Re: [PATCH] bpf: merge two CONFIG_BPF entries

[Yonghong Song <yonghong.song@linux.dev>]

On 2/4/24 3:37 PM, Masahiro Yamada wrote:
> On Mon, Feb 5, 2024 at 3:11 AM Yonghong Song <yonghong.song@linux.dev> wrote:
>>
>> On 2/3/24 11:56 PM, Masahiro Yamada wrote:
>>> 'config BPF' exists in both init/Kconfig and kernel/bpf/Kconfig.
>>>
>>> Commit b24abcff918a ("bpf, kconfig: Add consolidated menu entry for bpf
>>> with core options") added the second one to kernel/bpf/Kconfig instead
>>> of moving the existing one.
>>>
>>> Merge them together.
>>>
>>> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
>>> ---
>>>
>>>    init/Kconfig       | 5 -----
>>>    kernel/bpf/Kconfig | 1 +
>>>    2 files changed, 1 insertion(+), 5 deletions(-)
>>>
>>> diff --git a/init/Kconfig b/init/Kconfig
>>> index 8d4e836e1b6b..46ccad83a664 100644
>>> --- a/init/Kconfig
>>> +++ b/init/Kconfig
>>> @@ -1457,11 +1457,6 @@ config SYSCTL_ARCH_UNALIGN_ALLOW
>>>    config HAVE_PCSPKR_PLATFORM
>>>        bool
>>>
>>> -# interpreter that classic socket filters depend on
>>> -config BPF
>>> -     bool
>>> -     select CRYPTO_LIB_SHA1
>>> -
>>>    menuconfig EXPERT
>>>        bool "Configure standard kernel features (expert users)"
>>>        # Unhide debug options, to make the on-by-default options visible
>>> diff --git a/kernel/bpf/Kconfig b/kernel/bpf/Kconfig
>>> index 6a906ff93006..bc25f5098a25 100644
>>> --- a/kernel/bpf/Kconfig
>>> +++ b/kernel/bpf/Kconfig
>>> @@ -3,6 +3,7 @@
>>>    # BPF interpreter that, for example, classic socket filters depend on.
>>>    config BPF
>>>        bool
>>> +     select CRYPTO_LIB_SHA1
>> Currently, the kernel/bpf directory is guarded with CONFIG_BPF
>>     obj-$(CONFIG_BPF) += bpf/
>> in kernel/bpf/Makefile.
>
> Wrong.
>
> "in kernel/Makefile".
>
>
> Why is it related to this patch?

Sorry, my obvious mistake.

>
>
>
>> Your patch probably works since there are lots of some other BPF related
>> configurations which requires CONFIG_BPF. But maybe we sould
>> keep 'config BPF' in init/Kconfig and remove 'config BPF'
>> in kernel/bpf/Kconfig. This will be less confusing?
>
> Why?

The 'less confusing' part is just my initial feeling. I found
some CGROUP related configs are defined in init/Kconfig but not
under kernel/cgroup directory. So I thought 'config BPF' could
stay in init/Kconfig as well.

But I just did some other checking. For example, 'config NET'
is actually under 'net' directory. So probably you are right,
let us remove the one in init/Kconfig and use the one
in kernel/bpf/Kconfig.

So

Acked-by: Yonghong Song <yonghong.song@linux.dev>

>
>
>
>>>    # Used by archs to tell that they support BPF JIT compiler plus which
>>>    # flavour. Only one of the two can be selected for a specific arch since
>
>