[linux-2.6-block.git] / drivers / infiniband / hw / mlx5 / mem.c

/*
 * Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

#include <linux/module.h>
#include <rdma/ib_umem.h>
#include <rdma/ib_umem_odp.h>
#include "mlx5_ib.h"

/* @umem: umem object to scan
 * @addr: ib virtual address requested by the user
 * @max_page_shift: high limit for page_shift - 0 means no limit
 * @count: number of PAGE_SIZE pages covered by umem
 * @shift: page shift for the compound pages found in the region
 * @ncont: number of compund pages
 * @order: log2 of the number of compound pages
 */
void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr,
			unsigned long max_page_shift,
			int *count, int *shift,
			int *ncont, int *order)
{
	unsigned long tmp;
	unsigned long m;
	u64 base = ~0, p = 0;
	u64 len, pfn;
	int i = 0;
	struct scatterlist *sg;
	int entry;

	if (umem->is_odp) {
		struct ib_umem_odp *odp = to_ib_umem_odp(umem);
		unsigned int page_shift = odp->page_shift;

		*ncont = ib_umem_odp_num_pages(odp);
		*count = *ncont << (page_shift - PAGE_SHIFT);
		*shift = page_shift;
		if (order)
			*order = ilog2(roundup_pow_of_two(*ncont));

		return;
	}

	addr = addr >> PAGE_SHIFT;
	tmp = (unsigned long)addr;
	m = find_first_bit(&tmp, BITS_PER_LONG);
	if (max_page_shift)
		m = min_t(unsigned long, max_page_shift - PAGE_SHIFT, m);

	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
		len = sg_dma_len(sg) >> PAGE_SHIFT;
		pfn = sg_dma_address(sg) >> PAGE_SHIFT;
		if (base + p != pfn) {
			/* If either the offset or the new
			 * base are unaligned update m
			 */
			tmp = (unsigned long)(pfn | p);
			if (!IS_ALIGNED(tmp, 1 << m))
				m = find_first_bit(&tmp, BITS_PER_LONG);

			base = pfn;
			p = 0;
		}

		p += len;
		i += len;
	}

	if (i) {
		m = min_t(unsigned long, ilog2(roundup_pow_of_two(i)), m);

		if (order)
			*order = ilog2(roundup_pow_of_two(i) >> m);

		*ncont = DIV_ROUND_UP(i, (1 << m));
	} else {
		m  = 0;

		if (order)
			*order = 0;

		*ncont = 0;
	}
	*shift = PAGE_SHIFT + m;
	*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.
 *
 * dev - mlx5_ib device
 * umem - umem to use to fill the pages
 * page_shift - determines the page size used in the resulting array
 * offset - offset into the umem to start from,
 *          only implemented for ODP umems
 * num_pages - total number of pages to fill
 * pas - bus addresses array to fill
 * access_flags - access flags to set on all present pages.
		  use enum mlx5_ib_mtt_access_flags for this.
 */
void __mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
			    int page_shift, size_t offset, size_t num_pages,
			    __be64 *pas, int access_flags)
{
	int shift = page_shift - PAGE_SHIFT;
	int mask = (1 << shift) - 1;
	int i, k, idx;
	u64 cur = 0;
	u64 base;
	int len;
	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;
		base = sg_dma_address(sg);

		/* Skip elements below offset */
		if (i + len < offset << shift) {
			i += len;
			continue;
		}

		/* Skip pages below offset */
		if (i < offset << shift) {
			k = (offset << shift) - i;
			i = offset << shift;
		} else {
			k = 0;
		}

		for (; k < len; k++) {
			if (!(i & mask)) {
				cur = base + (k << PAGE_SHIFT);
				cur |= access_flags;
				idx = (i >> shift) - offset;

				pas[idx] = cpu_to_be64(cur);
				mlx5_ib_dbg(dev, "pas[%d] 0x%llx\n",
					    i >> shift, be64_to_cpu(pas[idx]));
			}
			i++;

			/* Stop after num_pages reached */
			if (i >> shift >= offset + num_pages)
				return;
		}
	}
}

void mlx5_ib_populate_pas(struct mlx5_ib_dev *dev, struct ib_umem *umem,
			  int page_shift, __be64 *pas, int access_flags)
{
	return __mlx5_ib_populate_pas(dev, umem, page_shift, 0,
				      ib_umem_num_pages(umem), pas,
				      access_flags);
}
int mlx5_ib_get_buf_offset(u64 addr, int page_shift, u32 *offset)
{
	u64 page_size;
	u64 page_mask;
	u64 off_size;
	u64 off_mask;
	u64 buf_off;

	page_size = (u64)1 << page_shift;
	page_mask = page_size - 1;
	buf_off = addr & page_mask;
	off_size = page_size >> 6;
	off_mask = off_size - 1;

	if (buf_off & off_mask)
		return -EINVAL;

