[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200115161917.GA24105@pi3.com.pl>
Date: Wed, 15 Jan 2020 17:19:17 +0100
From: Adam Zabrocki <pi3@....com.pl>
To: lkrg-users@...ts.openwall.com
Subject: Re: LKRG complains about battery being automatically
unload
Hi,
This might happen through the races which you've referred to. However, LKRG has
a logic to detect it and it does it here as well. You can see it in the logs:
"EXTRA MODULE IS THE SAME AS ON-GOING MODULE ACTIVITY EVENTS"
"!! MOST LIKELY SYSTEM IS STABLE !!"
This is just an information for the user that such situation happened. User can
decide to do more investigation if he likes. However it is not necessary.
This situation might also happened during normal run of the system after it has
been fully initialized. During the boot process it is more likely to see that
races since (as you also mentioned) execution is parallelized.
Btw. Do you think that these messages are ambiguous?
Thanks,
Adam
On Sat, Jan 11, 2020 at 09:04:01PM +0000, Patrick Schleizer wrote:
> This happened in a VirtualBox VM. Just saw this once.
>
>
> Jan 11 20:44:21 host kernel: battery: ACPI: Battery Slot [BAT0] (battery
> present)
> Jan 11 20:44:21 host kernel: [p_lkrg] ALERT !!! FOUND LESS[1] MODULES IN
> DB IN MODULE LIST[54] THAN IN CURRENT MODULE LIST[55]
> Jan 11 20:44:21 host kernel: [p_lkrg] EXTRA MODULE:
> name[battery] module at
> addr[00000000aae59f0b] module core[00000000d5bd1fc8] with size[0x2000]
> hash[0xcc42714bdd4f6be7]
> Jan 11 20:44:21 host kernel: [p_lkrg] ** EXTRA MODULE IS THE SAME AS
> ON-GOING MODULE ACTIVITY EVENTS **
> ** !! MOST LIKELY SYSTEM IS STABLE !! **
> Jan 11 20:44:21 host kernel: [p_lkrg] ALERT !!! FOUND LESS[1] MODULES IN
> DB IN KOBJ[54] THAN IN CURRENT KOBJ[55]
> Jan 11 20:44:21 host kernel: [p_lkrg] EXTRA MODULE:
> name[battery] module at
> addr[00000000aae59f0b] module core[00000000d5bd1fc8] with size[0x2000]
> hash[0xcc42714bdd4f6be7]
> Jan 11 20:44:21 host kernel: [p_lkrg] ** EXTRA MODULE IS THE SAME AS
> ON-GOING MODULE ACTIVITY EVENTS **
> ** !! MOST LIKELY SYSTEM IS STABLE !! **
>
>
> Not reproducible. Could be a race condition depending on when LKRG gets
> load.
>
> No idea why first the battery module was load and then automatically
> unload. I remember your LKRG speech (youtube video) where you mentioned
> "has to happen through kernel API". I guess if the battery module
> automatically unloads itself or something else uses `rmmod battery` then
> LKRG shouldn't even complain about this?
>
> Another task perhaps for /etc/modprobe.d/30_lkrg.conf configuration file?
>
> > softdep p_lkrg pre: battery
>
>
> Might be related / same as my earlier message:
>
> Possible to load LKRG through systemd-modules-load.service /
> /usr/lib/modules-load.d / /etc/modprobe.d as last module? -
> https://www.openwall.com/lists/lkrg-users/2020/01/11/1
>
> Does this mean it's better to load LKRG through a systemd unit file
> rather than systemd-modules-load.service?
>
> Kind regards,
> Patrick
--
pi3 (pi3ki31ny) - pi3 (at) itsec pl
http://pi3.com.pl
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
