user42 at zip.com.au
Fri Jan 23 09:39:36 CET 2004
Dennis Lubert <plasmahh at gmx.net> writes:
> configure:19193: gcc -g -O2 -m64 -mptr64 -Wa,-xarch=v9 -mcpu=v9 -c
> conftes1.c >&5
> && gcc -c -g -O2 -m64 -mptr64 -Wa,-xarch=v9 -mcpu=v9 conftes2.s >&5
> && gcc -g -O2
> -m64 -mptr64 -Wa,-xarch=v9 -mcpu=v9 conftes1.o conftes2.o >&5
> conftes2.o(.stab+0x14): relocation truncated to fit: R_SPARC_32 .text
> collect2: ld returned 1 exit status
Hmm. Doesn't look good. Hard to say what's wrong. It looks like all
the right -m64 and -Wa are there to get 64-bit mode. Does a "hello
world" program work with that gcc in 64-bits?
You can force the test with for instance
assuming it's a non-underscore system. But I'd guess if it has
trouble at that stage then it'll bomb later on too.
You might get more joy from ABI=32, though it'll run slower.
All followups to gmp-bugs at swox.com please.
More information about the gmp-bugs