linux-kernel - Re: fault injection: query if a fault has been injected?

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] [day] [month] [year] [list]

Message-ID: <CAC5umyiQZdAJY0GPYG18N9=g8v0kZVKE9KG-utWS-xrBhseG4Q@mail.gmail.com>
Date: Sun, 30 Nov 2025 21:07:21 +0900
From: Akinobu Mita <akinobu.mita@...il.com>
To: Jani Nikula <jani.nikula@...el.com>
Cc: linux-kernel@...r.kernel.org, 
	Juha-Pekka Heikkila <juhapekka.heikkila@...il.com>
Subject: Re: fault injection: query if a fault has been injected?

2025年11月27日(木) 21:49 Jani Nikula <jani.nikula@...el.com>:
>
>
> Hi Akinobu -
>
> The i915 driver has its own fault injection mechanism that we're in the
> process of migrating to the kernel generic fault injection.
>
> One feature we have is being able to check if a fault has been injected
> in the driver. We could then tone down certain error/warning messages to
> debug messages. The point being, CI could more trivially ignore the
> expected failures, while still failing on errors and warnings that are
> unrelated to the injected fault.
>
> Is there anything like that in the generic fault injection mechanism? Or
> do we have to figure all of this out in user space?

I took a quick look at the i915 driver's fault injection.

The fail-nth function may be useful for implementing inject_probe_failure.
The section "Systematic faults using fail-nth" in
Documentation/fault-injection/fault-injection.rst explains this.