Re: Regression due to d98849aff879 (dma-direct: handle DMA_ATTR_NO_KERNEL_MAPPING in common code)

[Christoph Hellwig]

On Tue, Aug 06, 2019 at 03:59:40PM +0000, Lendacky, Thomas wrote:
> As long as two different cookie types (page pointer for encrypted DMA
> and virtual address returned from page_address() for unencrypted DMA)
> is ok. I'm just not familiar with how the cookie is used in any other
> functions, if at all.

DMA_ATTR_NO_KERNEL_MAPPING is intended for memory never used in the
kernel, either because it is just a buffer for device that are too cheap
to enough dram, or because it is a buffer for userspace to device
communication that the kernel just mediates.