Re: [PATCH] x86: avoid unnecessary low zone allocation in AMD IOMMU's alloc_coherent

[Joerg Roedel <joerg.roedel@amd.com>]

On Wed, Sep 10, 2008 at 10:03:32PM +0900, FUJITA Tomonori wrote:

Ok, I see.

> Hmm, I'm not sure what code you look at. Here's dma_alloc_coherent()
> in tip/x86/iommu:
> 
> dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
> 		gfp_t gfp)
> {
> 	struct dma_mapping_ops *ops = get_dma_ops(dev);
> 	void *memory;
> 
> 	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
> 
> Surely we here clear the flag but...
> 
> 	if (dma_alloc_from_coherent(dev, size, dma_handle, &memory))
> 		return memory;
> 
> 	if (!dev) {
> 		dev = &x86_dma_fallback_dev;
> 		gfp |= GFP_DMA;
> 	}
> 
> we play with it here though (not happens with pci devices),
> 
> 	if (!dev->dma_mask)
> 		return NULL;
> 
> 	if (!ops->alloc_coherent)
> 		return NULL;
> 
> Then dma_alloc_coherent_gfp_flags() sets it again according to
> device->coherent_dma_mask and gfp before ops->alloc_coherent hook:
> 
> 	return ops->alloc_coherent(dev, size, dma_handle,
> 				   dma_alloc_coherent_gfp_flags(dev, gfp));
> 
> 
> This code can set up the exact same gfp flag for swiotbl and nommu as
> before.

So its possible that alloc_coherent is called with region specifiers in
the gfp flags. Can't we simply make the gfp hacks depend on
dma_ops->is_phys and avoid further gfp hacks in the hardware iommu
implementations?

Joerg

-- 
           |           AMD Saxony Limited Liability Company & Co. KG
 Operating |         Wilschdorfer Landstr. 101, 01109 Dresden, Germany
 System    |                  Register Court Dresden: HRA 4896
 Research  |              General Partner authorized to represent:
 Center    |             AMD Saxony LLC (Wilmington, Delaware, US)
           | General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy