Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: pahole + BTF was: Re: [Question] bpf: about a new
 'tools/bpf/bpf_dwarf2btf'
From:   Taeung Song <treeze.taeung@gmail.com>
To:     Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>,
        Daniel Borkmann <daniel@iogearbox.net>
Cc:     Alexei Starovoitov <ast@kernel.org>,
        Martin KaFai Lau <kafai@fb.com>, netdev@vger.kernel.org,
        Thomas Girard <thomas.g.girard@free.fr>,
        Martin Cermak <mcermak@redhat.com>,
        linux-kernel@vger.kernel.org, acme@kernel.org
References: <47cfb90e-a9f4-8127-2b89-0e20d9fda2f0@gmail.com>
 <20180725175231.GA2121@redhat.com>
 <35663517-26e1-c2a2-1554-5dd90be2747f@gmail.com>
Message-ID: <592c6325-23d0-771b-cc28-6a66842a1265@gmail.com>
Date:   Thu, 26 Jul 2018 04:21:31 +0900
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <35663517-26e1-c2a2-1554-5dd90be2747f@gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 07/26/2018 03:27 AM, Taeung Song wrote:
> Hi Arnaldo,
> 
> On 07/26/2018 02:52 AM, Arnaldo Carvalho de Melo wrote:
>> Em Thu, Jul 26, 2018 at 02:23:32AM +0900, Taeung Song escreveu:
>>> Hi,
>>>
>>> Building bpf programs with .BTF section,
>>> I thought it'd be better to convert dwarf info to .BTF by
>>> a new tool such as 'tools/bpf/bpf_dwarf2btf' instead of pahole
>>> in the future.
>>> Currently for bpf binary that have .BTF section,
>>> we need to use pahole from https://github.com/iamkafai/pahole/tree/btf
>>> with the command line such as "pahole -J bpf_prog.o".
>>> I think it is great but if implementing new 'bpf_dwarf2btf'
>>> (dwarf parsing + btf encoder code written by Martin KaFai Lau on
>>> the pahole project i.e. btf.h, btf_encoder.c, btf_encoder.h,
>>> libbtf.c, libbtf.h),
>>> BPF developers would more easily use functionalities based on BTF.
>>
>> What would be easier exactly? Not having to install a package but build
>> it from the kernel sources?
>>
>> Many kernel developers already have pahole installed for other uses, so
>> no need to install anything.
>>
> 
> Understood, but I think there are many non-kernel developers
> developing BPF programs and they mightn't have or use pahole.
> 
> So, if providing the 'dwarf2btf' feature on tools/bpf or tools/bpf/bpftool,
> non-kernel developers can also more easily build bpf prog with .BPF, no ?
> 

Or, if tools/lib/bpf/ have the 'dwarf2btf' feature,
I think BPF developers can just use bpf programs that have dwarf info
after compiling with clang '-g' and llc '-mattr=dwarfris', even though 
not using pahole.
Isn't it good way ?

>> BTW, Daniel, I just pushed to pahole's main repository at:
>>
>>    git://git.kernel.org/pub/scm/devel/pahole/pahole.git
>>
>> with the Martin's BTF patch, so no need to pull from the github one,
>> I'll tag v1.12 and announce the release so that distro package
>> maintainers can update their packages.
>>> What do you think about this ? Do you think this is needed ?
>>> Or, already implementing something like this ?
>>> If it is needed, I want to try to make 'tools/bpf/bpf_dwarf2btf'
>>> based on the pahole code. I'd appreciate it, if you reply to this
>>
>> The way Martin took advantage of the work done a long time ago to
>> support CTF out of the same DWARF reading codebase was really cool, not
>> that much work to do, just add a new format to pahole's codebase making
>> it more useful.
>>
> 
> I got it !
> 
> Thanks,
> Taeung
> 
>> I was just so far overly picky with testing it and kept leaving for
>> later to have a good documentation about testing it, vacation and perf
>> maintainership duties kept making this take like forever, grumble :-\
>>
>> - Arnaldo
>>
> 
> -- 
> oh.. :'(