Re: [PATCH 03/19] dma-iommu: don't use a scatterlist in iommu_dma_alloc

From: Christoph Hellwig
Date: Mon Feb 11 2019 - 11:00:54 EST

On Wed, Feb 06, 2019 at 03:28:28PM +0000, Robin Murphy wrote:
> Because if iommu_map() only gets called at PAGE_SIZE granularity, then the
> IOMMU PTEs will be created at PAGE_SIZE (or smaller) granularity, so any
> effort to get higher-order allocations matching larger IOMMU block sizes is
> wasted, and we may as well have just done this:
> for (i = 0; i < count; i++) {
> struct page *page = alloc_page(gfp);
> ...
> iommu_map(..., page_to_phys(page), PAGE_SIZE, ...);
> }

True. I've dropped this patch.

> Really, it's a shame we have to split huge pages for the CPU remap, since
> in the common case the CPU MMU will have a matching block size, but IIRC
> there was something in vmap() or thereabouts that explicitly chokes on
> them.

That just needs a volunteer to fix the implementation, as there is no
fundamental reason not to remap large pages.