[PATCH 3.16.y-ckt 012/180] net: bpf: reject invalid shifts

[Luis Henriques]

3.16.7-ckt24 -stable review patch.  If anyone has any objections, please let me know.

---8<------------------------------------------------------------

From: Rabin Vincent <rabin at rab.in>

commit 229394e8e62a4191d592842cf67e80c62a492937 upstream.

On ARM64, a BUG() is triggered in the eBPF JIT if a filter with a
constant shift that can't be encoded in the immediate field of the
UBFM/SBFM instructions is passed to the JIT.  Since these shifts
amounts, which are negative or >= regsize, are invalid, reject them in
the eBPF verifier and the classic BPF filter checker, for all
architectures.

Signed-off-by: Rabin Vincent <rabin at rab.in>
Acked-by: Alexei Starovoitov <ast at kernel.org>
Acked-by: Daniel Borkmann <daniel at iogearbox.net>
Signed-off-by: David S. Miller <davem at davemloft.net>
[ luis: backported to 3.16:
  - drop changes to eBPF verifier, only added in 3.18 kernel
  - function rename: bpf_check_classic() -> sk_chk_filter() ]
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 net/core/filter.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/net/core/filter.c b/net/core/filter.c
index 3139f966a178..dfc5f31dc5a1 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1251,6 +1251,11 @@ int sk_chk_filter(struct sock_filter *filter, unsigned int flen)
 			if (ftest->k == 0)
 				return -EINVAL;
 			break;
+		case BPF_ALU | BPF_LSH | BPF_K:
+		case BPF_ALU | BPF_RSH | BPF_K:
+			if (ftest->k >= 32)
+				return -EINVAL;
+			break;
 		case BPF_LD | BPF_MEM:
 		case BPF_LDX | BPF_MEM:
 		case BPF_ST: