From: Jan Kiszka <jan.kiszka@siemens.com>
To: Per Oberg <pero@wolfram.com>, xenomai@lists.linux.dev
Subject: Re: RTNet unable to handle kernel NULL pointer dereference at 0000000000000068
Date: Thu, 5 Oct 2023 16:21:27 +0200 [thread overview]
Message-ID: <773177ed-357d-48d5-af35-645ad192b209@siemens.com> (raw)
In-Reply-To: <1711728904.3619924.1696515557888.JavaMail.zimbra@wolfram.com>
On 05.10.23 16:19, Per Oberg wrote:
>
> ----- Den 4 okt 2023, på kl 16:15, Per Öberg pero@wolfram.com skrev:
>
>> ----- Den 4 okt 2023, på kl 15:57, Jan Kiszka jan.kiszka@siemens.com skrev:
>
>>> On 04.10.23 15:01, Per Oberg wrote:
>>>> I just had another hickup. This is 100 pct reproducible
>
>>>> To get it up and running I had to remove rtpacket,rttcp,rtudp,rtipv4 and then
>>>> forcibly remove rt_igb driver and then rtnet. This gave me another set of
>>>> errors, also listed below.
>
>>> What does "forcibly" mean here? Actually "rmmod --force"? Then you get
>>> what you deserve ;).
>
>> Yes, intentionally. I expected that the "rmmod --foce" would reek havok but I
>> thought that perhaphs the actual havok caused could give more pointers to the
>> actual problem.
>> I wanted to see if I could recover from the first crash (not the one from my
>> first email) without rebooting.
>
>>>> [15018.436248] ------------[ cut here ]------------
>>>> [15018.436249] [Xenomai] switching rtnet-stack to secondary mode after exception
>>>> #6 in kernel-space at 0xffffffffb837064b (pid 1438)
>>>> [15018.436258] WARNING: CPU: 0 PID: 1438 at
>>>> /usr/src/kernel/kernel/xenomai/rtdm/fd.c:299 __put_fd+0x26b/0x2c0
>
>>> You are not using the latest release, not even of your stable series.
>>> This is needlessly risky.
>
>> I am currently updating the kernel and xenomai libraries to see what that can
>> do. I could not, however, find any patches that looked relevant for the issue.
>
>>> There is some imbalance in the reference counter for the socket, and
>>> later on are complaints about leaking buffers in the pool. The real
>>> problem may rather be that you didn't shut down the interface properly -
>>> or that we don't do that when pulling the plugs in the order you describe.
>
>> To be clear, the
>> "[Xenomai] switching rtnet-stack to secondary mode after exception #6 in
>> kernel-space at 0xffffffffb837064b (pid 1438)"
>> happens after a fresh reboot, not something I kept alive with artifial life
>> support.
>
>>> Can you reproduce this rather generic issue over latest master as well?
>
>> I'm working on it. Hopefully not.
>
>>> Jan
>
>>> --
>>> Siemens AG, Technology
>>> Linux Expert Center
>
>> Per Öberg
>
> Now tested on Linux 5.10.191-dovetail and Xenomai 3.2.4.
> Cannot reproduce the crash anymore
>
Ok, that's good news. If you happen to identify missing patches in 3.1,
we can probably pick them for that version as well.
Jan
--
Siemens AG, Technology
Linux Expert Center
prev parent reply other threads:[~2023-10-05 14:21 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-04 8:45 RTNet unable to handle kernel NULL pointer dereference at 0000000000000068 Per Oberg
2023-10-04 9:42 ` Jan Kiszka
2023-10-04 13:01 ` Per Oberg
2023-10-04 13:57 ` Jan Kiszka
2023-10-04 14:15 ` Per Oberg
2023-10-05 14:19 ` Per Oberg
2023-10-05 14:21 ` Jan Kiszka [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=773177ed-357d-48d5-af35-645ad192b209@siemens.com \
--to=jan.kiszka@siemens.com \
--cc=pero@wolfram.com \
--cc=xenomai@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).