Messages in this thread | | | Date | Mon, 20 Oct 2008 09:46:03 +0900 | Subject | Re: [PATCH] intel-iommu: use coherent_dma_mask in alloc_coherent | From | FUJITA Tomonori <> |
| |
On Sat, 18 Oct 2008 14:46:56 +0100 David Woodhouse <dwmw2@infradead.org> wrote:
> On Wed, 2008-10-15 at 16:08 +0900, FUJITA Tomonori wrote: > > This patch fixes intel-iommu to use dev->coherent_dma_mask in > > alloc_coherent. Currently, intel-iommu uses dev->dma_mask in > > alloc_coherent but alloc_coherent is supposed to use > > coherent_dma_mask. It could break drivers that uses smaller > > coherent_dma_mask than dma_mask (though the current code works for the > > majority that use the same mask for coherent_dma_mask and dma_mask). > > I've fixed this up to cope with the fact that dma_mask can be bigger > than an 'unsigned long', and committed it to the iommu-2-6.git tree.
Hmm, 'unsigned long' is 64 bits on IA64 and X86_64. How can dma_mask be bigger than an 'unsigned long'?
| |