Finally have a feature request for openMSX :)

Pagina 5/5
1 | 2 | 3 | 4 |

Van wouter_

Champion (453)

afbeelding van wouter_

10-03-2013, 10:55

bluemsx wrote:

According to FS-A1FX/WX/WSX/ST/GT, PHC-70FD/70FD2, memory I/O (7FFC-7FFF) is shown as "FC FC FF 3F".
This is also mirrored to (BFFC-BFFF)

Ok, thanks for pointing out you also get these values on the FS-A1FX,.... machines. I've updated openMSX (I can only test myself on a real turboR machine).

bluemsx wrote:

If you impement to add Memory Mapped I/O "FC FF FC 3F", there is no need to replace rom.
But if not possible or wanted, rom must be replaced.

Actually on a turboR machine it's possible to read the content of the disk ROM in two different ways. One way is via slot 3-2 (where the 'actual' disk ROM is located). But another way is via slot 3-3 (where all ROM content can be viewed in a big 4MB ROM mapper). E.g. in the blueMSX rom files, compare the content of 'a1gtdos.rom' with the content of 'a1gtfirm.rom' at offset 0x60000. You'll see that the content is identical, except for the location of the FDC registers. (A real turboR machine only contains this one big ROM chip).

This shows that the actual ROM does not contain those "FC FC FF 3F" values. Because openMSX does not have a separate ROM image for the disk ROM (because it's anyway a subset of the big 'firmware' ROM), patching the ROM dump is not an option.

This means there is extra hardware that generates those "FC FC FF 3F" values at addresses 0x7FFC-0x7FFF. Anyone any idea what this is? Or that these locations always read these same values? There's nothing critical located at these addresses because disk emulation does seem to work fine by always returning fixed values (and even if returning "FF FF FF FF").

Van bluemsx

Rookie (32)

afbeelding van bluemsx

10-03-2013, 13:34

Anyway, if you can add I/O register with "FC FC FF 3F" in FS-A1FX/WX/WSX/ST/GT, PHC-70FD/70FD2, thanks.

I've got real dumped of PHC-35J (no FDD, no MSX Music, JIS 1st class Kanji)
I've checked Slot 3-2.
I'm supprised that all the contents from 4000-7FFB is 00h. (Not FFh)
7FFC-7FFFh is "FC FC FF 3F". This is also mirrored to "BFFC-BFFFh"

blueMSX team refused to accept this implemention ago.

If you want to check it, I can send it by PM.

>If you impement to add Memory Mapped I/O "FC FF FC 3F", there is no need to replace rom.
>But if not possible or wanted, rom must be replaced.

Sorry "FC FF FC 3F" --> "FC FC FF 3F"

I've checked openMSX support PHC-35J ROM. Not bad dumped.
No problem to run emulator.
But this is incomplete ROM set.

Van bluemsx

Rookie (32)

afbeelding van bluemsx

10-03-2013, 13:31

Memory Mapped I/O Register
FDC : TC8566AF based MSX machine.

(External FDD Unit)
FS-FD1A : (7FFA-7FFB, mirrored to BFFA-BFFB) -- 80 00

FS-A1F / FS-A1FM : (7FFA-7FFB, mirrored to BFFA-BFFB) -- 80 00

FS-A1FX / FS-A1WX / FS-A1WSX / PHC-70FD / PHC-70FD2 : (7FFA-7FFF, mirrored to BFFA-BFFF) -- 80 00 FC FC FF 3F

PHC-35J (7FFC-7FFF, mirrored to BFFC-BFFF) -- FC FC FF 3F (No FDD BIOS/ROM, but I/O Register is shown as)

--> 1) (7FF0-7FF1, mirrored to BFF0-BFF1) -- 0x 33 (x is Bank Number 0 - 3, so you can see the content beween 00 33 to 03 33)
--> 2) (7FF4-7FF5, mirrored to BFF4-BFF5) -- 80 00
--> 3) (7FFC-7FFF, mirrored to BFFC-BFFF) -- FC FC FF 3F

I know this can't be shown with normal access. (I think this seems to be copyprotected or disable to access with interrupt.)
To dump them, SAVEROM.COM or KSAVER.COM recommend.

Van tvalenca

Paladin (747)

afbeelding van tvalenca

19-03-2013, 16:55

Manuel wrote:

tvalenca: there is a GUI debugger, but I don't think anyone ever tried to compile it for Mac... Only for Windows and Linux.

What kind of crash do you get? Can you give some information?

Sorry for taking too long to answer... and sorry again because of the long answer you're going to read...

I tried again to run the openMSX_Debugger version that FiXato compiled and found out that it was complaining about the absence of something called QtCore... And then I remembered that in the past I searched upside down the internet and could not find it, so I just forgot about it and kept using blueMSX through Parallels (which I do own a copy, but I don't have any windows license anymore)... but it's just too unpratical! (well it could be worse if I had to run BrMSX through DOSBox, but if I wanted to keep it strictly legal, this is my only alternative!) And so, anytime I got inspired to write something to MSX I think twice because I can't use openMSX and have to start windows on Parallels (and get the memory bloated)

That's OK if openMSX relies on some very good pieces of software out there... really, you don't need to redo whats already done! But it would be nicer if it at least had its dependencies declared, say, in a readme file! Or somewhere right before the download page, above or below the download link! Or at least in a simple dialog message telling that I don't have some piece of software that is required to run, instead of some weird crashes complaining about some random library that was not loaded to find out that I need some (obscure?) 3rd party software...

Van Manuel

Ascended (17753)

afbeelding van Manuel

20-03-2013, 23:21

It's not documented because there is no official compile of the debugger for Mac at the moment... I didn't even remember that FiXato compiled it... But indeed, it's just here:

I suppose FiXato is the one to help here, as I don't know anything about it myself.

Van FiXato

Scribe (1694)

afbeelding van FiXato

21-03-2013, 00:30

You probably need or though I'm not too sure. Better ask mth.
It's been a long time since I set up the required libraries for it.

Van tvalenca

Paladin (747)

afbeelding van tvalenca

21-03-2013, 01:57

I think it has something related with qt4. I'll try later.

Pagina 5/5
1 | 2 | 3 | 4 |