linux-kernel - Re: [PATCH] amba: Drop builtin_amba

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Thu, 19 May 2022 11:53:55 +0100
From:   Arnd Bergmann <arnd@...db.de>
To:     Anshuman Khandual <anshuman.khandual@....com>
Cc:     Arnd Bergmann <arnd@...db.de>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will@...nel.org>,
        Russell King <linux@...linux.org.uk>,
        Wang Kefeng <wangkefeng.wang@...wei.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] amba: Drop builtin_amba_driver()

On Thu, May 19, 2022 at 4:08 AM Anshuman Khandual
<anshuman.khandual@....com> wrote:
> On 5/18/22 17:50, Arnd Bergmann wrote:
> > On Tue, May 17, 2022 at 7:29 AM Anshuman Khandual
> > <anshuman.khandual@....com> wrote:
> >>
> >> Drop builtin_amba_driver() which is not used anymore.
> >
> > The patch looks correct, but I don't see the purpose. Are you trying
> > to discourage
> > having amba drivers as built-in? Otherwise the next time we get an amba
> > driver that cannot be a loadable module, someone might want to add back the
> > same macro.
>
> Right, then it should be added back. But for now this is just dead code
> and not being used, hence why keep it ?

I don't care if the function is there or not, the interface does exactly what
one expects it to do, and it uses no space in the binary, but most likely
nobody would miss it if it's gone.

However, neither adding nor removing the function by itself to me hits the
threshold of doing a meaningful change that is an overall improvement, so
why send a patch?

The original patch that added the function was similar: the coresight drivers
were written to be loadable modules but Kconfig forced them to be built-in,
so Paul's cleanup to remove the module bits felt like completely pointless
churn that just made it harder for Kim to make them modular again.

If you want to remove the macro out of spite for the original change,
I'm sympathetic with that, but then put this in the patch description ;-)

        Arnd