Page 1 of 5

DSTB1 - official firmware tests

Posted: 05 May 2022 11:30
by exxos
I started this new thread because all my tests were using experimental firmware previously.

My test setup is with a H5 motherboard (no blitter) and using flashyclock for the TOS206 decoding.

Firmware as released.

IMG_0539.JPG

Tests done with GB6 with altram.prg (ran from AUTO) .

IMG_0540.JPG

I repeated this test with altram.prg run from desktop as @Badwolf pointed out, TOS seems to allocate *something* in alt-ram as the LED is flashing. I never heard TOS by default loading anything into alt-ram. So not sure if anyone else knows any more about this?

Looks like there are no change in GB6 tests.

IMG_0542.JPG

I tried running altrom.prg after altram.prg but no change. I thought then to just run altrom.prg but then the floppy goes nuts..

IMG_0541.JPG

But not sure if it's going to help with a external ROM decoder in the mix ?


YAARTTT seems ok ...

IMG_0543.JPG

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 12:29
by Badwolf
exxos wrote: 05 May 2022 11:30 I tried running altrom.prg after altram.prg but no change. I thought then to just run altrom.prg but then the floppy goes nuts..

But not sure if it's going to help with a external ROM decoder in the mix ?
Yep AltROM.PRG replaces AltRAM.PRG -- it'll simply refuse to run if run when AltRAM already active.

Not sure what's going on with the floppy, though. External decoder shouldn't make a difference as once AltROM has done its thing it takes over decoding for the CPU.

Interesting, though. That's the first time I've seen that. Run from AUTO again?

BW

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 12:38
by exxos
Badwolf wrote: 05 May 2022 12:29 Interesting, though. That's the first time I've seen that. Run from AUTO again?
Yep.

Something is amiss somewhere even before loading anything. I clean booted into TOS104 earlier and when I opened from floppy GB6.PRG icon was not correct. Though it only did that once and I have not been able to replicate that again yet :roll:

EDIT:

Actually same problem it seems as this post https://www.exxosforum.co.uk/forum/viewt ... =30#p84405

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 12:44
by Badwolf
exxos wrote: 05 May 2022 12:38 Something is amiss somewhere even before loading anything. I clean booted into TOS104 earlier and when I opened from floppy GB6.PRG icon was not correct. Though it only did that once and I have not been able to replicate that again yet :roll:

EDIT:

Actually same problem it seems as this post https://www.exxosforum.co.uk/forum/viewt ... =30#p84405
In TOS 1.04!?

WTF? :o

BW

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 12:50
by exxos
Badwolf wrote: 05 May 2022 12:44 In TOS 1.04!?

WTF? :o
WTF indeed. There must be something with the RAM tripping up as I have similar problems in the other thread with the experimental firmware :shrug:

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 12:57
by Badwolf
exxos wrote: 05 May 2022 12:50 WTF indeed. There must be something with the RAM tripping up as I have similar problems in the other thread with the experimental firmware :shrug:
So this is in 'stock' configuration under TOS1.04? That means doing 16MHz switching, but no AltRAM as, well it's TOS1.

Perhaps it's 'coming back' too quickly from DMA?

BW

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 13:02
by exxos
Badwolf wrote: 05 May 2022 12:57 So this is in 'stock' configuration under TOS1.04? That means doing 16MHz switching, but no AltRAM as, well it's TOS1.
Yep.
Badwolf wrote: 05 May 2022 12:57 Perhaps it's 'coming back' too quickly from DMA?
Not sure. I'll take a look at your code and see if I can see anything. It is somewhat difficult to follow someone else's program in a language you are not very familiar with tho :lol: :roll:

Though I may look at the firmware build I did a few days back aswell as that one seemed to basically "work".

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 13:11
by Badwolf
exxos wrote: 05 May 2022 13:02
Badwolf wrote: 05 May 2022 12:57 Perhaps it's 'coming back' too quickly from DMA?
Not sure. I'll take a look at your code and see if I can see anything. It is somewhat difficult to follow someone else's program in a language you are not very familiar with tho :lol: :roll:

Though I may look at the firmware build I did a few days back aswell as that one seemed to basically "work".
Cool. If you ignore all the SDRAM stuff, it's not that much logic.

We could perhaps add an AS_HOLDOFF term that blocks AS asserting until (say) two clock cycles after BGK goes high?

Something like

Code: Select all

wire [1:0] AS_HOLDOFF;
FDCP ff_asholdoff0( .D( ~BGK ), .C( ~CLK8 ), .CLR(1'b0), .PRE( ~BGK ), .Q( AS_HOLDOFF[0] ) );
FDCP ff_asholdoff1( .D( AS_HOLDOFF[0]  ), .C( ~CLK8 ), .CLR(1'b0), .PRE( ~BGK ), .Q( AS_HOLDOFF[1] ) );
...
assign AS = AS_HOLDOFF[1] | ( ... the other terms ... );
Only obvious thing I can think of.

BW

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 17:01
by Badwolf
exxos wrote: 05 May 2022 12:50
Badwolf wrote: 05 May 2022 12:44 In TOS 1.04!?

WTF? :o
WTF indeed. There must be something with the RAM tripping up as I have similar problems in the other thread with the experimental firmware :shrug:
Had another idea. I think I've set up the FRB cookie incorrectly. I don't remember exactly, but I don't think I'm TSRing -- the FRB memory is probably being returned to the OS & subject to being overwritten.

I may have been lucky that nothing critical happened, but I think altram.prg is fundamentally wrong in that regard & needs repair work.

I'll bet that's the problem you're seeing.

BW

Re: DSTB1 - official firmware tests

Posted: 05 May 2022 17:33
by exxos
Badwolf wrote: 05 May 2022 17:01 Had another idea. I think I've set up the FRB cookie incorrectly. I don't remember exactly, but I don't think I'm TSRing -- the FRB memory is probably being returned to the OS & subject to being overwritten.

I may have been lucky that nothing critical happened, but I think altram.prg is fundamentally wrong in that regard & needs repair work.

I'll bet that's the problem you're seeing.
Maybe but that died without even loading anything. I have tried EMUTOS a few times in the mix as well.. I will try EMUTOS on my current build just to rule that out.