[PATCH] powerpc: node-aware dma allocations

Make sure dma_alloc_coherent allocates memory from the local node.  This
is important on Cell where we avoid going through the slow cpu
interconnect.

Note:  I could only test this patch on Cell, it should be verified on
some pseries machine by those that have the hardware.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NPaul Mackerras <paulus@samba.org>

arch/powerpc/kernel/iommu.c

@@ -536,11 +536,12 @@ void iommu_unmap_single(struct iommu_table *tbl, dma_addr_t dma_handle,
  * to the dma address (mapping) of the first page.
  */
 void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
-		dma_addr_t *dma_handle, unsigned long mask, gfp_t flag)
+		dma_addr_t *dma_handle, unsigned long mask, gfp_t flag, int node)
 {
 	void *ret = NULL;
 	dma_addr_t mapping;
 	unsigned int npages, order;
+	struct page *page;
 
 	size = PAGE_ALIGN(size);
 	npages = size >> PAGE_SHIFT;
@@ -560,9 +561,10 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
 		return NULL;
 
 	/* Alloc enough pages (and possibly more) */
-	ret = (void *)__get_free_pages(flag, order);
-	if (!ret)
+	page = alloc_pages_node(flag, order, node);
+	if (!page)
 		return NULL;
+	ret = page_address(page);
 	memset(ret, 0, size);
 
 	/* Set up tces to cover the allocated range */
@@ -570,8 +572,8 @@ void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
 			      mask >> PAGE_SHIFT, order);
 	if (mapping == DMA_ERROR_CODE) {
 		free_pages((unsigned long)ret, order);
-		ret = NULL;
-	} else
+		return NULL;
+	}
 	*dma_handle = mapping;
 	return ret;
 }

arch/powerpc/kernel/pci_iommu.c

@@ -86,7 +86,8 @@ static void *pci_iommu_alloc_coherent(struct device *hwdev, size_t size,
 			   dma_addr_t *dma_handle, gfp_t flag)
 {
 	return iommu_alloc_coherent(device_to_table(hwdev), size, dma_handle,
-			device_to_mask(hwdev), flag);
+			device_to_mask(hwdev), flag,
+			pcibus_to_node(to_pci_dev(hwdev)->bus));
 }
 
 static void pci_iommu_free_coherent(struct device *hwdev, size_t size,

arch/powerpc/kernel/vio.c

@@ -394,7 +394,7 @@ static void *vio_alloc_coherent(struct device *dev, size_t size,
 			   dma_addr_t *dma_handle, gfp_t flag)
 {
 	return iommu_alloc_coherent(to_vio_dev(dev)->iommu_table, size,
-			dma_handle, ~0ul, flag);
+			dma_handle, ~0ul, flag, -1);
 }
 
 static void vio_free_coherent(struct device *dev, size_t size,

include/asm-powerpc/iommu.h

@@ -76,7 +76,8 @@ extern void iommu_unmap_sg(struct iommu_table *tbl, struct scatterlist *sglist,
 		int nelems, enum dma_data_direction direction);
 
 extern void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size,
-		dma_addr_t *dma_handle, unsigned long mask, gfp_t flag);
+		dma_addr_t *dma_handle, unsigned long mask,
+		gfp_t flag, int node);
 extern void iommu_free_coherent(struct iommu_table *tbl, size_t size,
 		void *vaddr, dma_addr_t dma_handle);
 extern dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr,