PCI floppy controller

[Christian Corti]

On Fri, 22 Apr 2022, "Maciej W. Rozycki" wrote:
> You can of course build a PCI FDD interface around the NEC uPD765 or an
> equivalent controller, but you can't make it compatible with existing PC
> software, because too much PC specifics has been embedded there around the
> 8237 DMA controller and DMA page registers at fixed port I/O locations,
> which is inherently incompatible with the PCI decoding model.

Yes, but I have never understood why you would need BIOS compatibility. 
Current operating systems have their own drivers. I mean, what would be 
the problem to write a, say, Linux driver for such a thing?
And you don't even need DMA for the FDC if you want to make it cheap. OTOH 
a small dedicated SRAM and supporting microcontroller on the PCI board 
would make up a great PCI floppy controller.

> A feasible solution is a SCSI FDD option, such as the DEC RX23 device
> (which is actually a whole embedded microcomputer built around an 8080 CPU
> and using an 8237 DMA controller, an 8259 interrupt controller, a uPD765
> floppy drive controller and a 5380 SCSI interface), which works as a
> removable drive with any single-ended parallel SCSI host adapter, e.g.:

I have a DaynaFile II, it was once an external SCSI 5.25" FDD with 
proprietary protocol, designed for crappy Apple computers.
I had disassembled the firmware and documented the command set. And I 
think that I have redrawn the schematics at that time. It is much 
more capable of what the Apple driver might have supported.
The hardware itself is just a small board with a DP5380, 80C31, a 32kB 
SRAM and 2793 FDC.
So everyone shouting for a solution, I recommend looking for existing ones 
and eventually (re)build one from scratch. I mean, today you just need the 
FDC and a microcontroller with integrated USB or whatever to communicate 
with the host.

(Note: of course this applies only to standard IBM formats)

Christian