Path: utzoo!mnetor!uunet!husc6!mailrus!nrl-cmf!ames!amdahl!nsc!voder!blia!ted
From: ted@blia.BLI.COM (Ted Marshall)
Newsgroups: comp.os.vms
Subject: IOSBs on the stack (was Re: 2 C questions)
Message-ID: <4500@blia.BLI.COM>
Date: 13 Apr 88 17:05:21 GMT
References: <281@cullsj.UUCP> <282@cullsj.UUCP>
Organization: Britton Lee, Los Gatos, CA
Lines: 21
Summary: IOSBs on the stack

In article <282@cullsj.UUCP>, jayz@cullsj.UUCP (Jay Zorzy) writes:
> Speaking of stacks, here's a common gotcha to watch out for:  If you're 
> issuing any asynchronous $QIOs (no wait, I/O completion handled by AST
> routine), make sure to declare your IOSB variables as static.  Otherwise,
> once the calling routine returns, the stack is reset; an IOSB on the stack
> may now point to never-never land.

This can bite you even if you use QIOW, IF you have an condition handler
which unwinds the stack. This bit me. In our VMS host code, the QIOW
read sits waiting for data to come back from the database server. If the
user types ^C, The AST routine signals a condition and the handler
unwinds the stack before cancelling the QIO (actually this was with our
private exception handler instead of the standard VMS condition handler,
but it's the same difference). I fixed that by making the IOSBs static.

Of course, be careful about making things static if the routine is recusive!

-- 
Ted Marshall       ...!ucbvax!mtxinu!blia!ted <or> mtxinu!blia!ted@Berkeley.EDU
Britton Lee, Inc., 14600 Winchester Blvd, Los Gatos, Ca 95030     (408)378-7000
The opinions expressed above are those of the poster and not his employer.