linux-kernel - Re: [PATCH v2] leds: lm3697: Rename struct into more appropriate name

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20201125112559.GD29328@amd>
Date:   Wed, 25 Nov 2020 12:25:59 +0100
From:   Pavel Machek <pavel@....cz>
To:     Dan Murphy <dmurphy@...com>
Cc:     Marek Behun <kabel@...ckhole.sk>,
        Gabriel David <ultracoolguy4@...tonmail.com>,
        linux-leds@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] leds: lm3697: Rename struct into more appropriate name

Hi!

> > From the perspective of Linux you see the two control banks as 2 LED
> >class devices (because you are setting brightness for control banks,
> >not for the LED strips).
> 
> The way Marek explains it is correct and the way I wrote the driver
> intially.  There is no direct control of the LEDs only controlling the 2
> banks.
> 
> As an example a device can put LED string 1 and 2 on a single bank to
> control the backlight for a display and put LED string 3 on a different bank
> to control the backlight of a keyboard. Like in the Droid and Droid 4
> devices.  2 strings illuminate the display backlight and 1 string
> illuminates the keyboard the display backlight can have a independent
> brightness then the keyboard.
> 
> To me the name of the structure does not impose any functional changes just
> semantic changes.  And it just makes it a bit more difficult to back port
> functional fixes as this patch would be made mandatory for cherry picking. 
> But I do not get many requests to back port this driver so it maybe be a
> moot point.

Ok, sorry for the confusion, and .. I believe the code can stay as-is.

Bank is single entity Linux controls, and it does not need to know how
many pins are really controlled on the hardware level.

It will be confusing one way or another.

Best regards,
									Pavel
-- 
http://www.livejournal.com/~pavelmachek

Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)