linux-kernel - Re: linux-next: build warning after merge of the leds-lj tree

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

Message-ID: <87wmreee6j.fsf@meer.lwn.net>
Date: Thu, 08 Feb 2024 16:20:20 -0700
From: Jonathan Corbet <corbet@....net>
To: Stephen Rothwell <sfr@...b.auug.org.au>, Florian Eckert <fe@....tdt.de>
Cc: Lee Jones <lee@...nel.org>, Linux
 Kernel Mailing List <linux-kernel@...r.kernel.org>, Linux Next Mailing List
 <linux-next@...r.kernel.org>
Subject: Re: linux-next: build warning after merge of the leds-lj tree

Stephen Rothwell <sfr@...b.auug.org.au> writes:

> Hi Florian,
>
> On Mon, 08 Jan 2024 08:47:07 +0100 Florian Eckert <fe@....tdt.de> wrote:
>>
>> Hello Stephen,
>> 
>> thanks for your hint
>> 
>> On 2024-01-05 07:33, Stephen Rothwell wrote:
>> > Hi all,
>> > 
>> > After merging the leds-lj tree, today's linux-next build (htmldocs)
>> > produced this warning:
>> > 
>> > Warning: /sys/class/leds/<led>/rx is defined 2 times:
>> > Documentation/ABI/testing/sysfs-class-led-trigger-tty:7
>> > Documentation/ABI/testing/sysfs-class-led-trigger-netdev:49
>> > Warning: /sys/class/leds/<led>/tx is defined 2 times:
>> > Documentation/ABI/testing/sysfs-class-led-trigger-tty:15
>> > Documentation/ABI/testing/sysfs-class-led-trigger-netdev:34  
>> 
>> The behavior of the tty trigger can be controlled via the Rx and Tx file.
>> If a value is set in Rx or Tx, the LED flashes when data is transmitted in
>> this direction. The same behavior is used for the netdev trigger.
>> I have therefore used the same pattern for the new tty trigger as well.
>> 
>> I didn't know that the names have to be unique!
>> 
>> I'm a bit at a loss as to what to do now. Should I put a prefix "tty_"
>> in front of the names so that we have "tty_rx", "tty_tx"?
>> 
>> If we do it this way, however, the general question arises as to whether
>> we do have to use a prefix everywhere! If new triggers are added, then the
>> names for a config file are already used up and anyone who then wants to use
>> the same name for an other trigger with the same config file because it describe
>> the same function must then work with a prefix!
>
> I think this is only a problem with the documentation system, not the
> actual sysfs file naming.  Maybe just adding a uniquifying bit to the
> "<led>" part will solve it.  Or maybe we need the tooling to be taught
> about placeholders in sysfs names (or maybe there is already a way).

So I finally remembered to look at this when I had a chance to...  yes,
it wants each ABI entry to be unique, and the ones listed here are not.
I *think* the easiest answer is to take a line like:

  What:		/sys/class/leds/<led>/rx

and turn it into something like:

  What:		/sys/class/leds/<netdev-led>/rx

..that makes the warning go away and, I think, conveys the information
just as well.  A bit kludgy, perhaps, but I don't really see anything
else there that could be used to disambiguate the names automatically.

Thanks,

jon