[net.micro.mac] uw, 128K, and SERD

jimb@amdcad.UUCP (Jim Budler) (07/26/85)

As I reported previously, uw works just fine with a 128K Mac, UNTIL I
installed SERD.  The BOMB!!! It works up to the point where it accesses
SERD (i.e. at the point where the alert saying the ROM driver would be
used), then the bomb ID=2.  Did I install SERD wrong ( in the system)?
Or doesn't SERD work with a 128K Mac?

Any information will be appreciated.  Thank You.

-- 
 Jim Budler
 Advanced Micro Devices, Inc.
 (408) 749-5806
 UUCPnet: {ucbvax,decwrl,ihnp4,allegra,intelca}!amdcad!jimb
 Compuserve:	72415,1200

"... Don't sue me, I'm just the piano player!...."

jdb@mordor.UUCP (John Bruner) (08/01/85)

> As I reported previously, uw works just fine with a 128K Mac, UNTIL I
> installed SERD.  The BOMB!!! It works up to the point where it accesses
> SERD (i.e. at the point where the alert saying the ROM driver would be
> used), then the bomb ID=2.  Did I install SERD wrong ( in the system)?
> Or doesn't SERD work with a 128K Mac?

I must confess that I'm not really surprised that the RAM serial
driver bombs out with a 128K Mac.  (However, I have received
responses from other 128K Mac owners who didn't have this problem.)
The reason: I wrote my own RAMSDOpen.

"uw" is written in Megamax C, and I've been waiting six weeks now
for the V2.1 upgrade that they originally were supposed to ship
mid-June.  (I called up a couple of weeks ago and was promised that
it would ship the day after my call.  It still hasn't arrived.)  The
V2.1 upgrade is supposed to include support for the RAM drivers.

I had the RAM serial driver resources because I purchased the
software supplement; however, Apple provided RAMSDOpen in a
".rel" file which was useless to my Megamax linker.  About the
third week in July (I think), I gave up waiting for Megamax V2.1
and decided to write my own RAMSDOpen.  Unfortunately, the
description of RAMSDOpen in the software supplement was nearly
useless, since it didn't tell me *how* RAMSDOpen worked, just
some of what it did.  I disassembled the first 0x80 or 0x90
bytes of SERD 1 and came up with a RAMSDOpen that seemed to work.

I know of three specific deficiencies in my RAMSDOpen.  First,
it always looks for SERD resource #1.  This causes trouble for
Lisa (oops, Mac XL) users.  Second, it doesn't check the
version number of the in-use serial driver, so if you normally
run using the RAM serial driver, it will attempt to reload it
when you open the modem port and it will return to the ROM
drivers unconditionally when the modem port is closed.  Third,
I didn't make any special checks for AppleTalk.

(Oh, BTW, I looked at the Sumac RAMSDOpen, but it didn't
work.  I think it is based upon an older version of the
SERD resources than the one I received with the 2/85
supplement.)

I plan to stop using my RAMSDOpen and use the one from
Megamax when it finally arrives.
-- 
  John Bruner (S-1 Project, Lawrence Livermore National Laboratory)
  MILNET: jdb@mordor [jdb@s1-c.ARPA]	(415) 422-0758
  UUCP: ...!ucbvax!dual!mordor!jdb 	...!seismo!mordor!jdb