PC Engines apu2 experiences

DaddyGo

@valnar said in PC Engines apu2 experiences:

when you update your apu2 from 2.4.5 --> 2.5.0.

Hi,

With that, I would wait a little longer

here,....you will find a lot of useful info if you even decide to upgrade...
https://teklager.se/en/knowledge-base/
https://teklager.se/en/knowledge-base/apu2-1-gigabit-throughput-pfsense/

+++edit:
and of course
https://pcengines.github.io/

Qinn

Smooth update to pfSense 2.5.1 CE

Qinn

Updated to bios coreboot v4.13.05.

FLOK

@qinn Fine !
I am still on 2.4.5_1.
Did you notice any performance degradation ?

Qinn

@flok No, so far not.

kevindd992002

I received an email notification about v4.14.0.2 and this IMPORTANT message was included:

To update the firmware and keep the runtime configuration unchanged please use the following command:
flashrom -p internal -w apuX_v4.14.0.2.rom --fmap -i COREBOOT
The persistent runtime configuration works only when migrating from versions v4.14.0.1 and later. The feature is not yet supported on apu1. Flashrom version needs to be v1.1 or newer.

I did the upgrade from v4.14.0.1 with that new command without any issues. However, for the longest time this is what I've been using:

flashrom -p internal -w apuX_v4.14.0.2.rom

What is the difference?

bigsy

@kevindd992002
I'm not by any means an expert on coreboot, but

flashrom -p internal -w apuX_v4.14.0.2.rom

will overwrite the entire flash chip with the new rom file.

flashrom -p internal -w apuX_v4.14.0.2.rom --fmap -i COREBOOT

The flash chip has various partitions defined in a 'flashmap' format and this command specifies which area to overwrite. This probably allows for user altered settings such as seabios boot order changes to persist and not get overwritten during the coreboot update, but I haven't tried it out.

kevindd992002

@bigsy said in PC Engines apu2 experiences:

@kevindd992002
I'm not by any means an expert on coreboot, but
flashrom -p internal -w apuX_v4.14.0.2.rom
will overwrite the entire flash chip with the new rom file.
flashrom -p internal -w apuX_v4.14.0.2.rom --fmap -i COREBOOT
The flash chip has various partitions defined in a 'flashmap' format and this command specifies which area to overwrite. This probably allows for user altered settings such as seabios boot order changes to persist and not get overwritten during the coreboot update, but I haven't tried it out.

That's what I thought. However, with the 1st command I have never seen any seabios setting get changed when I use it to update my APU2. I'm wondering if it's really better to use the 2nd command from this point moving forward.

bigsy

@kevindd992002 If you look at the changes to the pcEngines fork of SeaBIOS, the addition of 'support for persistent bootorder in FMAP region' has only been in the most recent release of SeaBIOS v1.14.0.1 and similarly for sortbootorder v4.6.21.

kevindd992002

@bigsy I see. So we're just preserving boot order here? Since I just have one boot device in my APU2, then it wouldn't matter to me.

kevindd992002

For the APU2C4, why is it that when I have a serial cable connected between its serial port and a computer, its USB ports don't work? I have a UPS plugged in to one of its USB ports and I use the NUT package in pfsense. As soon as I connect to the hardware through the serial port (using putty), the USB port does not work. Unplugging the serial cable fixes the issue and that's why I know it's the one causing it. This is happening for both of my APU2C4 boxes.

DaddyGo

@kevindd992002 said in PC Engines apu2 experiences:

connected between its serial port and a computer, its USB ports don't work?

Hmmmm, I haven't seen it yet, it is true we don't even connect to MOBO in this multiple way...

Have you tried to remove NUT? (of course temporarily)

kevindd992002

@daddygo said in PC Engines apu2 experiences:

@kevindd992002 said in PC Engines apu2 experiences:

connected between its serial port and a computer, its USB ports don't work?

Hmmmm, I haven't seen it yet, it is true we don't even connect to MOBO in this multiple way...

Have you tried to remove NUT? (of course temporarily)

I haven't yet but I see the usb disconnect message in the terminal itself so I figured it's a board issue. I also use the USB port only for NUT so removing it will be a moot point but I can still try for the sake of troubleshooting.

DaddyGo

@kevindd992002 said in PC Engines apu2 experiences:

but I can still try for the sake of troubleshooting.

Sure

Would it be worth doing to rule out the SW/Code issue (on NUT) in the background...

The NUT is a strange animal anyway and sometimes difficult to get to work..

It also handles USB and RS232 stuff, so I guess that's where the rub is

kevindd992002

@daddygo said in PC Engines apu2 experiences:

@kevindd992002 said in PC Engines apu2 experiences:

but I can still try for the sake of troubleshooting.

Sure

Would it be worth doing to rule out the SW/Code issue (on NUT) in the background...

The NUT is a strange animal anyway and sometimes difficult to get to work..

It also handles USB and RS232 stuff, so I guess that's where the rub is

I already uninstalled NUT and I soon as I connected to the serial port, I still see the issue:

ugen0.2: <Unknown > at usbus0 (disconnected)

So this is definitely an APU2C4 issue. Can anyone try this?

fireodo

@kevindd992002 said in PC Engines apu2 experiences:

So this is definitely an APU2C4 issue. Can anyone try this?

I dont have a APU2C4 (Only APU2C0) but I would suggest to try an older BIOS version.

kevindd992002

@fireodo said in PC Engines apu2 experiences:

@kevindd992002 said in PC Engines apu2 experiences:

So this is definitely an APU2C4 issue. Can anyone try this?

I dont have a APU2C4 (Only APU2C0) but I would suggest to try an older BIOS version.

I'm at v4.14.0.2 but this is not the first time I've tried this. I've tried this a year ago when it was still using an older BIOS version and it's the same issue.

DaddyGo

@kevindd992002 said in PC Engines apu2 experiences:

4.14.0.2 but this is not the first time I've tried this

You say, this is the case on several boxes (MOBO)
(so no HW issue I think, some PCB schematics or BIOS)

Worth a question for him:
https://github.com/miczyg1

BTW:
I can't check it either, because we only use apu24d4

kevindd992002

@daddygo said in PC Engines apu2 experiences:

@kevindd992002 said in PC Engines apu2 experiences:

4.14.0.2 but this is not the first time I've tried this

You say, this is the case on several boxes (MOBO)
(so no HW issue I think, some PCB schematics or BIOS)

Worth a question for him:
https://github.com/miczyg1

BTW:
I can't check it either, because we only use apu24d4

Just created an issue:

https://github.com/pcengines/pcengines.github.io/issues/107

VAMike

@kevindd992002 using the serial port is pretty normal, and what you're describing is not at all common. I'd probably first make sure of the power supply (marginal power supplies cause all sorts of issues), then look at whether the serial cable is dodgy/shorted/etc.