linux-kernel - Re: [PATCH v1 1/1] mux: consumer: Add dummy functions for !CONFIG

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: <c94b90f2-2b56-8209-df55-3bd7ca30f90d@axentia.se>
Date:   Mon, 10 Jul 2017 07:32:13 +0200
From:   Peter Rosin <peda@...ntia.se>
To:     "Kuppuswamy, Sathyanarayanan" <sathyaosid@...il.com>,
        sathyanarayanan.kuppuswamy@...ux.intel.com
Cc:     linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 1/1] mux: consumer: Add dummy functions for
 !CONFIG_MULTIPLEXER case

On 2017-07-09 01:04, Kuppuswamy, Sathyanarayanan wrote:
> Hi Peter,
> 
> On 7/8/2017 1:55 PM, Peter Rosin wrote:
>> On 2017-07-07 23:41, sathyanarayanan.kuppuswamy@...ux.intel.com wrote:
>>> From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@...ux.intel.com>
>>>
>>> Add dummy functions to avoid compile time issues when CONFIG_MULTIPLEXER
>>> is not enabled.
>> Hi!
>>
>> Consumers should "select MULTIPLEXER",
>   If their driver can't work without mux_* calls then  you can make it 
> compulsory. But its not always true.
>>   so this does not make sense.
>> Or do you have a driver that has an optional mux consumer?
> I came across this case when I was working on Intel USB MUX driver. I 
> think you know the history behind it. Although I am not planning to 
> merge that driver now, but I think the use case is still valid.

Yeah, it's a valid use case. But why add a facility that noone uses? Sure,
if there's an actual consumer that needs it. But there isn't...

See, I have spent considerable time taking stuff like this out in order to
get the thing merged at all. I even think I wrote dummy inlines like this
at some point (but I'm not sure if I actually wrote them and I don't think
I submitted them. But I did think about it, that's for sure). Anyway, I'm
not very happy about ballooning the core with support for non-essentials
just yet. Maybe my mind-set will change over time?

(And no, I don't *know* the history behind the "Intel USB MUX driver",
 I e.g. never saw the consumer code. And I have the feeling that stuff
 were discussed in other threads that I was not part of and some (most?)
 questions I asked about it was left unanswered.)

Cheers,
peda