Damnation

@bob_smith "Unknown Device" - It could be that this device is not your graphics card, is there anything listed under Display Adapters?

@Mov AX, 0xDEAD @Dietmar I was doing an experiment merging ntoskrn8.sys with the main kernel, since I wanted to add in extra Zw* functions that make use of internal KiSystemService function. I don't think this approach worked though.

@bob_smith Yeah, that would have something to do with the issue. Not sure about this one though, try disabling IRQ steering?

@d0minat0r 7B BSOD is a storage driver issue. So you're missing either an AHCI or NVME driver (IDE is included in XP so that should be fine) I noticed you're using x64 XP so you'll need the x64 versions of the drivers, x86 wont work in this case.

@bob_smith It is admittedly not as easy to get a DeviceID in 98SE as it is in XP. It's easiest just get the DeviceID when booted into XP, it won't have changed between the 2 OS's.

@d0minat0r Newer hardware doesn't use IDE for storage anymore, you'll need to use an XP AHCI or NVMe driver.

@SweetLow Do you think you might be able to figure out why Windows 98's ESDI_506.PDR driver fails to load when booting off of the SD card interface on rasteri's WeeCee and Eivind's Tinyllama 1 and 2. They both use this "RDC SD-IDE Host Controller" device that works fine in 2000 and XP, but 98 doesn't like it for some as yet undiscovered reason. Do you think you might be interested in solving this?

Whats the Device ID of your Xpress 200 graphics chip?

@sparty411 Try with just ACPI, AHCI and USB3 patches to get past 7B BSOD.

@TheFighterJetDude You should install the driver for that ? Universal Serial Bus (USB) controller - see if that fixes things.

@K4sum1 So yet another AMLILoadDDB() error, I thought we'd caught all of these. If you want to find the exact cause of this so it can be patched you'll need to setup a windbg session to trace the BSOD.

@SweetLow Where can I find these files? Can you share?

@Dietmar Sorry for not getting back to you sooner. Unfortunately, your delay acpi.sys did not show any change in behaviour for the USB 3.0 on threadripper. I do still think it's an ACPI issue though - my B550 Steel Legend has the exact same 149C controller in it and it works fine in XP on there. So something in my Threadripper DSDT is causing the issue. Not to worry though, I've worked around the issue with a separate USB 3.0 card so it doesn't bother me all that much now.

@Dietmar I'm curious to see if adding this kind of delay would help with the USB problem on my threadripper system. Can I get a 32-bit build from you to test this with?

@Dietmar Regarding only having IRQ 1-15 in XP. Don't you need to be using an APIC to get more than 16 IRQs?

@ISKra I know there is an XP driver for the E2-6110 out there on the internet. Sorry I can't remember where it was, but I'd imagine that driver would also work for XP. Google your DeviceID and see what you can find.

@George King It's the ACPI OS Version lie that daniel_k implemented originally.

@PPeti66x I think this might be because this code does not have the ACPI OS Version lie implemented which is what the 2021.02.02 acpi.sys implemented. @Dietmar can you compile a version with the version lie implemented for @PPeti66x to test?

@Dietmar @Mov AX, 0xDEAD Nice! acpi.sys v7 Solved both of my issues! No more A5 BSOD and those last 6 CPUs are now identified correctly! Thanks! I think this is good enough for @Outbreaker to put into Integral Edition for next release!

@Dietmar The most recent acpi.sys I have that is working for my system was built by @George King on the 4th December. so something changed after that point broke.

@Dietmar Thanks, Unfortunately your acpi.sys gave BSOD A5 0x11 0x08 I'm going to need to refresh my memory for how to properly setup KDNET - I have admittedly been very slack on this regard. If it helps for reference, the last working acpi.sys for me was compiled by @George King on 4 dec 2022. and the only issue with it was that processors 1A to 1F would not auto install - they did work with a brute force install but just had the name "processor".

@George King can you build v6 please?

@Mov AX, 0xDEAD If QEMU will let you try with more then 32 CPU cores and see if XP freezes - as my threadripper system freezes XP with 48 CPUs but is fine with 24 CPUs.

@Dietmar I tried your DDB hacked acpi.sys - unfortunately I got BSOD A5 0x000000D with it. @George King can you compile that pre-v6 code version and I'll see how that one goes?

@Mov AX, 0xDEAD @Dietmar Your most recent free build acpi.sys on my AMD system gives BSOD A5 0x11 0x8 I still have not yet setup KDNET on this system so no debug logs. edit: @George King's 4th Dec build still works just fine.