Commit b116c702 authored by Christoph Hellwig's avatar Christoph Hellwig Committed by Jason Gunthorpe
Browse files

RDMA/umem: Use ib_dma_max_seg_size instead of dma_get_max_seg_size 

RDMA ULPs must not call DMA mapping APIs directly but instead use the
ib_dma_* wrappers.

Fixes: 0c16d963 ("RDMA/umem: Move to allocate SG table from pages")
Link: https://lore.kernel.org/r/20201106181941.1878556-3-hch@lst.de


Reported-by: default avatarJason Gunthorpe <jgg@nvidia.com>
Signed-off-by: default avatarChristoph Hellwig <hch@lst.de>
Signed-off-by: default avatarJason Gunthorpe <jgg@nvidia.com>
parent fbb7dc5d

drivers/infiniband/core/umem.c

+4 −4
Original line number Diff line number Diff line
@@ -229,9 +229,9 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, 


		cur_base += ret * PAGE_SIZE;

		npages -= ret;

		sg = __sg_alloc_table_from_pages(

			&umem->sg_head, page_list, ret, 0, ret << PAGE_SHIFT,

			dma_get_max_seg_size(device->dma_device), sg, npages,

		sg = __sg_alloc_table_from_pages(&umem->sg_head, page_list, ret,

				0, ret << PAGE_SHIFT,

				ib_dma_max_seg_size(device), sg, npages,

				GFP_KERNEL);

		umem->sg_nents = umem->sg_head.nents;

		if (IS_ERR(sg)) {