Langsdorf, Mark
2007-Mar-01 19:17 UTC
[Xen-devel] [PATCH][retry 2] Make x86_64 swiotlb code to support dma_ops [2/2]
As per Keir''s and Jan''s suggestions, I''m simplifying my approach to getting GART into the kernel. This patch set makes the x86_64 swiotlb code work with dma_ops. The first patch creates the arch/x86_64/kernel/pci-dma-xen.c file based on the standard pci-dma.c. The second patch modifies arch/x86_64/kernel/pci-swiotlb-xen.c to work with dma_ops, and modifies arch/x86_64/kernel/pci-dma-xen.c slightly to be Xen-safe by copying over some functions from arch/i386/kernel/pci-dma-xen.c Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com> -Mark Langsdorf AMD, Inc. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Jan Beulich
2007-Mar-02 09:28 UTC
Re: [Xen-devel] [PATCH][retry 2] Make x86_64 swiotlb code to support dma_ops [2/2]
>The second patch modifies >arch/x86_64/kernel/pci-swiotlb-xen.c to work with >dma_ops, and modifies arch/x86_64/kernel/pci-dma-xen.c >slightly to be Xen-safe by copying over some functions >from arch/i386/kernel/pci-dma-xen.cThe removal of sync_single_range_for_{cpu,device} from swiotlb_dma_ops cannot work in my opinion, as this is not a no-op. You will need to either touch Xen''s swiotlb.c to add this function, or clone the i386 stubs that map dma_sync_single_range_for_ to dma_sync_single_for_ (perhaps into an ''else if'' in the existing functions in asm-x86_64/mach-xen/asm/dma-mapping.h. Jan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel