linux-kernel - Re: [PATCH 09/10] netfilter: Add message pragma for deprecated xt_*.h, ipt

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: <33fe32ae-44aa-433d-a77f-bb75e8957842@intel.com>
Date: Thu, 9 Jan 2025 11:22:49 +0100
From: Przemek Kitszel <przemyslaw.kitszel@...el.com>
To: Szőke Benjamin <egyszeregy@...email.hu>, Jozsef Kadlecsik
	<kadlec@...filter.org>
CC: <fw@...len.de>, <pablo@...filter.org>, <lorenzo@...nel.org>,
	<daniel@...earbox.net>, <leitao@...ian.org>, <amiculas@...co.com>,
	<davem@...emloft.net>, <dsahern@...nel.org>, <edumazet@...gle.com>,
	<kuba@...nel.org>, <pabeni@...hat.com>, <horms@...nel.org>,
	<netfilter-devel@...r.kernel.org>, <coreteam@...filter.org>,
	<linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>
Subject: Re: [PATCH 09/10] netfilter: Add message pragma for deprecated
 xt_*.h, ipt_*.h.

On 1/8/25 22:38, Szőke Benjamin wrote:
> 2025. 01. 08. 21:51 keltezéssel, Jozsef Kadlecsik írta:
>> On Tue, 7 Jan 2025, Szőke Benjamin wrote:
>>
>>> 2025. 01. 07. 20:39 keltezéssel, Jozsef Kadlecsik írta:
>>>> On Tue, 7 Jan 2025, egyszeregy@...email.hu wrote:
>>>>
>>>>> From: Benjamin Szőke <egyszeregy@...email.hu>
>>>>>
>>>>> Display information about deprecated xt_*.h, ipt_*.h files
>>>>> at compile time. Recommended to use header files with
>>>>> lowercase name format in the future.
>>>>
>>>> I still don't know whether adding the pragmas to notify about header
>>>> file deprecation is a good idea.
>>>
>>> Do you have any other ideas how can you display this information to the
>>> users/customers, that it is time to stop using the uppercase header
>>> files then they shall to use its merged lowercase named files instead in
>>> their userspace SW?
>>
>> Honestly, I don't know. What about Jan's clever idea of having the
>> clashing filenames with identical content, i.e.
>>
>> ipt_ttl.h:
>> #ifndef _IPT_TTL_H
>> #define _IPT_TTL_H
>> #include <linux/netfilter_ipv4/ipt_ttl_common.h>
>> #endif _IPT_TTL_H
>>
>> ipt_TTL.h:
>> #ifndef _IPT_TTL_H
>> #define _IPT_TTL_H
>> #include <linux/netfilter_ipv4/ipt_ttl_common.h>
>> #endif _IPT_TTL_H
>>
>> Would cloning such a repo on a case-insensitive filesystem produce errors
>> or would work just fine?
>>
> 
> What is this suggestion, in ipt_ttl.h and ipt_TTL.h really? How it can 
> solve and provide in compile or run-time information for the users about 
> the recomendded changes? (It seems to me that you are completely 
> misunderstanding the purpose of this message at this time.)

likely the uppercased names will be with us forever

> 
> 
> Listen carefully, this are the points/scope.
> 
> This patchset provide the following:
> - 1. Merge upper and lowercase named haeder files in UAPI netfilter.
> - 2. Merge upper and lowercase named source files in UAPI netfilter. 
> (uppercase named files can be removed)
> - 3. Keep the backward compatibility, there is no any breaking API 
> changes yet.
> - 4. Keep uppercase header files as just a "wrapper" for include same 
> lowercase header files.
> - 5. Provide a clear message for the UAPI's users that in the future 
> should have to use the lowercase named files instead.

6. lot's of drama too.

Please remember to add a proper versioning to your next revision, also
target to net-next.

> 
> Later, for example when Linux kernel goes to 7.0 version, uppercase 
> header files can be removed. Breaking API possibble when version of a SW 
> is incremented in major field. Before, in first patchset, UAPI users 

that would be correct for "semantic versioning", not used by the kernel

> were informed about what is better to use. So it can be a clear and slow 
> roadmap to solve case-insensitive filesystem issue on this files.
> 
> 
>> Best regards,
>> Jozsef
> 
>