Received-SPF: pass (google.com: domain of linux-kernel+bounces-76080-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-153-g7e3bb84806-fm-20240215.007-g7e3bb848
Precedence: bulk
MIME-Version: 1.0
Message-Id: <d419dc3f-ce85-4bae-bb93-05e0103a1c6c@app.fastmail.com>
In-Reply-To: 
 <CAK7LNAQG_dX8p0Lch-wg8cOVRQzaJUh2zFJL+3tgpcT8_iSTow@mail.gmail.com>
References: <20240221202655.2423854-1-jannh@google.com>
 <20240221202655.2423854-2-jannh@google.com>
 <CAK7LNAQG_dX8p0Lch-wg8cOVRQzaJUh2zFJL+3tgpcT8_iSTow@mail.gmail.com>
Date: Thu, 22 Feb 2024 09:20:26 +0100
From: "Arnd Bergmann" <arnd@arndb.de>
To: "Masahiro Yamada" <masahiroy@kernel.org>, "Jann Horn" <jannh@google.com>
Cc: "Andrew Morton" <akpm@linux-foundation.org>,
 "Nick Desaulniers" <ndesaulniers@google.com>,
 "Miguel Ojeda" <ojeda@kernel.org>, "Zhen Lei" <thunder.leizhen@huawei.com>,
 linux-kernel@vger.kernel.org, linux-kbuild@vger.kernel.org
Subject: Re: [PATCH 2/2] kallsyms: build faster by using .incbin
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable

On Thu, Feb 22, 2024, at 05:06, Masahiro Yamada wrote:
> On Thu, Feb 22, 2024 at 5:27=E2=80=AFAM Jann Horn <jannh@google.com> w=
rote:
>> This change reduces the time for an incremental kernel rebuild
>> (touch fs/ioctl.c, then re-run make) from 27.7s to 24.1s (medians
>> over 16 runs each) on my machine - saving around 3.6 seconds.

Nice!

..

> This reverts bea5b74504742f1b51b815bcaf9a70bddbc49ce3
>
> Somebody might struggle with debugging again, but I am not sure.
>
> Arnd?

So far, I have not needed it again, but it's only been a year.

> If the effort were "I invented a way to do kallsyms in
> one pass instead of three", it would be so much more attractive.
>
>
> I am not so sure if this grain of the optimization is exciting,
> but I confirmed that a few seconds were saved for the defconfig.
>
> I am neutral about this.

I think the time savings are worth it, especially since this
is going to help anyone building on large machines where
the compile stage is already optimized a lot but the link
stage is limited by single-thread performance.

      Arnd