[gfortran] Fix list directed reads of real number
Bud Davis
bdavis9659@comcast.net
Thu Dec 9 10:14:00 GMT 2004
On Thu, 2004-12-09 at 00:25, Steve Kargl wrote:
> The attached patch fixes a bug in the list directed read of
> several values from a single line. This patch has been
> bootstrapped and regression tested on i386-*-FreeBSD6.0.
> It actually fixes 4 failures in the testsuite.
>
> === gfortran Summary ===
>
> # of expected passes 4739
> # of unexpected failures 1
> # of unexpected successes 8
> # of expected failures 18
> # of unsupported tests 8
>
> These are the best results I've seen on FreeBSD. I've also
> include a test case for the testsuite.
>
>
> 2004-12-08 Steven G. Kargl <kargls@comcast.net>
>
> * io/list_read.c (read_real): Don't return early from function.
steve,
with your patch i get no change in the testsuite on i686/gnu/linux:
before:
=== gfortran Summary ===
# of expected passes 4757
# of unexpected failures 1
# of unexpected successes 5
# of expected failures 21
/usr/work/2004344/ibin/gcc/testsuite/../gfortran version 4.0.0 20041209 (experimental)
after:
=== gfortran Summary ===
# of expected passes 4757
# of unexpected failures 1
# of unexpected successes 5
# of expected failures 21
/home/work/gfortran/current/bin/gcc/testsuite/../gfortran version 4.0.0 20041209 (experimental)
i am running it on freebsd now, but it takes a few hours. real curious
as to what it fixes in those test failures, probably something new to be
discovered in chasing that down.
--bud
More information about the Fortran
mailing list