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: <20230526183401.2326121-1-irogers@google.com> <ZHWCPXKo2fFJmtlT@tassilo>
In-Reply-To: <ZHWCPXKo2fFJmtlT@tassilo>
From:   Ian Rogers <irogers@google.com>
Date:   Mon, 29 May 2023 23:54:37 -0700
Message-ID: <CAP-5=fXXykcwjXjd+W=8Eez79JhmuKhf1Z-1R_8+AD4XdY4a8A@mail.gmail.com>
Subject: Re: [PATCH v2 00/16] Address some perf memory/data size issues
To:     Andi Kleen <ak@linux.intel.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>,
        Namhyung Kim <namhyung@kernel.org>,
        Adrian Hunter <adrian.hunter@intel.com>,
        "Masami Hiramatsu (Google)" <mhiramat@kernel.org>,
        "Steven Rostedt (Google)" <rostedt@goodmis.org>,
        Ross Zwisler <zwisler@chromium.org>,
        Leo Yan <leo.yan@linaro.org>,
        Tiezhu Yang <yangtiezhu@loongson.cn>,
        Yang Jihong <yangjihong1@huawei.com>,
        Kan Liang <kan.liang@linux.intel.com>,
        Ravi Bangoria <ravi.bangoria@amd.com>,
        Sean Christopherson <seanjc@google.com>,
        K Prateek Nayak <kprateek.nayak@amd.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, May 29, 2023 at 9:57=E2=80=AFPM Andi Kleen <ak@linux.intel.com> wro=
te:
>
>
> FWIW I think the whole patchkit could be replaced with a one liner
> that disables THP for the BSS segment. I suspect that would be roughly
> equivalent for memory consumption because 4K pages that are never
> touched would never be allocated.
>
> -Andi

So, it is worth reading some of the comments on the code to see why a
wider clean up is worth it:
https://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git/tree/tools/p=
erf/util/path.c?h=3Dperf-tools#n7
"It's obviously not thread-safe. Sue me."
(a comment stemming back to the git origins of the code base)

BSS won't count toward file size, which the patches were primarily
going after - but checking the size numbers I have miscalculated from
reading size's output that I'm not familiar with. The numbers are
still improved, but I just see a 37kb saving, with 5kb more in
.rodata. Something but not much. .data.rel.ro is larger, which imo is
good, but those pages will still be dirtied so a mute point wrt file
size and memory overhead.

For huge pages I thought it was correct that things are aligned by max
page size which I thought on x86-64 was 2MB, so I tried:
EXTRA_LDFLAGS=3D"-z max-page-size=3D4096"
but it made no difference to anything, and with:
EXTRA_CFLAGS=3D"-Wl,-z,max-page-size=3D4096"
EXTRA_CXXFLAGS=3D"-Wl,-z,max-page-size=3D4096"
file size just got worse.

Thanks,
Ian