1/*
2 * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
3 * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
4 *
5 * This software is available to you under a choice of one of two
6 * licenses.  You may choose to be licensed under the terms of the GNU
7 * General Public License (GPL) Version 2, available from the file
8 * COPYING in the main directory of this source tree, or the
9 * OpenIB.org BSD license below:
10 *
11 *     Redistribution and use in source and binary forms, with or
12 *     without modification, are permitted provided that the following
13 *     conditions are met:
14 *
15 *      - Redistributions of source code must retain the above
16 *        copyright notice, this list of conditions and the following
17 *        disclaimer.
18 *
19 *      - Redistributions in binary form must reproduce the above
20 *        copyright notice, this list of conditions and the following
21 *        disclaimer in the documentation and/or other materials
22 *        provided with the distribution.
23 *
24 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
25 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
26 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
27 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
28 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
29 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
30 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
31 * SOFTWARE.
32 */
33
34#include <linux/mlx4/qp.h>
35#include <linux/mlx4/srq.h>
36#include <linux/slab.h>
37
38#include "mlx4_ib.h"
39#include <rdma/mlx4-abi.h>
40#include <rdma/uverbs_ioctl.h>
41
42static void *get_wqe(struct mlx4_ib_srq *srq, int n)
43{
44	return mlx4_buf_offset(&srq->buf, n << srq->msrq.wqe_shift);
45}
46
47static void mlx4_ib_srq_event(struct mlx4_srq *srq, enum mlx4_event type)
48{
49	struct ib_event event;
50	struct ib_srq *ibsrq = &to_mibsrq(srq)->ibsrq;
51
52	if (ibsrq->event_handler) {
53		event.device      = ibsrq->device;
54		event.element.srq = ibsrq;
55		switch (type) {
56		case MLX4_EVENT_TYPE_SRQ_LIMIT:
57			event.event = IB_EVENT_SRQ_LIMIT_REACHED;
58			break;
59		case MLX4_EVENT_TYPE_SRQ_CATAS_ERROR:
60			event.event = IB_EVENT_SRQ_ERR;
61			break;
62		default:
63			pr_warn("Unexpected event type %d "
64			       "on SRQ %06x\n", type, srq->srqn);
65			return;
66		}
67
68		ibsrq->event_handler(&event, ibsrq->srq_context);
69	}
70}
71
72int mlx4_ib_create_srq(struct ib_srq *ib_srq,
73		       struct ib_srq_init_attr *init_attr,
74		       struct ib_udata *udata)
75{
76	struct mlx4_ib_dev *dev = to_mdev(ib_srq->device);
77	struct mlx4_ib_ucontext *ucontext = rdma_udata_to_drv_context(
78		udata, struct mlx4_ib_ucontext, ibucontext);
79	struct mlx4_ib_srq *srq = to_msrq(ib_srq);
80	struct mlx4_wqe_srq_next_seg *next;
81	struct mlx4_wqe_data_seg *scatter;
82	u32 cqn;
83	u16 xrcdn;
84	int desc_size;
85	int buf_size;
86	int err;
87	int i;
88
89	/* Sanity check SRQ size before proceeding */
90	if (init_attr->attr.max_wr  >= dev->dev->caps.max_srq_wqes ||
91	    init_attr->attr.max_sge >  dev->dev->caps.max_srq_sge)
92		return -EINVAL;
93
94	mutex_init(&srq->mutex);
95	spin_lock_init(&srq->lock);
96	srq->msrq.max    = roundup_pow_of_two(init_attr->attr.max_wr + 1);
97	srq->msrq.max_gs = init_attr->attr.max_sge;
98
99	desc_size = max(32UL,
100			roundup_pow_of_two(sizeof (struct mlx4_wqe_srq_next_seg) +
101					   srq->msrq.max_gs *
