Commit cbe4b8f0 authored by Artemy Kovalyov's avatar Artemy Kovalyov Committed by Jason Gunthorpe
Browse files

IB/mlx5: Unify ODP MR code paths to allow extra flexibility 

Building MR translation table in the ODP case requires additional
flexibility, namely random access to DMA addresses. Make both direct and
indirect ODP MR use same code path, separated from the non-ODP MR code
path.

With the restructuring the correct page_shift is now used around
__mlx5_ib_populate_pas().

Fixes: d2183c6f ("RDMA/umem: Move page_shift from ib_umem to ib_odp_umem")
Link: https://lore.kernel.org/r/20191222124649.52300-2-leon@kernel.org


Signed-off-by: default avatarArtemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: default avatarLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: default avatarJason Gunthorpe <jgg@mellanox.com>
parent 4ad6429d

drivers/infiniband/hw/mlx5/mem.c

+0 −25
Original line number Diff line number Diff line
@@ -101,18 +101,6 @@ void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr, 
	*count = i;

}



static u64 umem_dma_to_mtt(dma_addr_t umem_dma)

{

	u64 mtt_entry = umem_dma & ODP_DMA_ADDR_MASK;



	if (umem_dma & ODP_READ_ALLOWED_BIT)

		mtt_entry |= MLX5_IB_MTT_READ;

	if (umem_dma & ODP_WRITE_ALLOWED_BIT)

		mtt_entry |= MLX5_IB_MTT_WRITE;



	return mtt_entry;

}



/*

 * Populate the given array with bus addresses from the umem.

 *
@@ -139,19 +127,6 @@ void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem, 
	struct scatterlist *sg;

	int entry;



	if (umem->is_odp) {

		WARN_ON(shift != 0);

		WARN_ON(access_flags != (MLX5_IB_MTT_READ | MLX5_IB_MTT_WRITE));



		for (i = 0; i < num_pages; ++i) {

			dma_addr_t pa =

				to_ib_umem_odp(umem)->dma_list[offset + i];



			pas[i] = cpu_to_be64(umem_dma_to_mtt(pa));

		}

		return;

	}



	i = 0;

	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {

		len = sg_dma_len(sg) >> PAGE_SHIFT;

drivers/infiniband/hw/mlx5/mlx5_ib.h

+4 −5
Original line number Diff line number Diff line
@@ -1276,8 +1276,8 @@ void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *ibdev); 
int __init mlx5_ib_odp_init(void);

void mlx5_ib_odp_cleanup(void);

void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent);

void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset,

			   size_t nentries, struct mlx5_ib_mr *mr, int flags);

void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries,

			   struct mlx5_ib_mr *mr, int flags);



int mlx5_ib_advise_mr_prefetch(struct ib_pd *pd,

			       enum ib_uverbs_advise_mr_advice advice,
@@ -1293,9 +1293,8 @@ static inline void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *ibdev) {} 
static inline int mlx5_ib_odp_init(void) { return 0; }

static inline void mlx5_ib_odp_cleanup(void)				    {}

static inline void mlx5_odp_init_mr_cache_entry(struct mlx5_cache_ent *ent) {}

static inline void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t offset,

					 size_t nentries, struct mlx5_ib_mr *mr,

					 int flags) {}

static inline void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries,

					 struct mlx5_ib_mr *mr, int flags) {}



static inline int

mlx5_ib_advise_mr_prefetch(struct ib_pd *pd,

drivers/infiniband/hw/mlx5/mr.c

+23 −35
Original line number Diff line number Diff line
@@ -868,36 +868,6 @@ static struct mlx5_ib_mr *alloc_mr_from_cache( 
	return mr;

}



static inline int populate_xlt(struct mlx5_ib_mr *mr, int idx, int npages,

			       void *xlt, int page_shift, size_t size,

			       int flags)

{

	struct mlx5_ib_dev *dev = mr->dev;

	struct ib_umem *umem = mr->umem;



	if (flags & MLX5_IB_UPD_XLT_INDIRECT) {

		if (!umr_can_use_indirect_mkey(dev))

			return -EPERM;

		mlx5_odp_populate_klm(xlt, idx, npages, mr, flags);

		return npages;

	}



	npages = min_t(size_t, npages, ib_umem_num_pages(umem) - idx);



	if (!(flags & MLX5_IB_UPD_XLT_ZAP)) {

		__mlx5_ib_populate_pas(dev, umem, page_shift,

				       idx, npages, xlt,

				       MLX5_IB_MTT_PRESENT);

		/* Clear padding after the pages

		 * brought from the umem.

