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
|
/* $NetBSD: bufq_readprio.c,v 1.16 2017/05/04 11:03:27 kamil Exp $ */
/* NetBSD: subr_disk.c,v 1.61 2004/09/25 03:30:44 thorpej Exp */
/*-
* Copyright (c) 1996, 1997, 1999, 2000 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
* NASA Ames Research Center.
*
* 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.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
*/
/*
* Copyright (c) 1982, 1986, 1988, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*
* 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. 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.
*
* @(#)ufs_disksubr.c 8.5 (Berkeley) 1/21/94
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: bufq_readprio.c,v 1.16 2017/05/04 11:03:27 kamil Exp $");
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/buf.h>
#include <sys/bufq.h>
#include <sys/bufq_impl.h>
#include <sys/kmem.h>
#include <sys/module.h>
/*
* Seek sort for disks.
*
* There are two queues. The first queue holds read requests; the second
* holds write requests. The read queue is first-come first-served; the
* write queue is sorted in ascendening block order.
* The read queue is processed first. After PRIO_READ_BURST consecutive
* read requests with non-empty write queue PRIO_WRITE_REQ requests from
* the write queue will be processed.
*/
#define PRIO_READ_BURST 48
#define PRIO_WRITE_REQ 16
struct bufq_prio {
TAILQ_HEAD(, buf) bq_read, bq_write; /* actual list of buffers */
struct buf *bq_write_next; /* next request in bq_write */
struct buf *bq_next; /* current request */
int bq_read_burst; /* # of consecutive reads */
};
static void bufq_readprio_init(struct bufq_state *);
static void bufq_prio_put(struct bufq_state *, struct buf *);
static struct buf *bufq_prio_get(struct bufq_state *, int);
BUFQ_DEFINE(readprio, 30, bufq_readprio_init);
static void
bufq_prio_put(struct bufq_state *bufq, struct buf *bp)
{
struct bufq_prio *prio = bufq_private(bufq);
struct buf *bq;
int sortby;
sortby = bufq->bq_flags & BUFQ_SORT_MASK;
/*
* If it's a read request append it to the list.
*/
if ((bp->b_flags & B_READ) == B_READ) {
TAILQ_INSERT_TAIL(&prio->bq_read, bp, b_actq);
return;
}
bq = TAILQ_FIRST(&prio->bq_write);
/*
* If the write list is empty, simply append it to the list.
*/
if (bq == NULL) {
TAILQ_INSERT_TAIL(&prio->bq_write, bp, b_actq);
prio->bq_write_next = bp;
return;
}
/*
* If we lie after the next request, insert after this request.
*/
if (buf_inorder(prio->bq_write_next, bp, sortby))
bq = prio->bq_write_next;
/*
* Search for the first request at a larger block number.
* We go before this request if it exists.
*/
while (bq != NULL && buf_inorder(bq, bp, sortby))
bq = TAILQ_NEXT(bq, b_actq);
if (bq != NULL)
TAILQ_INSERT_BEFORE(bq, bp, b_actq);
else
TAILQ_INSERT_TAIL(&prio->bq_write, bp, b_actq);
}
static struct buf *
bufq_prio_get(struct bufq_state *bufq, int remove)
{
struct bufq_prio *prio = bufq_private(bufq);
struct buf *bp;
/*
* If no current request, get next from the lists.
*/
if (prio->bq_next == NULL) {
/*
* If at least one list is empty, select the other.
*/
if (TAILQ_FIRST(&prio->bq_read) == NULL) {
prio->bq_next = prio->bq_write_next;
prio->bq_read_burst = 0;
} else if (prio->bq_write_next == NULL) {
bp = prio->bq_next = TAILQ_FIRST(&prio->bq_read);
prio->bq_read_burst = 0;
KASSERT((bp == NULL) ||
((bp->b_flags & B_READ) == B_READ));
} else {
/*
* Both list have requests. Select the read list up
* to PRIO_READ_BURST times, then select the write
* list PRIO_WRITE_REQ times.
*/
if (prio->bq_read_burst++ < PRIO_READ_BURST)
prio->bq_next = TAILQ_FIRST(&prio->bq_read);
else if (prio->bq_read_burst <
PRIO_READ_BURST + PRIO_WRITE_REQ)
prio->bq_next = prio->bq_write_next;
else {
prio->bq_next = TAILQ_FIRST(&prio->bq_read);
prio->bq_read_burst = 0;
}
}
}
bp = prio->bq_next;
if (bp != NULL && remove) {
if ((bp->b_flags & B_READ) == B_READ)
TAILQ_REMOVE(&prio->bq_read, bp, b_actq);
else {
/*
* Advance the write pointer before removing
* bp since it is actually prio->bq_write_next.
*/
prio->bq_write_next =
TAILQ_NEXT(prio->bq_write_next, b_actq);
TAILQ_REMOVE(&prio->bq_write, bp, b_actq);
if (prio->bq_write_next == NULL)
prio->bq_write_next =
TAILQ_FIRST(&prio->bq_write);
}
prio->bq_next = NULL;
}
return (bp);
}
static struct buf *
bufq_prio_cancel(struct bufq_state *bufq, struct buf *buf)
{
struct bufq_prio *prio = bufq_private(bufq);
struct buf *bq;
/* search read queue */
TAILQ_FOREACH(bq, &prio->bq_read, b_actq) {
if (bq == buf) {
TAILQ_REMOVE(&prio->bq_read, bq, b_actq);
/* force new section */
prio->bq_next = NULL;
return buf;
}
}
/* not found in read queue, search write queue */
TAILQ_FOREACH(bq, &prio->bq_write, b_actq) {
if (bq == buf) {
if (bq == prio->bq_write_next) {
/*
* Advance the write pointer before removing
* bp since it is actually prio->bq_write_next.
*/
prio->bq_write_next =
TAILQ_NEXT(prio->bq_write_next, b_actq);
TAILQ_REMOVE(&prio->bq_write, bq, b_actq);
if (prio->bq_write_next == NULL)
prio->bq_write_next =
TAILQ_FIRST(&prio->bq_write);
} else {
TAILQ_REMOVE(&prio->bq_write, bq, b_actq);
}
/* force new section */
prio->bq_next = NULL;
return buf;
}
}
/* still not found */
return NULL;
}
static void
bufq_prio_fini(struct bufq_state *bufq)
{
KASSERT(bufq->bq_private != NULL);
kmem_free(bufq->bq_private, sizeof(struct bufq_prio));
}
static void
bufq_readprio_init(struct bufq_state *bufq)
{
struct bufq_prio *prio;
bufq->bq_get = bufq_prio_get;
bufq->bq_put = bufq_prio_put;
bufq->bq_cancel = bufq_prio_cancel;
bufq->bq_fini = bufq_prio_fini;
bufq->bq_private = kmem_zalloc(sizeof(struct bufq_prio), KM_SLEEP);
prio = (struct bufq_prio *)bufq->bq_private;
TAILQ_INIT(&prio->bq_read);
TAILQ_INIT(&prio->bq_write);
}
MODULE(MODULE_CLASS_BUFQ, bufq_readprio, NULL);
static int
bufq_readprio_modcmd(modcmd_t cmd, void *opaque)
{
switch (cmd) {
case MODULE_CMD_INIT:
return bufq_register(&bufq_strat_readprio);
case MODULE_CMD_FINI:
return bufq_unregister(&bufq_strat_readprio);
default:
return ENOTTY;
}
}
|