102					   sizeof (struct mlx4_wqe_data_seg)));
103	srq->msrq.wqe_shift = ilog2(desc_size);
104
105	buf_size = srq->msrq.max * desc_size;
106
107	if (udata) {
108		struct mlx4_ib_create_srq ucmd;
109
110		if (ib_copy_from_udata(&ucmd, udata, sizeof(ucmd)))
111			return -EFAULT;
112
113		srq->umem =
114			ib_umem_get(ib_srq->device, ucmd.buf_addr, buf_size, 0);
115		if (IS_ERR(srq->umem))
116			return PTR_ERR(srq->umem);
117
118		err = mlx4_mtt_init(
119			dev->dev, ib_umem_num_dma_blocks(srq->umem, PAGE_SIZE),
120			PAGE_SHIFT, &srq->mtt);
121		if (err)
122			goto err_buf;
123
124		err = mlx4_ib_umem_write_mtt(dev, &srq->mtt, srq->umem);
125		if (err)
126			goto err_mtt;
127
128		err = mlx4_ib_db_map_user(udata, ucmd.db_addr, &srq->db);
129		if (err)
130			goto err_mtt;
131	} else {
132		err = mlx4_db_alloc(dev->dev, &srq->db, 0);
133		if (err)
134			return err;
135
136		*srq->db.db = 0;
137
138		if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2,
139				   &srq->buf)) {
140			err = -ENOMEM;
141			goto err_db;
142		}
143
144		srq->head    = 0;
145		srq->tail    = srq->msrq.max - 1;
146		srq->wqe_ctr = 0;
147
148		for (i = 0; i < srq->msrq.max; ++i) {
149			next = get_wqe(srq, i);
150			next->next_wqe_index =
151				cpu_to_be16((i + 1) & (srq->msrq.max - 1));
152
153			for (scatter = (void *) (next + 1);
154			     (void *) scatter < (void *) next + desc_size;
155			     ++scatter)
156				scatter->lkey = cpu_to_be32(MLX4_INVALID_LKEY);
157		}
158
159		err = mlx4_mtt_init(dev->dev, srq->buf.npages, srq->buf.page_shift,
160				    &srq->mtt);
161		if (err)
162			goto err_buf;
163
164		err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf);
165		if (err)
166			goto err_mtt;
167
168		srq->wrid = kvmalloc_array(srq->msrq.max,
169					   sizeof(u64), GFP_KERNEL);
170		if (!srq->wrid) {
171			err = -ENOMEM;
172			goto err_mtt;
173		}
174	}
175
176	cqn = ib_srq_has_cq(init_attr->srq_type) ?
177		to_mcq(init_attr->ext.cq)->mcq.cqn : 0;
178	xrcdn = (init_attr->srq_type == IB_SRQT_XRC) ?
179		to_mxrcd(init_attr->ext.xrc.xrcd)->xrcdn :
180		(u16) dev->dev->caps.reserved_xrcds;
181	err = mlx4_srq_alloc(dev->dev, to_mpd(ib_srq->pd)->pdn, cqn, xrcdn,
182			     &srq->mtt, srq->db.dma, &srq->msrq);
183	if (err)
184		goto err_wrid;
185
186	srq->msrq.event = mlx4_ib_srq_event;
187	srq->ibsrq.ext.xrc.srq_num = srq->msrq.srqn;
188
189	if (udata)
190		if (ib_copy_to_udata(udata, &srq->msrq.srqn, sizeof (__u32))) {
191			err = -EFAULT;
192			goto err_wrid;
193		}
194
195	init_attr->attr.max_wr = srq->msrq.max - 1;
196
197	return 0;
198
199err_wrid:
200	if (udata)
201		mlx4_ib_db_unmap_user(ucontext, &srq->db);
202	else
203		kvfree(srq->wrid);
204
205err_mtt:
206	mlx4_mtt_cleanup(dev->dev, &srq->mtt);
207
208err_buf:
209	if (!srq->umem)
210		mlx4_buf_free(dev->dev, buf_size, &srq->buf);
211	ib_umem_release(srq->umem);
212
213err_db:
214	if (!udata)
215		mlx4_db_free(dev->dev, &srq->db);
216
217	return err;
218}
219
220int mlx4_ib_modify_srq(struct ib_srq *ibsrq, struct ib_srq_attr *attr,
221		       enum ib_srq_attr_mask attr_mask, struct ib_udata *udata)
222{
223	struct mlx4_ib_dev *dev = to_mdev(ibsrq->device);
224	struct mlx4_ib_srq *srq = to_msrq(ibsrq);
225	int ret;
226
227	/* We don't support resizing SRQs (yet?) */
228	if (attr_mask & IB_SRQ_MAX_WR)
229		return -EINVAL;
230
231	if (attr_mask & IB_SRQ_LIMIT) {
232		if (attr->srq_limit >= srq->msrq.max)
233			return -EINVAL;
234
