linux-kernel - Re: [PATCH v2 01/12] mm/shmem: add `const` to lots of pointer parameters

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]

Message-ID: <2325a092-c810-4ae2-bf71-0a2c6fe361eb@redhat.com>
Date: Mon, 1 Sep 2025 12:40:02 +0200
From: David Hildenbrand <david@...hat.com>
To: Max Kellermann <max.kellermann@...os.com>
Cc: Kiryl Shutsemau <kirill@...temov.name>, akpm@...ux-foundation.org,
 axelrasmussen@...gle.com, yuanchu@...gle.com, willy@...radead.org,
 hughd@...gle.com, mhocko@...e.com, linux-kernel@...r.kernel.org,
 linux-mm@...ck.org, lorenzo.stoakes@...cle.com, Liam.Howlett@...cle.com,
 vbabka@...e.cz, rppt@...nel.org, surenb@...gle.com, vishal.moola@...il.com
Subject: Re: [PATCH v2 01/12] mm/shmem: add `const` to lots of pointer
 parameters

On 01.09.25 12:36, Max Kellermann wrote:
> On Mon, Sep 1, 2025 at 12:07 PM David Hildenbrand <david@...hat.com> wrote:
>>
>> On 01.09.25 12:00, Max Kellermann wrote:
>>> On Mon, Sep 1, 2025 at 11:53 AM David Hildenbrand <david@...hat.com> wrote:
>>>> I'm sorry, I have no time to argue about the basics of writing a patch
>>>> description. I even proposed a simple example of what we (multiple
>>>> reviewers) would expect as a bare minimum.
>>>
>>> But Lorenzo Stoakes and Mike Rappoport wanted much more than that.
>>
>> Sure, if it's not a simple "test" function as the one I commented on, it
>> might make sense to explain more why it is okay.
> 
> Lorenzo and Mike commented on the very same patch as you (i.e. 01/12).
> 
> I remember that you provided an example, and implementing that would
> have been easy - but it would not have been enough.

It would have been :)

See, Willy's patch made it clear that these are "test" functions. I 
incorporated that in my suggestion by using the term "test function".

For a "test" function (or a getter), it's trivial to see why we would 
want to have it const.

For other functions it's less clear, and might contradict to some plans 
we have (e.g., currently does not modify it but might in the future).

-- 
Cheers

David / dhildenb