Re: [PATCH bpf-next v4 01/10] xfrm: bpf: Move xfrm_interface_bpf.c to xfrm_bpf.c

[Alexei Starovoitov]

On Mon, Dec 4, 2023 at 12:56 PM Daniel Xu <dxu@xxxxxxxxx> wrote:
>
> This commit moves the contents of xfrm_interface_bpf.c into a new file,
> xfrm_bpf.c This is in preparation for adding more xfrm kfuncs. We'd like
> to keep all the bpf integrations in a single file.
>
> Signed-off-by: Daniel Xu <dxu@xxxxxxxxx>
> ---
>  net/xfrm/Makefile                             |  7 +------
>  net/xfrm/{xfrm_interface_bpf.c => xfrm_bpf.c} | 12 ++++++++----
>  2 files changed, 9 insertions(+), 10 deletions(-)
>  rename net/xfrm/{xfrm_interface_bpf.c => xfrm_bpf.c} (88%)
>
> diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
> index cd47f88921f5..29fff452280d 100644
> --- a/net/xfrm/Makefile
> +++ b/net/xfrm/Makefile
> @@ -5,12 +5,6 @@
>
>  xfrm_interface-$(CONFIG_XFRM_INTERFACE) += xfrm_interface_core.o
>
> -ifeq ($(CONFIG_XFRM_INTERFACE),m)
> -xfrm_interface-$(CONFIG_DEBUG_INFO_BTF_MODULES) += xfrm_interface_bpf.o
> -else ifeq ($(CONFIG_XFRM_INTERFACE),y)
> -xfrm_interface-$(CONFIG_DEBUG_INFO_BTF) += xfrm_interface_bpf.o
> -endif
> -
>  obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \
>                       xfrm_input.o xfrm_output.o \
>                       xfrm_sysctl.o xfrm_replay.o xfrm_device.o
> @@ -21,3 +15,4 @@ obj-$(CONFIG_XFRM_USER_COMPAT) += xfrm_compat.o
>  obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
>  obj-$(CONFIG_XFRM_INTERFACE) += xfrm_interface.o
>  obj-$(CONFIG_XFRM_ESPINTCP) += espintcp.o
> +obj-$(CONFIG_DEBUG_INFO_BTF) += xfrm_bpf.o
...
> +#if IS_BUILTIN(CONFIG_XFRM_INTERFACE) || \
> +    (IS_MODULE(CONFIG_XFRM_INTERFACE) && IS_ENABLED(CONFIG_DEBUG_INFO_BTF_MODULES))
> +
>  /* bpf_xfrm_info - XFRM metadata information
>   *
>   * Members:
> @@ -108,3 +110,5 @@ int __init register_xfrm_interface_bpf(void)
>         return register_btf_kfunc_id_set(BPF_PROG_TYPE_SCHED_CLS,
>                                          &xfrm_interface_kfunc_set);
>  }
> +
> +#endif /* xfrm interface */

imo the original approach was cleaner.
#ifdefs in .c should be avoided when possible.
But I'm not going to insist.

ipsec folks please ack the first 3 patches.