Linux beaglebone 4.0.0-bone0 - unexpected IRQ trap at vector 00

[Artur Festyn]

Hi all.

I am experiencing "unexpected IRQ trap at vector 00" problem on my BBB.

This suppose to be fixed in 3.14 version of kernel as per:

https://www.kernel.org/pub/linux/kernel/v3.0/ChangeLog-3.14.30

cpsw.c

Can anybody tell me if this patch has been implemented in 4.0.0-bone0?

Kind Regards

Artur

[Robert Nelson]

Did you look at the git history?

btw.. the stable rules state.. before a patch can hit stable, it has
to be applied to mainline..

Regards,

--
Robert Nelson
https://rcn-ee.com/

[Peter Hurley]

On 04/30/2015 11:00 AM, Artur Festyn wrote:
> Hi all.
> I am experiencing "unexpected IRQ trap at vector 00" problem on my BBB.

I've had this happen maybe a handful of times in 3 mos., so that's made
it difficult to track down.

How repeatable is this for you?

Regards,
Peter Hurley

[Jan Kinkazu]

Thank you very much for your answer Robert.

No I haven't looked anywhere, I am total newbie. 

All i know is how to compile a kernel using your build_deb.sh and install it.

I have no idea where is a git history.

I don't know how to check if this patch has been applied to mainline.

I can't even locate this patch anywhere to download :) 

I tried manually applied it just using vi, but it doesn't match cpsw.c downloaded by your scripts.

If it is not applied to 4.0.0-bone0 then I will wait and learn a little bit in a meantime.

Kind Regards

[Robert Nelson]

On Thu, Apr 30, 2015 at 12:21 PM, Jan Kinkazu <fes...@googlemail.com> wrote:
> Thank you very much for your answer Robert.
> No I haven't looked anywhere, I am total newbie.
> All i know is how to compile a kernel using your build_deb.sh and install
> it.
> I have no idea where is a git history.
> I don't know how to check if this patch has been applied to mainline.
> I can't even locate this patch anywhere to download :)
> I tried manually applied it just using vi, but it doesn't match cpsw.c
> downloaded by your scripts.
>
> If it is not applied to 4.0.0-bone0 then I will wait and learn a little bit
> in a meantime.

I checked earlier, it's part of v4.0.x..

What issue are you seeing?

[Jan Kinkazu]

I observed exactly the same behavior in kernel 3.8.13.

The only difference is, I was able to capture more using debug console.

here is the dump: http://pastebin.com/b73vnbmX

In 4.0.0-bone0 there is only: "unexpected IRQ trap at vector 00"

[Jan Kinkazu]

Great, thank you.

This means it is related to something else.

[Jan Kinkazu]

Hi Peter.

I replicate this as many times as I want.

I have got can0 interface and SIM908 connected to ttyO2 (GPS) and ttyO4 (GSM).

To achieve this all I need to do is:

Option 1:

run candump can0 in one session of putty

cat /dev/ttyO2 (gives GPS position) on second putty session 

Wait a little bit, sometimes minute, sometimes 10 or more

Option 2:

run candump can0 in one session of putty

start download of 20-50Mb file using modem on ttyO4 and wget.

Wait

I would say it is CANBUS related, but the result on debug console matches the patch pointed above:

"unexpected IRQ trap at vector 00"

Kind Regards

[Jan Kinkazu]

I found one more patch with same problem:

davinci_emac

https://kernel.googlesource.com/pub/scm/linux/kernel/git/horms/ipvs/+/cd2d6d33e2c5be653d10cdc8fcd7dcf0be28de50%5E%21/#F0

Is this one also applied to 4.0.0-bone0?

Sorry for all those questions, I am like a child in the fog.