Experimental Physics and
Industrial Control System

Tony Cox - (415)926-3105 <[email protected]> · Fri, 08 Dec 1995 10:57:21 PST

Rozelle writes:-

>Hi Greg,
>I built freegdct under Solaris.  The Makefile.start that is currently being distributed needs
>to be changed to 
>	@case "$(HOST_ARCH)" in \
>	sun4|solaris) \
>	
>instead of just sun4.  And you have to have a decent copy of interviews.
>I got mine from Jim Kowalski at Argonne.  He had modified it
>so that it worked with Solaris.

Note that the Alpha/Unix version of interviews from interviews.stanford.edu
is badly broken. In fact, it doesn't seem to work at all. I fixed the problems
with Interviews, essentially from scratch, at least as far as providing
sufficient functionality to get free_gdct working.

For those porting to other architectures (esp. 64 bit ones), look out for
the usual confusion between `int's and `long's. A wonderful `gotcha' in
C++ is that class definitions like this:-

	class rubbish : public otherRubbish {
		.....

		void someFunction( int, ....);
		....
	}

do not reveal the important fact that 'someFunction' may be declared to be 
virtual in a superclass. Hence, simply changing `int' to `long' completely
changes the operation of the program - silently, since C++ compilers see
these routines as completely different & most don't bother to inform you that
'someFunction' is masking a virtual function further up the class hierarchy.

InterViews is full of stuff like that. If anyone wants the (pseudo)working
Alpha/Unix Interviews, I'll be glad to provide...

Tony

--------------------------------------------------------------------------------
Dr Anthony D Cox
Computer Systems Specialist
Stanford Synchrotron Radiation Laboratory
Stanford Linear Accelerator Center
MS 69, Box 4349
Stanford CA 94305
[email protected]
--------------------------------------------------------------------------------

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System