[Coco] 6x09 read and write strobes.

John Kent jekent at optusnet.com.au
Sun Dec 4 18:54:38 EST 2011


Maybe fixed width font works a bit better for the timing diagram
John

On 5/12/2011 5:35 AM, John Kent wrote:
> Is Reset active high ?
> I assume it's to protect against RD & WR being asserted during reset, 
> in which case it should be Reset_n.
> WR should really be gated with E rather than Q as data out from the 
> CPU is valid on the falling edge of E.
> (I hope I have Q around the right way).
>
> I'm not sure what you mean by latching on the rising edge of WR.
>
> E --\_________/----------\__________/-----------\_______
> Q _____/-----------\__________/------------\__________/
> D =>xxxxxxxxxx<=========>xxxxxxxxx<==========>--------- Data from CPU 
> valid during E,
> RW xxxx\__________________/xx/--------------------\xxx
> RD _______________________________/------------\______
> WR ___________/----------\____________________________
> John.
>
> On 5/12/2011 2:12 AM, Phill Harvey-Smith wrote:
>> Hi all,
>>
>> I am currently developing a SD/MMC interface for the Dragon and CoCo, 
>> that uses a Xilinx CPLD to interface between the 6809 and the AVR on 
>> the board.
>>
>> I am however having some problems with random lockups / crashes that 
>> don't seem to be software related as I have tested the software in 
>> mess and it will run flawless ly there.
>>
>> The thing I am unsure about is the read and write strobes and which 
>> clocks they should be based upon.
>>
>> Currently in my Verilog I have :
>>
>> // Generate OE and WE signals, only do this if Reset is high !
>> assign RD            = E & RW & Reset;
>> assign WR            = Q & ~RW & Reset;
>> assign nRD            = ~RD;
>> assign nWR            = ~WR | RamWP;
>>
>> This way RD and WR are positive logic, so I am latiching into the 
>> CPLD on the rising edge of WR and outputting to the 6809 bus while RD 
>> is true. nRD and nWR are the read and write strobes for the onboard 
>> ROM and RAM, the RamWP is in a latched register and allows me to 
>> write protect the RAM.
>>
>> Is this the correct logic ?
>>
>> I have seen (on the DragonDos schematic, the read and write strobes 
>> for the WD2797 are similar to the ones above based off both clocks, 
>> however the write strobe for the drive/density select latch is based 
>> off E.
>>
>> Comments anyone ?
>>
>> Cheers.
>>
>> Phill.
>>
>

-- 
http://www.johnkent.com.au
http://members.optusnet.com.au/jekent




More information about the Coco mailing list