Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!lsuc!jimomura
From: jimomura@lsuc.UUCP
Newsgroups: comp.sys.m6809
Subject: Re: semantics of read()
Message-ID: <2200@lsuc.UUCP>
Date: Mon, 16-Nov-87 21:25:16 EST
Article-I.D.: lsuc.2200
Posted: Mon Nov 16 21:25:16 1987
Date-Received: Wed, 18-Nov-87 04:19:30 EST
References: <553@mcrware.UUCP>
Reply-To: jimomura@lsuc.UUCP (Jim Omura)
Organization: Consultant, Toronto
Lines: 45
Keywords: OS-9, Nuxi, read()
Summary: Updated 'gs_rdy.a'

In article <553@mcrware.UUCP> jejones@mcrware.UUCP (James Jones) writes:

...

>and for things such as terminal programs, file transfer programs, and the
>like in OS-9, one can make profitable use of the SS_SIG and SS_RDY setstat
>and getstat calls.  (I must admit to never having figured out why it is
>that not having any data ready is thought to be an error (E$NRdy), but them's
>the breaks.)  If you didn't grab Jim Omura's posting, you might want to.

     Actually, Mark Ackfield pointed out an error in my code which
could cause an occasional error due to a corrupted stack.  The
version I'm now using is as follows:

 psect gs_rdy_c,0,0,0,0,0
 nam gs_rdy_c
 ttl _gs_rdy
_gs_rdy: pshs u
 ldd #_1
 lbsr _stkcheck
 leas -2,s  Set stack pointer at "waiting"
 lda 7,s    load "path" -- actually ldd 6 and shift left 8.
 ldb #1
 os9 $8d    I$GetStt
 bcs errhand
 clra clear upper byte to make sense.
 bra cleanexit
errhand ldd #-1
cleanexit std 0,s Store the Result in    ""waiting"
* comment out next 2 lines to save program from
* crashing if interrupt occurs at this point:
* leas 2,s  Restore stack pointer.
* leas -2,s
 ldd 0,s
 leas 2,s
 puls u,pc
_1 equ -66
 endsect


Cheers -- Jim O.
-- 
Jim Omura, 2A King George's Drive, Toronto, (416) 652-3880
ihnp4!utzoo!lsuc!jimomura
Byte Information eXchange: jimomura