On Fri, Apr 03, 2020 at 11:02:13AM +0200, Tobias Burnus wrote:
> Those are all for the same switch statement;
>
> gomp_for contains 'tree clauses' and this clauses's '->code' is used
> to handle store 'enum omp_clauses_code' values in in gimple.{h,c}.
>
> I think adding this cast (and only this one) makes sense and it
> also silences a (clang) compiler warning.
That looks wrong to me. If *pc are OMP_CLAUSES, then it should use
OMP_CLAUSE_CODE (c) instead of TREE_CODE (c).
If something creates a tree that has TREE_CODE OMP_CLAUSE_LINEAR, then
that looks very suspicios (TREE_CODE should be OMP_CLAUSE).
> omp-grid.c – add cast to silence different enumeration types warning
>
> * omp-grid.c (grid_eliminate_combined_simd_part): Add cast
> to omp_clause_code to silence compiler warning.
>
> diff --git a/gcc/omp-grid.c b/gcc/omp-grid.c
> index b98e45de6a0..878977da2f9 100644
> --- a/gcc/omp-grid.c
> +++ b/gcc/omp-grid.c
> @@ -1058,21 +1058,21 @@ grid_eliminate_combined_simd_part (gomp_for *parloop)
> while (*tgt)
> tgt = &OMP_CLAUSE_CHAIN (*tgt);
>
> /* Copy over all clauses, except for linear clauses, which are turned into
> private clauses, and all other simd-specific clauses, which are
> ignored. */
> tree *pc = gimple_omp_for_clauses_ptr (simd);
> while (*pc)
> {
> tree c = *pc;
> - switch (TREE_CODE (c))
> + switch ((omp_clause_code) TREE_CODE (c))
> {
> case OMP_CLAUSE_LINEAR:
> {
> tree priv = build_omp_clause (UNKNOWN_LOCATION, OMP_CLAUSE_PRIVATE);
> OMP_CLAUSE_DECL (priv) = OMP_CLAUSE_DECL (c);
> OMP_CLAUSE_CHAIN (priv) = NULL;
> *tgt = priv;
> tgt = &OMP_CLAUSE_CHAIN (priv);
> pc = &OMP_CLAUSE_CHAIN (c);
> break;
Jakub
