Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!decwrl!sun!pitstop!sundc!seismo!uunet!mcvax!hp4nl!kunivv1!wn2!sommel
From: sommel@wn2.sci.kun.nl (Ron Sommeling)
Newsgroups: comp.sys.atari.st
Subject: Re: Mfpint()
Summary: BIOS/XBIOS is not reentrant
Message-ID: <336@wn2.sci.kun.nl>
Date: 17 Feb 89 10:52:32 GMT
References: <8902140121.AA03228@ucbvax.Berkeley.EDU> <825@per2.UUCP>
Organization: University of Nijmegen, The Netherlands.
Lines: 19

In article <825@per2.UUCP>, dag@per2.UUCP (Daniel A. Glasser) writes:
> 
> Although, technically, you can call BIOS and XBIOS functions from interrupt
> level routines, it is not a good idea in general.  It is better to set flags
> for the application "user-mode" routines to deal with.

No, you can't safely call any BIOS or XBIOS function from within an
interrupt. Take a look at the trap 13/14 trap-handler! Some registers
are copied to an area pointed to by saveptr and then saveptr is updated.
If an interrupt occurs after saving some registers in this area but
before saveptr is updated, then the saved registers will be overwritten
if you call BIOS/XBIOS routines in that interrupt. I can't see why
Atari decided to save the registers in the area pointed to by saveptr
instead of on the (supervisor)stack. If they did the latter then
BIOS/XBIOS would be reentrant.

						Ron Sommeling

email: sommel@wn2.sci.kun.nl