xref: /netbsd-src/sys/nfs/nfs_subs.c (revision 0dd5877adce57db949b16ae963e5a6831cccdfb6)
1 /*	$NetBSD: nfs_subs.c,v 1.100 2002/01/26 02:52:20 chs Exp $	*/
2 
3 /*
4  * Copyright (c) 1989, 1993
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley by
8  * Rick Macklem at The University of Guelph.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  * 3. All advertising materials mentioning features or use of this software
19  *    must display the following acknowledgement:
20  *	This product includes software developed by the University of
21  *	California, Berkeley and its contributors.
22  * 4. Neither the name of the University nor the names of its contributors
23  *    may be used to endorse or promote products derived from this software
24  *    without specific prior written permission.
25  *
26  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
27  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
30  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
31  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
32  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
33  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
34  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
35  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
36  * SUCH DAMAGE.
37  *
38  *	@(#)nfs_subs.c	8.8 (Berkeley) 5/22/95
39  */
40 
41 /*
42  * Copyright 2000 Wasabi Systems, Inc.
43  * All rights reserved.
44  *
45  * Written by Frank van der Linden for Wasabi Systems, Inc.
46  *
47  * Redistribution and use in source and binary forms, with or without
48  * modification, are permitted provided that the following conditions
49  * are met:
50  * 1. Redistributions of source code must retain the above copyright
51  *    notice, this list of conditions and the following disclaimer.
52  * 2. Redistributions in binary form must reproduce the above copyright
53  *    notice, this list of conditions and the following disclaimer in the
54  *    documentation and/or other materials provided with the distribution.
55  * 3. All advertising materials mentioning features or use of this software
56  *    must display the following acknowledgement:
57  *      This product includes software developed for the NetBSD Project by
58  *      Wasabi Systems, Inc.
59  * 4. The name of Wasabi Systems, Inc. may not be used to endorse
60  *    or promote products derived from this software without specific prior
61  *    written permission.
62  *
63  * THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``AS IS'' AND
64  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
65  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
66  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL WASABI SYSTEMS, INC
67  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
68  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
69  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
70  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
71  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
72  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
73  * POSSIBILITY OF SUCH DAMAGE.
74  */
75 
76 #include <sys/cdefs.h>
77 __KERNEL_RCSID(0, "$NetBSD: nfs_subs.c,v 1.100 2002/01/26 02:52:20 chs Exp $");
78 
79 #include "fs_nfs.h"
80 #include "opt_nfs.h"
81 #include "opt_nfsserver.h"
82 #include "opt_iso.h"
83 #include "opt_inet.h"
84 
85 /*
86  * These functions support the macros and help fiddle mbuf chains for
87  * the nfs op functions. They do things like create the rpc header and
88  * copy data between mbuf chains and uio lists.
89  */
90 #include <sys/param.h>
91 #include <sys/proc.h>
92 #include <sys/systm.h>
93 #include <sys/kernel.h>
94 #include <sys/mount.h>
95 #include <sys/vnode.h>
96 #include <sys/namei.h>
97 #include <sys/mbuf.h>
98 #include <sys/socket.h>
99 #include <sys/stat.h>
100 #include <sys/malloc.h>
101 #include <sys/filedesc.h>
102 #include <sys/time.h>
103 #include <sys/dirent.h>
104 
105 #include <uvm/uvm_extern.h>
106 
107 #include <nfs/rpcv2.h>
108 #include <nfs/nfsproto.h>
109 #include <nfs/nfsnode.h>
110 #include <nfs/nfs.h>
111 #include <nfs/xdr_subs.h>
112 #include <nfs/nfsm_subs.h>
113 #include <nfs/nfsmount.h>
114 #include <nfs/nqnfs.h>
115 #include <nfs/nfsrtt.h>
116 #include <nfs/nfs_var.h>
117 
118 #include <miscfs/specfs/specdev.h>
119 
120 #include <netinet/in.h>
121 #ifdef ISO
122 #include <netiso/iso.h>
123 #endif
124 
125 /*
126  * Data items converted to xdr at startup, since they are constant
127  * This is kinda hokey, but may save a little time doing byte swaps
128  */
129 u_int32_t nfs_xdrneg1;
130 u_int32_t rpc_call, rpc_vers, rpc_reply, rpc_msgdenied, rpc_autherr,
131 	rpc_mismatch, rpc_auth_unix, rpc_msgaccepted,
132 	rpc_auth_kerb;
133 u_int32_t nfs_prog, nqnfs_prog, nfs_true, nfs_false;
134 
135 /* And other global data */
136 static u_int32_t nfs_xid = 0;
137 const nfstype nfsv2_type[9] =
138 	{ NFNON, NFREG, NFDIR, NFBLK, NFCHR, NFLNK, NFNON, NFCHR, NFNON };
139 const nfstype nfsv3_type[9] =
140 	{ NFNON, NFREG, NFDIR, NFBLK, NFCHR, NFLNK, NFSOCK, NFFIFO, NFNON };
141 const enum vtype nv2tov_type[8] =
142 	{ VNON, VREG, VDIR, VBLK, VCHR, VLNK, VNON, VNON };
143 const enum vtype nv3tov_type[8] =
144 	{ VNON, VREG, VDIR, VBLK, VCHR, VLNK, VSOCK, VFIFO };
145 int nfs_ticks;
146 int nfs_commitsize;
147 
148 /* NFS client/server stats. */
149 struct nfsstats nfsstats;
150 
151 /*
152  * Mapping of old NFS Version 2 RPC numbers to generic numbers.
153  */
154 const int nfsv3_procid[NFS_NPROCS] = {
155 	NFSPROC_NULL,
156 	NFSPROC_GETATTR,
157 	NFSPROC_SETATTR,
158 	NFSPROC_NOOP,
159 	NFSPROC_LOOKUP,
160 	NFSPROC_READLINK,
161 	NFSPROC_READ,
162 	NFSPROC_NOOP,
163 	NFSPROC_WRITE,
164 	NFSPROC_CREATE,
165 	NFSPROC_REMOVE,
166 	NFSPROC_RENAME,
167 	NFSPROC_LINK,
168 	NFSPROC_SYMLINK,
169 	NFSPROC_MKDIR,
170 	NFSPROC_RMDIR,
171 	NFSPROC_READDIR,
172 	NFSPROC_FSSTAT,
173 	NFSPROC_NOOP,
174 	NFSPROC_NOOP,
175 	NFSPROC_NOOP,
176 	NFSPROC_NOOP,
177 	NFSPROC_NOOP,
178 	NFSPROC_NOOP,
179 	NFSPROC_NOOP,
180 	NFSPROC_NOOP
181 };
182 
183 /*
184  * and the reverse mapping from generic to Version 2 procedure numbers
185  */
186 const int nfsv2_procid[NFS_NPROCS] = {
187 	NFSV2PROC_NULL,
188 	NFSV2PROC_GETATTR,
189 	NFSV2PROC_SETATTR,
190 	NFSV2PROC_LOOKUP,
191 	NFSV2PROC_NOOP,
192 	NFSV2PROC_READLINK,
193 	NFSV2PROC_READ,
194 	NFSV2PROC_WRITE,
195 	NFSV2PROC_CREATE,
196 	NFSV2PROC_MKDIR,
197 	NFSV2PROC_SYMLINK,
198 	NFSV2PROC_CREATE,
199 	NFSV2PROC_REMOVE,
200 	NFSV2PROC_RMDIR,
201 	NFSV2PROC_RENAME,
202 	NFSV2PROC_LINK,
203 	NFSV2PROC_READDIR,
204 	NFSV2PROC_NOOP,
205 	NFSV2PROC_STATFS,
206 	NFSV2PROC_NOOP,
207 	NFSV2PROC_NOOP,
208 	NFSV2PROC_NOOP,
209 	NFSV2PROC_NOOP,
210 	NFSV2PROC_NOOP,
211 	NFSV2PROC_NOOP,
212 	NFSV2PROC_NOOP,
213 };
214 
215 /*
216  * Maps errno values to nfs error numbers.
217  * Use NFSERR_IO as the catch all for ones not specifically defined in
218  * RFC 1094.
219  */
220 static const u_char nfsrv_v2errmap[ELAST] = {
221   NFSERR_PERM,	NFSERR_NOENT,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
222   NFSERR_NXIO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
223   NFSERR_IO,	NFSERR_IO,	NFSERR_ACCES,	NFSERR_IO,	NFSERR_IO,
224   NFSERR_IO,	NFSERR_EXIST,	NFSERR_IO,	NFSERR_NODEV,	NFSERR_NOTDIR,
225   NFSERR_ISDIR,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
226   NFSERR_IO,	NFSERR_FBIG,	NFSERR_NOSPC,	NFSERR_IO,	NFSERR_ROFS,
227   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
228   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
229   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
230   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
231   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
232   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
233   NFSERR_IO,	NFSERR_IO,	NFSERR_NAMETOL,	NFSERR_IO,	NFSERR_IO,
234   NFSERR_NOTEMPTY, NFSERR_IO,	NFSERR_IO,	NFSERR_DQUOT,	NFSERR_STALE,
235   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
236   NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,	NFSERR_IO,
237   NFSERR_IO,	NFSERR_IO,
238 };
239 
240 /*
241  * Maps errno values to nfs error numbers.
242  * Although it is not obvious whether or not NFS clients really care if
243  * a returned error value is in the specified list for the procedure, the
244  * safest thing to do is filter them appropriately. For Version 2, the
245  * X/Open XNFS document is the only specification that defines error values
246  * for each RPC (The RFC simply lists all possible error values for all RPCs),
247  * so I have decided to not do this for Version 2.
248  * The first entry is the default error return and the rest are the valid
249  * errors for that RPC in increasing numeric order.
