Hi Andrew!
On 2024-11-06T15:27:19+0000, Andrew Stubbs <[email protected]> wrote:
> If requested, return the vectorization factor appropriate for the offload
> device, if any.
> --- a/gcc/omp-general.cc
> +++ b/gcc/omp-general.cc
> @@ -987,10 +987,11 @@ find_combined_omp_for (tree *tp, int *walk_subtrees,
> void *data)
> return NULL_TREE;
> }
>
> -/* Return maximum possible vectorization factor for the target. */
> +/* Return maximum possible vectorization factor for the target, or for
> + the OpenMP offload target if one exists. */
>
> poly_uint64
> -omp_max_vf (void)
> +omp_max_vf (bool offload)
> {
> if (!optimize
> || optimize_debug
> @@ -999,6 +1000,18 @@ omp_max_vf (void)
> && OPTION_SET_P (flag_tree_loop_vectorize)))
> return 1;
>
> + if (ENABLE_OFFLOADING && offload)
> + {
> + for (const char *c = getenv ("OFFLOAD_TARGET_NAMES"); c;)
> + {
> + if (startswith (c, "amdgcn"))
> + return ordered_max (64, omp_max_vf (false));
> + else if ((c = strchr (c, ':')))
> + c++;
> + }
> + /* Otherwise, fall through to host VF. */
> + }
> +
> auto_vector_modes modes;
> targetm.vectorize.autovectorize_vector_modes (&modes, true);
> if (!modes.is_empty ())
Maybe move the magic number "64" into a '#define' or similar, and clarify
(source code comment?) whether the choice of 64 is also correct for GCN
'TARGET_WAVE64_COMPAT' configurations (like gfx1100, which default to
'--param=gcn-preferred-vectorization-factor=32')?
Grüße
Thomas
