Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210111180609.713998-1-natechancellor@gmail.com>
 <CAK7LNAQ=38BUi-EG5v2UiuAF-BOsVe5BTd-=jVYHHHPD7ikS5A@mail.gmail.com>
 <20210111193400.GA1343746@ubuntu-m3-large-x86> <CAK7LNASZuWp=aPOCKo6QkdHwM5KG6MUv8305v3x-2yR7cKEX-w@mail.gmail.com>
 <20210111200010.GA3635011@ubuntu-m3-large-x86> <CAEf4BzaL18a2+j3EYaD7jcnbJzqwG2MuBxXR2iRZ3KV9Jwrj6w@mail.gmail.com>
 <CAEf4Bzbv6nrJNxbZAvFx4Djvf1zbWnrV_i90vPGHtV-W7Tz=bQ@mail.gmail.com> <20210113230739.GA22747@Ryzen-9-3900X.localdomain>
In-Reply-To: <20210113230739.GA22747@Ryzen-9-3900X.localdomain>
Reply-To: sedat.dilek@gmail.com
From:   Sedat Dilek <sedat.dilek@gmail.com>
Date:   Thu, 4 Feb 2021 08:22:01 +0100
Message-ID: <CA+icZUVfznOpAQK=6GWoF6XmzHyXjdUgNG5HeoQw3Dwb4wW9uA@mail.gmail.com>
Subject: Re: [PATCH] bpf: Hoise pahole version checks into Kconfig
To:     Nathan Chancellor <natechancellor@gmail.com>
Cc:     Andrii Nakryiko <andrii.nakryiko@gmail.com>,
        Masahiro Yamada <masahiroy@kernel.org>,
        Michal Marek <michal.lkml@markovi.net>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Andrii Nakryiko <andrii@kernel.org>,
        Martin KaFai Lau <kafai@fb.com>,
        Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
        John Fastabend <john.fastabend@gmail.com>,
        KP Singh <kpsingh@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Linux Kbuild mailing list <linux-kbuild@vger.kernel.org>,
        Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Jan 14, 2021 at 12:07 AM Nathan Chancellor
<natechancellor@gmail.com> wrote:
>
> On Wed, Jan 13, 2021 at 02:38:27PM -0800, Andrii Nakryiko wrote:
> > Hm.. Just saw Linus proposing using $(error-if) in Kconfig for an
> > unrelated issue ([0]). If we can make this work, then it would catch
> > such issue early on, yet won't have any downsides of hiding
> > CONFIG_DEBUG_INFO_BTF if pahole is too old. WDYT?
> >
> >   [0] https://lore.kernel.org/lkml/CAHk-=wh-+TMHPTFo1qs-MYyK7tZh-OQovA=pP3=e06aCVp6_kA@mail.gmail.com/
>
> Yes, I think that would be exactly what we want because DEBUG_INFO_BTF
> could cause the build to error if PAHOLE_VERSION is not >= 116. I will
> try to keep an eye on that thread to see how it goes then respin this
> based on anything that comes from it.
>

For BPF/pahole testing (see [1]) with CONFIG_DEBUG_INFO_DWARF5=y I did:

$ git diff
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index b0840d192e95..f15b37143165 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -272,7 +272,7 @@ config DEBUG_INFO_DWARF5
       bool "Generate DWARF Version 5 debuginfo"
       depends on GCC_VERSION >= 50000 || CC_IS_CLANG
       depends on CC_IS_GCC ||
$(success,$(srctree)/scripts/test_dwarf5_support.sh $(CC)
$(CLANG_FLAGS))
-       depends on !DEBUG_INFO_BTF
+       depends on !DEBUG_INFO_BTF || (DEBUG_INFO_BTF && PAHOLE_VERSION >= 120)
       help
         Generate DWARF v5 debug info. Requires binutils 2.35.2, gcc 5.0+ (gcc
         5.0+ accepts the -gdwarf-5 flag but only had partial support for some

Thanks again for that patch.

- Sedat -

[1] https://git.kernel.org/pub/scm/devel/pahole/pahole.git/log/?h=tmp.1.20