summaryrefslogtreecommitdiff
path: root/net/sunrpc/xprtrdma/fmr_ops.c
blob: cb25c89da6239154475d6c31736e328d13f19134 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
/*
 * Copyright (c) 2015 Oracle.  All rights reserved.
 * Copyright (c) 2003-2007 Network Appliance, Inc. All rights reserved.
 */

/* Lightweight memory registration using Fast Memory Regions (FMR).
 * Referred to sometimes as MTHCAFMR mode.
 *
 * FMR uses synchronous memory registration and deregistration.
 * FMR registration is known to be fast, but FMR deregistration
 * can take tens of usecs to complete.
 */

/* Normal operation
 *
 * A Memory Region is prepared for RDMA READ or WRITE using the
 * ib_map_phys_fmr verb (fmr_op_map). When the RDMA operation is
 * finished, the Memory Region is unmapped using the ib_unmap_fmr
 * verb (fmr_op_unmap).
 */

/* Transport recovery
 *
 * After a transport reconnect, fmr_op_map re-uses the MR already
 * allocated for the RPC, but generates a fresh rkey then maps the
 * MR again. This process is synchronous.
 */

#include "xprt_rdma.h"

#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)
# define RPCDBG_FACILITY	RPCDBG_TRANS
#endif

/* Maximum scatter/gather per FMR */
#define RPCRDMA_MAX_FMR_SGES	(64)

static int
fmr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep,
	    struct rpcrdma_create_data_internal *cdata)
{
	struct ib_device_attr *devattr = &ia->ri_devattr;
	struct ib_mr *mr;

	/* Obtain an lkey to use for the regbufs, which are
	 * protected from remote access.
	 */
	if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) {
		ia->ri_dma_lkey = ia->ri_device->local_dma_lkey;
	} else {
		mr = ib_get_dma_mr(ia->ri_pd, IB_ACCESS_LOCAL_WRITE);
		if (IS_ERR(mr)) {
			pr_err("%s: ib_get_dma_mr for failed with %lX\n",
			       __func__, PTR_ERR(mr));
			return -ENOMEM;
		}
		ia->ri_dma_lkey = ia->ri_dma_mr->lkey;
		ia->ri_dma_mr = mr;
	}

	return 0;
}

/* FMR mode conveys up to 64 pages of payload per chunk segment.
 */
static size_t
fmr_op_maxpages(struct rpcrdma_xprt *r_xprt)
{
	return min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS,
		     rpcrdma_max_segments(r_xprt) * RPCRDMA_MAX_FMR_SGES);
}

static int
fmr_op_init(struct rpcrdma_xprt *r_xprt)
{
	struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
	int mr_access_flags = IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ;
	struct ib_fmr_attr fmr_attr = {
		.max_pages	= RPCRDMA_MAX_FMR_SGES,
		.max_maps	= 1,
		.page_shift	= PAGE_SHIFT
	};
	struct ib_pd *pd = r_xprt->rx_ia.ri_pd;
	struct rpcrdma_mw *r;
	int i, rc;

	spin_lock_init(&buf->rb_mwlock);
	INIT_LIST_HEAD(&buf->rb_mws);
	INIT_LIST_HEAD(&buf->rb_all);

	i = max_t(int, RPCRDMA_MAX_DATA_SEGS / RPCRDMA_MAX_FMR_SGES, 1);
	i += 2;				/* head + tail */
	i *= buf->rb_max_requests;	/* one set for each RPC slot */
	dprintk("RPC:       %s: initalizing %d FMRs\n", __func__, i);

	rc = -ENOMEM;
	while (i--) {
		r = kzalloc(sizeof(*r), GFP_KERNEL);
		if (!r)
			goto out;

		r->r.fmr.physaddrs = kmalloc(RPCRDMA_MAX_FMR_SGES *
					     sizeof(u64), GFP_KERNEL);
		if (!r->r.fmr.physaddrs)
			goto out_free;

		r->r.fmr.fmr = ib_alloc_fmr(pd, mr_access_flags, &fmr_attr);
		if (IS_ERR(r->r.fmr.fmr))
			goto out_fmr_err;

		list_add(&r->mw_list, &buf->rb_mws);
		list_add(&r->mw_all, &buf->rb_all);
	}
	return 0;

out_fmr_err:
	rc = PTR_ERR(r->r.fmr.fmr);
	dprintk("RPC:       %s: ib_alloc_fmr status %i\n", __func__, rc);
	kfree(r->r.fmr.physaddrs);
out_free:
	kfree(r);
out:
	return rc;
}

