linux-kernel - Re: [PATCH 1/2] kobject: return error code if writing /sys/.../uevent fails

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: <YIrsn98AFX3y6SDT@kroah.com>
Date:   Thu, 29 Apr 2021 19:27:59 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Phillip Susi <phill@...susis.net>
Cc:     rafael@...nel.org, jeyu@...nel.org, linux-kernel@...r.kernel.org,
        msekleta@...hat.com, Peter Rajnoha <prajnoha@...hat.com>
Subject: Re: [PATCH 1/2] kobject: return error code if writing
 /sys/.../uevent fails

On Thu, Apr 29, 2021 at 09:40:08AM -0400, Phillip Susi wrote:
> Just an FYI, I've been tracking down a bug that was causing the debian
> bullseye installer to immediately crash and reboot the xen domU and
> bisected it to this commit.  It appears that the Xen Virtual Keyboard
> driver ( and at least one other, probably more ) were always broken and
> failed to trigger the udev event, but this was never noticed.  When this
> patch returned the error code, it caused the d-i init scripts triggering
> coldplug events to fail, which caused init to bail out, causing a kernel
> panic.
> 
> In the future, when fixing error returns like this, could you please add
> a big fat printk so that hopefully things that always should have failed
> but didn't and now do can get noticed and fixed more quickly.
> 
> https://bugzilla.kernel.org/show_bug.cgi?id=207695
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=983357
> 

Spaming kernel logs with things that userspace can trigger is generally
a bad idea.  But I understand the pain here, sorry to hear it.

Was the offending drivers ever fixed?

thanks,

greg k-h