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: <YYQadWbtdZ9Ff9N4@kernel.org> <YYQdKijyt20cBQik@kernel.org>
 <CAEf4BzYtq5Fru0_=Stih+Tjya3i29xG+RSF=4oOT7GbUwVRQaQ@mail.gmail.com>
 <YYQiXnUxlOoWMdwZ@kernel.org> <C940FF7A-A27F-4F56-8659-9365FC4A2EF7@fb.com>
In-Reply-To: <C940FF7A-A27F-4F56-8659-9365FC4A2EF7@fb.com>
From:   Namhyung Kim <namhyung@kernel.org>
Date:   Thu, 4 Nov 2021 13:33:20 -0700
Message-ID: <CAM9d7cg9rXKUdEsdGUBSemzSrwE8XsyQtjCM=zT+8P+gs10n=Q@mail.gmail.com>
Subject: Re: perf build broken looking for bpf/{libbpf,bpf}.h after merge with upstream
To:     Song Liu <songliubraving@fb.com>
Cc:     Arnaldo Carvalho de Melo <acme@kernel.org>,
        Andrii Nakryiko <andrii.nakryiko@gmail.com>,
        Jiri Olsa <jolsa@kernel.org>, bpf <bpf@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hello,

On Thu, Nov 4, 2021 at 11:13 AM Song Liu <songliubraving@fb.com> wrote:
>
>
>
> > On Nov 4, 2021, at 11:11 AM, Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> >
> > Em Thu, Nov 04, 2021 at 10:56:26AM -0700, Andrii Nakryiko escreveu:
> >> On Thu, Nov 4, 2021 at 10:49 AM Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com> wrote:
> >
> >>> Em Thu, Nov 04, 2021 at 02:37:57PM -0300, Arnaldo Carvalho de Melo escreveu:
> >>>>
> >>>> Hi Song,
> >>>>
> >>>>      I just did a merge with upstream and I'm getting this:
> >>>>
> >>>>  LINK    /tmp/build/perf/plugins/plugin_scsi.so
> >>>>  INSTALL trace_plugins
> >>>
> >>> To clarify, the command line to build perf that results in this problem
> >>> is:
> >>>
> >>>  make -k BUILD_BPF_SKEL=1 CORESIGHT=1 PYTHON=python3 O=/tmp/build/perf -C tools/perf install-bin
> >>
> >> Oh, I dropped CORESIGN and left BUILD_BPF_SKEL=1 and yeah, I see the
> >> build failure. I do think now that it's related to the recent Makefile
> >> revamp effort. Quentin, PTAL.
> >>
> >> On the side note, why BUILD_BPF_SKEL=1 is not a default, we might have
> >> caught this sooner. Is there any reason not to flip the default?
> >
> > I asked Song in the past about this, and asked again on another reply to
> > this thread, I think it should be the default.
> >
> > Song, Namhyung? You're the skel guys (so far) :-)
>
> Yeah, let's make it default.

Then it'd require 'clang' for the perf build.  Maybe we can check
the availability of the compiler and disable it back if not.

Thanks,
Namhyung