		 */

		memset(xlt + (npages * sizeof(struct mlx5_mtt)), 0,

		       size - npages * sizeof(struct mlx5_mtt));

	}



	return npages;

}



#define MLX5_MAX_UMR_CHUNK ((1 << (MLX5_MAX_UMR_SHIFT + 4)) - \

			    MLX5_UMR_MTT_ALIGNMENT)

#define MLX5_SPARE_UMR_CHUNK 0x10000
@@ -921,6 +891,7 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, 
	size_t pages_mapped = 0;

	size_t pages_to_map = 0;

	size_t pages_iter = 0;

	size_t size_to_map = 0;

	gfp_t gfp;

	bool use_emergency_page = false;
@@ -967,6 +938,15 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, 
		goto free_xlt;

	}



	if (mr->umem->is_odp) {

		if (!(flags & MLX5_IB_UPD_XLT_INDIRECT)) {

			struct ib_umem_odp *odp = to_ib_umem_odp(mr->umem);

			size_t max_pages = ib_umem_odp_num_pages(odp) - idx;



			pages_to_map = min_t(size_t, pages_to_map, max_pages);

		}

	}



	sg.addr = dma;

	sg.lkey = dev->umrc.pd->local_dma_lkey;
@@ -989,14 +969,22 @@ int mlx5_ib_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, 
	     pages_mapped < pages_to_map && !err;

	     pages_mapped += pages_iter, idx += pages_iter) {

		npages = min_t(int, pages_iter, pages_to_map - pages_mapped);

		size_to_map = npages * desc_size;

		dma_sync_single_for_cpu(ddev, dma, size, DMA_TO_DEVICE);

		npages = populate_xlt(mr, idx, npages, xlt,

				      page_shift, size, flags);



		if (mr->umem->is_odp) {

			mlx5_odp_populate_xlt(xlt, idx, npages, mr, flags);

		} else {

			__mlx5_ib_populate_pas(dev, mr->umem, page_shift, idx,

					       npages, xlt,

					       MLX5_IB_MTT_PRESENT);

			/* Clear padding after the pages

			 * brought from the umem.

			 */

			memset(xlt + size_to_map, 0, size - size_to_map);

		}

		dma_sync_single_for_device(ddev, dma, size, DMA_TO_DEVICE);



		sg.length = ALIGN(npages * desc_size,

				  MLX5_UMR_MTT_ALIGNMENT);

		sg.length = ALIGN(size_to_map, MLX5_UMR_MTT_ALIGNMENT);



		if (pages_mapped + pages_iter >= pages_to_map) {

			if (flags & MLX5_IB_UPD_XLT_ENABLE)

drivers/infiniband/hw/mlx5/odp.c

+40 −2
Original line number Diff line number Diff line
@@ -93,7 +93,7 @@ struct mlx5_pagefault { 


static u64 mlx5_imr_ksm_entries;



void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t idx, size_t nentries,

static void populate_klm(struct mlx5_klm *pklm, size_t idx, size_t nentries,

			struct mlx5_ib_mr *imr, int flags)

{

	struct mlx5_klm *end = pklm + nentries;
@@ -144,6 +144,44 @@ void mlx5_odp_populate_klm(struct mlx5_klm *pklm, size_t idx, size_t nentries, 
	}

}



static u64 umem_dma_to_mtt(dma_addr_t umem_dma)

{

	u64 mtt_entry = umem_dma & ODP_DMA_ADDR_MASK;



	if (umem_dma & ODP_READ_ALLOWED_BIT)

		mtt_entry |= MLX5_IB_MTT_READ;

	if (umem_dma & ODP_WRITE_ALLOWED_BIT)

		mtt_entry |= MLX5_IB_MTT_WRITE;



	return mtt_entry;

}



static void populate_mtt(__be64 *pas, size_t idx, size_t nentries,

			 struct mlx5_ib_mr *mr, int flags)

{

	struct ib_umem_odp *odp = to_ib_umem_odp(mr->umem);

	dma_addr_t pa;

	size_t i;



	if (flags & MLX5_IB_UPD_XLT_ZAP)

		return;



	for (i = 0; i < nentries; i++) {

		pa = odp->dma_list[idx + i];

		pas[i] = cpu_to_be64(umem_dma_to_mtt(pa));

	}

}



void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries,

			   struct mlx5_ib_mr *mr, int flags)

{

	if (flags & MLX5_IB_UPD_XLT_INDIRECT) {

		populate_klm(xlt, idx, nentries, mr, flags);

	} else {

		populate_mtt(xlt, idx, nentries, mr, flags);

	}

}



static void dma_fence_odp_mr(struct mlx5_ib_mr *mr)

{

	struct ib_umem_odp *odp = to_ib_umem_odp(mr->umem);

CRA Git | Maintained and supported by SUSTech CRA and CCSE