235		mutex_lock(&srq->mutex);
236		ret = mlx4_srq_arm(dev->dev, &srq->msrq, attr->srq_limit);
237		mutex_unlock(&srq->mutex);
238
239		if (ret)
240			return ret;
241	}
242
243	return 0;
244}
245
246int mlx4_ib_query_srq(struct ib_srq *ibsrq, struct ib_srq_attr *srq_attr)
247{
248	struct mlx4_ib_dev *dev = to_mdev(ibsrq->device);
249	struct mlx4_ib_srq *srq = to_msrq(ibsrq);
250	int ret;
251	int limit_watermark;
252
253	ret = mlx4_srq_query(dev->dev, &srq->msrq, &limit_watermark);
254	if (ret)
255		return ret;
256
257	srq_attr->srq_limit = limit_watermark;
258	srq_attr->max_wr    = srq->msrq.max - 1;
259	srq_attr->max_sge   = srq->msrq.max_gs;
260
261	return 0;
262}
263
264int mlx4_ib_destroy_srq(struct ib_srq *srq, struct ib_udata *udata)
265{
266	struct mlx4_ib_dev *dev = to_mdev(srq->device);
267	struct mlx4_ib_srq *msrq = to_msrq(srq);
268
269	mlx4_srq_free(dev->dev, &msrq->msrq);
270	mlx4_mtt_cleanup(dev->dev, &msrq->mtt);
271
272	if (udata) {
273		mlx4_ib_db_unmap_user(
274			rdma_udata_to_drv_context(
275				udata,
276				struct mlx4_ib_ucontext,
277				ibucontext),
278			&msrq->db);
279	} else {
280		kvfree(msrq->wrid);
281		mlx4_buf_free(dev->dev, msrq->msrq.max << msrq->msrq.wqe_shift,
282			      &msrq->buf);
283		mlx4_db_free(dev->dev, &msrq->db);
284	}
285	ib_umem_release(msrq->umem);
286	return 0;
287}
288
289void mlx4_ib_free_srq_wqe(struct mlx4_ib_srq *srq, int wqe_index)
290{
291	struct mlx4_wqe_srq_next_seg *next;
292
293	/* always called with interrupts disabled. */
294	spin_lock(&srq->lock);
295
296	next = get_wqe(srq, srq->tail);
297	next->next_wqe_index = cpu_to_be16(wqe_index);
298	srq->tail = wqe_index;
299
300	spin_unlock(&srq->lock);
301}
302
303int mlx4_ib_post_srq_recv(struct ib_srq *ibsrq, const struct ib_recv_wr *wr,
304			  const struct ib_recv_wr **bad_wr)
305{
306	struct mlx4_ib_srq *srq = to_msrq(ibsrq);
307	struct mlx4_wqe_srq_next_seg *next;
308	struct mlx4_wqe_data_seg *scat;
309	unsigned long flags;
310	int err = 0;
311	int nreq;
312	int i;
313	struct mlx4_ib_dev *mdev = to_mdev(ibsrq->device);
314
315	spin_lock_irqsave(&srq->lock, flags);
316	if (mdev->dev->persist->state & MLX4_DEVICE_STATE_INTERNAL_ERROR) {
317		err = -EIO;
318		*bad_wr = wr;
319		nreq = 0;
320		goto out;
321	}
322
323	for (nreq = 0; wr; ++nreq, wr = wr->next) {
324		if (unlikely(wr->num_sge > srq->msrq.max_gs)) {
325			err = -EINVAL;
326			*bad_wr = wr;
327			break;
328		}
329
330		if (unlikely(srq->head == srq->tail)) {
331			err = -ENOMEM;
332			*bad_wr = wr;
333			break;
334		}
335
336		srq->wrid[srq->head] = wr->wr_id;
337
338		next      = get_wqe(srq, srq->head);
339		srq->head = be16_to_cpu(next->next_wqe_index);
340		scat      = (struct mlx4_wqe_data_seg *) (next + 1);
341
342		for (i = 0; i < wr->num_sge; ++i) {
343			scat[i].byte_count = cpu_to_be32(wr->sg_list[i].length);
344			scat[i].lkey       = cpu_to_be32(wr->sg_list[i].lkey);
345			scat[i].addr       = cpu_to_be64(wr->sg_list[i].addr);
346		}
347
348		if (i < srq->msrq.max_gs) {
349			scat[i].byte_count = 0;
350			scat[i].lkey       = cpu_to_be32(MLX4_INVALID_LKEY);
351			scat[i].addr       = 0;
352		}
353	}
354
355	if (likely(nreq)) {
356		srq->wqe_ctr += nreq;
357
358		/*
359		 * Make sure that descriptors are written before
360		 * doorbell record.
361		 */
362		wmb();
363
364		*srq->db.db = cpu_to_be32(srq->wqe_ctr);
365	}
366out:
367
368	spin_unlock_irqrestore(&srq->lock, flags);
369
370	return err;
371}
372