VS: Overlooked something

From: <geir-oyvind.validalo_at_telenor.com>
Date: Thu, 4 Mar 1999 08:17:26 +0100

Kanskje du kan hjelpe, Jo Even?

Geir...

-----Opprinnelig melding-----
Fra: Michael Grove [mailto:mgrove_at_iwvisp.com]
Sendt: 3. mars 1999 19:33
Til: geir-oyvind.validalo_at_telenor.com
Emne: Overlooked something




Hi Geir,

After reading the users manual for the 040 and 060 a few
times, I tried disabling the cache and MDIS on the 060. This
made no difference. I disconnected the MI line, but then the
machine would no start up, so the AB does use the MI
function, at least partly.

What I overlooked was, when testing the 040 with the
diagnostic cart, I didn't perform all the test, especially
the IDE and SCSI test. When I ran those test with the 040
installed, the test failed just as they did with the 060
installed.

My concerns now are;

BCLK delay for CLKEN. May not be right for bus transfers.
The 060 can run, but when it tries to access another device,
the timing may not be within spec.
This I can fix with a scope........

The AB may do something to the bus signal because the 040
fires off the falling edge of PCLK and the 060 performs bus
operations on the rising edge of CLK. The IPL2 status on
reset can invoke an extra data hold state where the 060
holds the data for one additional CLK, but the AB could be
inserting an unknown there.
In the 060 users manual, this is covered in the Output Hold
Time Differences section.
Sure wish there was some data on the AB card...

misaligned operand handling (because of the 16 bit data bus)
This looks like there could be a potential problem, but this
part is over my head.

I am still puzzled over the machine not at least getting to
a white screen without the diagnostic cart. If the cart can
bring up the machine for testing, something must be
different in the way the TOS ROM initializes the system. The
cart must bypass some steps where the TOS ROM is failing.

Regards,
Mike
Received on to. mars 04 1999 - 17:31:50 CET