fill_gpu(arr, val) replaces allocate(arr, source=val) (serial host
zero-fill) with a do concurrent kernel. Generic over rank 1D/2D/3D;
uses lbound/ubound so any lower bound works (e.g. z_ref(0:nz,...)).
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:) | |||
| real(kind=wp), | intent(in) | :: | val |
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:,:) | |||
| real(kind=wp), | intent(in) | :: | val |
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:,:,:) | |||
| real(kind=wp), | intent(in) | :: | val |
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:) | |||
| real(kind=wp), | intent(in) | :: | val |
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:,:) | |||
| real(kind=wp), | intent(in) | :: | val |
| Type | Intent | Optional | Attributes | Name | ||
|---|---|---|---|---|---|---|
| real(kind=wp), | intent(inout) | :: | arr(:,:,:) | |||
| real(kind=wp), | intent(in) | :: | val |