🚀 go-pugleaf

More on my foray into compiling gcc-2.1 on a Data General AViiON,  The
original note is quoted below in full.  Also I have tried to put gcc-2.0
on a sun4 and am getting different results there.

On the AViiON, there are these lines in float.h.  I looked at enquire.c,
but could not figure out where to go.  My experience on the SUN makes me
wonder  if the new gcc is picking up incompatible header files.  I will
try to put gcc-2.1 on our sun, and get back to you on that one
separately.

FROM AViiON   excerpt from gcc-2.1 produced float.h

   /* Maximum double */
#define DBL_MAX 1.7976931348623150e+308
 
/* *** WARNING: Possibly bad output from printf above */
/*     expected value around 1.7976931348623165e308, bit pattern:
    01111111 11101111 11111111 11111111 11111111 11111111 11111111
11111111 */
/*     sscanf gave           inf, bit pattern:
    01111111 11110000 00000000 00000000 00000000 00000000 00000000
00000000 */
/*     difference= -inf */
 
   /* Maximum int x such that 10**x is a representable double */
#define DBL_MAX_10_EXP 308
 
   /* Number of base-FLT_RADIX digits in the significand of a long
double */
#define LDBL_MANT_DIG 53
   /* Number of decimal digits of precision in a long double */
#define LDBL_DIG 15
   /* Difference between 1.0 and the minimum long double greater than
1.0 */
#define LDBL_EPSILON 2.2204460492503131e-16L
   /* Minimum int x such that FLT_RADIX**(x-1) is a normalised long
double */
#define LDBL_MIN_EXP (-1021)
   /* Minimum normalised long double */
#define LDBL_MIN 2.2250738585072023e-308L
 
/* *** WARNING: Possibly bad output from printf above */
/*     expected value around 2.2250738585072010e-308, bit pattern:
    00000000 00010000 00000000 00000000 00000000 00000000 00000000
00000000 */
/*     sscanf gave           2.2250738585072019e-308, bit pattern:
    00000000 00001111 11111111 11111111 11111111 11111111 11111111
11111111 */
/*     difference= 4.9406564584124686e-324 */
 
   /* Minimum int x such that 10**x is a normalised long double */
#define LDBL_MIN_10_EXP (-307)
#define LDBL_MAX_EXP 1024that FLT_RADIX**(x-1) is a representable long
double */--More--(84%)
   /* Maximum long double */
#define LDBL_MAX 1.7976931348623150e+308L
 
/* *** WARNING: Possibly bad output from printf above */
/*     expected value around 1.7976931348623165e308, bit pattern:
    01111111 11101111 11111111 11111111 11111111 11111111 11111111
11111111 */
/*     sscanf gave           inf, bit pattern:
    01111111 11110000 00000000 00000000 00000000 00000000 00000000
00000000 */
/*     difference= -inf */
 
   /* Maximum int x such that 10**x is a representable long double */
#define LDBL_MAX_10_EXP 308


##################original bug report quoted in full #############

Hi:
I hesitate to send this because I am a non expert, and certainly have no
ideas about fixes.  There might be nothing wrong that cannot be fixed by
digging deeper than I have managed to get into
the READMEs and documentation.  Let me know if you can point me in the 
right direction.

I did detect an interest in such problems in README-bugs and I am
anxious(desperate) for a good C++ compiler, so here goes.

I am trying to install gcc-2.1.tar.Z (obtained today from alfred.carleton.ca)
on a Data General AViiON 310 running DGUX5.4.0 .  (5.4.1 should be in
the mail).  This is supposed to be SVR4 compliant.  It uses gcc 1.37 as
its C compiler.

I uncompressed and untarred gcc into its own directory gcc-2.1 and
within this directory I did the following from within the c shell:

eval `sde-target m88kbcs`
./configure m88k-dgux
make LANGUAGES=c
make stage1
make CC=stage1/gcc CFLAGS="-g -O -Bstage1/"
make stage2
make CC=stage2/gcc CFLAGS="-g -O -Bstage2/"    **************
sh
$  for file in *.o; do
>  cmp $file stage2/$file
>           done

The output from the final lines starting with the trailing parts of the
make with the *********'s appended is as follows:

        (cd tmpcopy; ar rc ../tmplibgcc.a *.o)
        rm -rf tmpcopy
        if [ -f /usr/bin/ranlib -o -f /bin/ranlib ] ; then ranlib
tmplibgcc.a; e
lse true; fi
        mv tmplibgcc.a libgcc.a
        cp ./enquire.c . > /dev/null 2>&1
*** Error code 2 (ignored)
        ./gcc  -D_M88KBCS_TARGET  -g -O -Bstage2/ -B./    -DNO_SC
-DNO_MEM -DNO_
STDDEF -O0 -I. -c enquire.c
./gcc: file path prefix `stage2/' never used
        ./gcc  -D_M88KBCS_TARGET  -g -O -Bstage2/ -B./ -I. -I.
-I./config -DCRT_
BEGIN \
          -finhibit-size-directive -g0 -c ./crtstuff.c -o crtbegin.o
./gcc: file path prefix `stage2/' never used
        ./gcc  -D_M88KBCS_TARGET  -g -O -Bstage2/ -B./ -I. -I.
-I./config -DCRT_
END \
          -finhibit-size-directive -g0 -c ./crtstuff.c -o crtend.o
./gcc: file path prefix `stage2/' never used
        ./gcc  -D_M88KBCS_TARGET  -g -O -Bstage2/ -B./  enquire.o -o
enquire
./gcc: file path prefix `stage2/' never used
        ./enquire -f > tmp-float.h
*** Error code 4 (ignored)
        mv tmp-float.h float.h
GRGS-65% sh
$  for file in *.o; do
>  cmp $file stage2/$file
>           done
aux-output.o stage2/aux-output.o differ: char 7, line 1
c-aux-info.o stage2/c-aux-info.o differ: char 7, line 1
c-common.o stage2/c-common.o differ: char 7, line 1
c-convert.o stage2/c-convert.o differ: char 7, line 1
c-decl.o stage2/c-decl.o differ: char 7, line 1
c-lang.o stage2/c-lang.o differ: char 7, line 1
c-lex.o stage2/c-lex.o differ: char 7, line 1
c-parse.o stage2/c-parse.o differ: char 7, line 1
c-typeck.o stage2/c-typeck.o differ: char 7, line 1
caller-save.o stage2/caller-save.o differ: char 7, line 1
calls.o stage2/calls.o differ: char 7, line 1
cccp.o stage2/cccp.o differ: char 7, line 1
cexp.o stage2/cexp.o differ: char 7, line 1
combine.o stage2/combine.o differ: char 7, line 1
cp-call.o stage2/cp-call.o differ: char 6, line 1
cp-class.o stage2/cp-class.o differ: char 6, line 1
cp-cvt.o stage2/cp-cvt.o differ: char 6, line 1
cp-decl.o stage2/cp-decl.o differ: char 6, line 1
cp-decl2.o stage2/cp-decl2.o differ: char 6, line 1
cp-edsel.o stage2/cp-edsel.o differ: char 6, line 1
cp-except.o stage2/cp-except.o differ: char 6, line 1
cp-expr.o stage2/cp-expr.o differ: char 6, line 1
cp-gc.o stage2/cp-gc.o differ: char 6, line 1
cp-init.o stage2/cp-init.o differ: char 6, line 1
cp-lex.o stage2/cp-lex.o differ: char 6, line 1
cp-method.o stage2/cp-method.o differ: char 6, line 1
cp-parse.o stage2/cp-parse.o differ: char 6, line 1
cp-pt.o stage2/cp-pt.o differ: char 6, line 1
cp-ptree.o stage2/cp-ptree.o differ: char 6, line 1
cp-search.o stage2/cp-search.o differ: char 6, line 1
cp-spew.o stage2/cp-spew.o differ: char 6, line 1
cp-tree.o stage2/cp-tree.o differ: char 6, line 1
cp-type2.o stage2/cp-type2.o differ: char 6, line 1
cp-typeck.o stage2/cp-typeck.o differ: char 6, line 1
cp-xref.o stage2/cp-xref.o differ: char 6, line 1
crtbegin.o stage2/crtbegin.o differ: char 6, line 1
crtend.o stage2/crtend.o differ: char 6, line 1
cse.o stage2/cse.o differ: char 7, line 1
dbxout.o stage2/dbxout.o differ: char 7, line 1
dwarfout.o stage2/dwarfout.o differ: char 7, line 1
emit-rtl.o stage2/emit-rtl.o differ: char 7, line 1
enquire.o stage2/enquire.o differ: char 6, line 1
explow.o stage2/explow.o differ: char 7, line 1
expmed.o stage2/expmed.o differ: char 7, line 1
expr.o stage2/expr.o differ: char 7, line 1
final.o stage2/final.o differ: char 7, line 1
flow.o stage2/flow.o differ: char 7, line 1
fold-const.o stage2/fold-const.o differ: char 7, line 1
function.o stage2/function.o differ: char 7, line 1
gcc.o stage2/gcc.o differ: char 6, line 1
genattr.o stage2/genattr.o differ: char 7, line 1
genattrtab.o stage2/genattrtab.o differ: char 7, line 1
gencodes.o stage2/gencodes.o differ: char 7, line 1
genconfig.o stage2/genconfig.o differ: char 7, line 1
genemit.o stage2/genemit.o differ: char 7, line 1
genextract.o stage2/genextract.o differ: char 7, line 1
genflags.o stage2/genflags.o differ: char 7, line 1
genoutput.o stage2/genoutput.o differ: char 7, line 1
genpeep.o stage2/genpeep.o differ: char 7, line 1
genrecog.o stage2/genrecog.o differ: char 7, line 1
getopt.o stage2/getopt.o differ: char 6, line 1
getopt1.o stage2/getopt1.o differ: char 6, line 1
getpwd.o stage2/getpwd.o differ: char 7, line 1
global-alloc.o stage2/global-alloc.o differ: char 7, line 1
insn-attrtab.o stage2/insn-attrtab.o differ: char 7, line 1
insn-emit.o stage2/insn-emit.o differ: char 7, line 1
insn-extract.o stage2/insn-extract.o differ: char 7, line 1
insn-output.o stage2/insn-output.o differ: char 7, line 1
insn-peep.o stage2/insn-peep.o differ: char 7, line 1
insn-recog.o stage2/insn-recog.o differ: char 7, line 1
integrate.o stage2/integrate.o differ: char 7, line 1
jump.o stage2/jump.o differ: char 7, line 1
local-alloc.o stage2/local-alloc.o differ: char 7, line 1
loop.o stage2/loop.o differ: char 7, line 1
objc-actions.o stage2/objc-actions.o differ: char 6, line 1
objc-parse.o stage2/objc-parse.o differ: char 6, line 1
obstack.o stage2/obstack.o differ: char 7, line 1
optabs.o stage2/optabs.o differ: char 7, line 1
print-rtl.o stage2/print-rtl.o differ: char 7, line 1
print-tree.o stage2/print-tree.o differ: char 7, line 1
protoize.o stage2/protoize.o differ: char 6, line 1
recog.o stage2/recog.o differ: char 7, line 1
reg-stack.o stage2/reg-stack.o differ: char 7, line 1
regclass.o stage2/regclass.o differ: char 7, line 1
reload.o stage2/reload.o differ: char 7, line 1
reload1.o stage2/reload1.o differ: char 7, line 1
reorg.o stage2/reorg.o differ: char 7, line 1
rtl.o stage2/rtl.o differ: char 7, line 1
rtlanal.o stage2/rtlanal.o differ: char 7, line 1
sched.o stage2/sched.o differ: char 7, line 1
sdbout.o stage2/sdbout.o differ: char 7, line 1
stmt.o stage2/stmt.o differ: char 7, line 1
stor-layout.o stage2/stor-layout.o differ: char 7, line 1
stupid.o stage2/stupid.o differ: char 7, line 1
toplev.o stage2/toplev.o differ: char 7, line 1
tree.o stage2/tree.o differ: char 7, line 1
unprotoize.o stage2/unprotoize.o differ: char 6, line 1
unroll.o stage2/unroll.o differ: char 7, line 1
varasm.o stage2/varasm.o differ: char 7, line 1
version.o stage2/version.o differ: char 7, line 1
xcoffout.o stage2/xcoffout.o differ: char 7, line 1
$              (^D to move mack to c-shell)
GRGS-66%

