bpf: fix incorrect sign extension in check_alu_op()
authorJann Horn <jannh@google.com>
Tue, 19 Dec 2017 04:11:54 +0000 (20:11 -0800)
committerDaniel Borkmann <daniel@iogearbox.net>
Thu, 21 Dec 2017 01:15:41 +0000 (02:15 +0100)
commit95a762e2c8c942780948091f8f2a4f32fce1ac6f
tree12c21df5c872c87723ea4931c7bffa57c4c8616a
parent4374f256ce8182019353c0c639bb8d0695b4c941
bpf: fix incorrect sign extension in check_alu_op()

Distinguish between
BPF_ALU64|BPF_MOV|BPF_K (load 32-bit immediate, sign-extended to 64-bit)
and BPF_ALU|BPF_MOV|BPF_K (load 32-bit immediate, zero-padded to 64-bit);
only perform sign extension in the first case.

Starting with v4.14, this is exploitable by unprivileged users as long as
the unprivileged_bpf_disabled sysctl isn't set.

Debian assigned CVE-2017-16995 for this issue.

v3:
 - add CVE number (Ben Hutchings)

Fixes: 484611357c19 ("bpf: allow access into map value arrays")
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
kernel/bpf/verifier.c