning, alignment)",
+		.insns = {
+			BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1,
+				    offsetof(struct __sk_buff, data)),
+			BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_1,
+				    offsetof(struct __sk_buff, data_end)),
+			BPF_LDX_MEM(BPF_W, BPF_REG_7, BPF_REG_1,
+				    offsetof(struct __sk_buff, mark)),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 14),
+			BPF_JMP_IMM(BPF_JGT, BPF_REG_7, 1, 4),
+			BPF_JMP_REG(BPF_JGT, BPF_REG_0, BPF_REG_3, 1),
+			BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, -4),
+			BPF_MOV64_IMM(BPF_REG_0, 0),
+			BPF_EXIT_INSN(),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_0, 1),
+			BPF_JMP_A(-6),
+		},
+		.errstr = "misaligned packet access off 2+15+-4 size 4",
+		.result = REJECT,
+		.prog_type = BPF_PROG_TYPE_SCHED_CLS,
+		.flags = F_LOAD_WITH_STRICT_ALIGNMENT,
+	},
 	{
 		"helper access to packet: test1, valid packet_ptr range",
 		.insns = {
@@ -3340,6 +3365,70 @@ static struct bpf_test tests[] = {
 		.result = ACCEPT,
 		.prog_type = BPF_PROG_TYPE_SCHED_CLS
 	},
+	{
+		"alu ops on ptr_to_map_value_or_null, 1",
+		.insns = {
+			BPF_MOV64_IMM(BPF_REG_1, 10),
+			BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
+			BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
+			BPF_LD_MAP_FD(BPF_REG_1, 0),
+			BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
+				     BPF_FUNC_map_lookup_elem),
+			BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, -2),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_4, 2),
+			BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
+			BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
+			BPF_EXIT_INSN(),
+		},
+		.fixup_map1 = { 4 },
+		.errstr = "R4 invalid mem access",
+		.result = REJECT,
+		.prog_type = BPF_PROG_TYPE_SCHED_CLS
+	},
+	{
+		"alu ops on ptr_to_map_value_or_null, 2",
+		.insns = {
+			BPF_MOV64_IMM(BPF_REG_1, 10),
+			BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
+			BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
+			BPF_LD_MAP_FD(BPF_REG_1, 0),
+			BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
+				     BPF_FUNC_map_lookup_elem),
+			BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
+			BPF_ALU64_IMM(BPF_AND, BPF_REG_4, -1),
+			BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
+			BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
+			BPF_EXIT_INSN(),
+		},
+		.fixup_map1 = { 4 },
+		.errstr = "R4 invalid mem access",
+		.result = REJECT,
+		.prog_type = BPF_PROG_TYPE_SCHED_CLS
+	},
+	{
+		"alu ops on ptr_to_map_value_or_null, 3",
+		.insns = {
+			BPF_MOV64_IMM(BPF_REG_1, 10),
+			BPF_STX_MEM(BPF_DW, BPF_REG_10, BPF_REG_1, -8),
+			BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+			BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
+			BPF_LD_MAP_FD(BPF_REG_1, 0),
+			BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
+				     BPF_FUNC_map_lookup_elem),
+			BPF_MOV64_REG(BPF_REG_4, BPF_REG_0),
+			BPF_ALU64_IMM(BPF_LSH, BPF_REG_4, 1),
+			BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 1),
+			BPF_ST_MEM(BPF_DW, BPF_REG_4, 0, 0),
+			BPF_EXIT_INSN(),
+		},
+		.fixup_map1 = { 4 },
+		.errstr = "R4 invalid mem access",
+		.result = REJECT,
+		.prog_type = BPF_PROG_TYPE_SCHED_CLS
+	},
 	{
 		"invalid memory access with multiple map_lookup_elem calls",
 		.insns = {
@@ -4937,7 +5026,149 @@ static struct bpf_test tests[] = {
 		.fixup_map_in_map = { 3 },
 		.errstr = "R1 type=map_value_or_null expected=map_ptr",
 		.result = REJECT,
-	}
+	},
+	{
+		"ld_abs: check calling conv, r1",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_1, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R1 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_abs: check calling conv, r2",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_2, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R2 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_abs: check calling conv, r3",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_3, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R3 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_abs: check calling conv, r4",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_4, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_4),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R4 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_abs: check calling conv, r5",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_5, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_5),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R5 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_abs: check calling conv, r7",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_7, 0),
+			BPF_LD_ABS(BPF_W, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_7),
+			BPF_EXIT_INSN(),
+		},
+		.result = ACCEPT,
+	},
+	{
+		"ld_ind: check calling conv, r1",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_1, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_1, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R1 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_ind: check calling conv, r2",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_2, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_2, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_2),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R2 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_ind: check calling conv, r3",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_3, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_3, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_3),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R3 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_ind: check calling conv, r4",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_4, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_4, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_4),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R4 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_ind: check calling conv, r5",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_5, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_5, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_5),
+			BPF_EXIT_INSN(),
+		},
+		.errstr = "R5 !read_ok",
+		.result = REJECT,
+	},
+	{
+		"ld_ind: check calling conv, r7",
+		.insns = {
+			BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
+			BPF_MOV64_IMM(BPF_REG_7, 1),
+			BPF_LD_IND(BPF_W, BPF_REG_7, -0x200000),
+			BPF_MOV64_REG(BPF_REG_0, BPF_REG_7),
+			BPF_EXIT_INSN(),
+		},
+		.result = ACCEPT,
+	},
 };
 
 static int probe_filter_length(const struct bpf_insn *fp)
@@ -5059,9 +5290,9 @@ static void do_test_single(struct bpf_test *test, bool unpriv,
 
 	do_test_fixup(test, prog, map_fds);
 
-	fd_prog = bpf_load_program(prog_type ? : BPF_PROG_TYPE_SOCKET_FILTER,
-				   prog, prog_len, "GPL", 0, bpf_vlog,
-				   sizeof(bpf_vlog));
+	fd_prog = bpf_verify_program(prog_type ? : BPF_PROG_TYPE_SOCKET_FILTER,
+				     prog, prog_len, test->flags & F_LOAD_WITH_STRICT_ALIGNMENT,
+				     "GPL", 0, bpf_vlog, sizeof(bpf_vlog));
 
 	expected_ret = unpriv && test->result_unpriv != UNDEF ?
 		       test->result_unpriv : test->result;