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
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
|
/*
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)vm_swap.c 8.5 (Berkeley) 2/17/94
* $Id: vm_swap.c,v 1.8 1994/10/09 07:35:18 davidg Exp $
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/buf.h>
#include <sys/conf.h>
#include <sys/proc.h>
#include <sys/namei.h>
#include <sys/dmap.h> /* XXX */
#include <sys/vnode.h>
#include <sys/file.h>
#include <sys/rlist.h>
#include <miscfs/specfs/specdev.h>
/*
* Indirect driver for multi-controller paging.
*/
int nswap, nswdev;
int vm_swap_size;
#ifdef SEQSWAP
int niswdev; /* number of interleaved swap devices */
int niswap; /* size of interleaved swap area */
#endif
int bswneeded;
vm_offset_t swapbkva; /* swap buffers kva */
/*
* Set up swap devices.
* Initialize linked list of free swap
* headers. These do not actually point
* to buffers, but rather to pages that
* are being swapped in and out.
*/
void
swapinit()
{
register struct proc *p = &proc0; /* XXX */
struct swdevt *swp;
int error;
/*
* Count swap devices, and adjust total swap space available.
* Some of the space will not be countable until later (dynamically
* configurable devices) and some of the counted space will not be
* available until a swapon() system call is issued, both usually
* happen when the system goes multi-user.
*
* If using NFS for swap, swdevt[0] will already be bdevvp'd. XXX
*/
#ifdef SEQSWAP
nswdev = niswdev = 0;
nswap = niswap = 0;
/*
* All interleaved devices must come first
*/
for (swp = swdevt; swp->sw_dev != NODEV || swp->sw_vp != NULL; swp++) {
if (swp->sw_flags & SW_SEQUENTIAL)
break;
niswdev++;
if (swp->sw_nblks > niswap)
niswap = swp->sw_nblks;
}
niswap = roundup(niswap, dmmax);
niswap *= niswdev;
if (swdevt[0].sw_vp == NULL &&
bdevvp(swdevt[0].sw_dev, &swdevt[0].sw_vp))
panic("swapvp");
/*
* The remainder must be sequential
*/
for ( ; swp->sw_dev != NODEV; swp++) {
if ((swp->sw_flags & SW_SEQUENTIAL) == 0)
panic("binit: mis-ordered swap devices");
nswdev++;
if (swp->sw_nblks > 0) {
if (swp->sw_nblks % dmmax)
swp->sw_nblks -= (swp->sw_nblks % dmmax);
nswap += swp->sw_nblks;
}
}
nswdev += niswdev;
if (nswdev == 0)
panic("swapinit");
nswap += niswap;
#else
nswdev = 0;
nswap = 0;
for (swp = swdevt; swp->sw_dev != NODEV || swp->sw_vp != NULL; swp++) {
nswdev++;
if (swp->sw_nblks > nswap)
nswap = swp->sw_nblks;
}
if (nswdev == 0)
panic("swapinit");
if (nswdev > 1)
nswap = ((nswap + dmmax - 1) / dmmax) * dmmax;
nswap *= nswdev;
if (swdevt[0].sw_vp == NULL &&
bdevvp(swdevt[0].sw_dev, &swdevt[0].sw_vp))
panic("swapvp");
#endif
/*
* If there is no swap configured, tell the user. We don't activate
* any swapspaces in the kernel. The user must explicitly use swapon
* to tell start each device.
*/
if (nswap == 0)
printf("WARNING: no swap space found\n");
}
void
swstrategy(bp)
register struct buf *bp;
{
int sz, off, seg, index;
register struct swdevt *sp;
struct vnode *vp;
#ifdef GENERIC
/*
* A mini-root gets copied into the front of the swap
* and we run over top of the swap area just long
* enough for us to do a mkfs and restor of the real
* root (sure beats rewriting standalone restor).
*/
#define MINIROOTSIZE 4096
if (rootdev == dumpdev)
bp->b_blkno += MINIROOTSIZE;
#endif
sz = howmany(bp->b_bcount, DEV_BSIZE);
if (bp->b_blkno + sz > nswap) {
bp->b_error = EINVAL;
bp->b_flags |= B_ERROR;
biodone(bp);
return;
}
if (nswdev > 1) {
#ifdef SEQSWAP
if (bp->b_blkno < niswap) {
if (niswdev > 1) {
off = bp->b_blkno % dmmax;
if (off+sz > dmmax) {
bp->b_error = EINVAL;
bp->b_flags |= B_ERROR;
biodone(bp);
return;
}
seg = bp->b_blkno / dmmax;
index = seg % niswdev;
seg /= niswdev;
bp->b_blkno = seg*dmmax + off;
} else
index = 0;
} else {
register struct swdevt *swp;
bp->b_blkno -= niswap;
for (index = niswdev, swp = &swdevt[niswdev];
swp->sw_dev != NODEV;
swp++, index++) {
if (bp->b_blkno < swp->sw_nblks)
break;
bp->b_blkno -= swp->sw_nblks;
}
if (swp->sw_dev == NODEV ||
bp->b_blkno+sz > swp->sw_nblks) {
bp->b_error = swp->sw_dev == NODEV ?
