linux-kernel - Re: [PATCH] prctl: Deprecate non PR_SET_MM

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, 4 Apr 2018 14:36:55 -0700
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Cyrill Gorcunov <gorcunov@...il.com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Andrey Vagin <avagin@...nvz.org>,
        Andrew Morton <akpm@...uxfoundation.org>,
        Pavel Emelyanov <xemul@...tuozzo.com>,
        Michael Kerrisk <mtk.manpages@...il.com>,
        Yang Shi <yang.shi@...ux.alibaba.com>,
        Michal Hocko <mhocko@...nel.org>
Subject: Re: [PATCH] prctl: Deprecate non PR_SET_MM_MAP operations

On 04/04/2018 02:29 PM, Cyrill Gorcunov wrote:
> On Wed, Apr 04, 2018 at 01:53:08PM -0700, Randy Dunlap wrote:
>>>
>>> At first this was plain warning without code removal but I've
>>> been advised that dropping it completely may be a better idea
>>> which I agree https://lkml.org/lkml/2018/4/4/31 Or you mean the
>>> warning message itsef? We aready have similars, for example
>>> in kernel/auditfilter.c
>>
>> The traditional way (eons ago) to deprecate something was to add a
>> printk() and then delete the feature a few years later.
>> Still, I have no objection to dropping that prctl.
>>
>> Sorry if I wasn't clear.  I was objecting to the "language", i.e., to the
>> word "deprecated."  Deprecated means frowned on, advised against, etc.
>> It does not mean "deleted."
> 
> True. I remember this rule of deprecation. But when I dropped the
> code I though which message to put here (or should I put it at
> all) and since "deprecated" was the first word came into mind
> I decided to grep sources, the result you see by its own :)
> 
>>
>>> printk(KERN_ERR "AUDIT_POSSIBLE is deprecated\n");
>>
>> Yeah, that one's wrong also.  :)
> 
> So, maybe just get rid of any warning message at all?

That would be OK since -EINVAL or something similar is being returned.

-- 
~Randy