I agree with this patch in principle. It fixes the problem. But I think
we should evaluate if we can change the dma_mask of the fallback_dev to
24 bit.

On Fri, Sep 05, 2008 at 05:58:47PM +0900, FUJITA Tomonori wrote:
quoted text
> We need to use __GFP_DMA for NULL device argument (fallback_dev) with
> pci-nommu. It's a hack for ISA (and some old code) so we need to use
> GFP_DMA.
> 
> Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
> ---
>  arch/x86/kernel/pci-nommu.c   |    5 +----
>  include/asm-x86/dma-mapping.h |    2 ++
>  2 files changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
> index 73853d3..526b2db 100644
> --- a/arch/x86/kernel/pci-nommu.c
> +++ b/arch/x86/kernel/pci-nommu.c
> @@ -83,7 +83,6 @@ nommu_alloc_coherent(struct device *hwdev, size_t size,
>  	if (hwdev->dma_mask == NULL)
>  		return NULL;
>  
> -	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
>  	gfp |= __GFP_ZERO;
>  
>  	dma_mask = hwdev->coherent_dma_mask;
> @@ -96,14 +95,12 @@ nommu_alloc_coherent(struct device *hwdev, size_t size,
>  	node = dev_to_node(hwdev);
>  
>  #ifdef CONFIG_X86_64
> -	if (dma_mask <= DMA_32BIT_MASK)
> +	if (dma_mask <= DMA_32BIT_MASK && !(gfp & GFP_DMA))
>  		gfp |= GFP_DMA32;
>  #endif
> -
>  	/* No alloc-free penalty for ISA devices */
>  	if (dma_mask == DMA_24BIT_MASK)
>  		gfp |= GFP_DMA;
> -

Please leave these spare lines where they are. They increase the
readability of the code imho.

quoted text
>  again:
>  	page = alloc_pages_node(node, gfp, get_order(size));
>  	if (!page)
> diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h
> index 3a9a6f5..de4495e 100644
> --- a/include/asm-x86/dma-mapping.h
> +++ b/include/asm-x86/dma-mapping.h
> @@ -246,6 +246,8 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
>  	struct dma_mapping_ops *ops = get_dma_ops(dev);
>  	void *memory;
>  
> +	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32);
> +
>  	if (dma_alloc_from_coherent(dev, size, dma_handle, &memory))
>  		return memory;
>  
> -- 
> 1.5.4.2
> 
> 

-- 
           |           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

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