	*offset = buf_off >> ilog2(off_size);
	return 0;
}
Commit	Line	Data
e126ba97	1	/*
6cf0a15f	2	* Copyright (c) 2013-2015, Mellanox Technologies. All rights reserved.
e126ba97 EC	3	*
	4	* This software is available to you under a choice of one of two
	5	* licenses. You may choose to be licensed under the terms of the GNU
	6	* General Public License (GPL) Version 2, available from the file
	7	* COPYING in the main directory of this source tree, or the
	8	* OpenIB.org BSD license below:
	9	*
	10	* Redistribution and use in source and binary forms, with or
	11	* without modification, are permitted provided that the following
	12	* conditions are met:
	13	*
	14	* - Redistributions of source code must retain the above
	15	* copyright notice, this list of conditions and the following
	16	* disclaimer.
	17	*
	18	* - Redistributions in binary form must reproduce the above
	19	* copyright notice, this list of conditions and the following
	20	* disclaimer in the documentation and/or other materials
	21	* provided with the distribution.
	22	*
	23	* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
	24	* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
	25	* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
	26	* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
	27	* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
	28	* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
	29	* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
	30	* SOFTWARE.
	31	*/
	32
	33	#include <linux/module.h>
	34	#include <rdma/ib_umem.h>
cc149f75	35	#include <rdma/ib_umem_odp.h>
e126ba97 EC	36	#include "mlx5_ib.h"
	37
	38	/* @umem: umem object to scan
	39	* @addr: ib virtual address requested by the user
762f899a	40	* @max_page_shift: high limit for page_shift - 0 means no limit
e126ba97 EC	41	* @count: number of PAGE_SIZE pages covered by umem
	42	* @shift: page shift for the compound pages found in the region
	43	* @ncont: number of compund pages
	44	* @order: log2 of the number of compound pages
	45	*/
762f899a MD	46	void mlx5_ib_cont_pages(struct ib_umem *umem, u64 addr,
	47	unsigned long max_page_shift,
	48	int count, int shift,
e126ba97 EC	49	int ncont, int order)
e126ba97 EC	50	{
e126ba97 EC	51	unsigned long tmp;
e126ba97 EC	52	unsigned long m;
d67bc5d4 IL	53	u64 base = ~0, p = 0;
	54	u64 len, pfn;
	55	int i = 0;
eeb8461e YH	56	struct scatterlist *sg;
eeb8461e YH	57	int entry;
e126ba97	58
597ecc5a	59	if (umem->is_odp) {
27b7fb1a JG	60	struct ib_umem_odp *odp = to_ib_umem_odp(umem);
27b7fb1a JG	61	unsigned int page_shift = odp->page_shift;
d2183c6f	62
27b7fb1a	63	*ncont = ib_umem_odp_num_pages(odp);
b2ac9188 AK	64	count = ncont << (page_shift - PAGE_SHIFT);
b2ac9188 AK	65	*shift = page_shift;
cc149f75	66	if (order)
b2ac9188	67	order = ilog2(roundup_pow_of_two(ncont));
cc149f75 HE	68
	69	return;
	70	}
	71
d2183c6f	72	addr = addr >> PAGE_SHIFT;
e126ba97	73	tmp = (unsigned long)addr;
fffd6873	74	m = find_first_bit(&tmp, BITS_PER_LONG);
762f899a	75	if (max_page_shift)
d2183c6f	76	m = min_t(unsigned long, max_page_shift - PAGE_SHIFT, m);
d67bc5d4	77
eeb8461e	78	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
d2183c6f JG	79	len = sg_dma_len(sg) >> PAGE_SHIFT;
d2183c6f JG	80	pfn = sg_dma_address(sg) >> PAGE_SHIFT;
d67bc5d4 IL	81	if (base + p != pfn) {
	82	/* If either the offset or the new
	83	* base are unaligned update m
	84	*/
	85	tmp = (unsigned long)(pfn \| p);
	86	if (!IS_ALIGNED(tmp, 1 << m))
	87	m = find_first_bit(&tmp, BITS_PER_LONG);
	88
	89	base = pfn;
	90	p = 0;
e126ba97	91	}
d67bc5d4 IL	92
	93	p += len;
	94	i += len;
eeb8461e	95	}
e126ba97 EC	96
	97	if (i) {
	98	m = min_t(unsigned long, ilog2(roundup_pow_of_two(i)), m);
	99
	100	if (order)
	101	*order = ilog2(roundup_pow_of_two(i) >> m);
