2008-10-23 14:15:34

by FUJITA Tomonori

Subject: [PATCH] x86: restore the old swiotlb alloc_coherent behavior

This restores the old swiotlb alloc_coherent behavior (before the
alloc_coherent rewrite):

http://lkml.org/lkml/2008/8/12/200

The old alloc_coherent avoids GFP_DMA allocation first and if the
allocated address is not fit for the device's coherent_dma_mask, then
dma_alloc_coherent does GFP_DMA allocation. If it fails,
alloc_coherent calls swiotlb_alloc_coherent (in short, we rarely used
swiotlb_alloc_coherent).

After the alloc_coherent rewrite, dma_alloc_coherent
(include/asm-x86/dma-mapping.h) directly calls swiotlb_alloc_coherent.
It means that we possibly can't handle a device having dma_masks >
24bit < 32bits since swiotlb_alloc_coherent doesn't have the above
GFP_DMA retry mechanism.

This patch fixes x86's swiotlb alloc_coherent to use the GFP_DMA retry
mechanism, which dma_generic_alloc_coherent() provides now
(pci-nommu.c and GART IOMMU driver also use
dma_generic_alloc_coherent).

Signed-off-by: FUJITA Tomonori <[email protected]>
---
arch/x86/kernel/pci-swiotlb_64.c | 14 +++++++++++++-
1 files changed, 13 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index c4ce033..3c539d1 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -18,9 +18,21 @@ swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
return swiotlb_map_single(hwdev, phys_to_virt(paddr), size, direction);
}

+static void *x86_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flags)
+{
+ void *vaddr;
+
+ vaddr = dma_generic_alloc_coherent(hwdev, size, dma_handle, flags);
+ if (vaddr)
+ return vaddr;
+
+ return swiotlb_alloc_coherent(hwdev, size, dma_handle, flags);
+}
+
struct dma_mapping_ops swiotlb_dma_ops = {
.mapping_error = swiotlb_dma_mapping_error,
- .alloc_coherent = swiotlb_alloc_coherent,
+ .alloc_coherent = x86_swiotlb_alloc_coherent,
.free_coherent = swiotlb_free_coherent,
.map_single = swiotlb_map_single_phys,
.unmap_single = swiotlb_unmap_single,
--
1.5.5.GIT

2008-10-23 14:23:18

by Ingo Molnar

Subject: Re: [PATCH] x86: restore the old swiotlb alloc_coherent behavior

* FUJITA Tomonori <[email protected]> wrote:

> This restores the old swiotlb alloc_coherent behavior (before the
> alloc_coherent rewrite):
>
> http://lkml.org/lkml/2008/8/12/200
>
> The old alloc_coherent avoids GFP_DMA allocation first and if the
> allocated address is not fit for the device's coherent_dma_mask, then
> dma_alloc_coherent does GFP_DMA allocation. If it fails,
> alloc_coherent calls swiotlb_alloc_coherent (in short, we rarely used
> swiotlb_alloc_coherent).
>
> After the alloc_coherent rewrite, dma_alloc_coherent
> (include/asm-x86/dma-mapping.h) directly calls swiotlb_alloc_coherent.
> It means that we possibly can't handle a device having dma_masks >
> 24bit < 32bits since swiotlb_alloc_coherent doesn't have the above
> GFP_DMA retry mechanism.
>
> This patch fixes x86's swiotlb alloc_coherent to use the GFP_DMA retry
> mechanism, which dma_generic_alloc_coherent() provides now
> (pci-nommu.c and GART IOMMU driver also use
> dma_generic_alloc_coherent).
>
> Signed-off-by: FUJITA Tomonori <[email protected]>
> ---
> arch/x86/kernel/pci-swiotlb_64.c | 14 +++++++++++++-
> 1 files changed, 13 insertions(+), 1 deletions(-)

applied to tip/core/urgent (to after the two other iommu related
patches) - thanks Fujita!

Ingo