home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.programmer
- Path: chang.unx.sas.com!walker
- From: walker@chang.unx.sas.com (Doug Walker)
- Subject: Re: SAS/C 6.56 bug?
- Sender: news@unx.sas.com (Noter of Newsworthy Events)
- Message-ID: <Dotwon.FGn@unx.sas.com>
- Date: Mon, 25 Mar 1996 15:07:35 GMT
- X-Nntp-Posting-Host: chang.unx.sas.com
- References: <DA0E86A1@infoservice.it> <4iunfr$laf@nic.wi.leidenuniv.nl>
- Organization: SAS Institute Inc.
-
- In article <4iunfr$laf@nic.wi.leidenuniv.nl>,
- J.T.Vermeulen <jvermeul@wi.leidenuniv.nl> wrote:
- >In article <DA0E86A1@infoservice.it>,
- >Paolo Menichetti <ik34p2m@infoservice.it> wrote:
- >>Hi,
- >>
- >>I think that SAS/C compiler have a bug in the SC command. I've installed
- >>VMM and when I try to compile with ERRORCONSOLE active (I use KingCon),
- >>the SC program run in stack overflow (StackMon show 1.6MB of stack used!).
- >>When I compile with NOERRORCONSOLE, it works fine. I run the compiler
- >>with the Build icon from Workbench.
- >>
- >>Solutions?
- >
- >No solution because there is no problem.
- >
- >SAS/C probably uses stack extension internally; if there is not enough
- >stack it does a StackSwap() and StackMon cannot track this. It just
- >compares the stack pointer to the process stack bottom, not to the current
- >stack. Of course the difference between the original location of the
- >process stack and the newly allocated stack can be just about anything,
- >so StackMon reports garbage.
- >
- >Doug, am I right?
- >
-
- Yes, SAS/C uses stack extension internally on some routines (mostly in
- the global optimizer, but I think some routines in the code generator.)
- It is certainly possible that the stack gets swapped during a compile.
- If your debugging tools don't handle this, they will falsely detect a
- stack overflow.
-
- --
- *****
- *|_o_o|\\ Doug Walker walker@unx.sas.com
- *|. o.| ||
- | o |// Any opinions are mine, not those of SAS Institute, Inc.
- ======
-