summaryrefslogtreecommitdiffstats
path: root/module/zcommon/zfs_uio.c
blob: a2c1b5c3aafe347bfedae18f3d2307da63cb7c73 (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
/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
 * Use is subject to license terms.
 */

/*	Copyright (c) 1984, 1986, 1987, 1988, 1989 AT&T	*/
/*	  All Rights Reserved	*/

/*
 * University Copyright- Copyright (c) 1982, 1986, 1988
 * The Regents of the University of California
 * All Rights Reserved
 *
 * University Acknowledgment- Portions of this document are derived from
 * software developed by the University of California, Berkeley, and its
 * contributors.
 */
/*
 * Copyright (c) 2015 by Chunwei Chen. All rights reserved.
 */

/*
 * The uio support from OpenSolaris has been added as a short term
 * work around.  The hope is to adopt native Linux type and drop the
 * use of uio's entirely.  Under Linux they only add overhead and
 * when possible we want to use native APIs for the ZPL layer.
 */
#ifdef _KERNEL

#include <sys/types.h>
#include <sys/uio_impl.h>
#include <sys/sysmacros.h>
#include <sys/strings.h>
#include <linux/kmap_compat.h>
#include <linux/uaccess.h>

/*
 * Move "n" bytes at byte address "p"; "rw" indicates the direction
 * of the move, and the I/O parameters are provided in "uio", which is
 * update to reflect the data which was moved.  Returns 0 on success or
 * a non-zero errno on failure.
 */
static int
uiomove_iov(void *p, size_t n, enum uio_rw rw, struct uio *uio)
{
	const struct iovec *iov = uio->uio_iov;
	size_t skip = uio->uio_skip;
	ulong_t cnt;

	while (n && uio->uio_resid) {
		cnt = MIN(iov->iov_len - skip, n);
		switch (uio->uio_segflg) {
		case UIO_USERSPACE:
		case UIO_USERISPACE:
			/*
			 * p = kernel data pointer
			 * iov->iov_base = user data pointer
			 */
			if (rw == UIO_READ) {
				if (copy_to_user(iov->iov_base+skip, p, cnt))
					return (EFAULT);
			} else {
				if (uio->uio_fault_disable) {
					if (!access_ok(VERIFY_READ,
					    (iov->iov_base + skip), cnt)) {
						return (EFAULT);
					}

					pagefault_disable();
					if (__copy_from_user_inatomic(p,
					    (iov->iov_base + skip), cnt)) {
						pagefault_enable();
						return (EFAULT);
					}
					pagefault_enable();
				} else {
					if (copy_from_user(p,
					    (iov->iov_base + skip), cnt))
						return (EFAULT);
				}
			}
			break;
		case UIO_SYSSPACE:
			if (rw == UIO_READ)
				bcopy(p, iov->iov_base + skip, cnt);
			else
				bcopy(iov->iov_base + skip, p, cnt);
			break;
		default:
			ASSERT(0);
		}
		skip += cnt;
		if (skip == iov->iov_len) {
			skip = 0;
			uio->uio_iov = (++iov);
			uio->uio_iovcnt--;
		}
		uio->uio_skip = skip;
		uio->uio_resid -= cnt;
		uio->uio_loffset += cnt;
		p = (caddr_t)p + cnt;
		n -= cnt;
	}
	return (0);
}

static int
uiomove_bvec(void *p, size_t n, enum uio_rw rw, struct uio *uio)
{
	const struct bio_vec *bv = uio->uio_bvec;
	size_t skip = uio->uio_skip;
	ulong_t cnt;

	while (n && uio->uio_resid) {
		void *paddr;
		cnt = MIN(bv->bv_len - skip, n);

		paddr = zfs_kmap_atomic(bv->bv_page, KM_USER1);
		if (rw == UIO_READ)
			bcopy(p, paddr + bv->bv_offset + skip, cnt);
		else
			bcopy(paddr + bv->bv_offset + skip, p, cnt);
		zfs_kunmap_atomic(paddr, KM_USER1);

