linux-kernel - Re: Exponential LED brightness Re: [PATCH v4 0/6] leds: lm3692x: Allow to set ovp and brigthness mode

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: <20200108124000.GA21245@bogon.m.sigxcpu.org>
Date:   Wed, 8 Jan 2020 13:40:00 +0100
From:   Guido Günther <agx@...xcpu.org>
To:     Pavel Machek <pavel@....cz>
Cc:     Jacek Anaszewski <jacek.anaszewski@...il.com>,
        Dan Murphy <dmurphy@...com>, Rob Herring <robh+dt@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        linux-leds@...r.kernel.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: Exponential LED brightness Re: [PATCH v4 0/6] leds: lm3692x:
 Allow to set ovp and brigthness mode

Hi Pavel,
On Tue, Jan 07, 2020 at 02:31:20PM +0100, Pavel Machek wrote:
> Hi!
> 
> > Overvoltage protection and brightness mode are currently hardcoded
> > as 29V and disabled in the driver. Make these configurable via DT.
> > 
> > This v4 moves the exponential brightness mode to the back of the series
> > as per Pavel's request:
> > 
> >   https://lore.kernel.org/linux-next/20200106103233.GA32426@amd/T/#m93270a9bf10b88e060f4e4cf5701c527476de985
> > 
> > The end result is identical and i've tested everything still works when
> > dropping the last to patches and checked compiltion via
> 
> Thank you. Applied 1-4 (with some reformatting of changelog, and
> led->LED).
> 
> Exponential mode:
> 
> We should decide if LEDs should be linear or not. Most LEDs are linear
> now, and we may want to make it part of the API. Additional advantage
> is that linear is "well defined". It is actually quite important for
> RGB LEDs, because you get wrong colors otherwise.
> 
> (Non-linear can have advantages, too... like needing less bits.)
> 
> So, my suggestion is to document LEDs as linear, and leave
> exponential->linear conversion to someone else.

That would mean doing a conversion in the kernel that can be done by the
chip. Would exposing non-linearity like in
/sys/class/backlight/<backlight>/scale be an option?
Cheers,
 -- Guido


> 
> Best regards,
> 
> 									Pavel
> -- 
> (english) http://www.livejournal.com/~pavelmachek
> (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html