home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.m6809
- Path: sparky!uunet!psinntp!telesci!kindred
- From: kindred@telesciences.com (David L Kindred (Dave))
- Subject: Re: Stack problem.. is it a hardware or software fault?
- In-Reply-To: steveh@tasman.cc.utas.edu.au's message of Tue, 8 Dec 1992 23:16:38 GMT
- Message-ID: <KINDRED.92Dec16211435@pyrite.telesciences.com>
- Sender: news@telesciences.com
- Nntp-Posting-Host: pyrite
- Organization: Telesciences CO Systems, Inc.
- References: <steveh.723856598@tasman>
- Date: Thu, 17 Dec 1992 02:14:35 GMT
- Lines: 35
-
- In article <steveh.723856598@tasman> steveh@tasman.cc.utas.edu.au (Steve Howell) writes:
-
- Hello peoples.. ive run into a major problem with one of our 6809
- designs, and cannot figure what exactly is going on... we have a 6809 that
- crashes if an interupt occurs (of either firq and irq), and that it fails
- to maintain its stack correctly.. to reproduce the error, if we push the B
- register with a value, then push a 16 bit X on top, then pull X off, then B off,
- then B will be corrupt!.. is it RAM.. nope.. Ram works and passes every test!
- Is it the IRQ hardware line or a device interupting?.. nope. using a clean
- tight, as well as know good devices, the interupt receives a cleanly shaped
- high to low pulse. Blam!.. still dies...
- Is it ROM?.. why? everything else works fine.. including a full on 16k
- sized program (which is IRQless and B register less) it has no problems....
-
- One thing to remember is that the stacking operations perform multiple
- consecutive memory accesses. Normal instruction processing will
- provide many, many dead cycles, and of course the alternation
- between RAM and ROM. I have seen memory designs (although they were
- usually using dynamic RAMs) that would not function properly for more
- than one or two consecutive cycles. This could easily be checked with
- the following code:
-
- pshs a,b,cc,dp,x,y,u
- puls a,b,cc,dp,x,y,u
-
- and check the register contents.
-
-
- - Dave
- --
-
- EMail: kindred@telesciences.com
- CI$: 72456,3226 (72456.3226@compuserve.com)
- Phone: +1 609 866 1000 x222
- Snail: TeleSciences C O Systems, 351 New Albany Rd, Moorestown, NJ 08057-1177
-