This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Status of bug 12366
- From: "Neilen Marais" <nmarais at snowisp dot com>
- To: <fortran at gcc dot gnu dot org>
- Date: Tue, 15 Mar 2005 11:27:44 +0200
- Subject: Status of bug 12366
Hi.
I'm busy porting some f90 code to linux using a recent CVS gcc, but I'm
getting an ICE on some code. I was about to submit a bug, but it seems
that what I found is a duplicate of bug 12366. My test case looks almost
exactly like the one described in bug 18379, which has been marked as a
duplicate of 12366.
It seems bug 12366 has been around for an awfully long time. Is this a
particularly difficult problem to solve, or is it dependant on some
other particularly hairy gcc problems being fixed? What I'd like to know
is if I should hold my breath for it being fixed soon, or if I should
just try to incorporate the workaround in the code I'm working on.
Anyway, to save a trip to bugzilla, the bug 18379 test case looks like
this:
while the code
PROGRAM TEST
IMPLICIT NONE
INTEGER :: I(3)
REAL(8) :: XTEST(5)
I = (/ 1, 3, 5 /)
XTEST(I) = 5.
WRITE(*,*) XTEST
END
works, the slightly modified one
PROGRAM TEST
IMPLICIT NONE
REAL(8) :: XTEST(5)
XTEST((/ 1, 3, 5 /)) = 5.
WRITE(*,*) XTEST
END
gives the error test.f:4: internal compiler error: in gfc_conv_ss_descriptor,
at fortran/trans-array.c:1264
Thanks
Neilen