250  */
251 static const short nfsv3err_null[] = {
252 	0,
253 	0,
254 };
255 
256 static const short nfsv3err_getattr[] = {
257 	NFSERR_IO,
258 	NFSERR_IO,
259 	NFSERR_STALE,
260 	NFSERR_BADHANDLE,
261 	NFSERR_SERVERFAULT,
262 	0,
263 };
264 
265 static const short nfsv3err_setattr[] = {
266 	NFSERR_IO,
267 	NFSERR_PERM,
268 	NFSERR_IO,
269 	NFSERR_ACCES,
270 	NFSERR_INVAL,
271 	NFSERR_NOSPC,
272 	NFSERR_ROFS,
273 	NFSERR_DQUOT,
274 	NFSERR_STALE,
275 	NFSERR_BADHANDLE,
276 	NFSERR_NOT_SYNC,
277 	NFSERR_SERVERFAULT,
278 	0,
279 };
280 
281 static const short nfsv3err_lookup[] = {
282 	NFSERR_IO,
283 	NFSERR_NOENT,
284 	NFSERR_IO,
285 	NFSERR_ACCES,
286 	NFSERR_NOTDIR,
287 	NFSERR_NAMETOL,
288 	NFSERR_STALE,
289 	NFSERR_BADHANDLE,
290 	NFSERR_SERVERFAULT,
291 	0,
292 };
293 
294 static const short nfsv3err_access[] = {
295 	NFSERR_IO,
296 	NFSERR_IO,
297 	NFSERR_STALE,
298 	NFSERR_BADHANDLE,
299 	NFSERR_SERVERFAULT,
300 	0,
301 };
302 
303 static const short nfsv3err_readlink[] = {
304 	NFSERR_IO,
305 	NFSERR_IO,
306 	NFSERR_ACCES,
307 	NFSERR_INVAL,
308 	NFSERR_STALE,
309 	NFSERR_BADHANDLE,
310 	NFSERR_NOTSUPP,
311 	NFSERR_SERVERFAULT,
312 	0,
313 };
314 
315 static const short nfsv3err_read[] = {
316 	NFSERR_IO,
317 	NFSERR_IO,
318 	NFSERR_NXIO,
319 	NFSERR_ACCES,
320 	NFSERR_INVAL,
321 	NFSERR_STALE,
322 	NFSERR_BADHANDLE,
323 	NFSERR_SERVERFAULT,
324 	NFSERR_JUKEBOX,
325 	0,
326 };
327 
328 static const short nfsv3err_write[] = {
329 	NFSERR_IO,
330 	NFSERR_IO,
331 	NFSERR_ACCES,
332 	NFSERR_INVAL,
333 	NFSERR_FBIG,
334 	NFSERR_NOSPC,
335 	NFSERR_ROFS,
336 	NFSERR_DQUOT,
337 	NFSERR_STALE,
338 	NFSERR_BADHANDLE,
339 	NFSERR_SERVERFAULT,
340 	NFSERR_JUKEBOX,
341 	0,
342 };
343 
344 static const short nfsv3err_create[] = {
345 	NFSERR_IO,
346 	NFSERR_IO,
347 	NFSERR_ACCES,
348 	NFSERR_EXIST,
349 	NFSERR_NOTDIR,
350 	NFSERR_NOSPC,
351 	NFSERR_ROFS,
352 	NFSERR_NAMETOL,
353 	NFSERR_DQUOT,
354 	NFSERR_STALE,
355 	NFSERR_BADHANDLE,
356 	NFSERR_NOTSUPP,
357 	NFSERR_SERVERFAULT,
358 	0,
359 };
360 
361 static const short nfsv3err_mkdir[] = {
362 	NFSERR_IO,
363 	NFSERR_IO,
364 	NFSERR_ACCES,
365 	NFSERR_EXIST,
366 	NFSERR_NOTDIR,
367 	NFSERR_NOSPC,
368 	NFSERR_ROFS,
369 	NFSERR_NAMETOL,
370 	NFSERR_DQUOT,
371 	NFSERR_STALE,
372 	NFSERR_BADHANDLE,
373 	NFSERR_NOTSUPP,
374 	NFSERR_SERVERFAULT,
375 	0,
376 };
377 
378 static const short nfsv3err_symlink[] = {
379 	NFSERR_IO,
380 	NFSERR_IO,
381 	NFSERR_ACCES,
382 	NFSERR_EXIST,
383 	NFSERR_NOTDIR,
384 	NFSERR_NOSPC,
385 	NFSERR_ROFS,
386 	NFSERR_NAMETOL,
387 	NFSERR_DQUOT,
388 	NFSERR_STALE,
389 	NFSERR_BADHANDLE,
390 	NFSERR_NOTSUPP,
391 	NFSERR_SERVERFAULT,
392 	0,
393 };
394 
395 static const short nfsv3err_mknod[] = {
396 	NFSERR_IO,
397 	NFSERR_IO,
398 	NFSERR_ACCES,
399 	NFSERR_EXIST,
400 	NFSERR_NOTDIR,
401 	NFSERR_NOSPC,
402 	NFSERR_ROFS,
403 	NFSERR_NAMETOL,
404 	NFSERR_DQUOT,
405 	NFSERR_STALE,
406 	NFSERR_BADHANDLE,
407 	NFSERR_NOTSUPP,
408 	NFSERR_SERVERFAULT,
409 	NFSERR_BADTYPE,
410 	0,
411 };
412 
413 static const short nfsv3err_remove[] = {
414 	NFSERR_IO,
415 	NFSERR_NOENT,
416 	NFSERR_IO,
417 	NFSERR_ACCES,
418 	NFSERR_NOTDIR,
419 	NFSERR_ROFS,
420 	NFSERR_NAMETOL,
421 	NFSERR_STALE,
422 	NFSERR_BADHANDLE,
423 	NFSERR_SERVERFAULT,
424 	0,
425 };
426 
427 static const short nfsv3err_rmdir[] = {
428 	NFSERR_IO,
429 	NFSERR_NOENT,
430 	NFSERR_IO,
431 	NFSERR_ACCES,
432 	NFSERR_EXIST,
433 	NFSERR_NOTDIR,
434 	NFSERR_INVAL,
435 	NFSERR_ROFS,
436 	NFSERR_NAMETOL,
437 	NFSERR_NOTEMPTY,
438 	NFSERR_STALE,
439 	NFSERR_BADHANDLE,
440 	NFSERR_NOTSUPP,
441 	NFSERR_SERVERFAULT,
442 	0,
443 };
444 
445 static const short nfsv3err_rename[] = {
446 	NFSERR_IO,
447 	NFSERR_NOENT,
448 	NFSERR_IO,
449 	NFSERR_ACCES,
450 	NFSERR_EXIST,
451 	NFSERR_XDEV,
452 	NFSERR_NOTDIR,
453 	NFSERR_ISDIR,
454 	NFSERR_INVAL,
455 	NFSERR_NOSPC,
456 	NFSERR_ROFS,
457 	NFSERR_MLINK,
458 	NFSERR_NAMETOL,
459 	NFSERR_NOTEMPTY,
460 	NFSERR_DQUOT,
461 	NFSERR_STALE,
462 	NFSERR_BADHANDLE,
463 	NFSERR_NOTSUPP,
464 	NFSERR_SERVERFAULT,
465 	0,
466 };
467 
468 static const short nfsv3err_link[] = {
469 	NFSERR_IO,
470 	NFSERR_IO,
471 	NFSERR_ACCES,
472 	NFSERR_EXIST,
473 	NFSERR_XDEV,
474 	NFSERR_NOTDIR,
475 	NFSERR_INVAL,
476 	NFSERR_NOSPC,
477 	NFSERR_ROFS,
478 	NFSERR_MLINK,
479 	NFSERR_NAMETOL,
480 	NFSERR_DQUOT,
481 	NFSERR_STALE,
482 	NFSERR_BADHANDLE,
483 	NFSERR_NOTSUPP,
484 	NFSERR_SERVERFAULT,
485 	0,
486 };
487 
488 static const short nfsv3err_readdir[] = {
489 	NFSERR_IO,
490 	NFSERR_IO,
491 	NFSERR_ACCES,
492 	NFSERR_NOTDIR,
493 	NFSERR_STALE,
494 	NFSERR_BADHANDLE,
495 	NFSERR_BAD_COOKIE,
496 	NFSERR_TOOSMALL,
497 	NFSERR_SERVERFAULT,
498 	0,
499 };
500 
501 static const short nfsv3err_readdirplus[] = {
502 	NFSERR_IO,
503 	NFSERR_IO,
504 	NFSERR_ACCES,
505 	NFSERR_NOTDIR,
506 	NFSERR_STALE,
507 	NFSERR_BADHANDLE,
508 	NFSERR_BAD_COOKIE,
509 	NFSERR_NOTSUPP,
510 	NFSERR_TOOSMALL,
511 	NFSERR_SERVERFAULT,
512 	0,
513 };
514 
515 static const short nfsv3err_fsstat[] = {
516 	NFSERR_IO,
517 	NFSERR_IO,
518 	NFSERR_STALE,
519 	NFSERR_BADHANDLE,
520 	NFSERR_SERVERFAULT,
521 	0,
522 };
523 
524 static const short nfsv3err_fsinfo[] = {
525 	NFSERR_STALE,
526 	NFSERR_STALE,
527 	NFSERR_BADHANDLE,
528 	NFSERR_SERVERFAULT,
529 	0,
530 };
531 
532 static const short nfsv3err_pathconf[] = {
533 	NFSERR_STALE,
534 	NFSERR_STALE,
535 	NFSERR_BADHANDLE,
536 	NFSERR_SERVERFAULT,
537 	0,
538 };
539 
540 static const short nfsv3err_commit[] = {
541 	NFSERR_IO,
542 	NFSERR_IO,
543 	NFSERR_STALE,
544 	NFSERR_BADHANDLE,
545 	NFSERR_SERVERFAULT,
546 	0,
547 };
548 
549 static const short * const nfsrv_v3errmap[] = {
550 	nfsv3err_null,
551 	nfsv3err_getattr,
552 	nfsv3err_setattr,
553 	nfsv3err_lookup,
554 	nfsv3err_access,
555 	nfsv3err_readlink,
556 	nfsv3err_read,
557 	nfsv3err_write,
558 	nfsv3err_create,
559 	nfsv3err_mkdir,
560 	nfsv3err_symlink,
561 	nfsv3err_mknod,
562 	nfsv3err_remove,
563 	nfsv3err_rmdir,
564 	nfsv3err_rename,
565 	nfsv3err_link,
566 	nfsv3err_readdir,
567 	nfsv3err_readdirplus,
568 	nfsv3err_fsstat,
569 	nfsv3err_fsinfo,
570 	nfsv3err_pathconf,
571 	nfsv3err_commit,
572 };
573 
574 extern struct nfsrtt nfsrtt;
575 extern time_t nqnfsstarttime;
576 extern int nqsrv_clockskew;
577 extern int nqsrv_writeslack;
578 extern int nqsrv_maxlease;
579 extern const int nqnfs_piggy[NFS_NPROCS];
580 extern struct nfsnodehashhead *nfsnodehashtbl;
581 extern u_long nfsnodehash;
582 
583 LIST_HEAD(nfsnodehashhead, nfsnode);
584 u_long nfsdirhashmask;
585 
586 int nfs_webnamei __P((struct nameidata *, struct vnode *, struct proc *));
587 
588 /*
589  * Create the header for an rpc request packet
590  * The hsiz is the size of the rest of the nfs request header.
591  * (just used to decide if a cluster is a good idea)
592  */
593 struct mbuf *
594 nfsm_reqh(vp, procid, hsiz, bposp)
595 	struct vnode *vp;
596 	u_long procid;
597 	int hsiz;
598 	caddr_t *bposp;
599 {
600 	struct mbuf *mb;
601 	caddr_t bpos;
602 	struct nfsmount *nmp;
603 #ifndef NFS_V2_ONLY
604 	u_int32_t *tl;
605 	struct mbuf *mb2;
606 	int nqflag;
607 #endif
608 
609 	MGET(mb, M_WAIT, MT_DATA);
610 	if (hsiz >= MINCLSIZE)
611 		MCLGET(mb, M_WAIT);
612 	mb->m_len = 0;
613 	bpos = mtod(mb, caddr_t);
614 
615 	/*
616 	 * For NQNFS, add lease request.
617 	 */
618 	if (vp) {
619 		nmp = VFSTONFS(vp->v_mount);
620 #ifndef NFS_V2_ONLY
621 		if (nmp->nm_flag & NFSMNT_NQNFS) {
622 			nqflag = NQNFS_NEEDLEASE(vp, procid);
623 			if (nqflag) {
624 				nfsm_build(tl, u_int32_t *, 2*NFSX_UNSIGNED);
625 				*tl++ = txdr_unsigned(nqflag);
626 				*tl = txdr_unsigned(nmp->nm_leaseterm);
627 			} else {
628 				nfsm_build(tl, u_int32_t *, NFSX_UNSIGNED);
629 				*tl = 0;
630 			}
631 		}
632 #endif
633 	}
634 	/* Finally, return values */
635 	*bposp = bpos;
636 	return (mb);
637 }
638 
639 /*
640  * Build the RPC header and fill in the authorization info.
641  * The authorization string argument is only used when the credentials
642  * come from outside of the kernel.
643  * Returns the head of the mbuf list.
644  */
645 struct mbuf *
646 nfsm_rpchead(cr, nmflag, procid, auth_type, auth_len, auth_str, verf_len,
647 	verf_str, mrest, mrest_len, mbp, xidp)
648 	struct ucred *cr;
649 	int nmflag;
650 	int procid;
651 	int auth_type;
652 	int auth_len;
653 	char *auth_str;
654 	int verf_len;
655 	char *verf_str;
656 	struct mbuf *mrest;
657 	int mrest_len;
658 	struct mbuf **mbp;
659 	u_int32_t *xidp;
660 {
661 	struct mbuf *mb;
662 	u_int32_t *tl;
663 	caddr_t bpos;
664 	int i;
665 	struct mbuf *mreq, *mb2;
666 	int siz, grpsiz, authsiz;
667 	struct timeval tv;
668 	static u_int32_t base;
669 
670 	authsiz = nfsm_rndup(auth_len);
671 	MGETHDR(mb, M_WAIT, MT_DATA);
672 	if ((authsiz + 10 * NFSX_UNSIGNED) >= MINCLSIZE) {
673 		MCLGET(mb, M_WAIT);
674 	} else if ((authsiz + 10 * NFSX_UNSIGNED) < MHLEN) {
675 		MH_ALIGN(mb, authsiz + 10 * NFSX_UNSIGNED);
676 	} else {
677 		MH_ALIGN(mb, 8 * NFSX_UNSIGNED);
678 	}
679 	mb->m_len = 0;
680 	mreq = mb;
681 	bpos = mtod(mb, caddr_t);
682 
683 	/*
684 	 * First the RPC header.
685 	 */
686 	nfsm_build(tl, u_int32_t *, 8 * NFSX_UNSIGNED);
687 
688 	/*
689 	 * derive initial xid from system time
690 	 * XXX time is invalid if root not yet mounted
691 	 */
692 	if (!base && (rootvp)) {
693 		microtime(&tv);
694 		base = tv.tv_sec << 12;
695 		nfs_xid = base;
696 	}
697 	/*
698 	 * Skip zero xid if it should ever happen.
699 	 */
700 	if (++nfs_xid == 0)
701 		nfs_xid++;
702 
703 	*tl++ = *xidp = txdr_unsigned(nfs_xid);
704 	*tl++ = rpc_call;
705 	*tl++ = rpc_vers;
706 	if (nmflag & NFSMNT_NQNFS) {
707 		*tl++ = txdr_unsigned(NQNFS_PROG);
708 		*tl++ = txdr_unsigned(NQNFS_VER3);
709 	} else {
710 		*tl++ = txdr_unsigned(NFS_PROG);
711 		if (nmflag & NFSMNT_NFSV3)
712 			*tl++ = txdr_unsigned(NFS_VER3);
713 		else
714 			*tl++ = txdr_unsigned(NFS_VER2);
715 	}
716 	if (nmflag & NFSMNT_NFSV3)
717 		*tl++ = txdr_unsigned(procid);
718 	else
719 		*tl++ = txdr_unsigned(nfsv2_procid[procid]);
720 
721 	/*
722 	 * And then the authorization cred.