		skip += cnt;
		if (skip == bv->bv_len) {
			skip = 0;
			uio->uio_bvec = (++bv);
			uio->uio_iovcnt--;
		}
		uio->uio_skip = skip;
		uio->uio_resid -= cnt;
		uio->uio_loffset += cnt;
		p = (caddr_t)p + cnt;
		n -= cnt;
	}
	return (0);
}

int
uiomove(void *p, size_t n, enum uio_rw rw, struct uio *uio)
{
	if (uio->uio_segflg != UIO_BVEC)
		return (uiomove_iov(p, n, rw, uio));
	else
		return (uiomove_bvec(p, n, rw, uio));
}
EXPORT_SYMBOL(uiomove);

#define	fuword8(uptr, vptr)	get_user((*vptr), (uptr))

/*
 * Fault in the pages of the first n bytes specified by the uio structure.
 * 1 byte in each page is touched and the uio struct is unmodified. Any
 * error will terminate the process as this is only a best attempt to get
 * the pages resident.
 */
int
uio_prefaultpages(ssize_t n, struct uio *uio)
{
	const struct iovec *iov;
	ulong_t cnt, incr;
	caddr_t p;
	uint8_t tmp;
	int iovcnt;
	size_t skip;

	/* no need to fault in kernel pages */
	switch (uio->uio_segflg) {
		case UIO_SYSSPACE:
		case UIO_BVEC:
			return (0);
		case UIO_USERSPACE:
		case UIO_USERISPACE:
			break;
		default:
			ASSERT(0);
	}

	iov = uio->uio_iov;
	iovcnt = uio->uio_iovcnt;
	skip = uio->uio_skip;

	for (; n > 0 && iovcnt > 0; iov++, iovcnt--, skip = 0) {
		cnt = MIN(iov->iov_len - skip, n);
		/* empty iov */
		if (cnt == 0)
			continue;
		n -= cnt;
		/*
		 * touch each page in this segment.
		 */
		p = iov->iov_base + skip;
		while (cnt) {
			if (fuword8((uint8_t *)p, &tmp))
				return (EFAULT);
			incr = MIN(cnt, PAGESIZE);
			p += incr;
			cnt -= incr;
		}
		/*
		 * touch the last byte in case it straddles a page.
		 */
		p--;
		if (fuword8((uint8_t *)p, &tmp))
			return (EFAULT);
	}

	return (0);
}
EXPORT_SYMBOL(uio_prefaultpages);

/*
 * same as uiomove() but doesn't modify uio structure.
 * return in cbytes how many bytes were copied.
 */
int
uiocopy(void *p, size_t n, enum uio_rw rw, struct uio *uio, size_t *cbytes)
{
	struct uio uio_copy;
	int ret;

	bcopy(uio, &uio_copy, sizeof (struct uio));
	ret = uiomove(p, n, rw, &uio_copy);
	*cbytes = uio->uio_resid - uio_copy.uio_resid;
	return (ret);
}
EXPORT_SYMBOL(uiocopy);

/*
 * Drop the next n chars out of *uiop.
 */
void
uioskip(uio_t *uiop, size_t n)
{
	if (n > uiop->uio_resid)
		return;

	uiop->uio_skip += n;
	if (uiop->uio_segflg != UIO_BVEC) {
		while (uiop->uio_iovcnt &&
		    uiop->uio_skip >= uiop->uio_iov->iov_len) {
			uiop->uio_skip -= uiop->uio_iov->iov_len;
			uiop->uio_iov++;
			uiop->uio_iovcnt--;
		}
	} else {
		while (uiop->uio_iovcnt &&
		    uiop->uio_skip >= uiop->uio_bvec->bv_len) {
			uiop->uio_skip -= uiop->uio_bvec->bv_len;
			uiop->uio_bvec++;
			uiop->uio_iovcnt--;
		}
	}
	uiop->uio_loffset += n;
	uiop->uio_resid -= n;
}
EXPORT_SYMBOL(uioskip);
#endif /* _KERNEL */