All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Ihor Solodrai" <ihor.solodrai@linux.dev>
To: "Alan Maguire" <alan.maguire@oracle.com>, acme@kernel.org
Cc: yonghong.song@linux.dev, dwarves@vger.kernel.org, ast@kernel.org,
	andrii@kernel.org, bpf@vger.kernel.org, daniel@iogearbox.net,
	song@kernel.org, eddyz87@gmail.com, olsajiri@gmail.com
Subject: Re: [PATCH v2 dwarves 1/2] dwarves: Add github actions to build, test
Date: Wed, 09 Apr 2025 15:52:03 +0000	[thread overview]
Message-ID: <8f5711430e4f9bb02fd06d3a5b00d46d8643fea1@linux.dev> (raw)
In-Reply-To: <290e2542-fc51-402f-b84b-00e1f2ca2bfa@oracle.com>

On 4/9/25 1:59 AM, Alan Maguire wrote:
> 
>>> [...]
>>> +
>>> +cat tools/testing/selftests/bpf/config \
>>> +    tools/testing/selftests/bpf/config.${INPUTS_ARCH} > .config
>>> +# this file might or might not exist depending on kernel version
>>> +if [[ -f tools/testing/selftests/bpf/config.vm ]]; then
>>> +	cat tools/testing/selftests/bpf/config.vm >> .config
>>> +fi
>>> +make olddefconfig && make prepare
>>> +grep PAHOLE .config
>>> +grep _BTF .config
>>
>> This looks like debugging code, but instead of removing it I think it
>> is useful to dump entire config to the output (hence job log) in case
>> something goes wrong. How about `cat .config` before make
>> olddefconfig?
>>
>
> Sounds good, but would doing it after "make olddefconfig" be more
> informative maybe since some additional values may be set?

I think as long as you know what is being printed it's ok. The config
after olddefconfig is the full expanded config, which is quite big.

> [...]
>
>>> diff --git a/.github/workflows/vmtest.yml b/.github/workflows/vmtest.yml
>>> new file mode 100644
>>> index 0000000..0f66eed
>>> --- /dev/null
>>> +++ b/.github/workflows/vmtest.yml
>>> @@ -0,0 +1,62 @@
>>> +name: 'Build kernel run selftests via vmtest'
>>> +
>>> +on:
>>> +  workflow_call:
>>> +    inputs:
>>> +      runs_on:
>>> +        required: true
>>> +        default: 'ubuntu-24.04'
>>> +        type: string
>>> +      arch:
>>> +        description: 'what arch to test'
>>> +        required: true
>>> +        default: 'x86_64'
>>> +        type: string
>>> +      kernel:
>>> +        description: 'kernel version or LATEST'
>>> +        required: true
>>> +        default: 'LATEST'
>>> +        type: string
>>> +      pahole:
>>> +        description: 'pahole rev or branch'
>>> +        required: false
>>> +        default: 'master'
>>> +        type: string
>>> +      llvm-version:
>>> +        description: 'llvm version'
>>> +        required: false
>>> +        default: '18'
>>> +        type: string
>>> +jobs:
>>> +  vmtest:
>>> +    name: pahole@${{ inputs.arch }}
>>> +    runs-on: ${{ inputs.runs_on }}
>>> +    steps:
>>> +
>>> +      - uses: actions/checkout@v4
>>> +
>>> +      - name: Setup environment
>>> +        uses: libbpf/ci/setup-build-env@v3
>>> +        with:
>>> +          pahole: ${{ inputs.pahole }}
>>> +          arch: ${{ inputs.arch }}
>>> +          llvm-version: ${{ inputs.llvm-version }}
>>
>> I think I mentioned it before, but libbpf/ci/setup-build-env checks
>> out and installs pahole too, which is unnecessary here. Have you tried
>> removing this step from the job?
>>
>> You should be able to reuse a piece of SETUP logic from
>> build-debian.sh to install pahole's dependencies. Although you kernel
>> build deps are needed too.
>>
>
> Yeah it's the latter that are needed I think.
>
>> I could make a change in libbpf/ci/setup-build-env to accept a special
>> `pahole` input value or check for env variable to NOT build pahole.
>> What do you think?
>
> That would be great! Something like "pahole: none"?

Yes, something like that. It's a small change. I'll let you know when
it's done.

>
> I'll probably try and land this more or less as-is as we're hoping to
> get 1.30 out the door this week, but definitely will follow up with
> builds with shared library libbpf etc. Thanks for taking a look!
>
> Alan

  reply	other threads:[~2025-04-09 15:52 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-01  9:24 [PATCH v2 dwarves 0/2] dwarves: Introduce github actions for CI Alan Maguire
2025-04-01  9:24 ` [PATCH v2 dwarves 1/2] dwarves: Add github actions to build, test Alan Maguire
2025-04-08 19:09   ` Ihor Solodrai
2025-04-09  8:59     ` Alan Maguire
2025-04-09 15:52       ` Ihor Solodrai [this message]
2025-04-17 23:23         ` Ihor Solodrai
2025-04-01  9:24 ` [PATCH v2 dwarves 2/2] dwarves: Fix clang warning about unused variable Alan Maguire
2025-04-10 18:04 ` [PATCH v2 dwarves 0/2] dwarves: Introduce github actions for CI Alan Maguire

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8f5711430e4f9bb02fd06d3a5b00d46d8643fea1@linux.dev \
    --to=ihor.solodrai@linux.dev \
    --cc=acme@kernel.org \
    --cc=alan.maguire@oracle.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=dwarves@vger.kernel.org \
    --cc=eddyz87@gmail.com \
    --cc=olsajiri@gmail.com \
    --cc=song@kernel.org \
    --cc=yonghong.song@linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.