[gnu.gcc.bug] non-obvious warning

gary%sun.mcs.clarkson.edu@MITVMA.MIT.EDU (10/22/88)

Why does the following program get warnings about floating point
number exceeding range of `double'.  0.0 seems to cause trouble.
------
Program:
------                        line numbers
                                 1
                                 2
double F(x) int x;{              3
    double r;                    4
    r = 5;                       5
                                 6
    if( r ==  0.0 )              7
        return 0.0;              8
    else                         9
        return 1.0;              10
}
----
Compiled with: gcc -v -Wall -c test.c >& $T
----
Error output:
----
 /usr/local/Gnu/lib/gcc-cpp -v -undef -D__GNU__ -D__GNUC__ -Dmc68000 -Dsun -Duni
x -Wall -D__HAVE_FPU__ -Dmc68020 test.c /tmp/cca01528.cpp
GNU CPP version 1.26
 /usr/local/Gnu/lib/gcc-cc1 /tmp/cca01528.cpp -quiet -dumpbase test.c -Wall -ver
sion -o /tmp/cca01528.s
In function F:
test.c:7: warning: floating point number exceeds range of `double'
test.c:8: warning: floating point number exceeds range of `double'
GNU C version 1.26 (68k, MIT syntax) compiled by GNU C version 1.26.
 as -mc68020 /tmp/cca01528.s -o test.o
gcc version 1.26
----

gary@SUN.MCS.CLARKSON.EDU (10/22/88)

Forgot some of the requested information:
This is run on a Sun 3/50, under Sun Unix 4.0.
I didn't install the compiler, so I don't know any other specifics.
If it is important, I can track it down.
-----
Why does the following program get warnings about floating point
number exceeding range of `double'.  0.0 seems to cause trouble.
------
Program:
------                        line numbers
                                 1
				 2
double F(x) int x;{		 3
    double r;			 4
    r = 5;			 5
				 6
    if( r ==  0.0 )		 7
	return 0.0;		 8
    else			 9
	return 1.0;		 10
}
----
Compiled with: gcc -v -Wall -c test.c >& $T
----
Error output:
----
 /usr/local/Gnu/lib/gcc-cpp -v -undef -D__GNU__ -D__GNUC__ -Dmc68000 -Dsun -Dunix -Wall -D__HAVE_FPU__ -Dmc68020 test.c /tmp/cca01528.cpp
GNU CPP version 1.26
 /usr/local/Gnu/lib/gcc-cc1 /tmp/cca01528.cpp -quiet -dumpbase test.c -Wall -version -o /tmp/cca01528.s
In function F:
test.c:7: warning: floating point number exceeds range of `double'
test.c:8: warning: floating point number exceeds range of `double'
GNU C version 1.26 (68k, MIT syntax) compiled by GNU C version 1.26.
 as -mc68020 /tmp/cca01528.s -o test.o
gcc version 1.26
----