First whdload slave (TBL-Eon)

[paraj]

Hi Guys,
Trying to do my first whdload install. Never done any installer stuff before, and I think for this one it's easier to just copy the ADFs from the demo distribution rather than imaging discs. I've tried it on my own 060 and in WinUAE and it seems to work, but maybe somebody could do a spot check/review and see if it's up to snuff?
Other than that I'm not sure what next steps are. Would I just send it via e-mail to Wepl? (JOTD has kindly helped with the other stuff I've been involved in, but figured it was time to grow up, and handle some things myself ).
Thanks in advance.

[jotd]

congrats!

It contains very interesting stuff like uae debug commands that I never really used. I'm going to check this out!

the "datetime" file is produced by wdate (or wdate.py for windows but you have to use my tools) in a a makefile that maybe you should provide.

Send the install at release*at*whdload.de

You already have a whdload dev id, that will make things easier

[jotd]

BTW some more remarks:

1. can you document the AKS winuae commands?
2. you could use a print to WinUAE debugger console by writing to specific $ffxxxx memory zones in winuae instead of using whdload log. Toni mentionned that some time ago, never could make it work but I must have done it wrong

[paraj]

Thanks for the feedback! I've added some more comments in the source code (how to build, what the AKS stuff is), and fixed a few things in the readme and shipped it off (so I can finally forget about it ).

The AKS stuff is from https://eab.abime.net/showthread.php?p=1513044 (apparently it can also be made to work in FS-UAE). I got really tired of manually toggling warp mode / waiting for specific part (and missing it). Really useful for non-interactive stuff, but note that it needs some uae stuff to be loaded amiga side to work (not an issue as long as a HD is attached).

Yes, logging would also be useful (if you don't need it on the real machine), I'll look into that next time I need to debug some stuff

[DanyPPC]

Tested on WinUAE and Blizzard 1260, all ok.

Thank you for this new EON slave

[Wepl]

Originally Posted by jotd

2. you could use a print to WinUAE debugger console by writing to specific $ffxxxx memory zones in winuae instead of using whdload log. Toni mentionned that some time ago, never could make it work but I must have done it wrong

winuae logging probably has less overhead, but resload_Log will also work on real hw
@paraj I saw your macros for logging, maybe these are useful to be added to whdmacros.i?

[paraj]

Originally Posted by Wepl

winuae logging probably has less overhead, but resload_Log will also work on real hw
@paraj I saw your macros for logging, maybe these are useful to be added to whdmacros.i?

I certainly like it, so if others find it useful (and it's compatible with the assemblers people use) I say go for it.
The log feature is indeed quite useful on real HW. Especially when it's other peoples hardware - used it quite a bit in test builds to narrow down various problems.

[Wepl]

Originally Posted by jotd

you could use a print to WinUAE debugger console by writing to specific $ffxxxx memory zones in winuae instead of using whdload log. Toni mentionned that some time ago, never could make it work but I must have done it wrong

attached my uaedebug.i used, interesting I found the cycle counter to measure time to execute certain code

[jotd]

I have a question here

UAECMD macro
.call\@:
        pea     0.w
        move.l  a0,-(sp)
        pea     .cmdend\@-.cmd\@
        pea     .cmd\@(pc)
        pea     -1.w
        pea     82.w
        jsr     $f0ff60
        lea     24(sp),sp
        bra     .done\@
.cmd\@: dc.b    \1
        dc.b    0
.cmdend\@:
        even
.done\@:
        endm

why pushing a0 in the stack BTW?

@Wapl thanks for the macro!

[paraj]

Originally Posted by jotd

why pushing a0 in the stack BTW?

Should have commented the macro since I forgot, but looked it up again and it's just an unused pointer. Normally it would point to where the result of the command is written, but since the last parameter (first pushed) indicating the number of bytes to write is 0, nothing is written.


Calls here:



https://github.com/tonioni/WinUAE/bl...aelib.cpp#L385


To here:



https://github.com/tonioni/WinUAE/bl...file.cpp#L8063


And yeah, thanks Wepl, will go into my toolbox.