🚀 go-pugleaf

Hello,

I have a strange problem with the gcc-1.40 on our DEC3100:
All compiles correct in stage1 and stage2 and everything seems to be o.k.,
but if I test the compiler compiling a quite large program, there is no
problem with producing the *.o files, but the linkling fails:

gcc  -o rap rap.o string.o    comint.o    debug.o     taskeditor.o  system.o glo
bals.o   filehnd.o   error.o     dag.o       table.o scanner.o   ccheck.o    par
ser.o    unparser.o  comdef.o comdec.o    cp.o        eval.o      termop.o    ru
les.o equ.o       goals.o     termination.o cnarrow.o
ld:
table.o: jump relocation out-of-range, bad object file produced, can't jump from
 0x41746c to 0x10006210 (abort)
table.o: jump relocation out-of-range, bad object file produced, can't jump from
 0x419740 to 0x10006210 (abort)
/public/lib/gnu/gcc/gcc-gnulib(_eprintf.o): jump relocation out-of-range, bad ob
ject file produced, can't jump from 0x43fd80 to 0x10006210 (abort)
/lib/libc.a(setjmp.o): jump relocation out-of-range, bad object file produced, c
an't jump from 0x441a84 to 0x10006210 (abort)
/lib/libc.a(setjmp.o): jump relocation out-of-range, bad object file produced, c
an't jump from 0x441b64 to 0x10006210 (abort)
/lib/libc.a(_exit.o): jump relocation out-of-range, bad object file produced, ca
n't jump from 0x4442e8 to 0x10006210 (abort)
/lib/libc.a(getsvc.o): jump relocation out-of-range, bad object file produced, c
an't jump from 0x4481a4 to 0x100068b0 (stat)
/lib/libc.a(res_init.o): jump relocation out-of-range, bad object file produced,
 can't jump from 0x44e308 to 0x100068b0 (stat)
make: *** [rap] Error 1

Can anyone give me a hint, how to solve this problem ?

Further information I can give is, that the  binaries of the stage2 and stage3
differ, the comparsion even fails, if I use the `ecoff-cmp' script !
The strip is impossible:
	strip: Error: c-convert.o contains relocation entries-- not stripped
	strip: Error: c-decl.o contains relocation entries-- not stripped
	strip: Error: c-parse.tab.o contains relocation entries-- not stripped
	...

It would be very kind, if you can post a solution or mail directly to me:

					Stefan Schmid
					FORWISS Passau, Germany

					sschmid@forwiss.uni-passau.de


-- 
Stefan Schmid			sschmid@forwiss.uni-passau.de
FORWISS Passau			Universitaet Passau