This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
compilation time of block data
- From: Bud Davis <bdavis9659 at sbcglobal dot net>
- To: fortran at gcc dot gnu dot org
- Date: Fri, 31 Mar 2006 11:12:13 -0800 (PST)
- Subject: compilation time of block data
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=sbcglobal.net; h=Message-ID:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=nql5X4CQnbZs/VJKLg72k1V+XdtTQtuqEfryu0PbNry7tJFp891WDjr7UL4j7t9+7uBQUrBU0OxjwK3ja5dcrU1oxFR2xd5VrIq4Dka4CLlMQ6XFBM5FYDmqxyAcOaSDTKUhz2EaE/WmbNCH5GT5akmFHmxFvYBsEZIUBZKIpC8= ;
this is a problem that we used to have with g77; and a
similar problem exists with gfortran.
i have some idle curiosity as to how good other
compilers do with this code. anyone on the list have
a few spare cycles to give this example a try with
some other compilers ?
gfortran takes a bit over 2 hours to compile this
example.
$ cat c.f
integer size / 250000 /
print*,' BLOCK DATA'
print*,' REAL R(',SIZE,')'
print*,' COMMON /ONE/ R'
print*,' DATA R / 0.500'
DO I = 1,SIZE-1
print*,' 1 ,',RAND() * 10000
ENDDO
print*,' 1/'
print*,' END'
end
$ gfortran c.f
$ ./a.out >d.f
$ time gfortran -c d.f
i plan on looking at this in gfortran, unless it is
too hard (or has recursice routines !!)
--bud