linux-leds.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Pavel Machek <pavel@ucw.cz>
To: Vadim Pasternak <vadimp@nvidia.com>
Cc: Marek Behun <marek.behun@nic.cz>,
	"jacek.anaszewski@gmail.com" <jacek.anaszewski@gmail.com>,
	"linux-leds@vger.kernel.org" <linux-leds@vger.kernel.org>
Subject: Re: [PATCH led-next 1/1] leds: mlxreg: Allow multi-instantiation of same name LED for modular systems
Date: Wed, 21 Oct 2020 10:33:34 +0200	[thread overview]
Message-ID: <20201021083334.GA29124@duo.ucw.cz> (raw)
In-Reply-To: <DM6PR12MB38989BFE9D1967C69E397A6BAF070@DM6PR12MB3898.namprd12.prod.outlook.com>

[-- Attachment #1: Type: text/plain, Size: 2794 bytes --]

Hi!

> > > But why you consider it as function enumerator?
> > > For example card48, card56 are two different devices of same type.
> > > Both have 'status' LED.
> > 
> > OK this is a fair point.
> > 
> > I was thinking such because in my mind I had this idea that for an ethernet
> > switch with interfaces lan0 - lan4 it would make sense to use the
> > LED_FUNCTION_LAN function with function enumerator. But thinking about
> > this now again makes me wonder if instead the lan0 - lan4 should be
> > devicenames instead, since normally they are network interface names.
> > 
> > Vadim, the reason why Pavel and I think that mlxreg (or mlxregN) is not
> > valid devicename part (although mlxreg has to stay since many users
> > already depend on it, as you say), is that the mlxreg name is not exposed
> > anywhere else in Linux from userspace point of view.
> > 
> > Devicename eth0 is okay, because it is network interface name.
> > Devicename sda would be okay, because everyone knows it is a block device
> > and you can access it via /dev/sda.
> > Devicename hci0 would be okay because it is bluetooth interface accessible
> > via hcitool.
> > Devicenames mtd0, kbd0, mouse0 would be okay, I think.
> > 
> > But mlxreg is not accessible via anything else in the system. Unless your
> > systems also have something like /dev/mlxreg, that is.
> > 
> > Do the LEDs on these cards only indicate status of the cards themselves as a
> > whole? Or are there LEDs on these cards dedicated to their peripherals? For
> > example if there is an ethernet port with LEDs on one of these cards, the
> > devicename part for these LEDs should be of the device of that ethernet
> > port, not mlxreg...
> 
> Hi Marek,
> 
> Each line card must have 'status' LED, indicating status of line card itself.
> User can set non-green in case some there are some alarms on different devices,
> equipped on this line card. It can be set blink during line card initialization.
> 
> Line card could be equipped with UID LED. User can set this LED in order to
> find physical location of line card. Sometimes it's hard to see the sticker on
> chassis.
> 
> Line card also equipped with per port LED, but those LEDs are handled by FW.
> 
> So, the device in this case is 'line card'.
> 
> In my previous reply I suggest name 'fru' stands for the filed replaceable unit.
> This is not something, that is exposed in '/dev', but it describes any replaceable
> unit within the system.

So.. you'd use the LED to locate right PCI card, or the LED would
indicate that whole card is failing, etc...?

Could we use pci00:1b.0 as the device name? (same as lspci). Probably
replace : with _...

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

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

  reply	other threads:[~2020-10-21  8:33 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-06 16:58 [PATCH led-next 1/1] leds: mlxreg: Allow multi-instantiation of same name LED for modular systems Vadim Pasternak
2020-10-06 23:15 ` Marek Behun
2020-10-07  6:07   ` Vadim Pasternak
2020-10-07 12:17     ` Marek Behun
2020-10-07 11:31 ` Pavel Machek
2020-10-07 12:20   ` Marek Behun
2020-10-08  6:16     ` Vadim Pasternak
2020-10-08  7:56       ` Pavel Machek
2020-10-08  8:47         ` Vadim Pasternak
2020-10-08  8:55           ` Marek Behun
2020-10-08  9:30             ` Vadim Pasternak
2020-10-08 10:05               ` Pavel Machek
2020-10-08 10:32                 ` Vadim Pasternak
2020-11-25 11:20                   ` Pavel Machek
2020-11-25 12:01                     ` Vadim Pasternak
2020-12-30 18:48                       ` Pavel Machek
2020-10-08 10:32               ` Marek Behun
2020-10-12 10:14                 ` Vadim Pasternak
2020-10-21  8:33                   ` Pavel Machek [this message]
2020-10-21 12:29                     ` Vadim Pasternak

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20201021083334.GA29124@duo.ucw.cz \
    --to=pavel@ucw.cz \
    --cc=jacek.anaszewski@gmail.com \
    --cc=linux-leds@vger.kernel.org \
    --cc=marek.behun@nic.cz \
    --cc=vadimp@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).