Commit cbf53074 authored by Will Deacon's avatar Will Deacon Committed by Christoph Hellwig
Browse files

swiotlb: Honour dma_alloc_coherent() alignment in swiotlb_alloc() 



core-api/dma-api-howto.rst states the following properties of
dma_alloc_coherent():

  | The CPU virtual address and the DMA address are both guaranteed to
  | be aligned to the smallest PAGE_SIZE order which is greater than or
  | equal to the requested size.

However, swiotlb_alloc() passes zero for the 'alloc_align_mask'
parameter of swiotlb_find_slots() and so this property is not upheld.
Instead, allocations larger than a page are aligned to PAGE_SIZE,

Calculate the mask corresponding to the page order suitable for holding
the allocation and pass that to swiotlb_find_slots().

Fixes: e81e99ba ("swiotlb: Support aligned swiotlb buffers")
Signed-off-by: default avatarWill Deacon <will@kernel.org>
Reviewed-by: default avatarMichael Kelley <mhklinux@outlook.com>
Reviewed-by: default avatarPetr Tesarik <petr.tesarik1@huawei-partners.com>
Tested-by: default avatarNicolin Chen <nicolinc@nvidia.com>
Tested-by: default avatarMichael Kelley <mhklinux@outlook.com>
Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
parent 823353b7

kernel/dma/swiotlb.c

+3 −1
Original line number Diff line number Diff line
@@ -1679,12 +1679,14 @@ struct page *swiotlb_alloc(struct device *dev, size_t size) 
	struct io_tlb_mem *mem = dev->dma_io_tlb_mem;

	struct io_tlb_pool *pool;

	phys_addr_t tlb_addr;

	unsigned int align;

	int index;



	if (!mem)

		return NULL;



	index = swiotlb_find_slots(dev, 0, size, 0, &pool);

	align = (1 << (get_order(size) + PAGE_SHIFT)) - 1;

	index = swiotlb_find_slots(dev, 0, size, align, &pool);

	if (index == -1)

		return NULL;