Re: Glibc2 (was Re: [HACKERS] PostgreSQL - the Linux of Databases...)

In message <Pine.LNX.3.96.980307231533.20654E-100000@shefu.redhat.com>, Cristian Gafton w
rites:
> On Sat, 7 Mar 1998, Thomas A. Szybist wrote:
>
> > > > Thanks Cristian! What was the secret? Was it the compiler, or -O setting, or ??
> > >
> > > You're welcome !
> > >
> > > I just got a newer snapshot and recompiled it.
> >
> > Of what? Glibc?
>
> No, postgresql, because that is what we were talking about...
> Despite the subject of the thread.
>
> Cristian
> --
> ----------------------------------------------------------------------
> Cristian Gafton   --   gafton@redhat.com   --   Red Hat Software, Inc.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>  UNIX is user friendly. It's just selective about who its friends are.
>
>
>
>
Sorry, I guess I lost track of the snapshots you were packaging.

I think __math.h is the file.  Here is small diff (cut & paste) from
2.0.5 to 2.0.7:

*** __math.h205 Sun Mar  8 00:00:29 1998
--- __math.h207 Tue Feb 24 16:34:38 1998


<snip>

    ***************
    *** 21,30 ****
      #ifndef __MATH_H
      #define __MATH_H      1

    ! #if defined __GNUG__ && \
    !     (__GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ <= 7))
    ! /* gcc 2.7.2 and 2.7.2.1 have problems with inlining `long double'
    !    functions so we disable this now.  */
      #undef __NO_MATH_INLINES
      #define __NO_MATH_INLINES
      #endif
    --- 21,29 ----
      #ifndef __MATH_H
      #define __MATH_H      1

    ! #if (__GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ <= 7))
    ! /* The gcc, version 2.7 or below, has problems with all this inlining
    !    code.  So disable it for this version of the compiler.  */
      #undef __NO_MATH_INLINES
      #define __NO_MATH_INLINES
      #endif
    ***************
    *** 484,492 ****


It seems __NO_MATH_INLINES must be defined.  2.0.5 tests for __GNUG__.
2.0.7 does not, but only checks for __GNUC_MINOR__ <= 7. This would
explain why 2.8.0 also fails.


Tom Szybist

szybist@boxhill.com