[net.unix-wizards] machine dependent #ifdef's

dave@uwvax.UUCP (Dave Cohrs) (09/18/85)

There are machine-dependent #ifdef's for most cpu's that 4.2 runs on (e.g.
#ifdef vax, or #ifdef pyr).  Does anyone know the ifdef for a Gould
9080 Powernode (it runs what they call UTX/32)?  I've tried 'strings' on cpp,
but it doesn't give any indication.

Thanks.

-- 
Dave Cohrs
(608) 262-1204
...!{harvard,ihnp4,seismo,topaz}!uwvax!dave
dave@wisc-romano.arpa

davy@pur-ee.UUCP (Dave Curry) (09/19/85)

The #ifdef for Gould machines is "#ifdef sel".  There's another one,
"selport", but I don't think its use is encouraged.

--Dave Curry
Purdue ECN

willcox@ccvaxa.UUCP (09/20/85)

To clarify, the define "sel" (derived from Systems Engineering Labs,
the former name of Gould's Computer Systems Div.) is used to identify
hardware dependencies.  The define "selport" identifies changes that
were made to the original BSD code that were not hardware dependencies.
These are enhancements or fixes for bugs that appeared in the code
as we got it from Berkeley.

You should use "ifdef sel", not "ifdef selport", although both are
currently defined by cpp.


David A. Willcox
Gould CSD-Urbana
1101 E. University Ave.
Urbana, IL 61801
217-384-8500
{decvax!pur-ee,ihnp4}!uiucdcs!ccvaxa!willcox

barry@ucla-cs.UUCP (09/21/85)

In article <313@uwvax.UUCP> dave@uwvax.UUCP (Dave Cohrs) writes:
>There are machine-dependent #ifdef's for most cpu's that 4.2 runs on (e.g.
>#ifdef vax, or #ifdef pyr).  Does anyone know the ifdef for a Gould
>9080 Powernode (it runs what they call UTX/32)?  I've tried 'strings' on cpp,
>but it doesn't give any indication.
>
>Thanks.
>
>-- 
>Dave Cohrs
>(608) 262-1204
>...!{harvard,ihnp4,seismo,topaz}!uwvax!dave
>dave@wisc-romano.arpa



#ifdef sel

Barry Lustig		   UCLA Center for Experimental Computer Science  
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ARPA:	barry@LOCUS.UCLA.EDU                                           ~
UUCP:	{ihnp4, cepu, sdcrdcf}!ucla-cs!barry                           ~
PHONE:	(213) 825-5170                                                 ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


-- 
Barry Lustig		   UCLA Center for Experimental Computer Science  
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
ARPA:	barry@ucla-locus.ARPA  (soon to be)  barry@LOCUS.UCLA.EDU      ~
UUCP:	{ihnp4, cepu, sdcrdcf}!ucla-cs!barry                           ~
PHONE:	(213) 825-5170                                                 ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

dave@uwvax.UUCP (Dave Cohrs) (09/21/85)

Thanks for all of the responses for the Gould UTX #ifdef.  Two were
reported, sel and selport.  The latter does not work on our system so I
assume it is being fazed out.  These exist for historical reasons (that's
such a wonderful excuse) -- they have to do with the name of the company
which Gould acquired.  Anyway, I also heard that Gould wants to do away
with this stuff completely and change it to gould or something.  Oh --
the BRL System V emulation stuff uses '#ifdef gould' instead of '#ifdef sel'.

[ With all of this unsubstantiated material, this should go in net.rumor ]

-- 
Dave Cohrs
(608) 262-1204
...!{harvard,ihnp4,seismo,topaz}!uwvax!dave
dave@wisc-romano.arpa