Is Anyone Running Fuzix on their RC2014?

[Sir Isaac]

As the title says I'm really curious about running Fuzix on my RC2014. Has anyone here got it up and running?

How do you Install it? If I run the makefile in Linux does that just generate a rom file to burn the rom chip?

I watched Dr. Baker's video on it but he doesn't talk too much about the install process.

http://www.smbaker.com/z80-retrocomputing-16-unix-on-rc2014

I think it would be great to have a unix like environment on the RC2014.

[codifies]

Yeah I'd be interested in this, (funnily enough stumbled on it myself just recently)

it would be nice to get something like this working with it too
(petit FAT) http://elm-chan.org/fsw/ff/00index_p.html

indeed it would be nice to have an 8KB rom with just a SD boot loader and bunch of utility routines
then you could choose a different language / OS by booting from a different SD card...

[phillip.stevens]

t would be nice to get something like this working with it too

(petit FAT) http://elmchan.org/fsw/ff/00index_p.html

I've finished the assembly diskio components which run under the ChaN FatFS system to provide an interface using the Intel PIO8255. AA is helping to get the ChaN FatFS code added to z88dk as a third party library.

TLDR; it won't be long before a full FatFS (including exFAT and LFN) is available for rc2014 via z88dk.

[Spencer Owen]

On 2 Sep 2017 13:53, "phillip.stevens" <phillip...@gmail.com> wrote:

TLDR; it won't be long before a full FatFS (including exFAT and LFN) is available for rc2014 via z88dk.

Oooh... that sounds exciting!

Spencer 

[Jay Cotton]

I am interested in this also.  I looked at the code base, and the tool chain to build it.  

Its a bear to compile all the stuff and get it running, but it might we worth a try in the

next few weeks.

I send a request to baker for a binary or a tarball of the code but, never heard back.

[phillip.stevens]

 

I've finished the assembly diskio components which run under the ChaN FatFS system to provide an interface using the Intel PIO8255. 

AA is helping to get the ChaN FatFS code added to z88dk as a third party library.

 

TLDR; it won't be long before a full FatFS (including exFAT and LFN if desired) is available for rc2014 via z88dk.

All the code for the FatFS diskio system is now working, and integrated into z88dk so it can be programmed from C.

First Proof of Life from the FatFS layer tonight. As the FatFS layer is expected to be unmodified by "users", there's nothing more to do.

I'm not sure if there is a PIO8255 IDE interface underway for the RC2014 but, if not, it would be a good time to start one.

If the hardware looks like this example, or on page 6 of attached schematic PDF, then the diskio layer already works.

I sourced code from both PJRC in 8051 mnemonics, and from Retroleum in Z80 mnemonics for an 8 bit interface.

Working between the two of them, together with the examples from the OSDev Wiki, I'm pretty happy with the outcome.

I think I need a little rest now.

Phillip

[Ed Brindley]

Hi Philip,

Looks exciting!

I was looking through the RomWBW source, I see that has support for IDE drives driven by the 8255 on the SBC-v2 and Zeta boards, but the port / signal mapping looks a bit different (https://github.com/wwarthen/RomWBW/blob/master/Source/HBIOS/ppide.asm) than what you've used on your board. Contemplating if an RC2014 PIO interface was to be made, which would be the best mapping to pick. I'd be tempted to go with the RomWBW PPIDE mappings as there are already 2 boards using that format, and could run RomWBW without changes(unless your mapping is better :))?

How easy is it to get 8255s? I could only see them for sale at Jameco and they were marked "refurbished".

Cheers,
Ed

[phillip.stevens]

Thanks, It is all working fine now. Probably not going to put FatFS into z88dk, because it is very configurable and any configuration decision would probably be the wrong one for most people. The diskio layer is very transportable onto any other 8255 interface, just by changing the pin defines.

I had a quick look at the ROMWBW mapping, and it is essentially similar. Port A and B for the I/O, and Port C for the control lines. The only difference is which control line is attached to which 8255 pin.

The important thing is that only three IDE interface control pins are NOT passed through an inverter. These are the address pins DA0, DA1 and DA2. All other pins have to pass through an inverter, so that they "rest" in high state. They need to rest in high, so that when the interface is swapped from Read to Write, and back, they don't reset the disk. The 8255 sets all pins to Logic 0 when it changes modes, this is a bug / feature of the chip.

I chose to use the three pins for the address lines that are also used for input in Mode 1 and Mode 2 of the 8255. These pins are not passed through the inverters, so they can actually be used as inputs and the interface remains general. If they were passed through inverters, they would be unusable as inputs to the 8255, and therefore Mode 1 and Mode 2 wouldn't work.

Bottom line is that the pin selection I've made is more general than that chosen by others, so that the IDE pinout can be used for other purposes, as well as just for a disk drive. I'm planning in the long term to use the 8255 as an extension interface for other boards or interfaces, using the IDE 44pin cable and 2mm pins because they're very neat and easy to obtain, and cheap.

At the end of the day it makes very little difference (if you don't care about generality for the 8255), because the pin selections are mapped by definition, and it doesn't make any real difference. The code will work on all the interfaces just by changing a define.

I got my 82c55s from Digikey. They're "off the rack" in PLCC format. Not DIL, but the socket is still through hole.

[A A]

On Sunday, September 10, 2017 at 6:06:59 AM UTC-6, phillip.stevens wrote:

Thanks, It is all working fine now. Probably not going to put FatFS into z88dk, because it is very configurable and any configuration decision would probably be the wrong one for most people.

There is a library configuration mechanism in z88dk so this shouldn't be an issue.  It will have to be taken apart though - one function per file - as the assembler is unable to separate individual functions in the same file into independently linkable entities at the current time.

[phillip.stevens]

I was looking through the RomWBW source, I see that has support for IDE drives driven by the 8255 on the SBC-v2 and Zeta boards, but the port / signal mapping looks a bit different (https://github.com/wwarthen/RomWBW/blob/master/Source/HBIOS/ppide.asm) than what you've used on your board. Contemplating if an RC2014 PIO interface was to be made, which would be the best mapping to pick. I'd be tempted to go with the RomWBW PPIDE mappings as there are already 2 boards using that format, and could run RomWBW without changes(unless your mapping is better :))?

 

I'm not sure if there is a PIO8255 IDE interface underway for the RC2014 but, if not, it would be a good time to start one.

If the hardware looks like this example, or on page 6 of attached schematic PDF, then the diskio layer already works.

 

Ed,

I should also add that it is my intention to get both Fuzix and RomWBW running on this interface and on the Z180 platform I'm using.

Now that basic z88dk support is there, with both the APU interface and the IDE interface is proven, I need to pay attention to the I2C and USB programming interfaces.

It would be great to see a 82c55 IDE pin-out interface on the RC2014. And, if I wasn't so deep in hardware projects, I'd build it myself.

I'd certainly want to have one if it was available, and I'd also like to see it support Mode 1 and Mode 2 if possible.

PC2, PC4, and PC6 are the 3 three special pins in Port C, that need to NOT go through the inverters for IDE, if the interface is to support other applications.

Cheers, Phillip

[Ed Brindley]

Hi Philip,

I know what you mean about having too many projects on the go :)

I'll have a go at laying out a RC2014 version of your YAZ180 interface over the next week or so. I need to get some other PCBs fabbed soon so can get this made at the same time.

Will post schematic / PCB layout here for review when I'm done.

Cheers,

Ed