linux-kernel - Re: [PATCH] thinkpad_acpi: Add support for keyboard backlight

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: <20160121085736.GP7192@pali>
Date:	Thu, 21 Jan 2016 09:57:36 +0100
From:	Pali Rohár <pali.rohar@...il.com>
To:	Pavel Machek <pavel@....cz>
Cc:	Henrique de Moraes Holschuh <hmh@....eng.br>,
	Darren Hart <dvhart@...radead.org>,
	ibm-acpi-devel@...ts.sourceforge.net,
	platform-driver-x86@...r.kernel.org, linux-kernel@...r.kernel.org,
	Fabio D'Urso <fabiodurso@...mail.it>
Subject: Re: [PATCH] thinkpad_acpi: Add support for keyboard backlight

On Wednesday 13 January 2016 20:10:30 Pavel Machek wrote:
> On Wed 2016-01-13 20:07:36, Pavel Machek wrote:
> > On Wed 2016-01-13 09:54:55, Pali Rohár wrote:
> > > On Tuesday 12 January 2016 22:58:04 Pavel Machek wrote:
> > > > Hi!
> > > > 
> 
> > > > Next question is.. apparently there are some keyboards that have
> > > > per-key RGB backlight... but maybe we can just call that "weird
> > > > enough" and ignore...
> > > 
> > > First we need to defines stable kernel ABI for keyboard backlight. And I
> > > suggest to use existing convention used by upower/console-kit and other
> > > userspace apps...
> > 
> > Hmm... I'm not sure that can be done. What were the masks used by
> > upower again? Will upower write to all 6 leds if we present them?
> 
> Got it...
> 
> It has function up_kbd_backlight_find() which do:
> 
>         /* find a led device that is a keyboard device */
> 	        while ((filename = g_dir_read_name (dir)) != NULL) {
> 		                if (g_strstr_len (filename, -1, "kbd_backlight") != NULL) {
> 		                        dir_path = g_build_filename ("/sys/class/leds",
> 		                                                    filename,
> 								    	        NULL);
> 		                        break;
> 								
> That suggests that it stops at the first matching device. Adding new
> "virtual" led controlling 6 physical leds would be ugly. So... new
> interface should be done.
> 
> 								Pavel

Yes, it would be ugly, but lp5523 is already ugly... it can accept
numeric value, trigger and also program in bytecode either via sysfs or
via request_firmware... Another virtual led control should not be
problem for this :-)

-- 
Pali Rohár
pali.rohar@...il.com