Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <9dec8409-6e69-4f14-dcd8-9f47e088ce17@redhat.com>
Date:   Wed, 29 Mar 2023 13:40:41 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.9.0
Subject: Re: [PATCH bpf-next] kallsyms: move module-related functions under
 correct configs
Content-Language: en-US
To:     Daniel Borkmann <daniel@iogearbox.net>,
        Luis Chamberlain <mcgrof@kernel.org>
Cc:     bpf@vger.kernel.org, linux-modules@vger.kernel.org,
        linux-kernel@vger.kernel.org, Alexei Starovoitov <ast@kernel.org>,
        John Fastabend <john.fastabend@gmail.com>,
        Andrii Nakryiko <andrii@kernel.org>,
        Martin KaFai Lau <martin.lau@linux.dev>,
        Song Liu <song@kernel.org>, Yonghong Song <yhs@fb.com>,
        KP Singh <kpsingh@kernel.org>,
        Stanislav Fomichev <sdf@google.com>,
        Hao Luo <haoluo@google.com>, Jiri Olsa <jolsa@kernel.org>,
        "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Jesper Dangaard Brouer <hawk@kernel.org>,
        Nathan Chancellor <nathan@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Tom Rix <trix@redhat.com>, kernel test robot <lkp@intel.com>
References: <20230327161251.1129511-1-vmalik@redhat.com>
 <ZCHWtptOwPPtUe+u@bombadil.infradead.org>
 <c076e249-705a-e1bb-c657-f80cd4f2145b@redhat.com>
 <ZCHuU4Wui7Dwmdm2@bombadil.infradead.org>
 <bfaefad6-692f-e687-ad55-05a43aa54883@iogearbox.net>
From:   Viktor Malik <vmalik@redhat.com>
In-Reply-To: <bfaefad6-692f-e687-ad55-05a43aa54883@iogearbox.net>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 3/28/23 10:23, Daniel Borkmann wrote:
> On 3/27/23 9:28 PM, Luis Chamberlain wrote:
>> On Mon, Mar 27, 2023 at 08:20:56PM +0200, Viktor Malik wrote:
>>> On 3/27/23 19:47, Luis Chamberlain wrote:
>>>> On Mon, Mar 27, 2023 at 06:12:51PM +0200, Viktor Malik wrote:
>>>>> Functions for searching module kallsyms should have non-empty
>>>>> definitions only if CONFIG_MODULES=y and CONFIG_KALLSYMS=y. Until now,
>>>>> only CONFIG_MODULES check was used for many of these, which may have
>>>>> caused complilation errors on some configs.
>>>>>
>>>>> This patch moves all relevant functions under the correct configs.
>>>>>
>>>>> Signed-off-by: Viktor Malik <vmalik@redhat.com>
>>>>> Reported-by: kernel test robot <lkp@intel.com>
>>>>> Link: https://lore.kernel.org/oe-kbuild-all/202303181535.RFDCnz3E-lkp@intel.com/
>>>>
>>>> Thanks Viktor!  Does this fix something from an existing commit? If so
>>>> which one?  The commit log should mention it.
>>>
>>> Ah, right, I forgot about that. The commit log is missing:
>>>
>>> Fixes: bd5314f8dd2d ("kallsyms, bpf: Move find_kallsyms_symbol_value out of internal header")
>>>
>>> I can post v2 but I'm also fine with maintainers applying the tag.
>>
>> That patch went through the bpf tree so its fix can go throug that tree.
>> So up to Daniel if he wants a new patch.
> 
> Fixing up is fine with me. Viktor, which config combinations did you test for this
> patch and under which architectures?
> 
> I suspect kbuild bot might still complain. For example, your patch moves
> dereference_module_function_descriptor() stub definition under !CONFIG_MODULES ||
> !CONFIG_KALLSYMS. Looking at ppc's dereference_module_function_descriptor()
> implementation (!CONFIG_PPC64_ELF_ABI_V2 case), it's build under CONFIG_MODULES,
> so you'll have two different implementations of the functions, the generic one
> then w/o __weak attribute.

Right, I didn't notice this one, thanks. Looks like
dereference_module_function_descriptor will need to stay in the original
place (under CONFIG_MODULES only). I'll cross-compile for arches that
redefine this function (powerpc, ia64, parisc) with relevant config
combinations and submit v2. The rest of the functions don't have
arch-dependent implementations, so they should be fine.

> 
> Also small nit, please fix the patch up wrt indentation to align with kernel coding
> style.

Ok, will do.

Thanks,
Viktor

> 
> Thanks,
> Daniel
>