ext2fs_readwrite.c revision 1.38 1 /* $NetBSD: ext2fs_readwrite.c,v 1.38 2005/11/02 12:39:00 yamt Exp $ */
2
3 /*-
4 * Copyright (c) 1993
5 * The Regents of the University of California. All rights reserved.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
15 * 3. Neither the name of the University nor the names of its contributors
16 * may be used to endorse or promote products derived from this software
17 * without specific prior written permission.
18 *
19 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * SUCH DAMAGE.
30 *
31 * @(#)ufs_readwrite.c 8.8 (Berkeley) 8/4/94
32 * Modified for ext2fs by Manuel Bouyer.
33 */
34
35 /*-
36 * Copyright (c) 1997 Manuel Bouyer.
37 *
38 * Redistribution and use in source and binary forms, with or without
39 * modification, are permitted provided that the following conditions
40 * are met:
41 * 1. Redistributions of source code must retain the above copyright
42 * notice, this list of conditions and the following disclaimer.
43 * 2. Redistributions in binary form must reproduce the above copyright
44 * notice, this list of conditions and the following disclaimer in the
45 * documentation and/or other materials provided with the distribution.
46 * 3. All advertising materials mentioning features or use of this software
47 * must display the following acknowledgement:
48 * This product includes software developed by Manuel Bouyer.
49 * 4. The name of the author may not be used to endorse or promote products
50 * derived from this software without specific prior written permission.
51 *
52 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
53 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
54 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
55 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
56 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
57 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
58 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
59 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
60 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
61 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
62 *
63 * @(#)ufs_readwrite.c 8.8 (Berkeley) 8/4/94
64 * Modified for ext2fs by Manuel Bouyer.
65 */
66
67 #include <sys/cdefs.h>
68 __KERNEL_RCSID(0, "$NetBSD: ext2fs_readwrite.c,v 1.38 2005/11/02 12:39:00 yamt Exp $");
69
70 #include <sys/param.h>
71 #include <sys/systm.h>
72 #include <sys/resourcevar.h>
73 #include <sys/kernel.h>
74 #include <sys/file.h>
75 #include <sys/stat.h>
76 #include <sys/buf.h>
77 #include <sys/proc.h>
78 #include <sys/mount.h>
79 #include <sys/vnode.h>
80 #include <sys/malloc.h>
81 #include <sys/signalvar.h>
82
83 #include <ufs/ufs/inode.h>
84 #include <ufs/ufs/ufsmount.h>
85 #include <ufs/ufs/ufs_extern.h>
86 #include <ufs/ext2fs/ext2fs.h>
87 #include <ufs/ext2fs/ext2fs_extern.h>
88
89
90 #define doclusterread 0 /* XXX underway */
91 #define doclusterwrite 0
92
93 /*
94 * Vnode op for reading.
95 */
96 /* ARGSUSED */
97 int
98 ext2fs_read(void *v)
99 {
100 struct vop_read_args /* {
101 struct vnode *a_vp;
102 struct uio *a_uio;
103 int a_ioflag;
104 struct ucred *a_cred;
105 } */ *ap = v;
106 struct vnode *vp;
107 struct inode *ip;
108 struct uio *uio;
109 struct m_ext2fs *fs;
110 struct buf *bp;
111 struct ufsmount *ump;
112 void *win;
113 vsize_t bytelen;
114 daddr_t lbn, nextlbn;
115 off_t bytesinfile;
116 long size, xfersize, blkoffset;
117 int error, flags;
118
119 vp = ap->a_vp;
120 ip = VTOI(vp);
121 ump = ip->i_ump;
122 uio = ap->a_uio;
123 error = 0;
124
125 #ifdef DIAGNOSTIC
126 if (uio->uio_rw != UIO_READ)
127 panic("%s: mode", "ext2fs_read");
128
129 if (vp->v_type == VLNK) {
130 if (ext2fs_size(ip) < ump->um_maxsymlinklen ||
131 (ump->um_maxsymlinklen == 0 && ip->i_e2fs_nblock == 0))
