Thread: Re: [HACKERS] Call for platforms

On Thu, Mar 22, 2001 at 07:58:04PM +0000, Patrick Welche wrote:
> On Fri, Mar 23, 2001 at 06:25:50AM +1100, Giles Lean wrote:
> >
> > > PS: AFAIK geometry-positive-zeros-bsd works for all NetBSD platforms - the
> > > above difference is only for i386 + fpu.
> >
> > It doesn't on NetBSD-1.5/alpha -- there geometry-positive-zeros is
> > correct.
>
> Sorry, that should have read:
>
> AFAIK geometry-positive-zeros works for all NetBSD platforms - the
> above difference is only for i386 + fpu.

Seems that following patch is needed.  Now It Works For Me (tm).
Giles, does the regress test now succed for you?

--
marko


Index: src/test/regress/resultmap
===================================================================
RCS file: /home/projects/pgsql/cvsroot/pgsql/src/test/regress/resultmap,v
retrieving revision 1.45
diff -u -r1.45 resultmap
--- src/test/regress/resultmap    2001/03/22 15:13:18    1.45
+++ src/test/regress/resultmap    2001/03/22 17:29:49
@@ -17,6 +17,7 @@
 geometry/.*-openbsd=geometry-positive-zeros-bsd
 geometry/.*-irix6=geometry-irix
 geometry/.*-netbsd=geometry-positive-zeros
+geometry/i.86-.*-netbsdelf1.5=geometry-positive-zeros-bsd
 geometry/.*-sysv5uw7.*:cc=geometry-uw7-cc
 geometry/.*-sysv5uw7.*:gcc=geometry-uw7-gcc
 geometry/alpha.*-dec-osf=geometry-alpha-precision

On Fri, Apr 13, 2001 at 01:25:45PM +0000, Thomas Lockhart wrote:
> > Did we decide that "most NetBSD/i386 users have fpus" in which case Marko's
> > patch should be applied?
>
> I'm unclear on what y'all mean by "i386 + fpu", especially since NetBSD
> seems to insist on calling every Intel processor a "i386".

History ;-)

> In this case,
> are you suggesting that this patch covers all NetBSD installations on
> every Intel processor from i386 + fpu forward to i486, i586, etc etc?

Yes! It's simply, if the peecee type thing has a fpu (as in the sysctl
machdep.fpu_present returns 1), then libm387.so is used, and you get
differences in the (from memory 44th insignificant figure?) otherwise it
just uses libm.so and you get what is currently correct in resultmap.

Cheers,

Patrick