linux-kernel - Re: [PATCH v2 4/4] hwmon: (adt7470) Use standard update

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]

Date:   Mon, 30 Aug 2021 20:59:24 +0000
From:   Chris Packham <Chris.Packham@...iedtelesis.co.nz>
To:     Guenter Roeck <linux@...ck-us.net>
CC:     "jdelvare@...e.com" <jdelvare@...e.com>,
        "linux-hwmon@...r.kernel.org" <linux-hwmon@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 4/4] hwmon: (adt7470) Use standard update_interval
 property


On 31/08/21 3:36 am, Guenter Roeck wrote:
> On 8/29/21 2:09 PM, Chris Packham wrote:
>>
>> On 28/08/21 9:29 am, Guenter Roeck wrote:
>>> On Thu, Aug 26, 2021 at 02:41:21PM +1200, Chris Packham wrote:
>>>> Instead of the non-standard auto_update_interval make use of the
>>>> update_interval property that is supported by the hwmon core.
>>>>
>>>> Signed-off-by: Chris Packham <chris.packham@...iedtelesis.co.nz>
>>>> ---
>>>>
>>>> Notes:
>>>>       I kind of anticipate a NAK on this because it affects the 
>>>> ABI. But I figured
>>>>       I'd run it past the ML to see if moving towards the hwmon 
>>>> core is worth the hit
>>>>       in ABI compatibility.
>>> I personally don't mind (most likely no one is using it anyway), but 
>>> let's
>>> wait until after the upcoming commit window closes to give people 
>>> time to
>>> complain.
>>
>> I know of one application using this sysfs entry. But it's our in-house
>> environmental monitoring code so if this gets merged I'll just update it
>> to use the new path.
>>
>> One thought I had was we could do both. i.e. have an entry that conforms
>> to the hwmon core and a backwards compatible entry that just aliases the
>> new path.
>>
> Now you almost convinced me to indeed reject this patch. The idea of 
> the new API
> is to simplify driver code, not to make it more complicated. If we 
> can't simplify
> the code, it is better to leave it alone.
Sold. I agree what I've just suggested is adding more complexity without 
much gain. If something does start to care about having a standard 
update_interval property we could resurrect this.
>
> Guenter