History

bpf: add per-insn complexity limit

malicious bpf program may try to force the verifier to remember a lot of distinct verifier states. Put a limit to number of per-insn 'struct bpf_verifier_state'. Note that hitting the limit doesn't reject the program. It potentially makes the verifier do more steps to analyze the program. It means that malicious programs will hit BPF_COMPLEXITY_LIMIT_INSNS sooner instead of spending cpu time walking long link list. The limit of BPF_COMPLEXITY_LIMIT_STATES==64 affects cilium progs with slight increase in number of "steps" it takes to successfully verify the programs: before after bpf_lb-DLB_L3.o 1940 1940 bpf_lb-DLB_L4.o 3089 3089 bpf_lb-DUNKNOWN.o 1065 1065 bpf_lxc-DDROP_ALL.o 28052 | 28162 bpf_lxc-DUNKNOWN.o 35487 | 35541 bpf_netdev.o 10864 10864 bpf_overlay.o 6643 6643 bpf_lcx_jit.o 38437 38437 But it also makes malicious program to be rejected in 0.4 seconds vs 6.5 Hence apply this limit to unprivileged programs only.

Alexei Starovoitov committed 7 years ago

Commit ceefbc91 parent 4f7b3e81 file · +6 −1Browse files

Unified · Split

kernel/bpf/verifier.c+6−1
@@ -175,6 +175,7 @@ struct bpf_verifier_stack_elem { · 174 unmodified lines
175175
176176 #define BPF_COMPLEXITY_LIMIT_INSNS 131072
177177 #define BPF_COMPLEXITY_LIMIT_STACK 1024
178+#define BPF_COMPLEXITY_LIMIT_STATES 64
178179
179180 #define BPF_MAP_PTR_UNPRIV 1UL
180181 #define BPF_MAP_PTR_POISON ((void *)((0xeB9FUL << 1) + \
@@ -5047,7 +5048,7 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx) · 4866 unmodified lines
50475048 struct bpf_verifier_state_list *new_sl;
50485049 struct bpf_verifier_state_list *sl;
50495050 struct bpf_verifier_state *cur = env->cur_state, *new;
5050− int i, j, err;
5051+ int i, j, err, states_cnt = 0;
50515052
50525053 sl = env->explored_states[insn_idx];
50535054 if (!sl)
@@ -5074,8 +5075,12 @@ static int is_state_visited(struct bpf_verifier_env *env, int insn_idx) · 20 unmodified lines
50745075 return 1;
50755076 }
50765077 sl = sl->next;
5078+ states_cnt++;
50775079 }
50785080
5081+ if (!env->allow_ptr_leaks && states_cnt > BPF_COMPLEXITY_LIMIT_STATES)
5082+ return 0;
5083+
50795084 /* there were no equivalent states, remember current one.
50805085 * technically the current state is not proven to be safe yet,
50815086 * but it will either reach outer most bpf_exit (which means it's safe)