132 panic("%s: short symlink", "ext2fs_read");
133 } else if (vp->v_type != VREG && vp->v_type != VDIR)
134 panic("%s: type %d", "ext2fs_read", vp->v_type);
135 #endif
136 fs = ip->i_e2fs;
137 if ((u_int64_t)uio->uio_offset > ump->um_maxfilesize)
138 return (EFBIG);
139 if (uio->uio_resid == 0)
140 return (0);
141 if (uio->uio_offset >= ext2fs_size(ip))
142 goto out;
143
144 if (vp->v_type == VREG) {
145 while (uio->uio_resid > 0) {
146 bytelen = MIN(ext2fs_size(ip) - uio->uio_offset,
147 uio->uio_resid);
148 if (bytelen == 0)
149 break;
150
151 win = ubc_alloc(&vp->v_uobj, uio->uio_offset,
152 &bytelen, UBC_READ);
153 error = uiomove(win, bytelen, uio);
154 flags = UBC_WANT_UNMAP(vp) ? UBC_UNMAP : 0;
155 ubc_release(win, flags);
156 if (error)
157 break;
158 }
159 goto out;
160 }
161
162 for (error = 0, bp = NULL; uio->uio_resid > 0; bp = NULL) {
163 bytesinfile = ext2fs_size(ip) - uio->uio_offset;
164 if (bytesinfile <= 0)
165 break;
166 lbn = lblkno(fs, uio->uio_offset);
167 nextlbn = lbn + 1;
168 size = fs->e2fs_bsize;
169 blkoffset = blkoff(fs, uio->uio_offset);
170 xfersize = fs->e2fs_bsize - blkoffset;
171 if (uio->uio_resid < xfersize)
172 xfersize = uio->uio_resid;
173 if (bytesinfile < xfersize)
174 xfersize = bytesinfile;
175
176 if (lblktosize(fs, nextlbn) >= ext2fs_size(ip))
177 error = bread(vp, lbn, size, NOCRED, &bp);
178 else {
179 int nextsize = fs->e2fs_bsize;
180 error = breadn(vp, lbn,
181 size, &nextlbn, &nextsize, 1, NOCRED, &bp);
182 }
183 if (error)
184 break;
185
186 /*
187 * We should only get non-zero b_resid when an I/O error
188 * has occurred, which should cause us to break above.
189 * However, if the short read did not cause an error,
190 * then we want to ensure that we do not uiomove bad
191 * or uninitialized data.
192 */
193 size -= bp->b_resid;
194 if (size < xfersize) {
195 if (size == 0)
196 break;
197 xfersize = size;
198 }
199 error = uiomove((char *)bp->b_data + blkoffset, xfersize, uio);
200 if (error)
201 break;
202 brelse(bp);
203 }
204 if (bp != NULL)
205 brelse(bp);
206
207 out:
208 if (!(vp->v_mount->mnt_flag & MNT_NOATIME)) {
209 ip->i_flag |= IN_ACCESS;
210 if ((ap->a_ioflag & IO_SYNC) == IO_SYNC)
211 error = ext2fs_update(vp, NULL, NULL, UPDATE_WAIT);
212 }
213 return (error);
214 }
215
216 /*
217 * Vnode op for writing.
218 */
219 int
220 ext2fs_write(void *v)
221 {
222 struct vop_write_args /* {
223 struct vnode *a_vp;
224 struct uio *a_uio;
225 int a_ioflag;
226 struct ucred *a_cred;
227 } */ *ap = v;
228 struct vnode *vp;
229 struct uio *uio;
230 struct inode *ip;
231 struct m_ext2fs *fs;
232 struct buf *bp;
233 struct proc *p;
234 struct ufsmount *ump;
235 daddr_t lbn;
236 off_t osize;
237 int blkoffset, error, flags, ioflag, resid, xfersize;
238 vsize_t bytelen;
239 void *win;
240 off_t oldoff = 0; /* XXX */
241 boolean_t async;
242 int extended = 0;
243
244 ioflag = ap->a_ioflag;
245 uio = ap->a_uio;
246 vp = ap->a_vp;
247 ip = VTOI(vp);
248 ump = ip->i_ump;
249 error = 0;
250
251 #ifdef DIAGNOSTIC
252 if (uio->uio_rw != UIO_WRITE)
253 panic("%s: mode", "ext2fs_write");
254 #endif
255
256 switch (vp->v_type) {
257 case VREG:
258 if (ioflag & IO_APPEND)
259 uio->uio_offset = ext2fs_size(ip);
260 if ((ip->i_e2fs_flags & EXT2_APPEND) &&
261 uio->uio_offset != ext2fs_size(ip))
262 return (EPERM);
263 /* FALLTHROUGH */
264 case VLNK:
265 break;
266 case VDIR:
267 if ((ioflag & IO_SYNC) == 0)
268 panic("%s: nonsync dir write", "ext2fs_write");
269 break;
270 default:
271 panic("%s: type", "ext2fs_write");
272 }
273
274 fs = ip->i_e2fs;
275 if (uio->uio_offset < 0 ||
276 (u_int64_t)uio->uio_offset + uio->uio_resid > ump->um_maxfilesize)
277 return (EFBIG);
278 /*
279 * Maybe this should be above the vnode op call, but so long as
280 * file servers have no limits, I don't think it matters.
