Re: [patch 27/36] Hexagon: Provide DMA implementation

[Arnd Bergmann]

On Wednesday 17 August 2011 11:35:24 Richard Kuo wrote:

> +static dma_addr_t hexagon_map_page(struct device *dev, struct page *page,
> +				   unsigned long offset, size_t size,
> +				   enum dma_data_direction dir,
> +				   struct dma_attrs *attrs)
> +{
> +	dma_addr_t bus = page_to_phys(page) + offset;
> +	WARN_ON(size == 0);
> +
> +	if (!check_addr("map_single", dev, bus, size))
> +		return bad_dma_address;
> +
> +	flush_dcache_range(PAGE_OFFSET + bus, PAGE_OFFSET + bus + size);
> +
> +	return bus;
> +}
> +
> +static void hexagon_sync_single_for_cpu(struct device *dev,
> +					dma_addr_t dma_handle, size_t size,
> +					enum dma_data_direction dir)
> +{
> +	mb();
> +}

Why do you need to flush the cache in map_page, but not in sync?
Normally when you have non-coherent DMA, you need to invalidate
the cache in sync_*_foc_cpu and flush it sync_*_for_device.

	Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/