Re: GCC-4 compiler bug

["Simon 'corecode' Schubert" <corecode@xxxxxxxxxxxx>]

Matthew D. Fuller wrote:
> > Important to note is that this only happens when compiling with
> > optimization (they claim).
>
>
> % cc -o test test.c
> % ./test
> X
> % cc -O -o test test.c
> % ./test
> X
> % cc -O2 -o test test.c
> % ./test
> %
>
> gcc version 4.2.1 20070719  [FreeBSD]

I can confirm this on ubuntu:

gcc 4.1 w/o optimization:
no overflow (= fail)

gcc 4.1 w/o opt && -fwrapv:
overflow (= good)

gcc 4.2 w/o opt:
overflow

gcc 4.2 w/ opt:
no overflow

gcc 4.2 w/ opt && -fwrapv:
overflow

So no matter whether we upgrade gcc or not, we should change the default optimization to behave as if -fwrapv was passed.  Or we emit warnings whenever this optimization is taking place.  Or both :)

cheers
 simon

--
Serve - BSD     +++  RENT this banner advert  +++    ASCII Ribbon   /"\
Work - Mac      +++  space for low €€€ NOW!1  +++      Campaign     \ /
Party Enjoy Relax   |   http://dragonflybsd.org      Against  HTML   \
Dude 2c 2 the max   !   http://golden-apple.biz       Mail + News   / \