723 	 */
724 	*tl++ = txdr_unsigned(auth_type);
725 	*tl = txdr_unsigned(authsiz);
726 	switch (auth_type) {
727 	case RPCAUTH_UNIX:
728 		nfsm_build(tl, u_int32_t *, auth_len);
729 		*tl++ = 0;		/* stamp ?? */
730 		*tl++ = 0;		/* NULL hostname */
731 		*tl++ = txdr_unsigned(cr->cr_uid);
732 		*tl++ = txdr_unsigned(cr->cr_gid);
733 		grpsiz = (auth_len >> 2) - 5;
734 		*tl++ = txdr_unsigned(grpsiz);
735 		for (i = 0; i < grpsiz; i++)
736 			*tl++ = txdr_unsigned(cr->cr_groups[i]);
737 		break;
738 	case RPCAUTH_KERB4:
739 		siz = auth_len;
740 		while (siz > 0) {
741 			if (M_TRAILINGSPACE(mb) == 0) {
742 				MGET(mb2, M_WAIT, MT_DATA);
743 				if (siz >= MINCLSIZE)
744 					MCLGET(mb2, M_WAIT);
745 				mb->m_next = mb2;
746 				mb = mb2;
747 				mb->m_len = 0;
748 				bpos = mtod(mb, caddr_t);
749 			}
750 			i = min(siz, M_TRAILINGSPACE(mb));
751 			memcpy(bpos, auth_str, i);
752 			mb->m_len += i;
753 			auth_str += i;
754 			bpos += i;
755 			siz -= i;
756 		}
757 		if ((siz = (nfsm_rndup(auth_len) - auth_len)) > 0) {
758 			for (i = 0; i < siz; i++)
759 				*bpos++ = '\0';
760 			mb->m_len += siz;
761 		}
762 		break;
763 	};
764 
765 	/*
766 	 * And the verifier...
767 	 */
768 	nfsm_build(tl, u_int32_t *, 2 * NFSX_UNSIGNED);
769 	if (verf_str) {
770 		*tl++ = txdr_unsigned(RPCAUTH_KERB4);
771 		*tl = txdr_unsigned(verf_len);
772 		siz = verf_len;
773 		while (siz > 0) {
774 			if (M_TRAILINGSPACE(mb) == 0) {
775 				MGET(mb2, M_WAIT, MT_DATA);
776 				if (siz >= MINCLSIZE)
777 					MCLGET(mb2, M_WAIT);
778 				mb->m_next = mb2;
779 				mb = mb2;
780 				mb->m_len = 0;
781 				bpos = mtod(mb, caddr_t);
782 			}
783 			i = min(siz, M_TRAILINGSPACE(mb));
784 			memcpy(bpos, verf_str, i);
785 			mb->m_len += i;
786 			verf_str += i;
787 			bpos += i;
788 			siz -= i;
789 		}
790 		if ((siz = (nfsm_rndup(verf_len) - verf_len)) > 0) {
791 			for (i = 0; i < siz; i++)
792 				*bpos++ = '\0';
793 			mb->m_len += siz;
794 		}
795 	} else {
796 		*tl++ = txdr_unsigned(RPCAUTH_NULL);
797 		*tl = 0;
798 	}
799 	mb->m_next = mrest;
800 	mreq->m_pkthdr.len = authsiz + 10 * NFSX_UNSIGNED + mrest_len;
801 	mreq->m_pkthdr.rcvif = (struct ifnet *)0;
802 	*mbp = mb;
803 	return (mreq);
804 }
805 
806 /*
807  * copies mbuf chain to the uio scatter/gather list
808  */
809 int
810 nfsm_mbuftouio(mrep, uiop, siz, dpos)
811 	struct mbuf **mrep;
812 	struct uio *uiop;
813 	int siz;
814 	caddr_t *dpos;
815 {
816 	char *mbufcp, *uiocp;
817 	int xfer, left, len;
818 	struct mbuf *mp;
819 	long uiosiz, rem;
820 	int error = 0;
821 
822 	mp = *mrep;
823 	mbufcp = *dpos;
824 	len = mtod(mp, caddr_t)+mp->m_len-mbufcp;
825 	rem = nfsm_rndup(siz)-siz;
826 	while (siz > 0) {
827 		if (uiop->uio_iovcnt <= 0 || uiop->uio_iov == NULL)
828 			return (EFBIG);
829 		left = uiop->uio_iov->iov_len;
830 		uiocp = uiop->uio_iov->iov_base;
831 		if (left > siz)
832 			left = siz;
833 		uiosiz = left;
834 		while (left > 0) {
835 			while (len == 0) {
836 				mp = mp->m_next;
837 				if (mp == NULL)
838 					return (EBADRPC);
839 				mbufcp = mtod(mp, caddr_t);
840 				len = mp->m_len;
841 			}
842 			xfer = (left > len) ? len : left;
843 #ifdef notdef
844 			/* Not Yet.. */
845 			if (uiop->uio_iov->iov_op != NULL)
846 				(*(uiop->uio_iov->iov_op))
847 				(mbufcp, uiocp, xfer);
848 			else
849 #endif
850 			if (uiop->uio_segflg == UIO_SYSSPACE)
851 				memcpy(uiocp, mbufcp, xfer);
852 			else
853 				copyout(mbufcp, uiocp, xfer);
854 			left -= xfer;
855 			len -= xfer;
856 			mbufcp += xfer;
857 			uiocp += xfer;
858 			uiop->uio_offset += xfer;
859 			uiop->uio_resid -= xfer;
860 		}
861 		if (uiop->uio_iov->iov_len <= siz) {
862 			uiop->uio_iovcnt--;
863 			uiop->uio_iov++;
864 		} else {
865 			uiop->uio_iov->iov_base =
866 			    (caddr_t)uiop->uio_iov->iov_base + uiosiz;
867 			uiop->uio_iov->iov_len -= uiosiz;
868 		}
869 		siz -= uiosiz;
870 	}
871 	*dpos = mbufcp;
872 	*mrep = mp;
873 	if (rem > 0) {
874 		if (len < rem)
875 			error = nfs_adv(mrep, dpos, rem, len);
876 		else
877 			*dpos += rem;
878 	}
879 	return (error);
880 }
881 
882 /*
883  * copies a uio scatter/gather list to an mbuf chain.
884  * NOTE: can ony handle iovcnt == 1
885  */
886 int
887 nfsm_uiotombuf(uiop, mq, siz, bpos)
888 	struct uio *uiop;
889 	struct mbuf **mq;
890 	int siz;
891 	caddr_t *bpos;
892 {
893 	char *uiocp;
894 	struct mbuf *mp, *mp2;
895 	int xfer, left, mlen;
896 	int uiosiz, clflg, rem;
897 	char *cp;
898 
899 #ifdef DIAGNOSTIC
900 	if (uiop->uio_iovcnt != 1)
901 		panic("nfsm_uiotombuf: iovcnt != 1");
902 #endif
903 
904 	if (siz > MLEN)		/* or should it >= MCLBYTES ?? */
905 		clflg = 1;
906 	else
907 		clflg = 0;
908 	rem = nfsm_rndup(siz)-siz;
909 	mp = mp2 = *mq;
910 	while (siz > 0) {
911 		left = uiop->uio_iov->iov_len;
912 		uiocp = uiop->uio_iov->iov_base;
913 		if (left > siz)
914 			left = siz;
915 		uiosiz = left;
916 		while (left > 0) {
917 			mlen = M_TRAILINGSPACE(mp);
918 			if (mlen == 0) {
919 				MGET(mp, M_WAIT, MT_DATA);
920 				if (clflg)
921 					MCLGET(mp, M_WAIT);
922 				mp->m_len = 0;
923 				mp2->m_next = mp;
924 				mp2 = mp;
925 				mlen = M_TRAILINGSPACE(mp);
926 			}
927 			xfer = (left > mlen) ? mlen : left;
928 #ifdef notdef
929 			/* Not Yet.. */
930 			if (uiop->uio_iov->iov_op != NULL)
931 				(*(uiop->uio_iov->iov_op))
932 				(uiocp, mtod(mp, caddr_t)+mp->m_len, xfer);
933 			else
934 #endif
935 			if (uiop->uio_segflg == UIO_SYSSPACE)
936 				memcpy(mtod(mp, caddr_t)+mp->m_len, uiocp, xfer);
937 			else
938 				copyin(uiocp, mtod(mp, caddr_t)+mp->m_len, xfer);
939 			mp->m_len += xfer;
940 			left -= xfer;
941 			uiocp += xfer;
942 			uiop->uio_offset += xfer;
943 			uiop->uio_resid -= xfer;
944 		}
945 		uiop->uio_iov->iov_base = (caddr_t)uiop->uio_iov->iov_base +
946 		    uiosiz;
947 		uiop->uio_iov->iov_len -= uiosiz;
948 		siz -= uiosiz;
949 	}
950 	if (rem > 0) {
951 		if (rem > M_TRAILINGSPACE(mp)) {
952 			MGET(mp, M_WAIT, MT_DATA);
953 			mp->m_len = 0;
954 			mp2->m_next = mp;
955 		}
956 		cp = mtod(mp, caddr_t)+mp->m_len;
957 		for (left = 0; left < rem; left++)
958 			*cp++ = '\0';
959 		mp->m_len += rem;
960 		*bpos = cp;
961 	} else
962 		*bpos = mtod(mp, caddr_t)+mp->m_len;
963 	*mq = mp;
964 	return (0);
965 }
966 
967 /*
968  * Get at least "siz" bytes of correctly aligned data.
969  * When called the mbuf pointers are not necessarily correct,
970  * dsosp points to what ought to be in m_data and left contains
971  * what ought to be in m_len.
972  * This is used by the macros nfsm_dissect and nfsm_dissecton for tough
973  * cases. (The macros use the vars. dpos and dpos2)
974  */
975 int
976 nfsm_disct(mdp, dposp, siz, left, cp2)
977 	struct mbuf **mdp;
978 	caddr_t *dposp;
979 	int siz;
980 	int left;
981 	caddr_t *cp2;
982 {
983 	struct mbuf *m1, *m2;
984 	struct mbuf *havebuf = NULL;
985 	caddr_t src = *dposp;
986 	caddr_t dst;
987 	int len;
988 
989 #ifdef DEBUG
990 	if (left < 0)
991 		panic("nfsm_disct: left < 0");
992 #endif
993 	m1 = *mdp;
994 	/*
995 	 * Skip through the mbuf chain looking for an mbuf with
996 	 * some data. If the first mbuf found has enough data
997 	 * and it is correctly aligned return it.
998 	 */
999 	while (left == 0) {
1000 		havebuf = m1;
1001 		*mdp = m1 = m1->m_next;
1002 		if (m1 == NULL)
1003 			return (EBADRPC);
1004 		src = mtod(m1, caddr_t);
1005 		left = m1->m_len;
1006 		/*
1007 		 * If we start a new mbuf and it is big enough
1008 		 * and correctly aligned just return it, don't
1009 		 * do any pull up.
1010 		 */
1011 		if (left >= siz && nfsm_aligned(src)) {
1012 			*cp2 = src;
1013 			*dposp = src + siz;
1014 			return (0);
1015 		}
1016 	}
1017 	if (m1->m_flags & M_EXT) {
1018 		if (havebuf) {
1019 			/* If the first mbuf with data has external data
1020 			 * and there is a previous empty mbuf use it
1021 			 * to move the data into.
1022 			 */
1023 			m2 = m1;
1024 			*mdp = m1 = havebuf;
1025 			if (m1->m_flags & M_EXT) {
1026 				MEXTREMOVE(m1);
1027 			}
1028 		} else {
1029 			/*
1030 			 * If the first mbuf has a external data
1031 			 * and there is no previous empty mbuf
1032 			 * allocate a new mbuf and move the external
1033 			 * data to the new mbuf. Also make the first
1034 			 * mbuf look empty.
1035 			 */
1036 			m2 = m_get(M_WAIT, MT_DATA);
1037 			m2->m_ext = m1->m_ext;
1038 			m2->m_data = src;
1039 			m2->m_len = left;
1040 			MCLADDREFERENCE(m1, m2);
1041 			MEXTREMOVE(m1);
1042 			m2->m_next = m1->m_next;
1043 			m1->m_next = m2;
1044 		}
1045 		m1->m_len = 0;
1046 		dst = m1->m_dat;
1047 	} else {
1048 		/*
1049 		 * If the first mbuf has no external data
1050 		 * move the data to the front of the mbuf.
1051 		 */
1052 		if ((dst = m1->m_dat) != src)
1053 			memmove(dst, src, left);
1054 		dst += left;
1055 		m1->m_len = left;
1056 		m2 = m1->m_next;
1057 	}
1058 	m1->m_flags &= ~M_PKTHDR;
1059 	*cp2 = m1->m_data = m1->m_dat;   /* data is at beginning of buffer */
1060 	*dposp = mtod(m1, caddr_t) + siz;
1061 	/*
1062 	 * Loop through mbufs pulling data up into first mbuf until
1063 	 * the first mbuf is full or there is no more data to
1064 	 * pullup.
1065 	 */
1066 	while ((len = (MLEN - m1->m_len)) != 0 && m2) {
1067 		if ((len = min(len, m2->m_len)) != 0)
1068 			memcpy(dst, m2->m_data, len);
1069 		m1->m_len += len;
1070 		dst += len;
1071 		m2->m_data += len;
1072 		m2->m_len -= len;
1073 		m2 = m2->m_next;
1074 	}
1075 	if (m1->m_len < siz)
1076 		return (EBADRPC);
1077 	return (0);
1078 }
1079 
1080 /*
1081  * Advance the position in the mbuf chain.
1082  */
1083 int
1084 nfs_adv(mdp, dposp, offs, left)
1085 	struct mbuf **mdp;
1086 	caddr_t *dposp;
1087 	int offs;
1088 	int left;
1089 {
1090 	struct mbuf *m;
1091 	int s;
1092 
1093 	m = *mdp;
1094 	s = left;
1095 	while (s < offs) {
1096 		offs -= s;
1097 		m = m->m_next;
1098 		if (m == NULL)
1099 			return (EBADRPC);
1100 		s = m->m_len;
1101 	}
1102 	*mdp = m;
1103 	*dposp = mtod(m, caddr_t)+offs;
1104 	return (0);
1105 }
1106 
1107 /*
1108  * Copy a string into mbufs for the hard cases...
1109  */
1110 int
1111 nfsm_strtmbuf(mb, bpos, cp, siz)
1112 	struct mbuf **mb;
1113 	char **bpos;
1114 	const char *cp;
1115 	long siz;
1116 {
1117 	struct mbuf *m1 = NULL, *m2;
1118 	long left, xfer, len, tlen;
1119 	u_int32_t *tl;
1120 	int putsize;
1121 
1122 	putsize = 1;
1123 	m2 = *mb;
1124 	left = M_TRAILINGSPACE(m2);
1125 	if (left > 0) {
1126 		tl = ((u_int32_t *)(*bpos));
1127 		*tl++ = txdr_unsigned(siz);
1128 		putsize = 0;
1129 		left -= NFSX_UNSIGNED;
1130 		m2->m_len += NFSX_UNSIGNED;
1131 		if (left > 0) {
1132 			memcpy((caddr_t) tl, cp, left);
1133 			siz -= left;
1134 			cp += left;
1135 			m2->m_len += left;
1136 			left = 0;
1137 		}
1138 	}
1139 	/* Loop around adding mbufs */
1140 	while (siz > 0) {
1141 		MGET(m1, M_WAIT, MT_DATA);
1142 		if (siz > MLEN)
1143 			MCLGET(m1, M_WAIT);
1144 		m1->m_len = NFSMSIZ(m1);
1145 		m2->m_next = m1;
1146 		m2 = m1;
1147 		tl = mtod(m1, u_int32_t *);
1148 		tlen = 0;
1149 		if (putsize) {
1150 			*tl++ = txdr_unsigned(siz);
1151 			m1->m_len -= NFSX_UNSIGNED;
1152 			tlen = NFSX_UNSIGNED;
1153 			putsize = 0;
1154 		}
1155 		if (siz < m1->m_len) {
1156 			len = nfsm_rndup(siz);
1157 			xfer = siz;
1158 			if (xfer < len)
1159 				*(tl+(xfer>>2)) = 0;
1160 		} else {
1161 			xfer = len = m1->m_len;
1162 		}
1163 		memcpy((caddr_t) tl, cp, xfer);
1164 		m1->m_len = len+tlen;
1165 		siz -= xfer;
1166 		cp += xfer;
1167 	}
1168 	*mb = m1;
1169 	*bpos = mtod(m1, caddr_t)+m1->m_len;
1170 	return (0);
1171 }
1172 
1173 /*
1174  * Directory caching routines. They work as follows:
1175  * - a cache is maintained per VDIR nfsnode.
1176  * - for each offset cookie that is exported to userspace, and can
1177  *   thus be thrown back at us as an offset to VOP_READDIR, store
1178  *   information in the cache.
1179  * - cached are:
1180  *   - cookie itself
1181  *   - blocknumber (essentially just a search key in the buffer cache)
1182  *   - entry number in block.
1183  *   - offset cookie of block in which this entry is stored
1184  *   - 32 bit cookie if NFSMNT_XLATECOOKIE is used.
1185  * - entries are looked up in a hash table
1186  * - also maintained is an LRU list of entries, used to determine
1187  *   which ones to delete if the cache grows too large.
1188  * - if 32 <-> 64 translation mode is requested for a filesystem,
1189  *   the cache also functions as a translation table
1190  * - in the translation case, invalidating the cache does not mean
1191  *   flushing it, but just marking entries as invalid, except for
1192  *   the <64bit cookie, 32bitcookie> pair which is still valid, to
1193  *   still be able to use the cache as a translation table.
1194  * - 32 bit cookies are uniquely created by combining the hash table
1195  *   entry value, and one generation count per hash table entry,
1196  *   incremented each time an entry is appended to the chain.
1197  * - the cache is invalidated each time a direcory is modified
1198  * - sanity checks are also done; if an entry in a block turns
1199  *   out not to have a matching cookie, the cache is invalidated
1200  *   and a new block starting from the wanted offset is fetched from
1201  *   the server.
1202  * - directory entries as read from the server are extended to contain
1203  *   the 64bit and, optionally, the 32bit cookies, for sanity checking
1204  *   the cache and exporting them to userspace through the cookie
1205  *   argument to VOP_READDIR.
1206  */
1207 
1208 u_long
1209 nfs_dirhash(off)
1210 	off_t off;
1211 {
1212 	int i;
1213 	char *cp = (char *)&off;
1214 	u_long sum = 0L;
1215 
1216 	for (i = 0 ; i < sizeof (off); i++)
1217 		sum += *cp++;
1218 
1219 	return sum;
1220 }
1221 
1222 void
1223 nfs_initdircache(vp)
1224 	struct vnode *vp;
1225 {
1226 	struct nfsnode *np = VTONFS(vp);
1227 	struct nfsmount *nmp = VFSTONFS(vp->v_mount);
1228 
1229 	np->n_dircachesize = 0;
1230 	np->n_dblkno = 1;
1231 	np->n_dircache = hashinit(NFS_DIRHASHSIZ, HASH_LIST, M_NFSDIROFF,
1232 	    M_WAITOK, &nfsdirhashmask);
1233 	TAILQ_INIT(&np->n_dirchain);
1234 	if (nmp->nm_flag & NFSMNT_XLATECOOKIE) {
1235 		MALLOC(np->n_dirgens, unsigned *,
1236 		    NFS_DIRHASHSIZ * sizeof (unsigned), M_NFSDIROFF,
1237 		    M_WAITOK);
1238 		memset((caddr_t)np->n_dirgens, 0,
1239 		    NFS_DIRHASHSIZ * sizeof (unsigned));
1240 	}
1241 }
1242 
1243 static struct nfsdircache dzero = {0, 0, {0, 0}, {0, 0}, 0, 0, 0};
1244 
1245 struct nfsdircache *
1246 nfs_searchdircache(vp, off, do32, hashent)
1247 	struct vnode *vp;
1248 	off_t off;
1249 	int do32;
1250 	int *hashent;
1251 {
1252 	struct nfsdirhashhead *ndhp;
1253 	struct nfsdircache *ndp = NULL;
1254 	struct nfsnode *np = VTONFS(vp);
1255 	unsigned ent;
1256 
1257 	/*
1258 	 * Zero is always a valid cookie.
1259 	 */
1260 	if (off == 0)
1261 		return &dzero;
1262 
1263 	/*
1264 	 * We use a 32bit cookie as search key, directly reconstruct
1265 	 * the hashentry. Else use the hashfunction.
1266 	 */
1267 	if (do32) {
1268 		ent = (u_int32_t)off >> 24;
1269 		if (ent >= NFS_DIRHASHSIZ)
1270 			return NULL;
1271 		ndhp = &np->n_dircache[ent];
1272 	} else {
1273 		ndhp = NFSDIRHASH(np, off);
1274 	}
1275 
1276 	if (hashent)
1277 		*hashent = (int)(ndhp - np->n_dircache);
1278 	if (do32) {
1279 		for (ndp = ndhp->lh_first; ndp; ndp = ndp->dc_hash.le_next) {
1280 			if (ndp->dc_cookie32 == (u_int32_t)off) {
1281 				/*
1282 				 * An invalidated entry will become the
1283 				 * start of a new block fetched from
1284 				 * the server.
1285 				 */
1286 				if (ndp->dc_blkno == -1) {
1287 					ndp->dc_blkcookie = ndp->dc_cookie;
1288 					ndp->dc_blkno = np->n_dblkno++;
1289 					ndp->dc_entry = 0;
1290 				}
1291 				break;
1292 			}
1293 		}
1294 	} else {
1295 		for (ndp = ndhp->lh_first; ndp; ndp = ndp->dc_hash.le_next)
1296 			if (ndp->dc_cookie == off)
1297 				break;
1298 	}
1299 	return ndp;
1300 }
1301 
1302 
1303 struct nfsdircache *
1304 nfs_enterdircache(vp, off, blkoff, en, blkno)
1305 	struct vnode *vp;
1306 	off_t off, blkoff;
1307 	daddr_t blkno;
1308 	int en;
1309 {
1310 	struct nfsnode *np = VTONFS(vp);
1311 	struct nfsdirhashhead *ndhp;
1312 	struct nfsdircache *ndp = NULL, *first;
1313 	struct nfsmount *nmp = VFSTONFS(vp->v_mount);
1314 	int hashent, gen, overwrite;
1315 
1316 	if (!np->n_dircache)
1317 		/*
1318 		 * XXX would like to do this in nfs_nget but vtype
1319 		 * isn't known at that time.
1320 		 */
1321 		nfs_initdircache(vp);
1322 
1323 	/*
1324 	 * XXX refuse entries for offset 0. amd(8) erroneously sets
1325 	 * cookie 0 for the '.' entry, making this necessary. This
1326 	 * isn't so bad, as 0 is a special case anyway.
1327 	 */
1328 	if (off == 0)
1329 		return &dzero;
1330 
1331 	ndp = nfs_searchdircache(vp, off, 0, &hashent);
1332 
1333 	if (ndp && ndp->dc_blkno != -1) {
1334 		/*
1335 		 * Overwriting an old entry. Check if it's the same.
1336 		 * If so, just return. If not, remove the old entry.
1337 		 */
1338 		if (ndp->dc_blkcookie == blkoff && ndp->dc_entry == en)
1339 			return ndp;
1340 		TAILQ_REMOVE(&np->n_dirchain, ndp, dc_chain);
1341 		LIST_REMOVE(ndp, dc_hash);
1342 		FREE(ndp, M_NFSDIROFF);
1343 		ndp = 0;
1344 	}
1345 
1346 	ndhp = &np->n_dircache[hashent];
1347 
1348 	if (!ndp) {
1349 		MALLOC(ndp, struct nfsdircache *, sizeof (*ndp), M_NFSDIROFF,
1350 		    M_WAITOK);
1351 		overwrite = 0;
1352 		if (nmp->nm_flag & NFSMNT_XLATECOOKIE) {
1353 			/*
1354 			 * We're allocating a new entry, so bump the
1355 			 * generation number.
1356 			 */
1357 			gen = ++np->n_dirgens[hashent];
1358 			if (gen == 0) {
1359 				np->n_dirgens[hashent]++;
1360 				gen++;
1361 			}
1362 			ndp->dc_cookie32 = (hashent << 24) | (gen & 0xffffff);
1363 		}
1364 	} else
1365 		overwrite = 1;
1366 
1367 	/*
1368 	 * If the entry number is 0, we are at the start of a new block, so
1369 	 * allocate a new blocknumber.
1370 	 */
1371 	if (en == 0)
1372 		ndp->dc_blkno = np->n_dblkno++;
1373 	else
1374 		ndp->dc_blkno = blkno;
1375 
1376 	ndp->dc_cookie = off;
1377 	ndp->dc_blkcookie = blkoff;
1378 	ndp->dc_entry = en;
1379 
1380 	if (overwrite)
1381 		return ndp;
1382 
1383 	/*
1384 	 * If the maximum directory cookie cache size has been reached
1385 	 * for this node, take one off the front. The idea is that
1386 	 * directories are typically read front-to-back once, so that
1387 	 * the oldest entries can be thrown away without much performance
1388 	 * loss.
1389 	 */
1390 	if (np->n_dircachesize == NFS_MAXDIRCACHE) {
1391 		first = np->n_dirchain.tqh_first;
1392 		TAILQ_REMOVE(&np->n_dirchain, first, dc_chain);
1393 		LIST_REMOVE(first, dc_hash);
1394 		FREE(first, M_NFSDIROFF);
1395 	} else
1396 		np->n_dircachesize++;
1397 
1398 	LIST_INSERT_HEAD(ndhp, ndp, dc_hash);
1399 	TAILQ_INSERT_TAIL(&np->n_dirchain, ndp, dc_chain);
1400 	return ndp;
1401 }
1402 
1403 void
1404 nfs_invaldircache(vp, forcefree)
1405 	struct vnode *vp;
1406 	int forcefree;
1407 {
1408 	struct nfsnode *np = VTONFS(vp);
1409 	struct nfsdircache *ndp = NULL;
1410 	struct nfsmount *nmp = VFSTONFS(vp->v_mount);
1411 
1412 #ifdef DIAGNOSTIC
1413 	if (vp->v_type != VDIR)
1414 		panic("nfs: invaldircache: not dir");
1415 #endif
1416 
1417 	if (!np->n_dircache)
1418 		return;
1419 
1420 	if (!(nmp->nm_flag & NFSMNT_XLATECOOKIE) || forcefree) {
1421 		while ((ndp = np->n_dirchain.tqh_first)) {
1422 			TAILQ_REMOVE(&np->n_dirchain, ndp, dc_chain);
1423 			LIST_REMOVE(ndp, dc_hash);
1424 			FREE(ndp, M_NFSDIROFF);
1425 		}
1426 		np->n_dircachesize = 0;
1427 		if (forcefree && np->n_dirgens) {
1428 			FREE(np->n_dirgens, M_NFSDIROFF);
1429 		}
1430 	} else {
1431 		for (ndp = np->n_dirchain.tqh_first; ndp;
1432 		    ndp = ndp->dc_chain.tqe_next)
1433 			ndp->dc_blkno = -1;
1434 	}
1435 
1436 	np->n_dblkno = 1;
1437 }
1438 
1439 /*
1440  * Called once before VFS init to initialize shared and
1441  * server-specific data structures.
1442  */
1443 void
1444 nfs_init()
1445 {
1446 	nfsrtt.pos = 0;
1447 	rpc_vers = txdr_unsigned(RPC_VER2);
1448 	rpc_call = txdr_unsigned(RPC_CALL);
1449 	rpc_reply = txdr_unsigned(RPC_REPLY);
1450 	rpc_msgdenied = txdr_unsigned(RPC_MSGDENIED);
1451 	rpc_msgaccepted = txdr_unsigned(RPC_MSGACCEPTED);
1452 	rpc_mismatch = txdr_unsigned(RPC_MISMATCH);
1453 	rpc_autherr = txdr_unsigned(RPC_AUTHERR);
1454 	rpc_auth_unix = txdr_unsigned(RPCAUTH_UNIX);
1455 	rpc_auth_kerb = txdr_unsigned(RPCAUTH_KERB4);
1456 	nfs_prog = txdr_unsigned(NFS_PROG);
1457 	nqnfs_prog = txdr_unsigned(NQNFS_PROG);
1458 	nfs_true = txdr_unsigned(TRUE);
1459 	nfs_false = txdr_unsigned(FALSE);
1460 	nfs_xdrneg1 = txdr_unsigned(-1);
1461 	nfs_ticks = (hz * NFS_TICKINTVL + 500) / 1000;
1462 	if (nfs_ticks < 1)
1463 		nfs_ticks = 1;
1464 #ifdef NFSSERVER
1465 	nfsrv_init(0);			/* Init server data structures */
1466 	nfsrv_initcache();		/* Init the server request cache */
1467 #endif /* NFSSERVER */
1468 
1469 #if defined(NFSSERVER) || !defined(NFS_V2_ONLY)
1470 	/*
1471 	 * Initialize the nqnfs data structures.
1472 	 */
1473 	if (nqnfsstarttime == 0) {
1474 		nqnfsstarttime = boottime.tv_sec + nqsrv_maxlease
1475 			+ nqsrv_clockskew + nqsrv_writeslack;
1476 		NQLOADNOVRAM(nqnfsstarttime);
1477 		CIRCLEQ_INIT(&nqtimerhead);
1478 		nqfhhashtbl = hashinit(NQLCHSZ, HASH_LIST, M_NQLEASE,
1479 		    M_WAITOK, &nqfhhash);
1480 	}
1481 #endif
1482 
1483 	/*
1484 	 * Initialize reply list and start timer
1485 	 */
1486 	TAILQ_INIT(&nfs_reqq);
1487 	nfs_timer(NULL);
1488 }
1489 
1490 #ifdef NFS
1491 /*
1492  * Called once at VFS init to initialize client-specific data structures.
1493  */
1494 void
1495 nfs_vfs_init()
1496 {
1497 	nfs_nhinit();			/* Init the nfsnode table */
1498 	nfs_commitsize = uvmexp.npages << (PAGE_SHIFT - 4);
1499 }
1500 
1501 void
1502 nfs_vfs_reinit()
1503 {
1504 	nfs_nhreinit();
1505 }
1506 
1507 void
1508 nfs_vfs_done()
1509 {
1510 	nfs_nhdone();
1511 }
1512 
1513 /*
1514  * Attribute cache routines.
1515  * nfs_loadattrcache() - loads or updates the cache contents from attributes
1516  *	that are on the mbuf list
1517  * nfs_getattrcache() - returns valid attributes if found in cache, returns
1518  *	error otherwise
1519  */
1520 
1521 /*
1522  * Load the attribute cache (that lives in the nfsnode entry) with
1523  * the values on the mbuf list and
1524  * Iff vap not NULL
1525  *    copy the attributes to *vaper
1526  */
1527 int
1528 nfsm_loadattrcache(vpp, mdp, dposp, vaper)
1529 	struct vnode **vpp;
1530 	struct mbuf **mdp;
1531 	caddr_t *dposp;
1532 	struct vattr *vaper;
1533 {
1534 	int32_t t1;
1535 	caddr_t cp2;
1536 	int error = 0;
1537 	struct mbuf *md;
1538 	int v3 = NFS_ISV3(*vpp);
1539 
1540 	md = *mdp;
1541 	t1 = (mtod(md, caddr_t) + md->m_len) - *dposp;
1542 	error = nfsm_disct(mdp, dposp, NFSX_FATTR(v3), t1, &cp2);
1543 	if (error)
1544 		return (error);
1545 	return nfs_loadattrcache(vpp, (struct nfs_fattr *)cp2, vaper);
1546 }
1547 
1548 int
1549 nfs_loadattrcache(vpp, fp, vaper)
1550 	struct vnode **vpp;
1551 	struct nfs_fattr *fp;
1552 	struct vattr *vaper;
1553 {
1554 	struct vnode *vp = *vpp;
1555 	struct vattr *vap;
1556 	int v3 = NFS_ISV3(vp);
1557 	enum vtype vtyp;
1558 	u_short vmode;
1559 	struct timespec mtime;
1560 	struct vnode *nvp;
1561 	int32_t rdev;
1562 	struct nfsnode *np;
1563 	extern int (**spec_nfsv2nodeop_p) __P((void *));
1564 
1565 	if (v3) {
1566 		vtyp = nfsv3tov_type(fp->fa_type);
1567 		vmode = fxdr_unsigned(u_short, fp->fa_mode);
1568 		rdev = makedev(fxdr_unsigned(u_int32_t, fp->fa3_rdev.specdata1),
1569 			fxdr_unsigned(u_int32_t, fp->fa3_rdev.specdata2));
1570 		fxdr_nfsv3time(&fp->fa3_mtime, &mtime);
1571 	} else {
1572 		vtyp = nfsv2tov_type(fp->fa_type);
1573 		vmode = fxdr_unsigned(u_short, fp->fa_mode);
1574 		if (vtyp == VNON || vtyp == VREG)
1575 			vtyp = IFTOVT(vmode);
1576 		rdev = fxdr_unsigned(int32_t, fp->fa2_rdev);
1577 		fxdr_nfsv2time(&fp->fa2_mtime, &mtime);
1578 
1579 		/*
1580 		 * Really ugly NFSv2 kludge.
1581 		 */
1582 		if (vtyp == VCHR && rdev == 0xffffffff)
1583 			vtyp = VFIFO;
1584 	}
1585 
1586 	/*
1587 	 * If v_type == VNON it is a new node, so fill in the v_type,
1588 	 * n_mtime fields. Check to see if it represents a special
1589 	 * device, and if so, check for a possible alias. Once the
1590 	 * correct vnode has been obtained, fill in the rest of the
1591 	 * information.
1592 	 */
1593 	np = VTONFS(vp);
1594 	if (vp->v_type == VNON) {
1595 		vp->v_type = vtyp;
1596 		if (vp->v_type == VFIFO) {
1597 			extern int (**fifo_nfsv2nodeop_p) __P((void *));
1598 			vp->v_op = fifo_nfsv2nodeop_p;
1599 		}
1600 		if (vp->v_type == VCHR || vp->v_type == VBLK) {
1601 			vp->v_op = spec_nfsv2nodeop_p;
1602 			nvp = checkalias(vp, (dev_t)rdev, vp->v_mount);
1603 			if (nvp) {
1604 				/*
1605 				 * Discard unneeded vnode, but save its nfsnode.
1606 				 * Since the nfsnode does not have a lock, its
1607 				 * vnode lock has to be carried over.
1608 				 */
1609 				/*
1610 				 * XXX is the old node sure to be locked here?
1611 				 */
1612 				KASSERT(lockstatus(&vp->v_lock) ==
1613 				    LK_EXCLUSIVE);
1614 				nvp->v_data = vp->v_data;
1615 				vp->v_data = NULL;
1616 				VOP_UNLOCK(vp, 0);
1617 				vp->v_op = spec_vnodeop_p;
1618 				vrele(vp);
1619 				vgone(vp);
1620 				lockmgr(&nvp->v_lock, LK_EXCLUSIVE,
1621 				    &nvp->v_interlock);
1622 				/*
1623 				 * Reinitialize aliased node.
1624 				 */
1625 				np->n_vnode = nvp;
1626 				*vpp = vp = nvp;
1627 			}
1628 		}
1629 		np->n_mtime = mtime.tv_sec;
1630 	}
1631 	vap = np->n_vattr;
1632 	vap->va_type = vtyp;
1633 	vap->va_mode = vmode & ALLPERMS;
1634 	vap->va_rdev = (dev_t)rdev;
1635 	vap->va_mtime = mtime;
1636 	vap->va_fsid = vp->v_mount->mnt_stat.f_fsid.val[0];
1637 	switch (vtyp) {
1638 	case VDIR:
1639 		vap->va_blocksize = NFS_DIRFRAGSIZ;
1640 		break;
1641 	case VBLK:
1642 		vap->va_blocksize = BLKDEV_IOSIZE;
1643 		break;
1644 	case VCHR:
1645 		vap->va_blocksize = MAXBSIZE;
1646 		break;
1647 	default:
1648 		vap->va_blocksize = v3 ? vp->v_mount->mnt_stat.f_iosize :
1649 		    fxdr_unsigned(int32_t, fp->fa2_blocksize);
1650 		break;
1651 	}
1652 	if (v3) {
1653 		vap->va_nlink = fxdr_unsigned(u_short, fp->fa_nlink);
1654 		vap->va_uid = fxdr_unsigned(uid_t, fp->fa_uid);
1655 		vap->va_gid = fxdr_unsigned(gid_t, fp->fa_gid);
1656 		vap->va_size = fxdr_hyper(&fp->fa3_size);
1657 		vap->va_bytes = fxdr_hyper(&fp->fa3_used);
1658 		vap->va_fileid = fxdr_unsigned(int32_t,
1659 		    fp->fa3_fileid.nfsuquad[1]);
1660 		fxdr_nfsv3time(&fp->fa3_atime, &vap->va_atime);
1661 		fxdr_nfsv3time(&fp->fa3_ctime, &vap->va_ctime);
1662 		vap->va_flags = 0;
1663 		vap->va_filerev = 0;
1664 	} else {
1665 		vap->va_nlink = fxdr_unsigned(u_short, fp->fa_nlink);
1666 		vap->va_uid = fxdr_unsigned(uid_t, fp->fa_uid);
1667 		vap->va_gid = fxdr_unsigned(gid_t, fp->fa_gid);
1668 		vap->va_size = fxdr_unsigned(u_int32_t, fp->fa2_size);
1669 		vap->va_bytes = fxdr_unsigned(int32_t, fp->fa2_blocks)
1670 		    * NFS_FABLKSIZE;
1671 		vap->va_fileid = fxdr_unsigned(int32_t, fp->fa2_fileid);
1672 		fxdr_nfsv2time(&fp->fa2_atime, &vap->va_atime);
1673 		vap->va_flags = 0;
1674 		vap->va_ctime.tv_sec = fxdr_unsigned(u_int32_t,
1675 		    fp->fa2_ctime.nfsv2_sec);
1676 		vap->va_ctime.tv_nsec = 0;
1677 		vap->va_gen = fxdr_unsigned(u_int32_t,fp->fa2_ctime.nfsv2_usec);
1678 		vap->va_filerev = 0;
1679 	}
1680 	if (vap->va_size != np->n_size) {
1681 		if ((np->n_flag & NMODIFIED) && vap->va_size < np->n_size) {
1682 			vap->va_size = np->n_size;
1683 		} else {
1684 			np->n_size = vap->va_size;
1685 			if (vap->va_type == VREG) {
1686 				uvm_vnp_setsize(vp, np->n_size);
1687 			}
1688 		}
1689 	}
1690 	np->n_attrstamp = time.tv_sec;
1691 	if (vaper != NULL) {
1692 		memcpy((caddr_t)vaper, (caddr_t)vap, sizeof(*vap));
1693 		if (np->n_flag & NCHG) {
1694 			if (np->n_flag & NACC)
1695 				vaper->va_atime = np->n_atim;
1696 			if (np->n_flag & NUPD)
1697 				vaper->va_mtime = np->n_mtim;
1698 		}
1699 	}
1700 	return (0);
1701 }
1702 
1703 /*
1704  * Check the time stamp
1705  * If the cache is valid, copy contents to *vap and return 0
1706  * otherwise return an error
1707  */
1708 int
1709 nfs_getattrcache(vp, vaper)
1710 	struct vnode *vp;
1711 	struct vattr *vaper;
1712 {
1713 	struct nfsnode *np = VTONFS(vp);
1714 	struct vattr *vap;
1715 
1716 	if ((time.tv_sec - np->n_attrstamp) >= NFS_ATTRTIMEO(np)) {
1717 		nfsstats.attrcache_misses++;
1718 		return (ENOENT);
1719 	}
1720 	nfsstats.attrcache_hits++;
1721 	vap = np->n_vattr;
1722 	if (vap->va_size != np->n_size) {
1723 		if (vap->va_type == VREG) {
1724 			if (np->n_flag & NMODIFIED) {
1725 				if (vap->va_size < np->n_size)
1726 					vap->va_size = np->n_size;
1727 				else
1728 					np->n_size = vap->va_size;
1729 			} else
1730 				np->n_size = vap->va_size;
1731 			uvm_vnp_setsize(vp, np->n_size);
1732 		} else
1733 			np->n_size = vap->va_size;
1734 	}
1735 	memcpy((caddr_t)vaper, (caddr_t)vap, sizeof(struct vattr));
1736 	if (np->n_flag & NCHG) {
1737 		if (np->n_flag & NACC)
1738 			vaper->va_atime = np->n_atim;
1739 		if (np->n_flag & NUPD)
1740 			vaper->va_mtime = np->n_mtim;
1741 	}
1742 	return (0);
1743 }
1744 
1745 /*
1746  * Heuristic to see if the server XDR encodes directory cookies or not.