	102
	103	*ncont = DIV_ROUND_UP(i, (1 << m));
	104	} else {
	105	m = 0;
	106
	107	if (order)
	108	*order = 0;
	109
	110	*ncont = 0;
	111	}
d2183c6f	112	*shift = PAGE_SHIFT + m;
e126ba97 EC	113	*count = i;
	114	}
	115
cc149f75 HE	116	static u64 umem_dma_to_mtt(dma_addr_t umem_dma)
	117	{
	118	u64 mtt_entry = umem_dma & ODP_DMA_ADDR_MASK;
	119
	120	if (umem_dma & ODP_READ_ALLOWED_BIT)
	121	mtt_entry \|= MLX5_IB_MTT_READ;
	122	if (umem_dma & ODP_WRITE_ALLOWED_BIT)
	123	mtt_entry \|= MLX5_IB_MTT_WRITE;
	124
	125	return mtt_entry;
	126	}
cc149f75 HE	127
	128	/*
	129	* Populate the given array with bus addresses from the umem.
	130	*
	131	* dev - mlx5_ib device
	132	* umem - umem to use to fill the pages
	133	* page_shift - determines the page size used in the resulting array
832a6b06 HE	134	* offset - offset into the umem to start from,
	135	* only implemented for ODP umems
	136	* num_pages - total number of pages to fill
cc149f75 HE	137	* pas - bus addresses array to fill
	138	* access_flags - access flags to set on all present pages.
	139	use enum mlx5_ib_mtt_access_flags for this.
	140	*/
832a6b06 HE	141	void __mlx5_ib_populate_pas(struct mlx5_ib_dev dev, struct ib_umem umem,
	142	int page_shift, size_t offset, size_t num_pages,
	143	__be64 *pas, int access_flags)
e126ba97	144	{
d2183c6f	145	int shift = page_shift - PAGE_SHIFT;
e126ba97	146	int mask = (1 << shift) - 1;
7d0cc6ed	147	int i, k, idx;
e126ba97 EC	148	u64 cur = 0;
	149	u64 base;
	150	int len;
eeb8461e YH	151	struct scatterlist *sg;
eeb8461e YH	152	int entry;
13859d5d	153
597ecc5a	154	if (umem->is_odp) {
cc149f75 HE	155	WARN_ON(shift != 0);
	156	WARN_ON(access_flags != (MLX5_IB_MTT_READ \| MLX5_IB_MTT_WRITE));
	157
	158	for (i = 0; i < num_pages; ++i) {
597ecc5a JG	159	dma_addr_t pa =
597ecc5a JG	160	to_ib_umem_odp(umem)->dma_list[offset + i];
cc149f75 HE	161
	162	pas[i] = cpu_to_be64(umem_dma_to_mtt(pa));
	163	}
	164	return;
	165	}
e126ba97 EC	166
e126ba97 EC	167	i = 0;
eeb8461e	168	for_each_sg(umem->sg_head.sgl, sg, umem->nmap, entry) {
d2183c6f	169	len = sg_dma_len(sg) >> PAGE_SHIFT;
eeb8461e	170	base = sg_dma_address(sg);
7d0cc6ed AK	171
	172	/* Skip elements below offset */
	173	if (i + len < offset << shift) {
	174	i += len;
	175	continue;
	176	}
	177
	178	/* Skip pages below offset */
	179	if (i < offset << shift) {
	180	k = (offset << shift) - i;
	181	i = offset << shift;
	182	} else {
	183	k = 0;
	184	}
	185
	186	for (; k < len; k++) {
eeb8461e	187	if (!(i & mask)) {
d2183c6f	188	cur = base + (k << PAGE_SHIFT);
cc149f75	189	cur \|= access_flags;
7d0cc6ed	190	idx = (i >> shift) - offset;
e126ba97	191
7d0cc6ed	192	pas[idx] = cpu_to_be64(cur);
eeb8461e	193	mlx5_ib_dbg(dev, "pas[%d] 0x%llx\n",
7d0cc6ed AK	194	i >> shift, be64_to_cpu(pas[idx]));
7d0cc6ed AK	195	}
eeb8461e	196	i++;
7d0cc6ed AK	197
	198	/* Stop after num_pages reached */
	199	if (i >> shift >= offset + num_pages)
	200	return;
e126ba97	201	}
eeb8461e	202	}
e126ba97 EC	203	}
e126ba97 EC	204
832a6b06 HE	205	void mlx5_ib_populate_pas(struct mlx5_ib_dev dev, struct ib_umem umem,
	206	int page_shift, __be64 *pas, int access_flags)
	207	{
	208	return __mlx5_ib_populate_pas(dev, umem, page_shift, 0,
	209	ib_umem_num_pages(umem), pas,
	210	access_flags);
	211	}
e126ba97 EC	212	int mlx5_ib_get_buf_offset(u64 addr, int page_shift, u32 *offset)
	213	{
	214	u64 page_size;
	215	u64 page_mask;
	216	u64 off_size;
	217	u64 off_mask;
	218	u64 buf_off;
	219
f241e749	220	page_size = (u64)1 << page_shift;
e126ba97 EC	221	page_mask = page_size - 1;
	222	buf_off = addr & page_mask;
	223	off_size = page_size >> 6;
	224	off_mask = off_size - 1;
	225
	226	if (buf_off & off_mask)
	227	return -EINVAL;
	228
	229	*offset = buf_off >> ilog2(off_size);
	230	return 0;
	231	}