From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jacek Anaszewski Subject: Re: [PATCH v2] Documentation: leds: Add description of LED Flash class extension Date: Mon, 16 Feb 2015 12:18:48 +0100 Message-ID: <54E1D218.6020000@samsung.com> References: <1423125773-22751-1-git-send-email-j.anaszewski@samsung.com> <20150211215723.GC11313@amd> <54DC671E.8010705@samsung.com> <20150212090756.GA30508@amd> <54DC7C1E.5020105@samsung.com> <20150212105417.GA10377@amd> <54DC8B6F.9080802@samsung.com> <20150212174444.GA9450@amd> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mailout1.w1.samsung.com ([210.118.77.11]:50254 "EHLO mailout1.w1.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932255AbbBPLSy (ORCPT ); Mon, 16 Feb 2015 06:18:54 -0500 Received: from eucpsbgm1.samsung.com (unknown [203.254.199.244]) by mailout1.w1.samsung.com (Oracle Communications Messaging Server 7u4-24.01(7.0.4.24.0) 64bit (built Nov 17 2011)) with ESMTP id <0NJV00H8F3MA4T10@mailout1.w1.samsung.com> for linux-leds@vger.kernel.org; Mon, 16 Feb 2015 11:22:58 +0000 (GMT) In-reply-to: <20150212174444.GA9450@amd> Sender: linux-leds-owner@vger.kernel.org List-Id: linux-leds@vger.kernel.org To: Pavel Machek Cc: linux-leds@vger.kernel.org, b.zolnierkie@samsung.com, cooloney@gmail.com, rpurdie@rpsys.net, sakari.ailus@iki.fi, s.nawrocki@samsung.com Hi Pavel, On 02/12/2015 06:44 PM, Pavel Machek wrote: > Hi! > >>>> You acked LED Flash class patch, didn't you? :) >>>> >>>> There are many attributes documented in the list fashion, e.g.: >>>> available_frequencies in the Documentation/ABI/testing/sysfs-class-devfreq. >>> >>> No, sorry, you can't do this. We have this parsing nightmare in /proc, >>> before, and we don't want it again. >>> >>>> If we changed this a bit it would be easily parsed with AWK: >>>> >>>> echo "0 none;1 max77693-led1;2 max77693-led2" | awk -F';' '{ for (i=1; >>>> i<=NF; i++) print $i}' | awk '{print $1": "$2}' >>> >>> sysfs is one entry per file. If someone screwed it up in devfreq, it >>> is not reason to screw it up here. (Space separated lists of integers >>> might be acceptable. Fixed strings with one marked by []s happen, >>> too. Maps between ints and names are not.) >> >> "Fixed strings with one marked by []s happen" - I don't quite >> get this. > > root@duo:~# cat /sys/power/state > freeze mem disk > root@duo:~# cat /sys/power/disk > [platform] shutdown reboot suspend > root@duo:~# > >> Could you propose the acceptable format then? >> The numbers alone are inconvenient to use, we need a human >> readable description next to them. Or some another way >> to obtain the name. > > Files like "0.active", "0.name", "1.active", "1.name"? > > Subdirectories 0/1 with "name" and "active" files? We have flash_sync_strobe attribute which allows to read currently chosen sync_led. I would simplify this: Let's assume there are two LED Flash class deviecs: - max77693-flash1 - max77693-flash2 Proposed design described by use cases: #cd /sys/class/leds/max77693-flash1 #cat available_flash_leds #[0.none] 1.max77693-flash2 #cat flash_sync_strobe #0.none #echo 1 > flash_sync_strobe #cat flash_sync_strobe #1.max77693-flash2 #cat available_sync_leds #0.none [1.max77693-flash2] -- Best Regards, Jacek Anaszewski