linux-kernel - Re: patch suggestion: Kconfig symbols

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]

Date:   Wed, 28 Jul 2021 15:55:45 -0700
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Joe Perches <joe@...ches.com>
Cc:     linux-kernel-mentees@...ts.linuxfoundation.org,
        LKML <linux-kernel@...r.kernel.org>,
        kernel-janitors <kernel-janitors@...r.kernel.org>
Subject: Re: patch suggestion: Kconfig symbols

On 7/28/21 2:30 PM, Joe Perches wrote:
> On 2021-07-28 12:41, Randy Dunlap wrote:
>> On 7/28/21 8:37 AM, Joe Perches wrote:
>>> On Mon, 2021-07-26 at 17:21 -0700, Randy Dunlap wrote:
>>>> Running scripts/checkkconfigsymbols.py reports several hundred (maybe thousand)
>>>> Kconfig symbols that are used questionably. Lots of these are false positives
>>>> but lots of the remainder could use some cleaning up.
>>> []
>>>> False positive example:
>>>>
>>>> XCHOFFLD_MEM
>>>> Referencing files: drivers/scsi/qla2xxx/qla_mbx.c
>>>> Similar symbols: OF_PMEM, CXL_MEM, CXL_PMEM
>>>>
>>>> The Referencing source file does this:
>>>> #define CONFIG_XCHOFFLD_MEM    0x3
>>>>
>>>> which is legitimate, so no change is needed.
>>>
>>> Legitimate is perhaps dubious.
>>>
>>> It might be better if Kconfig has exclusive use of CONFIG_<foo> naming so
>>> renaming all the other existing CONFIG_<foo> defines might be appropriate.
>>
>> I would prefer that as well -- maybe 15 years ago.
>> But I think it's too invasive to make that change now.
> 
> I do not think it's that invasive.
> 
> It's something that doesn't have to be done immediately either.
> 
> It's not too many macro defines and not too many uses of those defines.

OK :)

I'm not trying to prevent such a change.

Getting the exclusive use of CONFIG_symbols being dedicated to
Kconfig namespace is the bug hurdle (IMO).

-- 
~Randy