Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 5 Feb 2020 11:53:31 +0000
From:   Quentin Perret <qperret@google.com>
To:     Matthias Maennich <maennich@google.com>
Cc:     masahiroy@kernel.org, nico@fluxnic.net,
        linux-kernel@vger.kernel.org, linux-kbuild@vger.kernel.org,
        kernel-team@android.com, jeyu@kernel.org
Subject: Re: [PATCH v2] kbuild: allow symbol whitelisting with
 TRIM_UNUSED_KSYMS
Message-ID: <20200205115331.GA74296@google.com>
References: <20200129181541.105335-1-qperret@google.com>
 <20200131131508.GH102066@google.com>
 <20200131174055.GA8425@google.com>
 <20200204144415.GC42496@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200204144415.GC42496@google.com>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tuesday 04 Feb 2020 at 14:44:15 (+0000), 'Matthias Maennich' via kernel-team wrote:
> That definitely looks like I would expect that config option to work.

Good, thank you, I'll send an update.

And while at it, I realized we could actually pre-populate autoksyms.h
very early on with the content of the whitelist instead of having it
empty. That way, there's a whole lot of things that will be compiled
'correctly' from the start (that is, adjust_autoksyms.sh won't need to
re-compile them later on). And in fact, if all symbols used by in-tree
modules happen to already be on the whitelist, you'll build a trimmed
kernel in a single pass, which should improve build time a bit I hope.

I'll add something to v3 in addition to your suggestion.

Thanks,
Quentin