(Solved) Latest Debian Stable won't install on Asus Prime B450M-A2

[yankeeFarmer]

Hello....I don't know where to start here. I'm not new to Debian, been using it exclusively since Sarge in 2005, and installed it to literally dozens of desktops and laptops over the years, but this thing is kicking my butt.
Specs:
Asus Prime B450M-A2 mainboard
AMD Ryzen 5 4500 CPU
Radeon HD 5450 video card.
500Gig SSD and 1TB hardrive.

I can install Windows 10 without issue, but putting the Debian USB stick in, it boots into the initial screen where you pick the install method just fine, then, whatever is selected, it immediately starts throwing errors....screen after screen after screen of errors. It can't find the hard drives, it can't see, it can't find anything. After about 10 screens of complaints about everything from the mileage of my truck to the price of eggs in China, it finally shows the first install screen.....except it's frozen. I tried searching the forum for instances of this mainboard, but got nothing. A broader search also turned up nothing pertinent. Maybe I have to select some parameter during the install?

Things I tried:
Updated BIOS
Formatted USB drive as both MBR and the other one, which I don't recall ATM.
Turned off the trusted computing function in the BIOS
Cursed.
Wept.
Finally gave up and asked for help.

[Pinhead]

Try Trixie. It has a newer kernel.

[kalle123]

I can install Windows 10 without issue, but putting the Debian USB stick in, it boots into the initial screen where you pick the install method just fine, then, whatever is selected, it immediately starts throwing errors ....

Where are the error messages? Please show ......

[yankeeFarmer]

Try Trixie. It has a newer kernel.

Thank you. I will try that.

[yankeeFarmer]

I can install Windows 10 without issue, but putting the Debian USB stick in, it boots into the initial screen where you pick the install method just fine, then, whatever is selected, it immediately starts throwing errors ....

Where are the error messages? Please show ......

I will try to catch some with the camera, but right now, it's 2:18 AM and my attention is flagging....LOL

[kalle123]

Your AMD Ryzen 5 4500 CPU needs a BIOS Version 3211 or higher, but that was out 2021/08/26, so this should be OK.

[Hetzer]

Going with 13 will change nothing, I believe. My B550M (so newer by one generation) had no problems with Debian 12
Is the install media good? I've also stumbled with such behavior of installers and always happened to be it
I'd try to burn / flash (whenever ye use opticals or flash drives) installer on different media. Or make sure, in case of flash drives, that image is really fully written (Don't yank it out when it seems to be "ready", unmount it first and then pull it out)
And well, maybe it's worth checking if image file itself is okay?

[kalle123]

Have one of those here

Host: valley-forge Kernel: 6.1.0-11-amd64 arch: x86_64 bits: 64
Desktop: Xfce v: 4.18.1 Distro: Debian GNU/Linux 12 (bookworm)
Machine:
Type: Desktop Mobo: ASUSTeK model: PRIME B450-PLUS v: Rev X.0x
serial: <superuser required> UEFI: American Megatrends v: 4202
date: 06/16/2023
CPU:
Info: quad core model: AMD Ryzen 3 1200 bits: 64 type: MCP cache: L2: 2 MiB
Speed (MHz): avg: 1940 min/max: 1550/3100 cores: 1: 1380 2: 1380 3: 1550
4: 3450

No issues ...

[yankeeFarmer]

Ok, so I tried Trixie...no change.
Tried different USB drives....no change.
Burned iso to DVD and booted from that....no change.

Below is the screens of errors that show in order. After about 2 minutes or so it flashes the message "starting frame buffer", then it goes to the first setup screen where you select the language. About 75% of the time it freezes there. Sometimes, I can get a couple of screens in before it either freezes or gives some other error. Windows 10 still installs without issue.
The issue(s) appear to be kernel related as far as I can tell, but I haven't found any solutions. Swapping hardrives also has no effect. This is a real head-scratcher.

[yankeeFarmer]

It seems to definitely be a kernel issue, as I was able to install Debian 11 with no problem right up until disk partitioning, at which point it couldn't detect any disks. I guess I'm just SOL until the kernel catches up to my hardware.
I feel like I'm back in 2005 installing Etch (Testing) because Sarge didn't recognize most of my hardware.

[Ardouos]

What BIOS version are you running?
Have you tried factory resetting your BIOS/EFI?

Does the PC come with any peripherals? I would strip the pc to its core and see if the same error occurs. Even taking out DIMMs if you can.

[friendlysalmon88]

To avoid these kind of issues I keep my Work Station on cse/lagacy mode instead of UFEI as it allows me to have virtually painless installation and upgrade paths haven't had any where near the issues the the OP has mentioned. On the other hand I've also got a newer ThinkPad X1 Carbon on which I'm stuck using UFEI so I just account for a 512 MB esr/uefi partition at the beginning of my partitioning scheeme.

[yankeeFarmer]

What BIOS version are you running?
Have you tried factory resetting your BIOS/EFI?

Does the PC come with any peripherals? I would strip the pc to its core and see if the same error occurs. Even taking out DIMMs if you can.

I don't know the BIOS version, and I can't look it up right now, but it was the latest one for that board on the ASUS website. I have not tried resetting the BIOS...
I bought the board and cpu as a bundle. It's a home-built PC, with no peripherals hooked up yet.

[yankeeFarmer]

To avoid these kind of issues I keep my Work Station on cse/lagacy mode instead of UFEI as it allows me to have virtually painless installation and upgrade paths haven't had any where near the issues the the OP has mentioned. On the other hand I've also got a newer ThinkPad X1 Carbon on which I'm stuck using UFEI so I just account for a 512 MB esr/uefi partition at the beginning of my partitioning scheeme.

I've tried it with and without UEFI. No difference.

[FreewheelinFrank]

A web search suggests the error

Code: Select all

AMD-Vi: Event logged [IO_PAGE_FAULT

is related to IOMMU.

https://bbs.archlinux.org/viewtopic.php?id=275425

https://forums.linuxmint.com/viewtopic. ... 34#p792634

You could try changing IOMMU options in the BIOS.

As the errors mention snd_hda_intel, the ultimate cause could be an issue with that driver and the motherboard and kernel.

[NorthEast]

Couple of ideas: 1. change the sata cables; 2. try kernel option: libata.force=3.0Gbps
These were successful for some users online in the case of the error: "failed command: READ FPDMA QUEUED"
No guarantees, just reversible suggestions.

[yankeeFarmer]

A web search suggests the error

Code: Select all

AMD-Vi: Event logged [IO_PAGE_FAULT

is related to IOMMU.

https://bbs.archlinux.org/viewtopic.php?id=275425

https://forums.linuxmint.com/viewtopic. ... 34#p792634

You could try changing IOMMU options in the BIOS.

As the errors mention snd_hda_intel, the ultimate cause could be an issue with that driver and the motherboard and kernel.

Thank you! That was the problem. Disabling IOMMU in the BIOS eliminated all the errors except the first one, which previous research indicated didn't matter. It works perfectly now.

[FreewheelinFrank]

Thank you! That was the problem. Disabling IOMMU in the BIOS eliminated all the errors except the first one, which previous research indicated didn't matter. It works perfectly now.

You're welcome.

It might be worth checking if there is a bug report for this already and if not filing one.

https://wiki.debian.org/DebianKernelReportingBugs