>Number:         8663
>Category:       bin
>Synopsis:       g77 is buggy
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    bin-bug-people (Utility Bug People)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Oct 21 17:42:01 1999
>Last-Modified:
>Originator:     Dan McMahill
>Organization:
>Release:        1.4.1
>Environment:
NetBSD cowboy-burt 1.4.1 NetBSD 1.4.1 (COWBOY-BURT) #1: Sat Oct  2 21:08:08 EDT 1999     dan@cowboy-burt:/amd/dinah-moe/export/src/sys/arch/sparc/compile/COWBOY-BURT sparc
NetBSD dinah-moe 1.4.1 NetBSD 1.4.1 (DINAH-MOE) #0: Sat Oct  2 02:34:31 EDT 1999     dan@dinah-moe:/export/src/sys/arch/mac68k/compile/DINAH-MOE mac68k


>Description:
the in tree fortran compiler (f77, g77, fort77) part of egcs seems
to have some problems on sparc and m68k platforms.  The code compiles
but in the case of sparc, crashes, and in the case of m68k, doesn't
work right.



>How-To-Repeat:
download the math/scilab package.  Edit 'Makefile' and comment out the
line "CONFIGURE_ARGS+=        --with-f2c" and do 'make'.  This will build
scilab using 'f77' for the fortran compiler.  Then
cd work/scilab-2.4.1/tests
make tests
cd ../examples
make tests

on the sparc, some of the tests fail saying "error detected".  those
cases correspond to scilab crashing.  On the m68k, some of the tests
just fail to produce the correct output.

This doesn't happen on i386 (all tests pass).  Also, compiling with
f2c makes all tests pass.

I'm not 100% sure that this is  a bug in g77, although it seems likely.
It is also possible some of the fortran libraries are less than portable,
however since it works correctly on i386, I'm suspicious.
>Fix:
unknown

>Audit-Trail:
>Unformatted: