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
|
/*
* Copyright (c) 1989, 1993, 1995
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Poul-Henning Kamp of the FreeBSD Project.
*
* 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.
*
* @(#)vfs_cache.c 8.3 (Berkeley) 8/22/94
* $FreeBSD$
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/time.h>
#include <sys/mount.h>
#include <sys/vnode.h>
#include <sys/namei.h>
#include <sys/errno.h>
#include <sys/malloc.h>
#define MAXVNODEUSE 32
/*
* Name caching works as follows:
*
* Names found by directory scans are retained in a cache
* for future reference. It is managed LRU, so frequently
* used names will hang around. Cache is indexed by hash value
* obtained from (vp, name) where vp refers to the directory
* containing name.
*
* If it is a "negative" entry, (i.e. for a name that is known NOT to
* exist) the vnode pointer will be NULL.
*
* For simplicity (and economy of storage), names longer than
* a maximum length of NCHNAMLEN are not cached; they occur
* infrequently in any case, and are almost never of interest.
*
* Upon reaching the last segment of a path, if the reference
* is for DELETE, or NOCACHE is set (rewrite), and the
* name is located in the cache, it will be dropped.
*/
/*
* Structures associated with name cacheing.
*/
#define NCHHASH(dvp, cnp) \
(&nchashtbl[((dvp)->v_id + (cnp)->cn_hash) & nchash])
static LIST_HEAD(nchashhead, namecache) *nchashtbl; /* Hash Table */
static u_long nchash; /* size of hash table - 1 */
static int doingcache = 1; /* 1 => enable the cache */
SYSCTL_INT(_debug, OID_AUTO, vfscache, CTLFLAG_RW, &doingcache, 0, "");
static u_long numcache; /* number of cache entries allocated */
static TAILQ_HEAD(, namecache) nclruhead; /* LRU chain */
struct nchstats nchstats; /* cache effectiveness statistics */
#ifdef NCH_STATISTICS
u_long nchnbr;
#define NCHNBR(ncp) (ncp)->nc_nbr = ++nchnbr;
#define NCHHIT(ncp) (ncp)->nc_hits++
#else
#define NCHNBR(ncp)
#define NCHHIT(ncp)
#endif
/*
* Delete an entry from its hash list and move it to the front
* of the LRU list for immediate reuse.
*/
#define PURGE(ncp) { \
LIST_REMOVE(ncp, nc_hash); \
ncp->nc_hash.le_prev = 0; \
TAILQ_REMOVE(&nclruhead, ncp, nc_lru); \
TAILQ_INSERT_HEAD(&nclruhead, ncp, nc_lru); \
}
/*
* Move an entry that has been used to the tail of the LRU list
* so that it will be preserved for future use.
*/
#define TOUCH(ncp) { \
if (ncp->nc_lru.tqe_next != 0) { \
TAILQ_REMOVE(&nclruhead, ncp, nc_lru); \
TAILQ_INSERT_TAIL(&nclruhead, ncp, nc_lru); \
NCHNBR(ncp); \
} \
}
/*
* Lookup an entry in the cache
*
* We don't do this if the segment name is long, simply so the cache
* can avoid holding long names (which would either waste space, or
* add greatly to the complexity).
*
* Lookup is called with dvp pointing to the directory to search,
* cnp pointing to the name of the entry being sought. If the lookup
* succeeds, the vnode is returned in *vpp, and a status of -1 is
* returned. If the lookup determines that the name does not exist
* (negative cacheing), a status of ENOENT is returned. If the lookup
* fails, a status of zero is returned.
*/
int
cache_lookup(dvp, vpp, cnp)
struct vnode *dvp;
struct vnode **vpp;
struct componentname *cnp;
{
register struct namecache *ncp, *nnp;
register struct nchashhead *ncpp;
if (!doingcache) {
cnp->cn_flags &= ~MAKEENTRY;
return (0);
}
if (cnp->cn_namelen > NCHNAMLEN) {
nchstats.ncs_long++;
cnp->cn_flags &= ~MAKEENTRY;
return (0);
}
ncpp = NCHHASH(dvp, cnp);
for (ncp = ncpp->lh_first; ncp != 0; ncp = nnp) {
nnp = ncp->nc_hash.le_next;
/* If one of the vp's went stale, don't bother anymore. */
if ((ncp->nc_dvpid != ncp->nc_dvp->v_id) ||
(ncp->nc_vp && ncp->nc_vpid != ncp->nc_vp->v_id)) {
nchstats.ncs_falsehits++;
PURGE(ncp);
continue;
}
/* Now that we know the vp's to be valid, is it ours ? */
if (ncp->nc_dvp == dvp &&
ncp->nc_nlen == cnp->cn_namelen &&
!bcmp(ncp->nc_name, cnp->cn_nameptr, (u_int)ncp->nc_nlen))
break;
}
/* We failed to find an entry */
if (ncp == 0) {
nchstats.ncs_miss++;
return (0);
}
NCHHIT(ncp);
/* We don't want to have an entry, so dump it */
if ((cnp->cn_flags & MAKEENTRY) == 0) {
nchstats.ncs_badhits++;
PURGE(ncp);
return (0);
}
/* We found a "positive" match, return the vnode */
if (ncp->nc_vp) {
nchstats.ncs_goodhits++;
TOUCH(ncp);
*vpp = ncp->nc_vp;
if ((*vpp)->v_usage < MAXVNODEUSE)
(*vpp)->v_usage++;
return (-1);
}
/* We found a negative match, and want to create it, so purge */
if (cnp->cn_nameiop == CREATE) {
nchstats.ncs_badhits++;
PURGE(ncp);
return (0);
}
/*
* We found a "negative" match, ENOENT notifies client of this match.
* The nc_vpid field records whether this is a whiteout.
*/
nchstats.ncs_neghits++;
TOUCH(ncp);
cnp->cn_flags |= ncp->nc_vpid;
return (ENOENT);
}
/*
* Add an entry to the cache.
*/
void
cache_enter(dvp, vp, cnp)
struct vnode *dvp;
struct vnode *vp;
struct componentname *cnp;
{
register struct namecache *ncp;
register struct nchashhead *ncpp;
if (!doingcache)
return;
#ifdef DIAGNOSTIC
if (cnp->cn_namelen > NCHNAMLEN) {
printf("cache_enter: name too long");
return;
}
#endif
/*
* We allocate a new entry if we are less than the maximum
* allowed and the one at the front of the LRU list is in use.
* Otherwise we use the one at the front of the LRU list.
*/
if (numcache < desiredvnodes &&
((ncp = nclruhead.tqh_first) == NULL ||
ncp->nc_hash.le_prev != 0)) {
/* Add one more entry */
ncp = (struct namecache *)
malloc((u_long)sizeof *ncp, M_CACHE, M_WAITOK);
bzero((char *)ncp, sizeof *ncp);
numcache++;
} else if (ncp = nclruhead.tqh_first) {
/* reuse an old entry */
TAILQ_REMOVE(&nclruhead, ncp, nc_lru);
if (ncp->nc_hash.le_prev != 0) {
LIST_REMOVE(ncp, nc_hash);
ncp->nc_hash.le_prev = 0;
}
} else {
/* give up */
return;
}
/*
* Fill in cache info, if vp is NULL this is a "negative" cache entry.
* For negative entries, we have to record whether it is a whiteout.
* the whiteout flag is stored in the nc_vpid field which is
* otherwise unused.
*/
ncp->nc_vp = vp;
if (vp) {
ncp->nc_vpid = vp->v_id;
if (vp->v_usage < MAXVNODEUSE)
++vp->v_usage;
} else
ncp->nc_vpid = cnp->cn_flags & ISWHITEOUT;
ncp->nc_dvp = dvp;
ncp->nc_dvpid = dvp->v_id;
ncp->nc_nlen = cnp->cn_namelen;
bcopy(cnp->cn_nameptr, ncp->nc_name, (unsigned)ncp->nc_nlen);
TAILQ_INSERT_TAIL(&nclruhead, ncp, nc_lru);
ncpp = NCHHASH(dvp, cnp);
LIST_INSERT_HEAD(ncpp, ncp, nc_hash);
}
/*
* Name cache initialization, from vfs_init() when we are booting
*/
void
nchinit()
{
TAILQ_INIT(&nclruhead);
nchashtbl = phashinit(desiredvnodes, M_CACHE, &nchash);
}
/*
* Invalidate all entries to particular vnode.
*
* We actually just increment the v_id, that will do it. The stale entries
* will be purged by lookup as they get found. If the v_id wraps around, we
* need to ditch the entire cache, to avoid confusion. No valid vnode will
* ever have (v_id == 0).
*/
void
cache_purge(vp)
struct vnode *vp;
{
struct namecache *ncp;
struct nchashhead *ncpp;
static u_long nextvnodeid;
vp->v_id = ++nextvnodeid;
if (nextvnodeid != 0)
return;
for (ncpp = &nchashtbl[nchash - 1]; ncpp >= nchashtbl; ncpp--) {
while (ncp = ncpp->lh_first)
PURGE(ncp);
}
vp->v_id = ++nextvnodeid;
}
/*
* Flush all entries referencing a particular filesystem.
*
* Since we need to check it anyway, we will flush all the invalid
* entries at the same time.
*/
void
cache_purgevfs(mp)
struct mount *mp;
{
struct nchashhead *ncpp;
struct namecache *ncp, *nnp;
/* Scan hash tables for applicable entries */
for (ncpp = &nchashtbl[nchash - 1]; ncpp >= nchashtbl; ncpp--) {
for (ncp = ncpp->lh_first; ncp != 0; ncp = nnp) {
nnp = ncp->nc_hash.le_next;
if (ncp->nc_dvpid != ncp->nc_dvp->v_id ||
(ncp->nc_vp && ncp->nc_vpid != ncp->nc_vp->v_id) ||
ncp->nc_dvp->v_mount == mp) {
PURGE(ncp);
}
}
}
}
|