[Bug fortran/32140] [4.3 Regression] Miscompilation with -O1

["rguenther at suse dot de" <gcc-bugzilla@gcc.gnu.org>]

------- Comment #27 from rguenther at suse dot de  2007-06-20 14:07 -------
Subject: Re:  [4.3 Regression] Miscompilation with -O1

On Wed, 20 Jun 2007, jv244 at cam dot ac dot uk wrote:

> 
> 
> ------- Comment #24 from jv244 at cam dot ac dot uk  2007-06-20 14:03 -------
> (In reply to comment #22)
> > It would be nice if a fortraner could make the testcase not output to
> > stdout/err
> > but to /dev/null instead.
> > 
> > open(6,'/dev/null')
> > 
> > didn't work for me ;)
> > 
> > Just to make a testcase suitable for the testsuite.
> 
> The following testcase, for an unpatched compiler, aborts at -O1, works fine at
> -O0, and doesn't write to stdout/stderr... I guess that fits the testsuite ?

Yep, I'll use that.

Thanks,
Richard.

> MODULE TEST
> CONTAINS
> PURE FUNCTION s2a_3(s1,s2,s3) RESULT(a)
>     CHARACTER(LEN=*), INTENT(IN)             :: s1, s2, s3
>     CHARACTER(LEN=4), DIMENSION(3)        :: a
> 
>   a(1)=s1; a(2)=s2; a(3)=s3
> END FUNCTION
> END MODULE
> 
> USE TEST
> character(len=12) :: line
> write(line,'(3A4)') s2a_3("a","bb","ccc")
> IF (line.NE."a   bb  ccc") CALL ABORT()
> END
> 
> 
> 


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32140