This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug libfortran/70279] New: Crash duing compilation with NINT and 128 bits integer
- From: "francois.willot at ensmp dot fr" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 17 Mar 2016 15:30:49 +0000
- Subject: [Bug libfortran/70279] New: Crash duing compilation with NINT and 128 bits integer
- Auto-submitted: auto-generated
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70279
Bug ID: 70279
Summary: Crash duing compilation with NINT and 128 bits integer
Product: gcc
Version: 4.8.4
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: libfortran
Assignee: unassigned at gcc dot gnu.org
Reporter: francois.willot at ensmp dot fr
Target Milestone: ---
Compiling the code below makes gfortran 4.8.4 crash:
gfortran -Wall -O3 -pedantic-errors -pedantic -Werror -std=f2003 -fopenmp -cpp
-DMULTITHREADS test.f90
test.f90: In function âMAIN__â:
test.f90:17:0: internal compiler error: in build_round_expr, at
fortran/trans-intrinsic.c:394
PRINT *,NINT(a,i)
^
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-4.8/README.Bugs> for instructions.
The code is:
PROGRAM MAIN
IMPLICIT NONE
INTEGER, PARAMETER :: i=SELECTED_INT_KIND(30)
DOUBLE PRECISION :: a
a=0.0D0
PRINT *,NINT(a,i)
END
Instead, this code runs as expected
PROGRAM MAIN
IMPLICIT NONE
INTEGER, PARAMETER :: i=SELECTED_INT_KIND(30)
PRINT *,NINT(0.0D0,i),i
END
(It produces:
0 16
)