[Gs-code-review] Re: Proposed fix for SourceForge #508722

[L. Peter Deutsch]

> I'm not in favor. A broken memmove() is a very basic bug.

OK.

> I will, however, accept a patch that _detects_ this compiler bug and
> prints a warning about it, possibly for genarch.c.

Forrest Cahoon said he was looking for a simple test case.  I don't have
access to an OpenVMS system myself to develop one.

Ghostscript does not print warnings for other environmental bugs.  For that
matter, neither does 'configure'.  It simply works around them.  Example:
the detection of the gcc code generation bug that decides what switches to
write into cc.tr.

The cc.tr approach can be applied to this bug, if OpenVMS has something
equivalent to the Unix `...` construct so that genarch can write either
/OPTIMIZE or /NOOPTIMIZE into a file that is then incorporated into the
compiler command line.  If an OpenVMS expert can tell me how to do that,
that's what I recommend.

> If the code works without -O, that should be the recommended workaround
> for now.

I will submit a patch that removes optimization from the OpenVMS builds.

-- 

L. Peter Deutsch     |    Aladdin Enterprises   |  203 Santa Margarita Ave.
ghost at aladdin.com    |  http://www.aladdin.com  |  Menlo Park, CA 94025

	The future of software is at http://www.opensource.org