Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) client-ip=23.128.96.34;
MIME-Version: 1.0
References: <20230914211948.814999-1-irogers@google.com>
In-Reply-To: <20230914211948.814999-1-irogers@google.com>
From:   Namhyung Kim <namhyung@kernel.org>
Date:   Mon, 18 Sep 2023 16:40:15 -0700
Message-ID: <CAM9d7cgNbRs3LJh_AjqAnRkJzsTxrGr_yqVK-urtoS-B2k1S=w@mail.gmail.com>
Subject: Re: [PATCH v1 0/5] Enable BPF skeletons by default
To:     Ian Rogers <irogers@google.com>
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        Mark Rutland <mark.rutland@arm.com>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        Jiri Olsa <jolsa@kernel.org>,
        Adrian Hunter <adrian.hunter@intel.com>,
        Nick Terrell <terrelln@fb.com>,
        Nathan Chancellor <nathan@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Tom Rix <trix@redhat.com>, Andrii Nakryiko <andrii@kernel.org>,
        Tiezhu Yang <yangtiezhu@loongson.cn>,
        James Clark <james.clark@arm.com>,
        Kajol Jain <kjain@linux.ibm.com>,
        Patrice Duroux <patrice.duroux@gmail.com>,
        Athira Rajeev <atrajeev@linux.vnet.ibm.com>,
        linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
        bpf@vger.kernel.org, llvm@lists.linux.dev
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Precedence: bulk

On Thu, Sep 14, 2023 at 2:20 PM Ian Rogers <irogers@google.com> wrote:
>
> Enable BPF skeletons by default but warn don't fail if they can't be
> supported. This was the intended behavior for Linux 6.4 but it caused
> an issue captured in this thread:
> https://lore.kernel.org/lkml/20230503211801.897735-1-acme@kernel.org/
>
> This issue isn't repeated here as the previous issue related to
> generating vmlinux.h, which is no longer performed by default as a
> checked-in vmlinux.h is used instead.
>
> Unlike with those changes, the BUILD_BPF_SKEL is kept and setting it
> to 0 disables BPF skeletons. Also, rather than fail the build due to a
> missed dependency, dependencies are checked and BPF skeletons disabled
> if they aren't present.
>
> Some related commits:
> b7a2d774c9c5 perf build: Add ability to build with a generated vmlinux.h
> a887466562b4 perf bpf skels: Stop using vmlinux.h generated from BTF, use subset of used structs + CO-RE
> a2af0f6b8ef7 perf build: Add system include paths to BPF builds
> 5be6cecda080 perf bpf skels: Make vmlinux.h use bpf.h and perf_event.h in source directory
> 9a2d5178b9d5 Revert "perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL"
> a887466562b4 perf bpf skels: Stop using vmlinux.h generated from BTF, use subset of used structs + CO-RE
> 1d7966547e11 perf build: Add warning for when vmlinux.h generation fails
> a980755beb5a perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL
>
> Ian Rogers (5):
>   perf version: Add status of bpf skeletons
>   perf build: Default BUILD_BPF_SKEL, warn/disable for missing deps
>   perf test: Update build test for changed BPF skeleton defaults
>   perf test: Ensure EXTRA_TESTS is covered in build test
>   perf test: Detect off-cpu support from build options

Tested-by: Namhyung Kim <namhyung@kernel.org>

Thanks,
Namhyung

>
>  tools/perf/Makefile.config              | 78 ++++++++++++++++---------
>  tools/perf/Makefile.perf                |  8 +--
>  tools/perf/builtin-version.c            |  1 +
>  tools/perf/tests/make                   |  7 ++-
>  tools/perf/tests/shell/record_offcpu.sh |  2 +-
>  5 files changed, 59 insertions(+), 37 deletions(-)
>
> --
> 2.42.0.459.ge4e396fd5e-goog
>