  There were no other suspicious parts (to my eye) from the make
output.
##################################################
in the process of the install i.e.:
  make CC=stage2/gcc install

the following lines appear:

        cp ./byteorder.h .
cp: ./byteorder.h and ./byteorder.h are identical
*** Error code 2 (ignored)

which do not worry me much (should they?)

###################################################
when I try to compile a program (which I wrote so it could be horrible,
but it compiles and runs fine under gcc 1.37 as delivered with the
AViiON and other C compilers), I get the following error:

 symbol                             in file
___iob                              felinprp.o
ld fatal: Symbol referencing errors. No output written to a.out

I do not reference anything called iob.  A grep -i iob *   comes up
empty.

#####################################################
The above is essentially the same as my attempts 2 weeks ago with
gcc-2.0.

Questions:
1. Does thie above mean anything to you?
2. Have I done the right steps?
3. Is there anything else I can try?

I would be happy to put some effort into this to help, but remember I am
somewhat of a neophyte here, and it takes 1 hour or more to do a make.

Any help advice or even an acknowledgement of this mail will be greatly
appreciated.

Thank you and gnu for all the good things that are out there for us.
 
Dave Greenberg
-- 
_______________________________________________________________________
|David A. Greenberg                    email dgreenbe@sable.bio.dfo.ca|
|Bedford Institute of Oceanography     OMNET c/o J.LODER              | 
|P.O. Box 1006                         phone (902) 426-2431           | 
|Dartmouth, Nova Scotia                fax   (902) 426-7827           | 
|CANADA B2Y-4A2                        home  (902) 865-7511           |
_______________________________________________________________________