bpf: fix precision bit propagation for BPF_ST instructions (b3b50f05) · Commits · 戴 / test

When backtracking instructions to propagate precision bit for registers and stack slots, one class of instructions (BPF_ST) weren't handled causing extra stack slots to be propagated into parent state. Parent state might not have that much stack allocated, though, which causes warning on invalid stack slot usage. This patch adds handling of BPF_ST instructions: BPF_MEM | <size> | BPF_ST: *(size *) (dst_reg + off) = imm32 Reported-by:

<syzbot+4da3ff23081bafe74fc2@syzkaller.appspotmail.com> Fixes: ("bpf: precise scalar_value tracking") Cc: Alexei Starovoitov <ast@fb.com> Signed-off-by:

Andrii Nakryiko <andriin@fb.com> Signed-off-by:

Daniel Borkmann <daniel@iogearbox.net>

kernel/bpf/verifier.c

+4 −7

Original line number	Diff line number	Diff line
		@@ -1519,9 +1519,9 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx,
		return -EFAULT;
		}
		*stack_mask \|= 1ull << spi;
		} else if (class == BPF_STX) {
		} else if (class == BPF_STX \|\| class == BPF_ST) {
		if (*reg_mask & dreg)
		/* stx shouldn't be using _scalar_ dst_reg
		/* stx & st shouldn't be using _scalar_ dst_reg
		* to access memory. It means backtracking
		* encountered a case of pointer subtraction.
		*/
		@@ -1540,6 +1540,7 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx,
		if (!(*stack_mask & (1ull << spi)))
		return 0;
		*stack_mask &= ~(1ull << spi);
		if (class == BPF_STX)
		*reg_mask \|= sreg;
		} else if (class == BPF_JMP \|\| class == BPF_JMP32) {
		if (opcode == BPF_CALL) {
		@@ -1569,10 +1570,6 @@ static int backtrack_insn(struct bpf_verifier_env *env, int idx,
		if (mode == BPF_IND \|\| mode == BPF_ABS)
		/* to be analyzed */
		return -ENOTSUPP;
		} else if (class == BPF_ST) {
		if (*reg_mask & dreg)
		/* likely pointer subtraction */
		return -ENOTSUPP;
		}
		return 0;
		}

Admin message