ENODEV : EINVAL;
bp->b_flags |= B_ERROR;
biodone(bp);
return;
}
}
#else
off = bp->b_blkno % dmmax;
if (off+sz > dmmax) {
bp->b_error = EINVAL;
bp->b_flags |= B_ERROR;
biodone(bp);
return;
}
seg = bp->b_blkno / dmmax;
index = seg % nswdev;
seg /= nswdev;
bp->b_blkno = seg*dmmax + off;
#endif
} else
index = 0;
sp = &swdevt[index];
if ((bp->b_dev = sp->sw_dev) == NODEV)
panic("swstrategy");
if (sp->sw_vp == NULL) {
bp->b_error = ENODEV;
bp->b_flags |= B_ERROR;
biodone(bp);
return;
}
VHOLD(sp->sw_vp);
if ((bp->b_flags & B_READ) == 0) {
vp = bp->b_vp;
if (vp) {
vp->v_numoutput--;
if ((vp->v_flag & VBWAIT) && vp->v_numoutput <= 0) {
vp->v_flag &= ~VBWAIT;
wakeup((caddr_t)&vp->v_numoutput);
}
}
sp->sw_vp->v_numoutput++;
}
if (bp->b_vp != NULL)
brelvp(bp);
bp->b_vp = sp->sw_vp;
VOP_STRATEGY(bp);
}
/*
* System call swapon(name) enables swapping on device name,
* which must be in the swdevsw. Return EBUSY
* if already swapping on this device.
*/
struct swapon_args {
char *name;
};
/* ARGSUSED */
int
swapon(p, uap, retval)
struct proc *p;
struct swapon_args *uap;
int *retval;
{
register struct vnode *vp;
register struct swdevt *sp;
dev_t dev;
int error;
struct nameidata nd;
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return (error);
NDINIT(&nd, LOOKUP, FOLLOW, UIO_USERSPACE, uap->name, p);
error = namei(&nd);
if (error)
return (error);
vp = nd.ni_vp;
if (vp->v_type != VBLK) {
vrele(vp);
return (ENOTBLK);
}
dev = (dev_t)vp->v_rdev;
if (major(dev) >= nblkdev) {
vrele(vp);
return (ENXIO);
}
for (sp = &swdevt[0]; sp->sw_dev != NODEV; sp++) {
if (sp->sw_dev == dev) {
if (sp->sw_flags & SW_FREED) {
vrele(vp);
return (EBUSY);
}
sp->sw_vp = vp;
error = swfree(p, sp - swdevt);
if (error) {
vrele(vp);
return (error);
}
return (0);
}
#ifdef SEQSWAP
/*
* If we have reached a non-freed sequential device without
* finding what we are looking for, it is an error.
* That is because all interleaved devices must come first
* and sequential devices must be freed in order.
*/
if ((sp->sw_flags & (SW_SEQUENTIAL|SW_FREED)) == SW_SEQUENTIAL)
break;
#endif
}
vrele(vp);
return (EINVAL);
}
/*
* Swfree(index) frees the index'th portion of the swap map.
* Each of the nswdev devices provides 1/nswdev'th of the swap
* space, which is laid out with blocks of dmmax pages circularly
* among the devices.
*/
int
swfree(p, index)
struct proc *p;
int index;
{
register struct swdevt *sp;
register swblk_t vsbase;
register long blk;
struct vnode *vp;
register swblk_t dvbase;
register int nblks;
int error;
sp = &swdevt[index];
vp = sp->sw_vp;
error = VOP_OPEN(vp, FREAD|FWRITE, p->p_ucred, p);
if (error)
return (error);
sp->sw_flags |= SW_FREED;
nblks = sp->sw_nblks;
/*
* Some devices may not exist til after boot time.
* If so, their nblk count will be 0.
*/
if (nblks <= 0) {
int perdev;
dev_t dev = sp->sw_dev;
if (bdevsw[major(dev)].d_psize == 0 ||
(nblks = (*bdevsw[major(dev)].d_psize)(dev)) == -1) {
(void) VOP_CLOSE(vp, FREAD|FWRITE, p->p_ucred, p);
sp->sw_flags &= ~SW_FREED;
return (ENXIO);
}
#ifdef SEQSWAP
if (index < niswdev) {
perdev = niswap / niswdev;
if (nblks > perdev)
nblks = perdev;
} else {
if (nblks % dmmax)
nblks -= (nblks % dmmax);
nswap += nblks;
}
#else
perdev = nswap / nswdev;
if (nblks > perdev)
nblks = perdev;
#endif
sp->sw_nblks = nblks;
}
if (nblks == 0) {
(void) VOP_CLOSE(vp, FREAD|FWRITE, p->p_ucred, p);
sp->sw_flags &= ~SW_FREED;
return (0); /* XXX error? */
}
#ifdef SEQSWAP
if (sp->sw_flags & SW_SEQUENTIAL) {
register struct swdevt *swp;
blk = niswap;
for (swp = &swdevt[niswdev]; swp != sp; swp++)
blk += swp->sw_nblks;
rlist_free(&swaplist, blk, blk + nblks - 1);
vm_swap_size += nblks;
return (0);
}
#endif
for (dvbase = 0; dvbase < nblks; dvbase += dmmax) {
blk = nblks - dvbase;
#ifdef SEQSWAP
if ((vsbase = index*dmmax + dvbase*niswdev) >= niswap)
panic("swfree");
#else
if ((vsbase = index*dmmax + dvbase*nswdev) >= nswap)
panic("swfree");
#endif
if (blk > dmmax)
blk = dmmax;
/* XXX -- we need to exclude the first cluster as above */
/* but for now, this will work fine... */
rlist_free(&swaplist, vsbase, vsbase + blk - 1);
vm_swap_size += blk;
}
return (0);
}
|