static int
__fmr_unmap(struct rpcrdma_mw *r)
{
	LIST_HEAD(l);

	list_add(&r->r.fmr.fmr->list, &l);
	return ib_unmap_fmr(&l);
}

/* Use the ib_map_phys_fmr() verb to register a memory region
 * for remote access via RDMA READ or RDMA WRITE.
 */
static int
fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
	   int nsegs, bool writing)
{
	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
	struct ib_device *device = ia->ri_device;
	enum dma_data_direction direction = rpcrdma_data_dir(writing);
	struct rpcrdma_mr_seg *seg1 = seg;
	int len, pageoff, i, rc;
	struct rpcrdma_mw *mw;

	mw = seg1->rl_mw;
	seg1->rl_mw = NULL;
	if (!mw) {
		mw = rpcrdma_get_mw(r_xprt);
		if (!mw)
			return -ENOMEM;
	} else {
		/* this is a retransmit; generate a fresh rkey */
		rc = __fmr_unmap(mw);
		if (rc)
			return rc;
	}

	pageoff = offset_in_page(seg1->mr_offset);
	seg1->mr_offset -= pageoff;	/* start of page */
	seg1->mr_len += pageoff;
	len = -pageoff;
	if (nsegs > RPCRDMA_MAX_FMR_SGES)
		nsegs = RPCRDMA_MAX_FMR_SGES;
	for (i = 0; i < nsegs;) {
		rpcrdma_map_one(device, seg, direction);
		mw->r.fmr.physaddrs[i] = seg->mr_dma;
		len += seg->mr_len;
		++seg;
		++i;
		/* Check for holes */
		if ((i < nsegs && offset_in_page(seg->mr_offset)) ||
		    offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len))
			break;
	}

	rc = ib_map_phys_fmr(mw->r.fmr.fmr, mw->r.fmr.physaddrs,
			     i, seg1->mr_dma);
	if (rc)
		goto out_maperr;

	seg1->rl_mw = mw;
	seg1->mr_rkey = mw->r.fmr.fmr->rkey;
	seg1->mr_base = seg1->mr_dma + pageoff;
	seg1->mr_nsegs = i;
	seg1->mr_len = len;
	return i;

out_maperr:
	dprintk("RPC:       %s: ib_map_phys_fmr %u@0x%llx+%i (%d) status %i\n",
		__func__, len, (unsigned long long)seg1->mr_dma,
		pageoff, i, rc);
	while (i--)
		rpcrdma_unmap_one(device, --seg);
	return rc;
}

/* Use the ib_unmap_fmr() verb to prevent further remote
 * access via RDMA READ or RDMA WRITE.
 */
static int
fmr_op_unmap(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg)
{
	struct rpcrdma_ia *ia = &r_xprt->rx_ia;
	struct rpcrdma_mr_seg *seg1 = seg;
	struct rpcrdma_mw *mw = seg1->rl_mw;
	int rc, nsegs = seg->mr_nsegs;

	dprintk("RPC:       %s: FMR %p\n", __func__, mw);

	seg1->rl_mw = NULL;
	while (seg1->mr_nsegs--)
		rpcrdma_unmap_one(ia->ri_device, seg++);
	rc = __fmr_unmap(mw);
	if (rc)
		goto out_err;
	rpcrdma_put_mw(r_xprt, mw);
	return nsegs;

out_err:
	/* The FMR is abandoned, but remains in rb_all. fmr_op_destroy
	 * will attempt to release it when the transport is destroyed.
	 */
	dprintk("RPC:       %s: ib_unmap_fmr status %i\n", __func__, rc);
	return nsegs;
}

static void
fmr_op_destroy(struct rpcrdma_buffer *buf)
{
	struct rpcrdma_mw *r;
	int rc;

	while (!list_empty(&buf->rb_all)) {
		r = list_entry(buf->rb_all.next, struct rpcrdma_mw, mw_all);
		list_del(&r->mw_all);
		kfree(r->r.fmr.physaddrs);

		rc = ib_dealloc_fmr(r->r.fmr.fmr);
		if (rc)
			dprintk("RPC:       %s: ib_dealloc_fmr failed %i\n",
				__func__, rc);

		kfree(r);
	}
}

const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
	.ro_map				= fmr_op_map,
	.ro_unmap			= fmr_op_unmap,
	.ro_open			= fmr_op_open,
	.ro_maxpages			= fmr_op_maxpages,
	.ro_init			= fmr_op_init,
	.ro_destroy			= fmr_op_destroy,
	.ro_displayname			= "fmr",
};