Re: [PATCH 13/31] nds32: DMA mapping API

[Arnd Bergmann]

On Wed, Nov 8, 2017 at 6:55 AM, Greentime Hu <green.hu@xxxxxxxxx> wrote:

> +static void consistent_sync(void *vaddr, size_t size, int direction)
> +{
> +       unsigned long start = (unsigned long)vaddr;
> +       unsigned long end = start + size;
> +
> +       switch (direction) {
> +       case DMA_FROM_DEVICE:   /* invalidate only */
> +               cpu_dma_inval_range(start, end);
> +               break;
> +       case DMA_TO_DEVICE:     /* writeback only */
> +               cpu_dma_wb_range(start, end);
> +               break;
> +       case DMA_BIDIRECTIONAL: /* writeback and invalidate */
> +               cpu_dma_wbinval_range(start, end);
> +               break;
> +       default:
> +               BUG();
> +       }
> +}

> +
> +static void
> +nds32_dma_sync_single_for_cpu(struct device *dev, dma_addr_t handle,
> +                             size_t size, enum dma_data_direction dir)
> +{
> +       consistent_sync((void *)dma_to_virt(dev, handle), size, dir);
> +}
> +
> +static void
> +nds32_dma_sync_single_for_device(struct device *dev, dma_addr_t handle,
> +                                size_t size, enum dma_data_direction dir)
> +{
> +       consistent_sync((void *)dma_to_virt(dev, handle), size, dir);
> +}

You do the same cache operations for _to_cpu and _to_device, which
usually works,
but is more expensive than you need. It's better to take the ownership into
account and only do what you need.

     Arnd