281 */
282 p = uio->uio_procp;
283 if (vp->v_type == VREG && p &&
284 uio->uio_offset + uio->uio_resid >
285 p->p_rlimit[RLIMIT_FSIZE].rlim_cur) {
286 psignal(p, SIGXFSZ);
287 return (EFBIG);
288 }
289 if (uio->uio_resid == 0)
290 return (0);
291
292 async = vp->v_mount->mnt_flag & MNT_ASYNC;
293 resid = uio->uio_resid;
294 osize = ext2fs_size(ip);
295
296 if (vp->v_type == VREG) {
297 while (uio->uio_resid > 0) {
298 oldoff = uio->uio_offset;
299 blkoffset = blkoff(fs, uio->uio_offset);
300 bytelen = MIN(fs->e2fs_bsize - blkoffset,
301 uio->uio_resid);
302
303 error = ufs_balloc_range(vp, uio->uio_offset,
304 bytelen, ap->a_cred, 0);
305 if (error)
306 break;
307 win = ubc_alloc(&vp->v_uobj, uio->uio_offset,
308 &bytelen, UBC_WRITE);
309 error = uiomove(win, bytelen, uio);
310 flags = UBC_WANT_UNMAP(vp) ? UBC_UNMAP : 0;
311 ubc_release(win, flags);
312 if (error)
313 break;
314
315 /*
316 * update UVM's notion of the size now that we've
317 * copied the data into the vnode's pages.
318 */
319
320 if (vp->v_size < uio->uio_offset) {
321 uvm_vnp_setsize(vp, uio->uio_offset);
322 extended = 1;
323 }
324
325 /*
326 * flush what we just wrote if necessary.
327 * XXXUBC simplistic async flushing.
328 */
329
330 if (!async && oldoff >> 16 != uio->uio_offset >> 16) {
331 simple_lock(&vp->v_interlock);
332 error = VOP_PUTPAGES(vp, (oldoff >> 16) << 16,
333 (uio->uio_offset >> 16) << 16, PGO_CLEANIT);
334 }
335 }
336 if (error == 0 && ioflag & IO_SYNC) {
337 simple_lock(&vp->v_interlock);
338 error = VOP_PUTPAGES(vp, trunc_page(oldoff),
339 round_page(blkroundup(fs, uio->uio_offset)),
340 PGO_CLEANIT | PGO_SYNCIO);
341 }
342
343 goto out;
344 }
345
346 flags = ioflag & IO_SYNC ? B_SYNC : 0;
347 for (error = 0; uio->uio_resid > 0;) {
348 lbn = lblkno(fs, uio->uio_offset);
349 blkoffset = blkoff(fs, uio->uio_offset);
350 xfersize = MIN(fs->e2fs_bsize - blkoffset, uio->uio_resid);
351 if (xfersize < fs->e2fs_bsize)
352 flags |= B_CLRBUF;
353 else
354 flags &= ~B_CLRBUF;
355 error = ext2fs_balloc(ip,
356 lbn, blkoffset + xfersize, ap->a_cred, &bp, flags);
357 if (error)
358 break;
359 if (ext2fs_size(ip) < uio->uio_offset + xfersize) {
360 error = ext2fs_setsize(ip, uio->uio_offset + xfersize);
361 if (error)
362 break;
363 }
364 error = uiomove((char *)bp->b_data + blkoffset, xfersize, uio);
365
366 /*
367 * update UVM's notion of the size now that we've
368 * copied the data into the vnode's pages.
369 */
370
371 if (vp->v_size < uio->uio_offset) {
372 uvm_vnp_setsize(vp, uio->uio_offset);
373 extended = 1;
374 }
375
376 if (ioflag & IO_SYNC)
377 (void)bwrite(bp);
378 else if (xfersize + blkoffset == fs->e2fs_bsize)
379 bawrite(bp);
380 else
381 bdwrite(bp);
382 if (error || xfersize == 0)
383 break;
384 }
385
386 /*
387 * If we successfully wrote any data, and we are not the superuser
388 * we clear the setuid and setgid bits as a precaution against
389 * tampering.
390 */
391
392 out:
393 ip->i_flag |= IN_CHANGE | IN_UPDATE;
394 if (resid > uio->uio_resid && ap->a_cred && ap->a_cred->cr_uid != 0)
395 ip->i_e2fs_mode &= ~(ISUID | ISGID);
396 if (resid > uio->uio_resid)
397 VN_KNOTE(vp, NOTE_WRITE | (extended ? NOTE_EXTEND : 0));
398 if (error) {
399 (void) ext2fs_truncate(vp, osize, ioflag & IO_SYNC, ap->a_cred,
400 uio->uio_procp);
401 uio->uio_offset -= resid - uio->uio_resid;
402 uio->uio_resid = resid;
403 } else if (resid > uio->uio_resid && (ioflag & IO_SYNC) == IO_SYNC)
404 error = ext2fs_update(vp, NULL, NULL, UPDATE_WAIT);
405 KASSERT(vp->v_size == ext2fs_size(ip));
406 return (error);
407 }
408