[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [N8VEM-S100:6199] Re: 80386 CPU Board



I at a loss to explain what is going on Gary. Particularly since you can get back to the z80 from PM. The only thing different there is possibly some address lines.
My only suggestion would be to exactly follow signal lines with a logic probe. Set the SMB  to switching mode using an output bit. Check the TMA0 line on the bus. Not the board LED (it's behind a gate). Then from the 80386 monitor output the same bit to the same port. Does the TMA0 line return high. If not board problem. If yes z80 board problem.

John Monahan (mon...@vitasoft.org)

On Feb 5, 2015 6:16 AM, "Gary Kaufman" <geka...@gmail.com> wrote:
Still no success with the 80386 --> Z80 problem.

Spent a while last nite trying various options.  I switched U52 to a 74F00, U55 to a 74LS244 and added the 47pf cap per the prior discussion with David Fry. 
I also simplified the system to SMBv2, Propeller Display, 16mb SRAM, MSDOSv2 Support, Z80 and 80386 boards.
I've swapped all of the 74LS74's (and also tried 74ALS74's) without any improvement.

I'm bypassing the monitor commands, using QIED and QOEE,01 // QOEE,00 to swap.

K4 on the SMBv2 is jumpered 2-3 and P36 jumpered 1-2 and 3-4.  (although I've tried every other combination).  With this setting the QIED swaps 80386<-->Z80 in both directions and the TMA LEDs remain off.
Interestingly If I've accessed the 80386 board with an input from port EDh and go to protected mode I can return to the Z80 board using QOEE,00 - but this doesn't work in non-protected mode. 

Using QOEE,01 or Monitor "W" or "O" commands will switch Z80-->80386 and the TMA0 LED lights up - but then a reset is the only way back to the Z80 (eg, QIED, QOEE,00 won't work and the TMA0 let remains on).

When I add a 6502 board set up to access with TMA1 It works fine with the SBMv2, as does the 8088 board set up for TMA0.

I can post a picture of the 80386 board tonight to confirm the other jumper settings.

- Gary



On Tuesday, February 3, 2015 at 1:38:14 PM UTC-5, Gary Kaufman wrote:
John -

Thanks so much for taking the time.  I'm obviously doing something wrong - but didn't have any luck last nite.  The input port EDh switching works fine, but the TMA don't.

I read thru the docs this morning and have a few other ideas to try tonight. 

If all else fails I'll post some pictures of my SMBv2, 80386 and Z80 boards and see if someone can spot a jumper problem.

- Gary

On Monday, February 2, 2015 at 10:38:54 PM UTC-5, monahanz wrote:

Hi Gary, I took a break from this “80486 project” and took a look at the switching back issue with the 80386.  With both my boards I have no problem switching back to the Z80 using the SMB V2 with its K4 jumper 2-3. This utilizes the output bits 0,1,2 & 3 to select the TMA lines 1,2,3 &4.  I use TMA0 for the 8086 family.

Jumpering the SMB V2 K4 1-2  and no P36 jumper,  does not bring up the 80386.  Adding the P36 jumper 1-2 then brings up the 80386, and allows a return to the Z80.   However the TMA0 line stays low!  If you also jumper P36 3-4, the TMA0 line goes low when the 80386 is in control.  However its “Z” command does not switch back to the Z80.

 

However if you remove the K4 jumper on the SMB V2, the 80386 onboard port behaves correctly i.e. “Z” switches it back to the Z80 and TMA0 goes high again.

So….

If you are using the SMB V2, best to jumper K4 2-3.  This will be the mode I will use anyway for future CPU boards (up to 15 on the S100 bus!)

If you use the 80386 onboard port, remove the K4 jumper on the SMB.

 

Have not tried the original SMB, but I suspect removing its jumper would be the same.

 

Let me know if this helps

 

John

 

PS I have not added Dave’s  modes in all of the above.

 

 

From: n8ve...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of monahanz
Sent: Monday, February 2, 2015 4:24 PM
To: n8ve...@googlegroups.com
Cc: mon...@vitasoft.org
Subject: Re: [N8VEM-S100:6186] Re: 80386 CPU Board

 


My error  Gary, the V1.49 is the most up to date version I have been using.   Most recent changes are just for PM testing BTW. The core 8086 stuff really has not changed in a while.

 

I stuck in hardware currently. Could you carefully checkout the 80386 switching code.  I seem to remember I used the old and new ways to switch TMA 0 line.  Only the output mode will work for the SMB V2.  With a logic probe see if you are getting signals.

 

John

 

 

 


On Monday, February 2, 2015 at 8:52:34 AM UTC-8, Gary Kaufman wrote:

Thanks for checking, I'll try tonight when I get home from work.

I believe I've tried both with 1-2 and 1-2,3-4 jumpered.  I've also tried direct port access to EDh and EEh.

The latest monitor I could find on your site was V1.49

- Gary

On Monday, February 2, 2015 at 11:34:53 AM UTC-5, monahanz wrote:

Gary, just checked my board.  I can get back to Z80 from 80386 monitor (V1.84) with the “Z” command if the P36 jumper is in 1-2 position. However NOT SO if it’s in 2-3  position.  The latter is utilizing the SMB V2 to pull down TMA1 rather than the 80386 onboard port.  I will look into it, but meanwhile check software for both CPU’s. I bet there is a conflict with either the (old) IN Port option to switch CPU’s  or the newer Output a bit method.

 

John

 

 

From: n8ve...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of Gary Kaufman
Sent: Monday, February 2, 2015 5:27 AM
To: n8ve...@googlegroups.com
Subject: [N8VEM-S100:6179] Re: 80386 CPU Board

 

David and John -

I've read thru the '386 page dozens of times!

I have the SMB and can switch to the '386 board thru QIED, or output QOEE,01.  Also the "W" and "O" commands both work.  Unfortunately when I use "Z" to return the system hangs.
I can go back to the Z80 from the 6502, 8088 or 80286 boards fine. 
I have P36 with 1-2 and 3-4 jumpered.

I do think it's likely a jumpering issue, but I'll also swap out all of the 74LS74's again once my Mouser order arrives.  I've had a few flakey ones recently.

- Gary

On Sunday, February 1, 2015 at 11:02:55 PM UTC-5, monahanz wrote:

Gary I have written up a few notes/tricks on hardware debugging these types of CPU boards:-

 

John

 

 


On Sunday, February 1, 2015 at 8:24:54 AM UTC-8, Gary Kaufman wrote:

I completed the 80386 board yesterday, and after some confusion (on my
part) getting the monitor configured and jumpered correctly on the
MS-DOS support board it's running nicely at 32mhz using an AMD 80386-33
chip.

I'm running into two issues, and wonder if anyone has encountered them.

When I try to return to the Z80 using the "Z" command it hangs.

I've tried using a TX486DLC/E-33GA but it hangs after printing the
initial monitor prompt  (at both 16 and 32mhz).

Thanks!

- Gary

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.