linux/arch/arm64/net
Xi Wang d63903bbc3 arm64: bpf: fix endianness conversion bugs
Upper bits should be zeroed in endianness conversion:

- even when there's no need to change endianness (i.e., BPF_FROM_BE
  on big endian or BPF_FROM_LE on little endian);

- after rev16.

This patch fixes such bugs by emitting extra instructions to clear
upper bits.

Cc: Zi Shen Lim <zlim.lnx@gmail.com>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Fixes: e54bcde3d6 ("arm64: eBPF JIT compiler")
Cc: <stable@vger.kernel.org> # 3.18+
Signed-off-by: Xi Wang <xi.wang@gmail.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
2015-06-26 14:15:39 +01:00
..
bpf_jit_comp.c arm64: bpf: fix endianness conversion bugs 2015-06-26 14:15:39 +01:00
bpf_jit.h arm64: bpf: fix endianness conversion bugs 2015-06-26 14:15:39 +01:00
Makefile arm64: eBPF JIT compiler 2014-09-08 14:39:21 +01:00