1747  * it is not supposed to, but a lot of servers may do this. Also, since
1748  * most/all servers will implement V2 as well, it is expected that they
1749  * may return just 32 bits worth of cookie information, so we need to
1750  * find out in which 32 bits this information is available. We do this
1751  * to avoid trouble with emulated binaries that can't handle 64 bit
1752  * directory offsets.
1753  */
1754 
1755 void
1756 nfs_cookieheuristic(vp, flagp, p, cred)
1757 	struct vnode *vp;
1758 	int *flagp;
1759 	struct proc *p;
1760 	struct ucred *cred;
1761 {
1762 	struct uio auio;
1763 	struct iovec aiov;
1764 	caddr_t buf, cp;
1765 	struct dirent *dp;
1766 	off_t *cookies = NULL, *cop;
1767 	int error, eof, nc, len;
1768 
1769 	MALLOC(buf, caddr_t, NFS_DIRFRAGSIZ, M_TEMP, M_WAITOK);
1770 
1771 	aiov.iov_base = buf;
1772 	aiov.iov_len = NFS_DIRFRAGSIZ;
1773 	auio.uio_iov = &aiov;
1774 	auio.uio_iovcnt = 1;
1775 	auio.uio_rw = UIO_READ;
1776 	auio.uio_segflg = UIO_SYSSPACE;
1777 	auio.uio_procp = p;
1778 	auio.uio_resid = NFS_DIRFRAGSIZ;
1779 	auio.uio_offset = 0;
1780 
1781 	error = VOP_READDIR(vp, &auio, cred, &eof, &cookies, &nc);
1782 
1783 	len = NFS_DIRFRAGSIZ - auio.uio_resid;
1784 	if (error || len == 0) {
1785 		FREE(buf, M_TEMP);
1786 		if (cookies)
1787 			free(cookies, M_TEMP);
1788 		return;
1789 	}
1790 
1791 	/*
1792 	 * Find the first valid entry and look at its offset cookie.
1793 	 */
1794 
1795 	cp = buf;
1796 	for (cop = cookies; len > 0; len -= dp->d_reclen) {
1797 		dp = (struct dirent *)cp;
1798 		if (dp->d_fileno != 0 && len >= dp->d_reclen) {
1799 			if ((*cop >> 32) != 0 && (*cop & 0xffffffffLL) == 0) {
1800 				*flagp |= NFSMNT_SWAPCOOKIE;
1801 				nfs_invaldircache(vp, 0);
1802 				nfs_vinvalbuf(vp, 0, cred, p, 1);
1803 			}
1804 			break;
1805 		}
1806 		cop++;
1807 		cp += dp->d_reclen;
1808 	}
1809 
1810 	FREE(buf, M_TEMP);
1811 	free(cookies, M_TEMP);
1812 }
1813 #endif /* NFS */
1814 
1815 /*
1816  * Set up nameidata for a lookup() call and do it.
1817  *
1818  * If pubflag is set, this call is done for a lookup operation on the
1819  * public filehandle. In that case we allow crossing mountpoints and
1820  * absolute pathnames. However, the caller is expected to check that
1821  * the lookup result is within the public fs, and deny access if
1822  * it is not.
1823  */
1824 int
1825 nfs_namei(ndp, fhp, len, slp, nam, mdp, dposp, retdirp, p, kerbflag, pubflag)
1826 	struct nameidata *ndp;
1827 	fhandle_t *fhp;
1828 	int len;
1829 	struct nfssvc_sock *slp;
1830 	struct mbuf *nam;
1831 	struct mbuf **mdp;
1832 	caddr_t *dposp;
1833 	struct vnode **retdirp;
1834 	struct proc *p;
1835 	int kerbflag, pubflag;
1836 {
1837 	int i, rem;
1838 	struct mbuf *md;
1839 	char *fromcp, *tocp, *cp;
1840 	struct iovec aiov;
1841 	struct uio auio;
1842 	struct vnode *dp;
1843 	int error, rdonly, linklen;
1844 	struct componentname *cnp = &ndp->ni_cnd;
1845 
1846 	*retdirp = (struct vnode *)0;
1847 
1848 	if ((len + 1) > MAXPATHLEN)
1849 		return (ENAMETOOLONG);
1850 	cnp->cn_pnbuf = PNBUF_GET();
1851 
1852 	/*
1853 	 * Copy the name from the mbuf list to ndp->ni_pnbuf
1854 	 * and set the various ndp fields appropriately.
1855 	 */
1856 	fromcp = *dposp;
1857 	tocp = cnp->cn_pnbuf;
1858 	md = *mdp;
1859 	rem = mtod(md, caddr_t) + md->m_len - fromcp;
1860 	for (i = 0; i < len; i++) {
1861 		while (rem == 0) {
1862 			md = md->m_next;
1863 			if (md == NULL) {
1864 				error = EBADRPC;
1865 				goto out;
1866 			}
1867 			fromcp = mtod(md, caddr_t);
1868 			rem = md->m_len;
1869 		}
1870 		if (*fromcp == '\0' || (!pubflag && *fromcp == '/')) {
1871 			error = EACCES;
1872 			goto out;
1873 		}
1874 		*tocp++ = *fromcp++;
1875 		rem--;
1876 	}
1877 	*tocp = '\0';
1878 	*mdp = md;
1879 	*dposp = fromcp;
1880 	len = nfsm_rndup(len)-len;
1881 	if (len > 0) {
1882 		if (rem >= len)
1883 			*dposp += len;
1884 		else if ((error = nfs_adv(mdp, dposp, len, rem)) != 0)
1885 			goto out;
1886 	}
1887 
1888 	/*
1889 	 * Extract and set starting directory.
1890 	 */
1891 	error = nfsrv_fhtovp(fhp, FALSE, &dp, ndp->ni_cnd.cn_cred, slp,
1892 	    nam, &rdonly, kerbflag, pubflag);
1893 	if (error)
1894 		goto out;
1895 	if (dp->v_type != VDIR) {
1896 		vrele(dp);
1897 		error = ENOTDIR;
1898 		goto out;
1899 	}
1900 
1901 	if (rdonly)
1902 		cnp->cn_flags |= RDONLY;
1903 
1904 	*retdirp = dp;
1905 
1906 	if (pubflag) {
1907 		/*
1908 		 * Oh joy. For WebNFS, handle those pesky '%' escapes,
1909 		 * and the 'native path' indicator.
1910 		 */
1911 		cp = PNBUF_GET();
1912 		fromcp = cnp->cn_pnbuf;
1913 		tocp = cp;
1914 		if ((unsigned char)*fromcp >= WEBNFS_SPECCHAR_START) {
1915 			switch ((unsigned char)*fromcp) {
1916 			case WEBNFS_NATIVE_CHAR:
1917 				/*
1918 				 * 'Native' path for us is the same
1919 				 * as a path according to the NFS spec,
1920 				 * just skip the escape char.
1921 				 */
1922 				fromcp++;
1923 				break;
1924 			/*
1925 			 * More may be added in the future, range 0x80-0xff
1926 			 */
1927 			default:
1928 				error = EIO;
1929 				FREE(cp, M_NAMEI);
1930 				goto out;
1931 			}
1932 		}
1933 		/*
1934 		 * Translate the '%' escapes, URL-style.
1935 		 */
1936 		while (*fromcp != '\0') {
1937 			if (*fromcp == WEBNFS_ESC_CHAR) {
1938 				if (fromcp[1] != '\0' && fromcp[2] != '\0') {
1939 					fromcp++;
1940 					*tocp++ = HEXSTRTOI(fromcp);
1941 					fromcp += 2;
1942 					continue;
1943 				} else {
1944 					error = ENOENT;
1945 					FREE(cp, M_NAMEI);
1946 					goto out;
1947 				}
1948 			} else
1949 				*tocp++ = *fromcp++;
1950 		}
1951 		*tocp = '\0';
1952 		PNBUF_PUT(cnp->cn_pnbuf);
1953 		cnp->cn_pnbuf = cp;
1954 	}
1955 
1956 	ndp->ni_pathlen = (tocp - cnp->cn_pnbuf) + 1;
1957 	ndp->ni_segflg = UIO_SYSSPACE;
1958 	ndp->ni_rootdir = rootvnode;
1959 
1960 	if (pubflag) {
1961 		ndp->ni_loopcnt = 0;
1962 		if (cnp->cn_pnbuf[0] == '/')
1963 			dp = rootvnode;
1964 	} else {
1965 		cnp->cn_flags |= NOCROSSMOUNT;
1966 	}
1967 
1968 	cnp->cn_proc = p;
1969 	VREF(dp);
1970 
1971     for (;;) {
1972 	cnp->cn_nameptr = cnp->cn_pnbuf;
1973 	ndp->ni_startdir = dp;
1974 	/*
1975 	 * And call lookup() to do the real work
1976 	 */
1977 	error = lookup(ndp);
1978 	if (error) {
1979 		PNBUF_PUT(cnp->cn_pnbuf);
1980 		return (error);
1981 	}
1982 	/*
1983 	 * Check for encountering a symbolic link
1984 	 */
1985 	if ((cnp->cn_flags & ISSYMLINK) == 0) {
1986 		if (cnp->cn_flags & (SAVENAME | SAVESTART))
1987 			cnp->cn_flags |= HASBUF;
1988 		else
1989 			PNBUF_PUT(cnp->cn_pnbuf);
1990 		return (0);
1991 	} else {
1992 		if ((cnp->cn_flags & LOCKPARENT) && (cnp->cn_flags & ISLASTCN))
1993 			VOP_UNLOCK(ndp->ni_dvp, 0);
1994 		if (!pubflag) {
1995 			error = EINVAL;
1996 			break;
1997 		}
1998 
1999 		if (ndp->ni_loopcnt++ >= MAXSYMLINKS) {
2000 			error = ELOOP;
2001 			break;
2002 		}
2003 		if (ndp->ni_vp->v_mount->mnt_flag & MNT_SYMPERM) {
2004 			error = VOP_ACCESS(ndp->ni_vp, VEXEC, cnp->cn_cred,
2005 			    cnp->cn_proc);
2006 			if (error != 0)
2007 				break;
2008 		}
2009 		if (ndp->ni_pathlen > 1)
2010 			cp = PNBUF_GET();
2011 		else
2012 			cp = cnp->cn_pnbuf;
2013 		aiov.iov_base = cp;
2014 		aiov.iov_len = MAXPATHLEN;
2015 		auio.uio_iov = &aiov;
2016 		auio.uio_iovcnt = 1;
2017 		auio.uio_offset = 0;
2018 		auio.uio_rw = UIO_READ;
2019 		auio.uio_segflg = UIO_SYSSPACE;
2020 		auio.uio_procp = (struct proc *)0;
2021 		auio.uio_resid = MAXPATHLEN;
2022 		error = VOP_READLINK(ndp->ni_vp, &auio, cnp->cn_cred);
2023 		if (error) {
2024 		badlink:
2025 			if (ndp->ni_pathlen > 1)
2026 				PNBUF_PUT(cp);
2027 			break;
2028 		}
2029 		linklen = MAXPATHLEN - auio.uio_resid;
2030 		if (linklen == 0) {
2031 			error = ENOENT;
2032 			goto badlink;
2033 		}
2034 		if (linklen + ndp->ni_pathlen >= MAXPATHLEN) {
2035 			error = ENAMETOOLONG;
2036 			goto badlink;
2037 		}
2038 		if (ndp->ni_pathlen > 1) {
2039 			memcpy(cp + linklen, ndp->ni_next, ndp->ni_pathlen);
2040 			PNBUF_PUT(cnp->cn_pnbuf);
2041 			cnp->cn_pnbuf = cp;
2042 		} else
2043 			cnp->cn_pnbuf[linklen] = '\0';
2044 		ndp->ni_pathlen += linklen;
2045 		vput(ndp->ni_vp);
2046 		dp = ndp->ni_dvp;
2047 		/*
2048 		 * Check if root directory should replace current directory.
