@tyukok You can share some disk benchmarks here:
Reliable Windows 98 storage benchmark

tyukok wrote on 2020-03-07, 16:14:

there was no DMA checkbox. And when I switched over to onboard SATA controller, there was no checkbox either

I should've read README.TXT better:

AHCI, by design, uses DMA. DMA is always enabled and the DMA Option will not be visible in the Device Manager Settings.

I just found a new review of MSI MB with PCI & ISA slots for LGA1151 platform (H110 chipset)
https://www.tomshardware.com/news/spectras-ne … -1992-pci-slots
It seem's nothing surprising, just another overpriced industrial MB. I wonder if ISA DMA would work there. I was not able to find that ISA bridge chip it use, some hi-res photo would help to figure out...

A thread was made of this board earlier today. I do wonder what bridge chip it uses, so it could be determined if there's ISA DMA or not.

Mentioned thread:
MS-98L9 V2.0 , a new 1151MB with an ISA slot

These is some datasheet:
https://b2b.rosch-computer.de/shop/file.php?a … plication%2Fpdf
Where is this - 1 x ISA Slot (Windows 7 32-bit Supported Only), because machine officialy supports only Win7 to Win10, its probably about it.. not about that it needs special driver, but i could be wrong.. Win10 probablyditched ISA same as AGP.. i never installed on ISA MB anything newer than Win7, maybe someone did.

Where is the dedicated thread?

AFAIK Win10 dropped support of ISA subsystem from the kernel so ISA card may not work there but it could work in DOS.
The datasheet has better photo, from this I can read:
IT8892E - PCIe to PCI Bridge
F81866AD - LPC based superIO with 6x UART, LPT, FDC...
F85226AD - https://www.alldatasheet.com/datasheet-pdf/pd … TEK/F85226.html Yeah! the LPC2ISA bridge, not PCI2ISA as on 99% of other MBs. So there's a chance ISA sound would work there. So anyone has enough courage to waste his money and test it? 😀 I already wasted >300$ for Commell P4LA years ago (that turned out to have wrong bridge without ISA DMA...)

Ok, nice, same bridge as my C2D board. This is exciting as the bridge definitely supports ISA DMA, as long as LDRQ is connected ~
My pockets are empty though, but if I am gonna upgrade, that's a board I would want to get 🤣.

I really don't think that board will support ISA DMA but I could be wrong. I looked through the manual the other day and the BIOS doesn't seem to have any settings for reserving IRQ or DMA or anything else having to do with legacy support.

Wouldn't the motherboard chipset also have to have support for ISA DMA through the bridge chip for it to be supported?

What chipset does your C2D motherboard have?

LPC bus has everything necessary, what does matter is IO ports configuration and a custom utility will be needed for that. DMAs aren't used by anything on the board though IRQ assignments may be on the arbitrary side by the BIOS but that can still be fixed by reprogramming the chipset.
EDIT: actually you were onto something, I got the H110 datasheet and LPC bus lacks the LDRQ signal, thus DMA will not be possible even if the bridge supports it.

My C2D board is G41+ICH7 based, RUBY-9719VG2AR

Thanks for your attention. Damn intel! I tracked chipset datasheets and the LDRQ was removed in 1xx series. The latest with LDRQ are H/P9x and X99. Do you have an idea how to overcome this issue with your FPGA/CPLD solution? Could they use some proprietary logic for this? I very doubt.. 🙁

Lack of LDRQ means that DMA controller should be finally deleted from the chipset, which means it should be possible to create your own over PCI or PCI-E and no longer need LPC for anything except for SERIRQ signal. FPGA will be necessary though, CPLD will be too small for such a stunt. PCI-E puts big latency on IO operations and the result may be fairly wonky, it is all untreaded territory and not something I'll venture into anytime soon.

EDIT: I have been looking at how IRQs work for PCI-E and you cannot actually target any specific IRQ with it. There's two mechanisms : the legacy INTA/B/C/D mechanism of normal PCI and "message signaled interrupts" and those can only be used with APIC inside the CPU.

PCI interrupts have arbitrary allocation by whatever BIOS sets things up and there's only 4 of them that are shared by other cards.

Message interrupts don't involve the legacy interrupt controller at all. I don't know if it is possible to program the PIC to trigger an interrupt in software, if so things would be possible with a software layer that sets up APIC which handlers then program PIC to generate an interrupt which is seen and handled by whatever DOS game etc. I'm not sure if the latency is unworkable or not.