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;
MIME-Version: 1.0
References: <20220407125224.310255-1-jolsa@kernel.org> <20220407125224.310255-5-jolsa@kernel.org>
 <CAEf4BzbE1n3Lie+tWTzN69RQUWgjxePorxRr9J8CuiQVUfy-kA@mail.gmail.com>
 <20220412094923.0abe90955e5db486b7bca279@kernel.org> <20220413124419.002abd87@rorschach.local.home>
 <CAEf4BzaA+vr6V24dG7JCHHmedp2TcJv4ZnuKB=zXzuOpi-QYFg@mail.gmail.com> <20220413125906.1689c3e2@rorschach.local.home>
In-Reply-To: <20220413125906.1689c3e2@rorschach.local.home>
From:   Andrii Nakryiko <andrii.nakryiko@gmail.com>
Date:   Wed, 13 Apr 2022 12:02:22 -0700
Message-ID: <CAEf4BzY0_39BMyFWH1VdqLAWUSuy4GSWPogqUNi-F0YkrhQ5=w@mail.gmail.com>
Subject: Re: [RFC bpf-next 4/4] selftests/bpf: Add attach bench test
To:     Steven Rostedt <rostedt@goodmis.org>
Cc:     Masami Hiramatsu <mhiramat@kernel.org>,
        Jiri Olsa <jolsa@kernel.org>,
        Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Andrii Nakryiko <andrii@kernel.org>,
        Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>,
        lkml <linux-kernel@vger.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@chromium.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Apr 13, 2022 at 9:59 AM Steven Rostedt <rostedt@goodmis.org> wrote:
>
> On Wed, 13 Apr 2022 09:45:52 -0700
> Andrii Nakryiko <andrii.nakryiko@gmail.com> wrote:
>
> > > Did you only use the "notrace" on the prototype? I see the semicolon at
> > > the end of your comment. It only affects the actual function itself,
> > > not the prototype.
> >
> > notrace is both on declaration and on definition, see kernel/bpf/trampoline.c:
>
> OK. Note, it only needs to be on the function, the prototype doesn't do
> anything. But that shouldn't be the issue.
>
> >
> > void notrace __bpf_tramp_exit(struct bpf_tramp_image *tr)
> > {
> >         percpu_ref_put(&tr->pcref);
> > }
> >
>
> What compiler are you using? as this seems to be a compiler bug.
> Because it's not ftrace that picks what functions to trace, but the
> compiler itself.

I build my local kernel with

$ gcc --version
gcc (GCC) 11.1.1 20210623 (Red Hat 11.1.1-6)


But we have the same issue in our production kernels which are most
probably built with some other version of GCC, but I don't know which
one.


>
> -- Steve
>
>