Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rutgers!cbmvax!fred
From: fred@cbmvax.commodore.com (Fred Bowen)
Newsgroups: comp.sys.cbm
Subject: Re: C128 Development package
Keywords: "P" error
Message-ID: <13943@cbmvax.commodore.com>
Date: 22 Aug 90 17:03:13 GMT
References: <13160@mentor.cc.purdue.edu>
Reply-To: fred@cbmvax (Fred Bowen)
Organization: Commodore, West Chester, PA
Lines: 40

In article <13160@mentor.cc.purdue.edu> piskacrj@mentor.cc.purdue.edu writes:
>
>   I have had only one problem with the assemble.  After using the
>   computer for long periods of time, the assembler starts to report
>   a "P" error.  The manual states that compiler is confused or the
>   disk is bad.  I know that the disk is fine.  If I shut the computer
>   off for an hour and re-assemble the source code, it works fine.
>   If I shut the computer off and immediately turn it back on, I
>   still get the same errors.  If I shut it off for a short period
>   of time and try again it works, but the "P" error will occur again.
>
>   If I put the source code on my 1750 REU using the ramdos, I still
>   get the same errors.

Yup, sounds like a heat problem, albeit a very strange one.  The 'P'hase
error simply means the value of a label was different between the two
passes of the assembler.  This is usually due the programmer changing
something he oughtn't, but since the symbol table is stored on disk it
could mean a disk media or I/O problem.  Since you wisely eliminated the
disk (and consequently disk I/O) as the culprit, it can only mean that
the contents of RAM (or the REU) where the symbols are stored is changing.

You probably aren't using the memory that's bad with other programs.  But
the problem is still very strange, and I would suspect something other than
the RAM as the problem, like probably the VIC chip which is responsible for
refreshing the RAM.

>   Does anyone know the compuserve mail address for the commodore 
>   online help person?

Dunno.  I have a Compuserve account (76702,702) which I don't use- better
to send email here I suppose, although I've not been reading net news very
much lately.  Too busy.

-- 
Fred Bowen			uucp:	{uunet|rutgers|pyramid}!cbmvax!fred
				domain:	fred@cbmvax.commodore.com
				tele:	215 431-9100

Commodore Electronics, Ltd.,  1200 Wilson Drive,  West Chester,  PA,  19380