bpf: Remove the unnecessary insn buffer comparison
authorHaiyue Wang <haiyue.wang@intel.com>
Sun, 8 Jan 2023 15:12:57 +0000 (23:12 +0800)
committerDaniel Borkmann <daniel@iogearbox.net>
Tue, 10 Jan 2023 22:09:29 +0000 (23:09 +0100)
The variable 'insn' is initialized to 'insn_buf' without being changed, only
some helper macros are defined, so the insn buffer comparison is unnecessary.
Just remove it. This missed removal back in 2377b81de527 ("bpf: split shared
bpf_tcp_sock and bpf_sock_ops implementation").

Signed-off-by: Haiyue Wang <haiyue.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/bpf/20230108151258.96570-1-haiyue.wang@intel.com
net/core/filter.c

index ab81129..d9befa6 100644 (file)
@@ -6847,9 +6847,6 @@ u32 bpf_tcp_sock_convert_ctx_access(enum bpf_access_type type,
                                        FIELD));                        \
        } while (0)
 
-       if (insn > insn_buf)
-               return insn - insn_buf;
-
        switch (si->off) {
        case offsetof(struct bpf_tcp_sock, rtt_min):
                BUILD_BUG_ON(sizeof_field(struct tcp_sock, rtt_min) !=
@@ -10147,9 +10144,6 @@ static u32 sock_ops_convert_ctx_access(enum bpf_access_type type,
                        SOCK_OPS_GET_FIELD(BPF_FIELD, OBJ_FIELD, OBJ);        \
        } while (0)
 
-       if (insn > insn_buf)
-               return insn - insn_buf;
-
        switch (si->off) {
        case offsetof(struct bpf_sock_ops, op):
                *insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_sock_ops_kern,