linux-kernel - Re: [PATCH v2 4/8] selftests/mm: Add -Wunused family of flags

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

Message-ID: <88ea592c-c6ff-4a19-a366-eadeca66b039@arm.com>
Date: Thu, 21 Aug 2025 12:43:12 +0200
From: Kevin Brodsky <kevin.brodsky@....com>
To: Muhammad Usama Anjum <usama.anjum@...labora.com>,
 Andrew Morton <akpm@...ux-foundation.org>,
 David Hildenbrand <david@...hat.com>,
 Lorenzo Stoakes <lorenzo.stoakes@...cle.com>,
 "Liam R. Howlett" <Liam.Howlett@...cle.com>, Vlastimil Babka
 <vbabka@...e.cz>, Mike Rapoport <rppt@...nel.org>,
 Suren Baghdasaryan <surenb@...gle.com>, Michal Hocko <mhocko@...e.com>,
 Jason Gunthorpe <jgg@...pe.ca>, John Hubbard <jhubbard@...dia.com>,
 Peter Xu <peterx@...hat.com>, Leon Romanovsky <leon@...nel.org>,
 Zi Yan <ziy@...dia.com>, Baolin Wang <baolin.wang@...ux.alibaba.com>,
 Nico Pache <npache@...hat.com>, Ryan Roberts <ryan.roberts@....com>,
 Dev Jain <dev.jain@....com>, Barry Song <baohua@...nel.org>,
 Paul Walmsley <paul.walmsley@...ive.com>, Palmer Dabbelt
 <palmer@...belt.com>, Albert Ou <aou@...s.berkeley.edu>,
 Alexandre Ghiti <alex@...ti.fr>, linux-kselftest@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-mm@...ck.org,
 linux-riscv@...ts.infradead.org, Shuah Khan <shuah@...nel.org>
Cc: kernel@...labora.com
Subject: Re: [PATCH v2 4/8] selftests/mm: Add -Wunused family of flags

On 21/08/2025 08:28, Muhammad Usama Anjum wrote:
>> As to -Wunused-parameter I am frankly not convinced it's worth the
>> hassle. We're getting 90 lines changed in patch 6-8 just to mark
>> parameters as unused, in other words noise to keep the compiler happy.
>> It is not enabled by default in the kernel proper precisely because it
>> is so noisy when callbacks are involved.
>>
>> Patch 5 is clearly an improvement, but I'd rather take it without
>> actually enabling -Wunused-parameter. The rest of this patch isn't that
>> useful either IMHO.
> Patch 5 removes genuinely unused parameters flagged by the compiler. If we
> drop the -Wunused-parameter option, however, new unused parameters will
> continue to creep in with future patches. The goal of enabling this warning
> is to surface such issues early so developers can address them during
> development, rather than later during review or debugging.
>
> Long term, I’d like us to rely more on compiler and static analysis just like
> kernel to catch these kinds of problems proactively, instead of waiting until
> they’re reported or someone fixes them later. While it may feel like noise
> initially, this is largely a one-time cleanup—once done, developers will
> simply fix warnings as they arise, keeping the codebase cleaner going forward.

Agreed on the general principle, but I think the hassle is just too big
for what we're getting in return here (see also Andrew's reply). New
code may also introduce a bunch of unused parameters for legitimate
reasons and it's easy to imagine contributors ignoring such seemingly
harmless/irrelevant warnings instead of sprinkling __unused all over. My
feeling is that unused parameters are expected to be allowed in the
kernel and it isn't helpful to go against that expectation in just a
small subset of kselftests.

- Kevin