Path: utzoo!attcan!uunet!lll-winken!lll-lcc!ncis.llnl.gov!helios.ee.lbl.gov!pasteur!ucbvax!WPI.BITNET!GREYELF
From: GREYELF@WPI.BITNET
Newsgroups: comp.sys.apple
Subject: brun
Message-ID: <8902171716.AA02247@wpi>
Date: 17 Feb 89 17:16:25 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 28

>>And no, brunning is NOT the same as bloading and calling, due to the vectors
>>dos sets up.

>But only if you specify an alternate position and/or length...

>UUCP: :uunet!rosevax, amdahl!bungia, chinet, killer:!orbit!pnet51!shawn
>INET: shawn@pnet51.cts.com

NO, NO, NO.  DOS sets up an additional vector that records the fact that
a BRUN command was done.  Somewhere in DOS the brun command is executed
from assembly language and the return address for the JSR is stored on the
stack.  When you call a routine in memory the memory access is handled
by BASIC instead and corresponds to a JSR instruction.

Some programs were meant to be BRUN and either save their entry information,
or exit to BASIC via a JSR $3D0.  Binary programs that don't
end with a JSR to $3D0 would work fine if called, but if BRUN
will hang.

e.g. LOADER.OBJ0.

+------------------------------------------------------------------+
|    Michael J Pender Jr         snail mail: Box 1942 c/o W.P.I.   |
|   Bitnet: greyelf@wpi.bitnet               100 Institute Rd.     |
| Internet: greyelf@wpi.wpi.edu              Worcester, Ma 01609   |
| User #20 at H.U.G.E. Bit Bucket                                  |
| Disclaimer:  Who me?  I was in Mexico buying drugs at the time!  |
+------------------------------------------------------------------+