2049 		 */
2050 		if (cnp->cn_pnbuf[0] == '/') {
2051 			vrele(dp);
2052 			dp = ndp->ni_rootdir;
2053 			VREF(dp);
2054 		}
2055 	}
2056    }
2057 	vrele(ndp->ni_dvp);
2058 	vput(ndp->ni_vp);
2059 	ndp->ni_vp = NULL;
2060 out:
2061 	PNBUF_PUT(cnp->cn_pnbuf);
2062 	return (error);
2063 }
2064 
2065 /*
2066  * A fiddled version of m_adj() that ensures null fill to a long
2067  * boundary and only trims off the back end
2068  */
2069 void
2070 nfsm_adj(mp, len, nul)
2071 	struct mbuf *mp;
2072 	int len;
2073 	int nul;
2074 {
2075 	struct mbuf *m;
2076 	int count, i;
2077 	char *cp;
2078 
2079 	/*
2080 	 * Trim from tail.  Scan the mbuf chain,
2081 	 * calculating its length and finding the last mbuf.
2082 	 * If the adjustment only affects this mbuf, then just
2083 	 * adjust and return.  Otherwise, rescan and truncate
2084 	 * after the remaining size.
2085 	 */
2086 	count = 0;
2087 	m = mp;
2088 	for (;;) {
2089 		count += m->m_len;
2090 		if (m->m_next == (struct mbuf *)0)
2091 			break;
2092 		m = m->m_next;
2093 	}
2094 	if (m->m_len > len) {
2095 		m->m_len -= len;
2096 		if (nul > 0) {
2097 			cp = mtod(m, caddr_t)+m->m_len-nul;
2098 			for (i = 0; i < nul; i++)
2099 				*cp++ = '\0';
2100 		}
2101 		return;
2102 	}
2103 	count -= len;
2104 	if (count < 0)
2105 		count = 0;
2106 	/*
2107 	 * Correct length for chain is "count".
2108 	 * Find the mbuf with last data, adjust its length,
2109 	 * and toss data from remaining mbufs on chain.
2110 	 */
2111 	for (m = mp; m; m = m->m_next) {
2112 		if (m->m_len >= count) {
2113 			m->m_len = count;
2114 			if (nul > 0) {
2115 				cp = mtod(m, caddr_t)+m->m_len-nul;
2116 				for (i = 0; i < nul; i++)
2117 					*cp++ = '\0';
2118 			}
2119 			break;
2120 		}
2121 		count -= m->m_len;
2122 	}
2123 	for (m = m->m_next;m;m = m->m_next)
2124 		m->m_len = 0;
2125 }
2126 
2127 /*
2128  * Make these functions instead of macros, so that the kernel text size
2129  * doesn't get too big...
2130  */
2131 void
2132 nfsm_srvwcc(nfsd, before_ret, before_vap, after_ret, after_vap, mbp, bposp)
2133 	struct nfsrv_descript *nfsd;
2134 	int before_ret;
2135 	struct vattr *before_vap;
2136 	int after_ret;
2137 	struct vattr *after_vap;
2138 	struct mbuf **mbp;
2139 	char **bposp;
2140 {
2141 	struct mbuf *mb = *mbp, *mb2;
2142 	char *bpos = *bposp;
2143 	u_int32_t *tl;
2144 
2145 	if (before_ret) {
2146 		nfsm_build(tl, u_int32_t *, NFSX_UNSIGNED);
2147 		*tl = nfs_false;
2148 	} else {
2149 		nfsm_build(tl, u_int32_t *, 7 * NFSX_UNSIGNED);
2150 		*tl++ = nfs_true;
2151 		txdr_hyper(before_vap->va_size, tl);
2152 		tl += 2;
2153 		txdr_nfsv3time(&(before_vap->va_mtime), tl);
2154 		tl += 2;
2155 		txdr_nfsv3time(&(before_vap->va_ctime), tl);
2156 	}
2157 	*bposp = bpos;
2158 	*mbp = mb;
2159 	nfsm_srvpostopattr(nfsd, after_ret, after_vap, mbp, bposp);
2160 }
2161 
2162 void
2163 nfsm_srvpostopattr(nfsd, after_ret, after_vap, mbp, bposp)
2164 	struct nfsrv_descript *nfsd;
2165 	int after_ret;
2166 	struct vattr *after_vap;
2167 	struct mbuf **mbp;
2168 	char **bposp;
2169 {
2170 	struct mbuf *mb = *mbp, *mb2;
2171 	char *bpos = *bposp;
2172 	u_int32_t *tl;
2173 	struct nfs_fattr *fp;
2174 
2175 	if (after_ret) {
2176 		nfsm_build(tl, u_int32_t *, NFSX_UNSIGNED);
2177 		*tl = nfs_false;
2178 	} else {
2179 		nfsm_build(tl, u_int32_t *, NFSX_UNSIGNED + NFSX_V3FATTR);
2180 		*tl++ = nfs_true;
2181 		fp = (struct nfs_fattr *)tl;
2182 		nfsm_srvfattr(nfsd, after_vap, fp);
2183 	}
2184 	*mbp = mb;
2185 	*bposp = bpos;
2186 }
2187 
2188 void
2189 nfsm_srvfattr(nfsd, vap, fp)
2190 	struct nfsrv_descript *nfsd;
2191 	struct vattr *vap;
2192 	struct nfs_fattr *fp;
2193 {
2194 
2195 	fp->fa_nlink = txdr_unsigned(vap->va_nlink);
2196 	fp->fa_uid = txdr_unsigned(vap->va_uid);
2197 	fp->fa_gid = txdr_unsigned(vap->va_gid);
2198 	if (nfsd->nd_flag & ND_NFSV3) {
2199 		fp->fa_type = vtonfsv3_type(vap->va_type);
2200 		fp->fa_mode = vtonfsv3_mode(vap->va_mode);
2201 		txdr_hyper(vap->va_size, &fp->fa3_size);
2202 		txdr_hyper(vap->va_bytes, &fp->fa3_used);
2203 		fp->fa3_rdev.specdata1 = txdr_unsigned(major(vap->va_rdev));
2204 		fp->fa3_rdev.specdata2 = txdr_unsigned(minor(vap->va_rdev));
2205 		fp->fa3_fsid.nfsuquad[0] = 0;
2206 		fp->fa3_fsid.nfsuquad[1] = txdr_unsigned(vap->va_fsid);
2207 		fp->fa3_fileid.nfsuquad[0] = 0;
2208 		fp->fa3_fileid.nfsuquad[1] = txdr_unsigned(vap->va_fileid);
2209 		txdr_nfsv3time(&vap->va_atime, &fp->fa3_atime);
2210 		txdr_nfsv3time(&vap->va_mtime, &fp->fa3_mtime);
2211 		txdr_nfsv3time(&vap->va_ctime, &fp->fa3_ctime);
2212 	} else {
2213 		fp->fa_type = vtonfsv2_type(vap->va_type);
2214 		fp->fa_mode = vtonfsv2_mode(vap->va_type, vap->va_mode);
2215 		fp->fa2_size = txdr_unsigned(vap->va_size);
2216 		fp->fa2_blocksize = txdr_unsigned(vap->va_blocksize);
2217 		if (vap->va_type == VFIFO)
2218 			fp->fa2_rdev = 0xffffffff;
2219 		else
2220 			fp->fa2_rdev = txdr_unsigned(vap->va_rdev);
2221 		fp->fa2_blocks = txdr_unsigned(vap->va_bytes / NFS_FABLKSIZE);
2222 		fp->fa2_fsid = txdr_unsigned(vap->va_fsid);
2223 		fp->fa2_fileid = txdr_unsigned(vap->va_fileid);
2224 		txdr_nfsv2time(&vap->va_atime, &fp->fa2_atime);
2225 		txdr_nfsv2time(&vap->va_mtime, &fp->fa2_mtime);
2226 		txdr_nfsv2time(&vap->va_ctime, &fp->fa2_ctime);
2227 	}
2228 }
2229 
2230 /*
2231  * nfsrv_fhtovp() - convert a fh to a vnode ptr (optionally locked)
2232  * 	- look up fsid in mount list (if not found ret error)
2233  *	- get vp and export rights by calling VFS_FHTOVP()
2234  *	- if cred->cr_uid == 0 or MNT_EXPORTANON set it to credanon
2235  *	- if not lockflag unlock it with VOP_UNLOCK()
2236  */
2237 int
2238 nfsrv_fhtovp(fhp, lockflag, vpp, cred, slp, nam, rdonlyp, kerbflag, pubflag)
2239 	fhandle_t *fhp;
2240 	int lockflag;
2241 	struct vnode **vpp;
2242 	struct ucred *cred;
2243 	struct nfssvc_sock *slp;
2244 	struct mbuf *nam;
2245 	int *rdonlyp;
2246 	int kerbflag;
2247 {
2248 	struct mount *mp;
2249 	int i;
2250 	struct ucred *credanon;
2251 	int error, exflags;
2252 	struct sockaddr_in *saddr;
2253 
2254 	*vpp = (struct vnode *)0;
2255 
2256 	if (nfs_ispublicfh(fhp)) {
2257 		if (!pubflag || !nfs_pub.np_valid)
2258 			return (ESTALE);
2259 		fhp = &nfs_pub.np_handle;
2260 	}
2261 
2262 	mp = vfs_getvfs(&fhp->fh_fsid);
2263 	if (!mp)
2264 		return (ESTALE);
2265 	error = VFS_CHECKEXP(mp, nam, &exflags, &credanon);
2266 	if (error)
2267 		return (error);
2268 	error = VFS_FHTOVP(mp, &fhp->fh_fid, vpp);
2269 	if (error)
2270 		return (error);
2271 
2272 	if (!(exflags & (MNT_EXNORESPORT|MNT_EXPUBLIC))) {
2273 		saddr = mtod(nam, struct sockaddr_in *);
2274 		if ((saddr->sin_family == AF_INET) &&
2275 		    ntohs(saddr->sin_port) >= IPPORT_RESERVED) {
2276 			vput(*vpp);
2277 			return (NFSERR_AUTHERR | AUTH_TOOWEAK);
2278 		}
2279 #ifdef INET6
2280 		if ((saddr->sin_family == AF_INET6) &&
2281 		    ntohs(saddr->sin_port) >= IPV6PORT_RESERVED) {
2282 			vput(*vpp);
2283 			return (NFSERR_AUTHERR | AUTH_TOOWEAK);
2284 		}
2285 #endif
2286 	}
2287 	/*
2288 	 * Check/setup credentials.
2289 	 */
2290 	if (exflags & MNT_EXKERB) {
2291 		if (!kerbflag) {
2292 			vput(*vpp);
2293 			return (NFSERR_AUTHERR | AUTH_TOOWEAK);
2294 		}
2295 	} else if (kerbflag) {
2296 		vput(*vpp);
2297 		return (NFSERR_AUTHERR | AUTH_TOOWEAK);
2298 	} else if (cred->cr_uid == 0 || (exflags & MNT_EXPORTANON)) {
2299 		cred->cr_uid = credanon->cr_uid;
2300 		cred->cr_gid = credanon->cr_gid;
2301 		for (i = 0; i < credanon->cr_ngroups && i < NGROUPS; i++)
2302 			cred->cr_groups[i] = credanon->cr_groups[i];
2303 		cred->cr_ngroups = i;
2304 	}
2305 	if (exflags & MNT_EXRDONLY)
2306 		*rdonlyp = 1;
2307 	else
2308 		*rdonlyp = 0;
2309 	if (!lockflag)
2310 		VOP_UNLOCK(*vpp, 0);
2311 	return (0);
2312 }
2313 
2314 /*
2315  * WebNFS: check if a filehandle is a public filehandle. For v3, this
2316  * means a length of 0, for v2 it means all zeroes. nfsm_srvmtofh has
2317  * transformed this to all zeroes in both cases, so check for it.
2318  */
2319 int
2320 nfs_ispublicfh(fhp)
2321 	fhandle_t *fhp;
2322 {
2323 	char *cp = (char *)fhp;
2324 	int i;
2325 
2326 	for (i = 0; i < NFSX_V3FH; i++)
2327 		if (*cp++ != 0)
2328 			return (FALSE);
2329 	return (TRUE);
2330 }
2331 
2332 /*
2333  * This function compares two net addresses by family and returns TRUE
2334  * if they are the same host.
2335  * If there is any doubt, return FALSE.
