8085 Address Decoding

[Adrian Graham]

On 26/01/2017 01:09, "Alexis Kotlowy" <thrashbarg at kaput.homeunix.org> wrote:

>> My Executel ROM disassembly puts this block of code between 0x0F38 and
>> 0x0F3E:
>> 
>> 0F38                            L0F38:
>> 0F38 : 7B                "{"        [4]        mov    a,e
>> 0F39 : 95                " "        [4]        sub    l
>> 0F3A : 5F                "_"        [4]        mov    e,a
>> 0F3B : 7A                "z"        [4]        mov    a,d
>> 0F3C : 9C                " "        [4]        sbb    h
>> 0F3D : 57                "W"        [4]        mov    d,a
>> 0F3E : C9                " "        [10]       ret
>> 
>> Does that make sense? I might not have a full disassembly since L0F38 is
>> only called from two places and neither of them jump back to 0x0FB2 but I
>> know that doesn't necessarily mean it's wrong.
>> 
>> Cheers,
>> 
> 
> Hi Adrian,
> 
> The instruction at 0x0F3E is RET, so it's going to read two bytes off
> the stack and jump to it. This is what you're seeing at 0xF0B2 and
> 0xF0B3. I'd suggest you look at how RAM is addressed at the top of
> memory space and see if it's correct.

I was afraid someone might say that. Whilst I don't have a memory map for
this system (or any other docs apart from the ones I'm writing myself) I did
notice last night that CAS-1 was missing. RAM is laid out like a CBM PET
with 16x 4116 DRAMs split into lower and upper blocks. RAS and CAS-0 are
pulsing.

More probing tonight then!

-- 
Adrian/Witchy
Binary Dinosaurs creator/curator
Www.binarydinosaurs.co.uk - the UK's biggest private home computer
collection?