netdev - Re: [PATCH net-next 01/15] devlink: add value check to devlink_info_version

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: <20a047ba-6b99-22d9-93e0-de7b4ed60b34@gmail.com>
Date: Fri, 11 Apr 2025 12:11:42 +0100
From: Edward Cree <ecree.xilinx@...il.com>
To: "Nelson, Shannon" <shannon.nelson@....com>,
 Jakub Kicinski <kuba@...nel.org>,
 "Jagielski, Jedrzej" <jedrzej.jagielski@...el.com>
Cc: "Nguyen, Anthony L" <anthony.l.nguyen@...el.com>,
 "davem@...emloft.net" <davem@...emloft.net>,
 "pabeni@...hat.com" <pabeni@...hat.com>, "Dumazet, Eric"
 <edumazet@...gle.com>, "andrew+netdev@...n.ch" <andrew+netdev@...n.ch>,
 "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
 "Kitszel, Przemyslaw" <przemyslaw.kitszel@...el.com>,
 "jiri@...nulli.us" <jiri@...nulli.us>, "horms@...nel.org"
 <horms@...nel.org>, "corbet@....net" <corbet@....net>,
 "linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
 Kalesh AP <kalesh-anakkur.purayil@...adcom.com>,
 "R, Bharath" <bharath.r@...el.com>
Subject: Re: [PATCH net-next 01/15] devlink: add value check to
 devlink_info_version_put()

On 09/04/2025 18:25, Nelson, Shannon wrote:
> On 4/9/2025 7:39 AM, Jakub Kicinski wrote:
>>
>> On Wed, 9 Apr 2025 14:14:23 +0000 Jagielski, Jedrzej wrote:
>>> No insisting on that but should empty entry be really presented to the user?
>>> Especially unintentionally? Actually it's exposing some driver's shortcomings.
>>> That means the output was not properly validated so imho there's no point in
>>> printing it.
>>
>> +1, FWIW, I don't see the point of outputting keys without values.
> 
> Because I like to see hints that something might be wrong, rather than hiding them.

+1 to this.  Failures should be noisy.  Time you care most about these
 data is when something *is* wrong and you're trying to debug it.
AFAICT the argument on the other side is "it makes the driver look bad",
 which has (expletive)-all to do with engineering.
Value often comes from firmware, anyway, in which case driver's (& core's)
 job is to be a dumb pipe, not go around 'validating' things.