summaryrefslogtreecommitdiffstats
path: root/net/sunrpc/xprtrdma/fmr_ops.c
blob: 29fc84c7ff98ff49b272deeb3b1a46759b794ca2 (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
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
// SPDX-License-Identifier: GPL-2.0
/*
 * 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).
 */

#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)

/* Access mode of externally registered pages */
enum {
	RPCRDMA_FMR_ACCESS_FLAGS	= IB_ACCESS_REMOTE_WRITE |
					  IB_ACCESS_REMOTE_READ,
};

bool
fmr_is_supported(struct rpcrdma_ia *ia)
{
	if (!ia->ri_device->alloc_fmr) {
		pr_info("rpcrdma: 'fmr' mode is not supported by device %s\n",
			ia->ri_device->name);
		return false;
	}
	return true;
}

static int
fmr_op_init_mr(struct rpcrdma_ia *ia, struct rpcrdma_mw *mw)
{
	static struct ib_fmr_attr fmr_attr = {
		.max_pages	= RPCRDMA_MAX_FMR_SGES,
		.max_maps	= 1,
		.page_shift	= PAGE_SHIFT
	};

	mw->fmr.fm_physaddrs = kcalloc(RPCRDMA_MAX_FMR_SGES,
				       sizeof(u64), GFP_KERNEL);
	if (!mw->fmr.fm_physaddrs)
		goto out_free;

	mw->mw_sg = kcalloc(RPCRDMA_MAX_FMR_SGES,
			    sizeof(*mw->mw_sg), GFP_KERNEL);
	if (!mw->mw_sg)
		goto out_free;

	sg_init_table(mw->mw_sg, RPCRDMA_MAX_FMR_SGES);

	mw->fmr.fm_mr = ib_alloc_fmr(ia->ri_pd, RPCRDMA_FMR_ACCESS_FLAGS,
				     &fmr_attr);
	if (IS_ERR(mw->fmr.fm_mr))
		goto out_fmr_err;

	return 0;

out_fmr_err:
	dprintk("RPC:       %s: ib_alloc_fmr returned %ld\n", __func__,
		PTR_ERR(mw->fmr.fm_mr));

out_free:
	kfree(mw->mw_sg);
	kfree(mw->fmr.fm_physaddrs);
	return -ENOMEM;
}

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

	list_add(&mw->fmr.fm_mr->list, &l);
	rc = ib_unmap_fmr(&l);
	list_del(&mw->fmr.fm_mr->list);
	return rc;
}

static void
fmr_op_release_mr(struct rpcrdma_mw *r)
{
	LIST_HEAD(unmap_list);
	int rc;

	/* Ensure MW is not on any rl_registered list */
	if (!list_empty(&r->mw_list))
		list_del(&r->mw_list);

	kfree(r->fmr.fm_physaddrs);
	kfree(r->mw_sg);

	/* In case this one was left mapped, try to unmap it
	 * to prevent dealloc_fmr from failing with EBUSY
	 */
	rc = __fmr_unmap(r);
	if (rc)
		pr_err("rpcrdma: final ib_unmap_fmr for %p failed %i\n",
		       r, rc);

	rc = ib_dealloc_fmr(r->fmr.fm_mr);
	if (rc)
		pr_err("rpcrdma: final ib_dealloc_fmr for %p returned %i\n",
		       r, rc);

	kfree(r);
}

/* Reset of a single FMR.
 */
static void
fmr_op_recover_mr(struct rpcrdma_mw *mw)
{
	struct rpcrdma_xprt *r_xprt = mw->mw_xprt;
	int rc;

	/* ORDER: invalidate first */
	rc = __fmr_unmap(mw);

	/* ORDER: then DMA unmap */
	ib_dma_unmap_sg(r_xprt->rx_ia.ri_device,
			mw->mw_sg, mw->mw_nents, mw->mw_dir);
	if (rc)
		goto out_release;

	rpcrdma_put_mw(r_xprt, mw);
	r_xprt->rx_stats.mrs_recovered++;
	return;

out_release:
	pr_err("rpcrdma: FMR reset failed (%d), %p released\n", rc, mw);
	r_xprt->rx_stats.mrs_orphaned++;

	spin_lock(&r_xprt->rx_buf.rb_mwlock);
	list_del(&mw->mw_all);
	spin_unlock(&r_xprt->rx_buf.rb_mwlock);

	fmr_op_release_mr(mw);
}

static int
fmr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep,
	    struct rpcrdma_create_data_internal *cdata)
{
	ia->ri_max_segs = max_t(unsigned int, 1, RPCRDMA_MAX_DATA_SEGS /
				RPCRDMA_MAX_FMR_SGES);
	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_HDR_SEGS * RPCRDMA_MAX_FMR_SGES);
}

