Date:   Thu, 23 Dec 2021 22:47:55 -0800
From:   Christoph Hellwig <hch@infradead.org>
To:     Vimal Agrawal <avimalin@gmail.com>
Cc:     Christoph Hellwig <hch@infradead.org>,
        Luis Chamberlain <mcgrof@kernel.org>,
        Vimal Agrawal <vimal.Agrawal@sophos.com>,
        Masahiro Yamada <masahiroy@kernel.org>,
        Michal Marek <michal.lkml@markovi.net>,
        Nick Desaulniers <ndesaulniers@google.com>,
        Jan Beulich <JBeulich@suse.com>, Jeff Mahoney <jeffm@suse.com>,
        Sam Ravnborg <sam@ravnborg.org>, linux-kbuild@vger.kernel.org,
        jeyu@kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] kernel/module.c: heuristic enhancement when
 INSTALL_MOD_STRIP= "--strip-unneeded" is used
Message-ID: <YcVtG26b/sO9k7ox@infradead.org>
References: <YcJZWiQ407ZxMM+y@bombadil.infradead.org>
 <20211222132332.7817-1-vimal.agrawal@sophos.com>
 <YcRRQCMZFepB/hzX@infradead.org>
 <CALkUMdRxTm6STT4CncTuvQ9hM_bez+B91TsuenEj71KPxFgMsg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <CALkUMdRxTm6STT4CncTuvQ9hM_bez+B91TsuenEj71KPxFgMsg@mail.gmail.com>
Precedence: bulk

On Thu, Dec 23, 2021 at 04:39:15PM +0530, Vimal Agrawal wrote:
> Hi Christoph,
> 
> On Thu, Dec 23, 2021 at 4:06 PM Christoph Hellwig <hch@infradead.org> wrote:
> >
> > We never build modules with that ёption, so this is completely pointless.
> >
> we use openwrt for build and packaging and it has been using this
> option for long.
> 
> kbuild documentation says the following for INSTALL_MOD_STRIP:
> If this variable is specified, it will cause modules to be stripped
> after they are installed. If INSTALL_MOD_STRIP is ‘1’, then the
> default option –strip-debug will be used. Otherwise, the
> INSTALL_MOD_STRIP value will be used as the option(s) to the strip
> command.
> 
> So if kbuild does not support INSTALL_MOD_STRIP=--strip-unneeded
> option then we should call out what it supports and should not even
> allow what is not supported. We don't know what other options others
> may be using but if we allow it then we should support it and it
> should not behave erratic just because someone is using a
> non-recommended option.

I don't think we can support passing arbitrary linker options and
expects things to work.  If we want to support --strip-unneeded
it needs a good rationale and be added as a direct config option.