[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: Christoph Hellwig <hch@lst.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>

diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index 4eba60a328907d753cbcb0ab5ab2736552e3a8e1..cbb794556d023e1a4f5172eece21cd9966d99a4f 100644 (file)
--- a/arch/powerpc/kernel/iommu.c
+++ b/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 */
@@ -570,9 +572,9 @@ 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
-               *dma_handle = mapping;
+               return NULL;
+       }
+       *dma_handle = mapping;
        return ret;
 }
 

diff --git a/arch/powerpc/kernel/pci_iommu.c b/arch/powerpc/kernel/pci_iommu.c
index 48aa82d01624f7f20e63e3c0203f05cc44f05d76..7fb4cca021be57e3e311cda66935e12f1237ce50 100644 (file)
--- a/arch/powerpc/kernel/pci_iommu.c
+++ b/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,

diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c
index fe92727aca3040f828452e97fc0ff1d92bf4644f..e746686d48b8eb34f673654fa5bb82ece331ed59 100644 (file)
--- a/arch/powerpc/kernel/vio.c
+++ b/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,

diff --git a/include/asm-powerpc/iommu.h b/include/asm-powerpc/iommu.h
index 18ca29e9105a5d92f1e302d49f411f1f83229159..9065f6c972a43d38c200cf90e326217eb0e137a5 100644 (file)
--- a/include/asm-powerpc/iommu.h
+++ b/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,