bpf: allow extended BPF programs access skb fields

introduce user accessible mirror of in-kernel 'struct sk_buff':
struct __sk_buff {
    __u32 len;
    __u32 pkt_type;
    __u32 mark;
    __u32 queue_mapping;
};

bpf programs can do:

int bpf_prog(struct __sk_buff *skb)
{
    __u32 var = skb->pkt_type;

which will be compiled to bpf assembler as:

dst_reg = *(u32 *)(src_reg + 4) // 4 == offsetof(struct __sk_buff, pkt_type)

bpf verifier will check validity of access and will convert it to:

dst_reg = *(u8 *)(src_reg + offsetof(struct sk_buff, __pkt_type_offset))
dst_reg &= 7

since skb->pkt_type is a bitfield.

Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 137 insertions, 17 deletions

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 669719ccc9ee..ea75c654af1b 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -519,7 +519,7 @@ static int bpf_prog_load(union bpf_attr *attr)
                 goto free_prog;
 
         /* run eBPF verifier */
-        err = bpf_check(prog, attr);
+        err = bpf_check(&prog, attr);
         if (err < 0)
                 goto free_used_maps;
 
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index e6b522496250..c22ebd36fa4b 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -1620,11 +1620,10 @@ static int do_check(struct verifier_env *env)
                                 return err;
 
                 } else if (class == BPF_LDX) {
-                        if (BPF_MODE(insn->code) != BPF_MEM ||
-                            insn->imm != 0) {
-                                verbose("BPF_LDX uses reserved fields\n");
-                                return -EINVAL;
-                        }
+                        enum bpf_reg_type src_reg_type;
+
+                        /* check for reserved fields is already done */
+
                         /* check src operand */
                         err = check_reg_arg(regs, insn->src_reg, SRC_OP);
                         if (err)
@@ -1643,6 +1642,29 @@ static int do_check(struct verifier_env *env)
                         if (err)
                                 return err;
 
+                        src_reg_type = regs[insn->src_reg].type;
+
+                        if (insn->imm == 0 && BPF_SIZE(insn->code) == BPF_W) {
+                                /* saw a valid insn
+                                 * dst_reg = *(u32 *)(src_reg + off)
+                                 * use reserved 'imm' field to mark this insn
+                                 */
+                                insn->imm = src_reg_type;
+
+                        } else if (src_reg_type != insn->imm &&
+                                   (src_reg_type == PTR_TO_CTX ||
+                                    insn->imm == PTR_TO_CTX)) {
+                                /* ABuser program is trying to use the same insn
+                                 * dst_reg = *(u32*) (src_reg + off)
+                                 * with different pointer types:
+                                 * src_reg == ctx in one branch and
+                                 * src_reg == stack|map in some other branch.
+                                 * Reject it.
+                                 */
+                                verbose("same insn cannot be used with different pointers\n");
+                                return -EINVAL;
+                        }
+
                 } else if (class == BPF_STX) {
                         if (BPF_MODE(insn->code) == BPF_XADD) {
                                 err = check_xadd(env, insn);
@@ -1790,6 +1812,13 @@ static int replace_map_fd_with_map_ptr(struct verifier_env *env)
         int i, j;
 
         for (i = 0; i < insn_cnt; i++, insn++) {
+                if (BPF_CLASS(insn->code) == BPF_LDX &&
+                    (BPF_MODE(insn->code) != BPF_MEM ||
+                     insn->imm != 0)) {
+                        verbose("BPF_LDX uses reserved fields\n");
+                        return -EINVAL;
+                }
+
                 if (insn[0].code == (BPF_LD | BPF_IMM | BPF_DW)) {
                         struct bpf_map *map;
                         struct fd f;
@@ -1881,6 +1910,92 @@ static void convert_pseudo_ld_imm64(struct verifier_env *env)
                         insn->src_reg = 0;
 }
 
+static void adjust_branches(struct bpf_prog *prog, int pos, int delta)
+{
+        struct bpf_insn *insn = prog->insnsi;
+        int insn_cnt = prog->len;
+        int i;
+
+        for (i = 0; i < insn_cnt; i++, insn++) {
+                if (BPF_CLASS(insn->code) != BPF_JMP ||
+                    BPF_OP(insn->code) == BPF_CALL ||
+                    BPF_OP(insn->code) == BPF_EXIT)
+                        continue;
+
+                /* adjust offset of jmps if necessary */
+                if (i < pos && i + insn->off + 1 > pos)
+                        insn->off += delta;
+                else if (i > pos && i + insn->off + 1 < pos)
+                        insn->off -= delta;
+        }
+}
+
+/* convert load instructions that access fields of 'struct __sk_buff'
+ * into sequence of instructions that access fields of 'struct sk_buff'
+ */
+static int convert_ctx_accesses(struct verifier_env *env)
+{
+        struct bpf_insn *insn = env->prog->insnsi;
+        int insn_cnt = env->prog->len;
+        struct bpf_insn insn_buf[16];
+        struct bpf_prog *new_prog;
+        u32 cnt;
+        int i;
+
+        if (!env->prog->aux->ops->convert_ctx_access)
+                return 0;
+
+        for (i = 0; i < insn_cnt; i++, insn++) {
+                if (insn->code != (BPF_LDX | BPF_MEM | BPF_W))
+                        continue;
+
+                if (insn->imm != PTR_TO_CTX) {
+                        /* clear internal mark */
+                        insn->imm = 0;
+                        continue;
+                }
+
+                cnt = env->prog->aux->ops->
+                        convert_ctx_access(insn->dst_reg, insn->src_reg,
+                                           insn->off, insn_buf);
+                if (cnt == 0 || cnt >= ARRAY_SIZE(insn_buf)) {
+                        verbose("bpf verifier is misconfigured\n");
+                        return -EINVAL;
+                }
+
+                if (cnt == 1) {
+                        memcpy(insn, insn_buf, sizeof(*insn));
+                        continue;
+                }
+
+                /* several new insns need to be inserted. Make room for them */
+                insn_cnt += cnt - 1;
+                new_prog = bpf_prog_realloc(env->prog,
+                                            bpf_prog_size(insn_cnt),
+                                            GFP_USER);
+                if (!new_prog)
+                        return -ENOMEM;
+
+                new_prog->len = insn_cnt;
+
+                memmove(new_prog->insnsi + i + cnt, new_prog->insns + i + 1,
+                        sizeof(*insn) * (insn_cnt - i - cnt));
+
+                /* copy substitute insns in place of load instruction */
+                memcpy(new_prog->insnsi + i, insn_buf, sizeof(*insn) * cnt);
+
+                /* adjust branches in the whole program */
+                adjust_branches(new_prog, i, cnt - 1);
+
+                /* keep walking new program and skip insns we just inserted */
+                env->prog = new_prog;
+                insn = new_prog->insnsi + i + cnt - 1;
+                i += cnt - 1;
+        }
+
+        return 0;
+}
+
 static void free_states(struct verifier_env *env)
 {
         struct verifier_state_list *sl, *sln;
@@ -1903,13 +2018,13 @@ static void free_states(struct verifier_env *env)
         kfree(env->explored_states);
 }
 
-int bpf_check(struct bpf_prog *prog, union bpf_attr *attr)
+int bpf_check(struct bpf_prog **prog, union bpf_attr *attr)
 {
         char __user *log_ubuf = NULL;
         struct verifier_env *env;
         int ret = -EINVAL;