1 /* $NetBSD: ext2fs_readwrite.c,v 1.34 2004/11/14 19:42:13 christos 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 1.34 2004/11/14 19:42:13 christos 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(v) 99 void *v; 100 { 101 struct vop_read_args /* { 102 struct vnode *a_vp; 103 struct uio *a_uio; 104 int a_ioflag; 105 struct ucred *a_cred; 106 } */ *ap = v; 107 struct vnode *vp; 108 struct inode *ip; 109 struct uio *uio; 110 struct m_ext2fs *fs; 111 struct buf *bp; 112 void *win; 113 vsize_t bytelen; 114 daddr_t lbn, nextlbn; 115 off_t bytesinfile; 116 long size, xfersize, blkoffset; 117 int error; 118 struct ufsmount *ump; 119 120 vp = ap->a_vp; 121 ip = VTOI(vp); 122 ump = ip->i_ump; 123 uio = ap->a_uio; 124 error = 0; 125 126 #ifdef DIAGNOSTIC 127 if (uio->uio_rw != UIO_READ) 128 panic("%s: mode", "ext2fs_read"); 129 130 if (vp->v_type == VLNK) { 131 if (ip->i_e2fs_size < ump->um_maxsymlinklen || 132 (ump->um_maxsymlinklen == 0 && ip->i_e2fs_nblock == 0)) 133 panic("%s: short symlink", "ext2fs_read"); 134 } else if (vp->v_type != VREG && vp->v_type != VDIR) 135 panic("%s: type %d", "ext2fs_read", vp->v_type); 136 #endif 137 fs = ip->i_e2fs; 138 if ((u_int64_t)uio->uio_offset > ump->um_maxfilesize) 139 return (EFBIG); 140 if (uio->uio_resid == 0) 141 return (0); 142 if (uio->uio_offset >= ip->i_e2fs_size) 143 goto out; 144 145 if (vp->v_type == VREG) { 146 while (uio->uio_resid > 0) { 147 bytelen = MIN(ip->i_e2fs_size - uio->uio_offset, 148 uio->uio_resid); 149 if (bytelen == 0) 150 break; 151 152 win = ubc_alloc(&vp->v_uobj, uio->uio_offset, 153 &bytelen, UBC_READ); 154 error = uiomove(win, bytelen, uio); 155 ubc_release(win, 0); 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 = ip->i_e2fs_size - 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) >= ip->i_e2fs_size) 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 = VOP_UPDATE(vp, NULL, NULL, UPDATE_WAIT); 212 } 213 return (error); 214 } 215 216 /* 217 * Vnode op for writing. 218 */ 219 int 220 ext2fs_write(v) 221 void *v; 222 { 223 struct vop_write_args /* { 224 struct vnode *a_vp; 225 struct uio *a_uio; 226 int a_ioflag; 227 struct ucred *a_cred; 228 } */ *ap = v; 229 struct vnode *vp; 230 struct uio *uio; 231 struct inode *ip; 232 struct m_ext2fs *fs; 233 struct buf *bp; 234 struct proc *p; 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; 241 boolean_t async; 242 int extended = 0; 243 struct ufsmount *ump; 244 245 ioflag = ap->a_ioflag; 246 uio = ap->a_uio; 247 vp = ap->a_vp; 248 ip = VTOI(vp); 249 ump = ip->i_ump; 250 error = 0; 251 252 #ifdef DIAGNOSTIC 253 if (uio->uio_rw != UIO_WRITE) 254 panic("%s: mode", "ext2fs_write"); 255 #endif 256 257 switch (vp->v_type) { 258 case VREG: 259 if (ioflag & IO_APPEND) 260 uio->uio_offset = ip->i_e2fs_size; 261 if ((ip->i_e2fs_flags & EXT2_APPEND) && 262 uio->uio_offset != ip->i_e2fs_size) 263 return (EPERM); 264 /* FALLTHROUGH */ 265 case VLNK: 266 break; 267 case VDIR: 268 if ((ioflag & IO_SYNC) == 0) 269 panic("%s: nonsync dir write", "ext2fs_write"); 270 break; 271 default: 272 panic("%s: type", "ext2fs_write"); 273 } 274 275 fs = ip->i_e2fs; 276 if (uio->uio_offset < 0 || 277 (u_int64_t)uio->uio_offset + uio->uio_resid > ump->um_maxfilesize) 278 return (EFBIG); 279 /* 280 * Maybe this should be above the vnode op call, but so long as 281 * file servers have no limits, I don't think it matters. 282 */ 283 p = uio->uio_procp; 284 if (vp->v_type == VREG && p && 285 uio->uio_offset + uio->uio_resid > 286 p->p_rlimit[RLIMIT_FSIZE].rlim_cur) { 287 psignal(p, SIGXFSZ); 288 return (EFBIG); 289 } 290 if (uio->uio_resid == 0) 291 return (0); 292 293 async = vp->v_mount->mnt_flag & MNT_ASYNC; 294 resid = uio->uio_resid; 295 osize = ip->i_e2fs_size; 296 297 if (vp->v_type == VREG) { 298 while (uio->uio_resid > 0) { 299 oldoff = uio->uio_offset; 300 blkoffset = blkoff(fs, uio->uio_offset); 301 bytelen = MIN(fs->e2fs_bsize - blkoffset, 302 uio->uio_resid); 303 304 error = ufs_balloc_range(vp, uio->uio_offset, 305 bytelen, ap->a_cred, 0); 306 if (error) 307 break; 308 win = ubc_alloc(&vp->v_uobj, uio->uio_offset, 309 &bytelen, UBC_WRITE); 310 error = uiomove(win, bytelen, uio); 311 ubc_release(win, 0); 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 (ip->i_e2fs_size < uio->uio_offset + xfersize) { 360 ip->i_e2fs_size = uio->uio_offset + xfersize; 361 } 362 error = uiomove((char *)bp->b_data + blkoffset, xfersize, uio); 363 364 /* 365 * update UVM's notion of the size now that we've 366 * copied the data into the vnode's pages. 367 */ 368 369 if (vp->v_size < uio->uio_offset) { 370 uvm_vnp_setsize(vp, uio->uio_offset); 371 extended = 1; 372 } 373 374 if (ioflag & IO_SYNC) 375 (void)bwrite(bp); 376 else if (xfersize + blkoffset == fs->e2fs_bsize) 377 bawrite(bp); 378 else 379 bdwrite(bp); 380 if (error || xfersize == 0) 381 break; 382 } 383 384 /* 385 * If we successfully wrote any data, and we are not the superuser 386 * we clear the setuid and setgid bits as a precaution against 387 * tampering. 388 */ 389 390 out: 391 ip->i_flag |= IN_CHANGE | IN_UPDATE; 392 if (resid > uio->uio_resid && ap->a_cred && ap->a_cred->cr_uid != 0) 393 ip->i_e2fs_mode &= ~(ISUID | ISGID); 394 if (resid > uio->uio_resid) 395 VN_KNOTE(vp, NOTE_WRITE | (extended ? NOTE_EXTEND : 0)); 396 if (error) { 397 (void) VOP_TRUNCATE(vp, osize, ioflag & IO_SYNC, ap->a_cred, 398 uio->uio_procp); 399 uio->uio_offset -= resid - uio->uio_resid; 400 uio->uio_resid = resid; 401 } else if (resid > uio->uio_resid && (ioflag & IO_SYNC) == IO_SYNC) 402 error = VOP_UPDATE(vp, NULL, NULL, UPDATE_WAIT); 403 KASSERT(vp->v_size == ip->i_e2fs_size); 404 return (error); 405 } 406