PCENGINES APU[1-5] Coreboot SeaBIOS Open Source Firmware

[newsense]

Hi miczyg

Any chance I could remotely identify the chip running on this APU1C4 ? Dmidecode isn't helpful

Code: [Select]

flashrom -w apu1v4.9.0.2.rom -p internal
flashrom v1.0 on FreeBSD 11.2-RELEASE-p6 (amd64)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 4, resolution: 70ns).
coreboot table found at 0xdffdf000.
Found chipset "AMD SB7x0/SB8x0/SB9x0".
Enabling flash write... OK.
Found Macronix flash chip "MX25L1605" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605A/MX25L1606E/MX25L1608E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605D/MX25L1608D/MX25L1673E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Multiple flash chip definitions match the detected chip(s): "MX25L1605", "MX25L1605A/MX25L1606E/MX25L1608E", "MX25L1605D/MX25L1608D/MX25L1673E"
Please specify which chip definition to use with the -c <chipname> option.


[0xFelix]

According to the APU2 documentation on Github it should be the second definition.

See https://github.com/pcengines/apu2-documentation/blob/master/docs/firmware_flashing.md#corebootrom-flashing

[newsense]

Thank you 0xFelix, I didn't find that bit prior to posting. Everything worked without issues.

[Mks]

Hi all.

Just for my understanding.

Is there a possible to update the BIOS from an running OpnSense or is it required to boot the system with an USB-Stick and Tinycore?

Thanks

[0xFelix]

I just upgraded my APU2C4 from a running opnsense.

Just login to the console, install flashrom, download the firmware and flash it.

Code: [Select]

pkg install flashrom
curl LINKTOFIRMWARE --output coreboot.rom
flashrom -w coreboot.rom -p internal

On my APU flashrom complained that the board id did not match. You can append ":boardmismatch=force" to the command though. After flashing I did a power reset and everything works fine.

[Mks]

Great Thanks!

[miczyg]

Hello Dear OPNsense community,

Signatures and hashes for all previous firmware release are now available: https://pcengines.github.io/

Hi miczyg

Any chance I could remotely identify the chip running on this APU1C4 ? Dmidecode isn't helpful

Code: [Select]

flashrom -w apu1v4.9.0.2.rom -p internal
flashrom v1.0 on FreeBSD 11.2-RELEASE-p6 (amd64)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 4, resolution: 70ns).
coreboot table found at 0xdffdf000.
Found chipset "AMD SB7x0/SB8x0/SB9x0".
Enabling flash write... OK.
Found Macronix flash chip "MX25L1605" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605A/MX25L1606E/MX25L1608E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605D/MX25L1608D/MX25L1673E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Multiple flash chip definitions match the detected chip(s): "MX25L1605", "MX25L1605A/MX25L1606E/MX25L1608E", "MX25L1605D/MX25L1608D/MX25L1673E"
Please specify which chip definition to use with the -c <chipname> option.


Hi newsense,

Yes this is a common problem with apu1, since few Macronix SPI chips have the same JEDEC ID and flashrom can not distinguish them. Try passing:

-c "MX25L1605A/MX25L1606E/MX25L1608E"

to flashrom as an argument and it will be ok. We have mentioned it here too: https://github.com/pcengines/apu2-documentation/blob/master/docs/firmware_flashing.md#corebootrom-flashing

[newsense]

Hello Dear OPNsense community,

Signatures and hashes for all previous firmware release are now available: https://pcengines.github.io/

Hi miczyg

Any chance I could remotely identify the chip running on this APU1C4 ? Dmidecode isn't helpful

Code: [Select]

flashrom -w apu1v4.9.0.2.rom -p internal
flashrom v1.0 on FreeBSD 11.2-RELEASE-p6 (amd64)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 4, resolution: 70ns).
coreboot table found at 0xdffdf000.
Found chipset "AMD SB7x0/SB8x0/SB9x0".
Enabling flash write... OK.
Found Macronix flash chip "MX25L1605" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605A/MX25L1606E/MX25L1608E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Found Macronix flash chip "MX25L1605D/MX25L1608D/MX25L1673E" (2048 kB, SPI) mapped at physical address 0x00000000ffe00000.
Multiple flash chip definitions match the detected chip(s): "MX25L1605", "MX25L1605A/MX25L1606E/MX25L1608E", "MX25L1605D/MX25L1608D/MX25L1673E"
Please specify which chip definition to use with the -c <chipname> option.


Hi newsense,

Yes this is a common problem with apu1, since few Macronix SPI chips have the same JEDEC ID and flashrom can not distinguish them. Try passing:

-c "MX25L1605A/MX25L1606E/MX25L1608E"

to flashrom as an argument and it will be ok. We have mentioned it here too: https://github.com/pcengines/apu2-documentation/blob/master/docs/firmware_flashing.md#corebootrom-flashing

Thank you miczyg, I already updated two APU1s without issues from their factory version, both using the OPNsense live method I described a few post back and from a fully deployed production running FW - following the steps in the Live method described followed by the reboot command when the upgrade was succesful.

[lattera]

I'm getting the attached error when attempting to install OPNsense 19.1 on an APU3 running BIOS firmware v4.9.0.2.

[lattera]

To get around this, I installed 18.7, then followed the upgrade procedure to get me to 19.1.2.

[newsense]

I'm getting the attached error when attempting to install OPNsense 19.1 on an APU3 running BIOS firmware v4.9.0.2.

That screams more like an error reading the USB stick to me, I wouldn't have pinned it on the BIOS. I've booted successfully APUs 3 and 4 with the live 19.1 on the factory BIOS and performed the upgrade to v4.9.0.2 successfully followed by reboot and install from the same (rather lousy) USB stick. I've always used the method described a few posts back.

Was there any other indication it might be BIOS related in your case ?

[tillsense]

Hi all,

for interested 4.9.0.3 is available.

cheers
till

[newsense]

Successfully upgraded APU1,3,4 to v4.9.0.3

[miczyg]

Hello all,

We have validated installation of OPNSense 19.1 on most commonly used storage media. Enjoy the OS status update:

https://pcengines.github.io/

Cheers,
Michał

[tillsense]

Hi all,

for interested 4.9.0.4 is available.

cheers
till