2336  * The AF_INET family is handled as a special case so that address mbufs
2337  * don't need to be saved to store "struct in_addr", which is only 4 bytes.
2338  */
2339 int
2340 netaddr_match(family, haddr, nam)
2341 	int family;
2342 	union nethostaddr *haddr;
2343 	struct mbuf *nam;
2344 {
2345 	struct sockaddr_in *inetaddr;
2346 
2347 	switch (family) {
2348 	case AF_INET:
2349 		inetaddr = mtod(nam, struct sockaddr_in *);
2350 		if (inetaddr->sin_family == AF_INET &&
2351 		    inetaddr->sin_addr.s_addr == haddr->had_inetaddr)
2352 			return (1);
2353 		break;
2354 #ifdef INET6
2355 	case AF_INET6:
2356 	    {
2357 		struct sockaddr_in6 *sin6_1, *sin6_2;
2358 
2359 		sin6_1 = mtod(nam, struct sockaddr_in6 *);
2360 		sin6_2 = mtod(haddr->had_nam, struct sockaddr_in6 *);
2361 		if (sin6_1->sin6_family == AF_INET6 &&
2362 		    IN6_ARE_ADDR_EQUAL(&sin6_1->sin6_addr, &sin6_2->sin6_addr))
2363 			return 1;
2364 	    }
2365 #endif
2366 #ifdef ISO
2367 	case AF_ISO:
2368 	    {
2369 		struct sockaddr_iso *isoaddr1, *isoaddr2;
2370 
2371 		isoaddr1 = mtod(nam, struct sockaddr_iso *);
2372 		isoaddr2 = mtod(haddr->had_nam, struct sockaddr_iso *);
2373 		if (isoaddr1->siso_family == AF_ISO &&
2374 		    isoaddr1->siso_nlen > 0 &&
2375 		    isoaddr1->siso_nlen == isoaddr2->siso_nlen &&
2376 		    SAME_ISOADDR(isoaddr1, isoaddr2))
2377 			return (1);
2378 		break;
2379 	    }
2380 #endif	/* ISO */
2381 	default:
2382 		break;
2383 	};
2384 	return (0);
2385 }
2386 
2387 /*
2388  * The write verifier has changed (probably due to a server reboot), so all
2389  * B_NEEDCOMMIT blocks will have to be written again. Since they are on the
2390  * dirty block list as B_DELWRI, all this takes is clearing the B_NEEDCOMMIT
2391  * flag. Once done the new write verifier can be set for the mount point.
2392  */
2393 void
2394 nfs_clearcommit(mp)
2395 	struct mount *mp;
2396 {
2397 	struct vnode *vp;
2398 	struct nfsnode *np;
2399 	struct vm_page *pg;
2400 	int s;
2401 
2402 	s = splbio();
2403 	LIST_FOREACH(vp, &mp->mnt_vnodelist, v_mntvnodes) {
2404 		KASSERT(vp->v_mount == mp);
2405 		if (vp->v_type == VNON)
2406 			continue;
2407 		np = VTONFS(vp);
2408 		np->n_pushlo = np->n_pushhi = np->n_pushedlo =
2409 		    np->n_pushedhi = 0;
2410 		np->n_commitflags &=
2411 		    ~(NFS_COMMIT_PUSH_VALID | NFS_COMMIT_PUSHED_VALID);
2412 		simple_lock(&vp->v_uobj.vmobjlock);
2413 		TAILQ_FOREACH(pg, &vp->v_uobj.memq, listq) {
2414 			pg->flags &= ~PG_NEEDCOMMIT;
2415 		}
2416 		simple_unlock(&vp->v_uobj.vmobjlock);
2417 	}
2418 	splx(s);
2419 }
2420 
2421 void
2422 nfs_merge_commit_ranges(vp)
2423 	struct vnode *vp;
2424 {
2425 	struct nfsnode *np = VTONFS(vp);
2426 
2427 	if (!(np->n_commitflags & NFS_COMMIT_PUSHED_VALID)) {
2428 		np->n_pushedlo = np->n_pushlo;
2429 		np->n_pushedhi = np->n_pushhi;
2430 		np->n_commitflags |= NFS_COMMIT_PUSHED_VALID;
2431 	} else {
2432 		if (np->n_pushlo < np->n_pushedlo)
2433 			np->n_pushedlo = np->n_pushlo;
2434 		if (np->n_pushhi > np->n_pushedhi)
2435 			np->n_pushedhi = np->n_pushhi;
2436 	}
2437 
2438 	np->n_pushlo = np->n_pushhi = 0;
2439 	np->n_commitflags &= ~NFS_COMMIT_PUSH_VALID;
2440 
2441 #ifdef fvdl_debug
2442 	printf("merge: committed: %u - %u\n", (unsigned)np->n_pushedlo,
2443 	    (unsigned)np->n_pushedhi);
2444 #endif
2445 }
2446 
2447 int
2448 nfs_in_committed_range(vp, off, len)
2449 	struct vnode *vp;
2450 	off_t off, len;
2451 {
2452 	struct nfsnode *np = VTONFS(vp);
2453 	off_t lo, hi;
2454 
2455 	if (!(np->n_commitflags & NFS_COMMIT_PUSHED_VALID))
2456 		return 0;
2457 	lo = off;
2458 	hi = lo + len;
2459 
2460 	return (lo >= np->n_pushedlo && hi <= np->n_pushedhi);
2461 }
2462 
2463 int
2464 nfs_in_tobecommitted_range(vp, off, len)
2465 	struct vnode *vp;
2466 	off_t off, len;
2467 {
2468 	struct nfsnode *np = VTONFS(vp);
2469 	off_t lo, hi;
2470 
2471 	if (!(np->n_commitflags & NFS_COMMIT_PUSH_VALID))
2472 		return 0;
2473 	lo = off;
2474 	hi = lo + len;
2475 
2476 	return (lo >= np->n_pushlo && hi <= np->n_pushhi);
2477 }
2478 
2479 void
2480 nfs_add_committed_range(vp, off, len)
2481 	struct vnode *vp;
2482 	off_t off, len;
2483 {
2484 	struct nfsnode *np = VTONFS(vp);
2485 	off_t lo, hi;
2486 
2487 	lo = off;
2488 	hi = lo + len;
2489 
2490 	if (!(np->n_commitflags & NFS_COMMIT_PUSHED_VALID)) {
2491 		np->n_pushedlo = lo;
2492 		np->n_pushedhi = hi;
2493 		np->n_commitflags |= NFS_COMMIT_PUSHED_VALID;
2494 	} else {
2495 		if (hi > np->n_pushedhi)
2496 			np->n_pushedhi = hi;
2497 		if (lo < np->n_pushedlo)
2498 			np->n_pushedlo = lo;
2499 	}
2500 #ifdef fvdl_debug
2501 	printf("add: committed: %u - %u\n", (unsigned)np->n_pushedlo,
2502 	    (unsigned)np->n_pushedhi);
2503 #endif
2504 }
2505 
2506 void
2507 nfs_del_committed_range(vp, off, len)
2508 	struct vnode *vp;
2509 	off_t off, len;
2510 {
2511 	struct nfsnode *np = VTONFS(vp);
2512 	off_t lo, hi;
2513 
2514 	if (!(np->n_commitflags & NFS_COMMIT_PUSHED_VALID))
2515 		return;
2516 
2517 	lo = off;
2518 	hi = lo + len;
2519 
2520 	if (lo > np->n_pushedhi || hi < np->n_pushedlo)
2521 		return;
2522 	if (lo <= np->n_pushedlo)
2523 		np->n_pushedlo = hi;
2524 	else if (hi >= np->n_pushedhi)
2525 		np->n_pushedhi = lo;
2526 	else {
2527 		/*
2528 		 * XXX There's only one range. If the deleted range
2529 		 * is in the middle, pick the largest of the
2530 		 * contiguous ranges that it leaves.
2531 		 */
2532 		if ((np->n_pushedlo - lo) > (hi - np->n_pushedhi))
2533 			np->n_pushedhi = lo;
2534 		else
2535 			np->n_pushedlo = hi;
2536 	}
2537 #ifdef fvdl_debug
2538 	printf("del: committed: %u - %u\n", (unsigned)np->n_pushedlo,
2539 	    (unsigned)np->n_pushedhi);
2540 #endif
2541 }
2542 
2543 void
2544 nfs_add_tobecommitted_range(vp, off, len)
2545 	struct vnode *vp;
2546 	off_t off, len;
2547 {
2548 	struct nfsnode *np = VTONFS(vp);
2549 	off_t lo, hi;
2550 
2551 	lo = off;
2552 	hi = lo + len;
2553 
2554 	if (!(np->n_commitflags & NFS_COMMIT_PUSH_VALID)) {
2555 		np->n_pushlo = lo;
2556 		np->n_pushhi = hi;
2557 		np->n_commitflags |= NFS_COMMIT_PUSH_VALID;
2558 	} else {
2559 		if (lo < np->n_pushlo)
2560 			np->n_pushlo = lo;
2561 		if (hi > np->n_pushhi)
2562 			np->n_pushhi = hi;
2563 	}
2564 #ifdef fvdl_debug
2565 	printf("add: tobecommitted: %u - %u\n", (unsigned)np->n_pushlo,
2566 	    (unsigned)np->n_pushhi);
2567 #endif
2568 }
2569 
2570 void
2571 nfs_del_tobecommitted_range(vp, off, len)
2572 	struct vnode *vp;
2573 	off_t off, len;
2574 {
2575 	struct nfsnode *np = VTONFS(vp);
2576 	off_t lo, hi;
2577 
2578 	if (!(np->n_commitflags & NFS_COMMIT_PUSH_VALID))
2579 		return;
2580 
2581 	lo = off;
2582 	hi = lo + len;
2583 
2584 	if (lo > np->n_pushhi || hi < np->n_pushlo)
2585 		return;
2586 
2587 	if (lo <= np->n_pushlo)
2588 		np->n_pushlo = hi;
2589 	else if (hi >= np->n_pushhi)
2590 		np->n_pushhi = lo;
2591 	else {
2592 		/*
2593 		 * XXX There's only one range. If the deleted range
2594 		 * is in the middle, pick the largest of the
2595 		 * contiguous ranges that it leaves.
2596 		 */
2597 		if ((np->n_pushlo - lo) > (hi - np->n_pushhi))
2598 			np->n_pushhi = lo;
2599 		else
2600 			np->n_pushlo = hi;
2601 	}
2602 #ifdef fvdl_debug
2603 	printf("del: tobecommitted: %u - %u\n", (unsigned)np->n_pushlo,
2604 	    (unsigned)np->n_pushhi);
2605 #endif
2606 }
2607 
2608 /*
2609  * Map errnos to NFS error numbers. For Version 3 also filter out error
2610  * numbers not specified for the associated procedure.
2611  */
2612 int
2613 nfsrv_errmap(nd, err)
2614 	struct nfsrv_descript *nd;
2615 	int err;
2616 {
2617 	const short *defaulterrp, *errp;
2618 
2619 	if (nd->nd_flag & ND_NFSV3) {
2620 	    if (nd->nd_procnum <= NFSPROC_COMMIT) {
2621 		errp = defaulterrp = nfsrv_v3errmap[nd->nd_procnum];
2622 		while (*++errp) {
2623 			if (*errp == err)
2624 				return (err);
2625 			else if (*errp > err)
2626 				break;
2627 		}
2628 		return ((int)*defaulterrp);
2629 	    } else
2630 		return (err & 0xffff);
2631 	}
2632 	if (err <= ELAST)
2633 		return ((int)nfsrv_v2errmap[err - 1]);
2634 	return (NFSERR_IO);
2635 }
2636 
2637 /*
2638  * Sort the group list in increasing numerical order.
2639  * (Insertion sort by Chris Torek, who was grossed out by the bubble sort
2640  *  that used to be here.)
2641  */
2642 void
2643 nfsrvw_sort(list, num)
2644         gid_t *list;
2645         int num;
2646 {
2647 	int i, j;
2648 	gid_t v;
2649 
2650 	/* Insertion sort. */
2651 	for (i = 1; i < num; i++) {
2652 		v = list[i];
2653 		/* find correct slot for value v, moving others up */
2654 		for (j = i; --j >= 0 && v < list[j];)
2655 			list[j + 1] = list[j];
2656 		list[j + 1] = v;
2657 	}
2658 }
2659 
2660 /*
2661  * copy credentials making sure that the result can be compared with memcmp().
2662  */
2663 void
2664 nfsrv_setcred(incred, outcred)
2665 	struct ucred *incred, *outcred;
2666 {
2667 	int i;
2668 
2669 	memset((caddr_t)outcred, 0, sizeof (struct ucred));
2670 	outcred->cr_ref = 1;
2671 	outcred->cr_uid = incred->cr_uid;
2672 	outcred->cr_gid = incred->cr_gid;
2673 	outcred->cr_ngroups = incred->cr_ngroups;
2674 	for (i = 0; i < incred->cr_ngroups; i++)
2675 		outcred->cr_groups[i] = incred->cr_groups[i];
2676 	nfsrvw_sort(outcred->cr_groups, outcred->cr_ngroups);
2677 }
2678