/* Use the ib_map_phys_fmr() verb to register a memory region
 * for remote access via RDMA READ or RDMA WRITE.
 */
static struct rpcrdma_mr_seg *
fmr_op_map(struct rpcrdma_xprt *r_xprt, struct rpcrdma_mr_seg *seg,
	   int nsegs, bool writing, struct rpcrdma_mw **out)
{
	struct rpcrdma_mr_seg *seg1 = seg;
	int len, pageoff, i, rc;
	struct rpcrdma_mw *mw;
	u64 *dma_pages;

	mw = rpcrdma_get_mw(r_xprt);
	if (!mw)
		return ERR_PTR(-ENOBUFS);

	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;) {
		if (seg->mr_page)
			sg_set_page(&mw->mw_sg[i],
				    seg->mr_page,
				    seg->mr_len,
				    offset_in_page(seg->mr_offset));
		else
			sg_set_buf(&mw->mw_sg[i], seg->mr_offset,
				   seg->mr_len);
		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;
	}
	mw->mw_dir = rpcrdma_data_dir(writing);

	mw->mw_nents = ib_dma_map_sg(r_xprt->rx_ia.ri_device,
				     mw->mw_sg, i, mw->mw_dir);
	if (!mw->mw_nents)
		goto out_dmamap_err;

	for (i = 0, dma_pages = mw->fmr.fm_physaddrs; i < mw->mw_nents; i++)
		dma_pages[i] = sg_dma_address(&mw->mw_sg[i]);
	rc = ib_map_phys_fmr(mw->fmr.fm_mr, dma_pages, mw->mw_nents,
			     dma_pages[0]);
	if (rc)
		goto out_maperr;

	mw->mw_handle = mw->fmr.fm_mr->rkey;
	mw->mw_length = len;
	mw->mw_offset = dma_pages[0] + pageoff;

	*out = mw;
	return seg;

out_dmamap_err:
	pr_err("rpcrdma: failed to DMA map sg %p sg_nents %d\n",
	       mw->mw_sg, i);
	rpcrdma_put_mw(r_xprt, mw);
	return ERR_PTR(-EIO);

out_maperr:
	pr_err("rpcrdma: ib_map_phys_fmr %u@0x%llx+%i (%d) status %i\n",
	       len, (unsigned long long)dma_pages[0],
	       pageoff, mw->mw_nents, rc);
	ib_dma_unmap_sg(r_xprt->rx_ia.ri_device,
			mw->mw_sg, mw->mw_nents, mw->mw_dir);
	rpcrdma_put_mw(r_xprt, mw);
	return ERR_PTR(-EIO);
}

/* Invalidate all memory regions that were registered for "req".
 *
 * Sleeps until it is safe for the host CPU to access the
 * previously mapped memory regions.
 *
 * Caller ensures that @mws is not empty before the call. This
 * function empties the list.
 */
static void
fmr_op_unmap_sync(struct rpcrdma_xprt *r_xprt, struct list_head *mws)
{
	struct rpcrdma_mw *mw;
	LIST_HEAD(unmap_list);
	int rc;

	/* ORDER: Invalidate all of the req's MRs first
	 *
	 * ib_unmap_fmr() is slow, so use a single call instead
	 * of one call per mapped FMR.
	 */
	list_for_each_entry(mw, mws, mw_list) {
		dprintk("RPC:       %s: unmapping fmr %p\n",
			__func__, &mw->fmr);
		list_add_tail(&mw->fmr.fm_mr->list, &unmap_list);
	}
	r_xprt->rx_stats.local_inv_needed++;
	rc = ib_unmap_fmr(&unmap_list);
	if (rc)
		goto out_reset;

	/* ORDER: Now DMA unmap all of the req's MRs, and return
	 * them to the free MW list.
	 */
	while (!list_empty(mws)) {
		mw = rpcrdma_pop_mw(mws);
		dprintk("RPC:       %s: DMA unmapping fmr %p\n",
			__func__, &mw->fmr);
		list_del(&mw->fmr.fm_mr->list);
		ib_dma_unmap_sg(r_xprt->rx_ia.ri_device,
				mw->mw_sg, mw->mw_nents, mw->mw_dir);
		rpcrdma_put_mw(r_xprt, mw);
	}

	return;

out_reset:
	pr_err("rpcrdma: ib_unmap_fmr failed (%i)\n", rc);

	while (!list_empty(mws)) {
		mw = rpcrdma_pop_mw(mws);
		list_del(&mw->fmr.fm_mr->list);
		fmr_op_recover_mr(mw);
	}
}

const struct rpcrdma_memreg_ops rpcrdma_fmr_memreg_ops = {
	.ro_map				= fmr_op_map,
	.ro_unmap_sync			= fmr_op_unmap_sync,
	.ro_recover_mr			= fmr_op_recover_mr,
	.ro_open			= fmr_op_open,
	.ro_maxpages			= fmr_op_maxpages,
	.ro_init_mr			= fmr_op_init_mr,
	.ro_release_mr			= fmr_op_release_mr,
	.ro_displayname			= "fmr",
	.ro_send_w_inv_ok		= 0,
};