linux-kernel - Re: [PATCH] soundwire: debugfs: use controller id instead of link

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

Message-ID: <6d4d4a6b-f28c-81db-4e67-2b5b94116fa4@linux.intel.com>
Date:   Tue, 19 Jan 2021 13:09:38 -0600
From:   Pierre-Louis Bossart <pierre-louis.bossart@...ux.intel.com>
To:     Srinivas Kandagatla <srinivas.kandagatla@...aro.org>,
        Vinod Koul <vkoul@...nel.org>
Cc:     alsa-devel@...a-project.org, gregkh@...uxfoundation.org,
        linux-kernel@...r.kernel.org, sanyog.r.kale@...el.com,
        yung-chuan.liao@...ux.intel.com
Subject: Re: [PATCH] soundwire: debugfs: use controller id instead of link_id


>>>> link_id can be zero and if we have multiple controller instances
>>>> in a system like Qualcomm debugfs will end-up with duplicate namespace
>>>> resulting in incorrect debugfs entries.
>>>>
>>>> Using id should give a unique debugfs directory entry and should fix 
>>>> below
>>>> warning too.
>>>> "debugfs: Directory 'master-0' with parent 'soundwire' already 
>>>> present!"
>>>
>>> Yeah id is guaranteed to be unique so this will work.
>>>
>>> Applied, thanks
>>
>> This patch is a no-op for Intel, but I am not convinced it's the right 
>> fix or the definitions are not consistent.
> 
> It depends if the intention is to represent full Hierarchy in debugfs, 
> then I agree. Its was consistent even before!

Indeed, we don't currently have a notion of controller in debugfs.

> currently we have
> /sys/kernel/debug/soundwire/master-*
> 
> Are you suggesting that we have something like this:
> 
> /sys/kernel/debug/soundwire/xyz-controller/master-<LINK-ID> ??

Yes this is what I was thinking about.

> /sys/kernel/debug/soundwire/xyz-controller/master-<LINK-ID>/xyz-slave/master-<LINK-ID> 
> ??

This would be for a bridge which to the best of my knowledge hasn't been 
implemented by anyone (clocking and command/control timing issues).

> Or may be something much simpler like:
> 
> /sys/kernel/debug/soundwire/master-<ID>.<LINK_ID>

the bus->id is an IDA, which is useful for to avoid conflicts, but it's 
not really meaningful for debugfs. I remember seeing a case where we had 
links 2 and 4 enabled, and the bus->id were 0 and 1, a completely 
artificial value that doesn't really help in debugging.