A3000 / 040 hangs when anything is run with DCACHE on

  • Thread starter Thread starter t-bus
  • Start date Start date
  • Replies Replies 7
  • Views Views 1899
T

t-bus

Member
Joined
Jun 30, 2013
Posts
77
Country
USA
Region
Pacific Northwest
I recently bought a PPS Mercury accelerator from another member, and installed it in my 3K.

It seems to work, until I turn the DCache on via SetPatch or Sysinfo, and then it'll hang after basically the next thing I do (machine won't boot at all with my existing sequence because it runs setpatch at the beginning)-- sometime after which point the machine hard-locks.

I'm running 3.1 roms and 3.1 dos / wb.

I doubt the dcache is damaged (could it be?)- I assume I am missing something elementary, like a library or setpatch update.

again, this happens even when I boot without startup-sequence. :S

EDIT: For instance, if I type SetPatch, then LoadWB, the system hangs.

Turning on DCACHE in sysinfo, then pressing the "memory" button in the gui produces a hang.

Turning on dcache via the CPU command, then opening aexplorer produces a hang.

If I loadwb, then enable the cache, the machine will hang after I click something.

If the dcache is on, the machine just wont run.

Thanks!

-
 
Last edited:
Most likely reason is missing LIBS:68040.library. SetPatch should automatically enable both caches if 68040 support is available.

Check if SetPatch reports "68040 support library loaded".
 
Toni.Wilen said:
Most likely reason is missing LIBS:68040.library. SetPatch should automatically enable both caches if 68040 support is available.

Check if SetPatch reports "68040 support library loaded".
Click to expand...

It does say that, yes. I have a LIBS:68040.library provided with the 3.1 install.

In fact, even the install media for 3.1 hangs on running the startup sequence unless I boot with both mouse buttons down and turn off CPU caches.
 
Try different version of 040 library. Same thing happens with 3640 when wrong one is used.
 
Last edited:
no dice.... Still freezes up. Could it be an issue with the motherboard ram? The accel. Ram checks as good.

or perhaps the 'fastrom' feature....
 
Pretty sure if you install an accelerator in the A3000, you can only use page mode RAM, not static column. If you have static column RAM on the motherboard, I would suspect this.
 
rcade said:
Pretty sure if you install an accelerator in the A3000, you can only use page mode RAM, not static column. If you have static column RAM on the motherboard, I would suspect this.
Click to expand...

Rcade was close. It was related to the static column ram on the board, however you can use it it would appear!

The PPI mercury had a jumper for disabling off-board (to motherboard) burst access to ram. In the errata for the manual, they mention the manual had the setting's polarity wrong. I overlooked the eratta... Moved a jumper to limit burst ram access to on-accellerator ram and now all is well (except for some stability issues I suspect are related to the FAST slot connector)

I've hit the connector with deoxit, and am posting from the 3K right now!



EDIT: So this issue can be considered solved, and I've opened another thread regarding stability issues on disk access. It seems the troubleshooting never ends on these.
Thanks for the help, Rcade.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom