| Message ID | 4D694C37.6040205@net-b.de |
|---|---|
| State | New |
| Headers | show |
On Sat, Feb 26, 2011 at 07:53:43PM +0100, Tobias Burnus wrote: > The patch is rather obvious. Without, one gets an ICE as the function is > not resolved but BT_UNKNOWN in > > !$omp task if(omp_get_num_threads()> 0) > > Build and regtested on x86-64-linux. > OK for the trunk? What about GCC 4.4 and 4.5? > Unless Jakub has some objection, I think the patch can be committed to all three.
On Sat, Feb 26, 2011 at 10:56:21AM -0800, Steve Kargl wrote: > On Sat, Feb 26, 2011 at 07:53:43PM +0100, Tobias Burnus wrote: > > The patch is rather obvious. Without, one gets an ICE as the function is > > not resolved but BT_UNKNOWN in > > > > !$omp task if(omp_get_num_threads()> 0) > > > > Build and regtested on x86-64-linux. > > OK for the trunk? What about GCC 4.4 and 4.5? > > > > Unless Jakub has some objection, I think the patch > can be committed to all three. Yeah. Thanks. Jakub
2011-02-26 Tobias Burnus <burnus@net-b.de> PR fortran/47886 * openmp.c (gfc_resolve_omp_directive): Resolve if() condition of OpenMP's task. diff --git a/gcc/fortran/openmp.c b/gcc/fortran/openmp.c index 24e32eb..2b70c69 100644 --- a/gcc/fortran/openmp.c +++ b/gcc/fortran/openmp.c @@ -1573,6 +1573,7 @@ gfc_resolve_omp_directive (gfc_code *code, gfc_namespace *ns ATTRIBUTE_UNUSED) case EXEC_OMP_PARALLEL_SECTIONS: case EXEC_OMP_SECTIONS: case EXEC_OMP_SINGLE: + case EXEC_OMP_TASK: if (code->ext.omp_clauses) resolve_omp_clauses (code); break;
The patch is rather obvious. Without, one gets an ICE as the function is not resolved but BT_UNKNOWN in !$omp task if(omp_get_num_threads()> 0) Build and regtested on x86-64-linux. OK for the trunk? What about GCC 4.4 and 4.5? Tobias