dd/dd.c

   1 /*-
   2  * Copyright (c) 1991, 1993, 1994
   3  *      The Regents of the University of California.  All rights reserved.
   4  *
   5  * This code is derived from software contributed to Berkeley by
   6  * Keith Muller of the University of California, San Diego and Lance
   7  * Visser of Convex Computer Corporation.
   8  *
   9  * Redistribution and use in source and binary forms, with or without
  10  * modification, are permitted provided that the following conditions
  11  * are met:
  12  * 1. Redistributions of source code must retain the above copyright
  13  *    notice, this list of conditions and the following disclaimer.
  14  * 2. Redistributions in binary form must reproduce the above copyright
  15  *    notice, this list of conditions and the following disclaimer in the
  16  *    documentation and/or other materials provided with the distribution.
  17  * 3. All advertising materials mentioning features or use of this software
  18  *    must display the following acknowledgement:
  19  *      This product includes software developed by the University of
  20  *      California, Berkeley and its contributors.
  21  * 4. Neither the name of the University nor the names of its contributors
  22  *    may be used to endorse or promote products derived from this software
  23  *    without specific prior written permission.
  24  *
  25  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  26  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  27  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  28  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  29  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  30  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  31  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  32  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  33  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  34  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  35  * SUCH DAMAGE.
  36  */
  37
  38 #ifndef lint
  39 static char const copyright[] =
  40 "@(#) Copyright (c) 1991, 1993, 1994\n\
  41         The Regents of the University of California.  All rights reserved.\n";
  42 #endif /* not lint */
  43
  44 #ifndef lint
  45 #if 0
  46 static char sccsid[] = "@(#)dd.c        8.5 (Berkeley) 4/2/94";
  47 #endif
  48 static const char rcsid[] =
  49   "$FreeBSD: src/bin/dd/dd.c,v 1.36 2002/03/07 14:00:33 markm Exp $";
  50 #endif /* not lint */
  51
  52 #include <sys/param.h>
  53 #include <sys/stat.h>
  54 #include <sys/conf.h>
  55 #include <sys/filio.h>
  56 #include <sys/time.h>
  57
  58 #ifdef __APPLE__
  59 #include <sys/ioctl.h>
  60 #else
  61 #include <sys/disklabel.h>
  62 #endif
  63
  64 #include <ctype.h>
  65 #include <err.h>
  66 #include <errno.h>
  67 #include <fcntl.h>
  68 #include <locale.h>
  69 #include <stdio.h>
  70 #include <stdlib.h>
  71 #include <string.h>
  72 #include <unistd.h>
  73
  74 #include "dd.h"
  75 #include "extern.h"
  76
  77 static void dd_close(void);
  78 static void dd_in(void);
  79 static void getfdtype(IO *);
  80 static void setup(void);
  81
  82 IO      in, out;                /* input/output state */
  83 STAT    st;                     /* statistics */
  84 void    (*cfunc)(void);         /* conversion function */
  85 quad_t  cpy_cnt;                /* # of blocks to copy */
  86 off_t   pending = 0;            /* pending seek if sparse */
  87 u_int   ddflags;                /* conversion options */
  88 size_t  cbsz;                   /* conversion block size */
  89 quad_t  files_cnt = 1;          /* # of files to copy */
  90 const   u_char *ctab;           /* conversion table */
  91
  92 int
  93 main(int argc, char *argv[])
  94 {
  95         (void)setlocale(LC_CTYPE, "");
  96         jcl(argv);
  97         setup();
  98
  99         (void)signal(SIGINFO, summaryx);
 100         (void)signal(SIGINT, terminate);
 101
 102         atexit(summary);
 103
 104         while (files_cnt--)
 105                 dd_in();
 106
 107         dd_close();
 108         exit(0);
 109 }
 110
 111 static void
 112 setup(void)
 113 {
 114         u_int cnt;
 115         struct timeval tv;
 116
 117         if (in.name == NULL) {
 118                 in.name = "stdin";
 119                 in.fd = STDIN_FILENO;
 120         } else {
 121                 in.fd = open(in.name, O_RDONLY, 0);
 122                 if (in.fd == -1)
 123                         err(1, "%s", in.name);
 124         }
 125
 126         getfdtype(&in);
 127
 128         if (files_cnt > 1 && !(in.flags & ISTAPE))
 129                 errx(1, "files is not supported for non-tape devices");
 130
 131         if (out.name == NULL) {
 132                 /* No way to check for read access here. */
 133                 out.fd = STDOUT_FILENO;
 134                 out.name = "stdout";
 135         } else {
 136 #define OFLAGS \
 137     (O_CREAT | (ddflags & (C_SEEK | C_NOTRUNC) ? 0 : O_TRUNC))
 138                 out.fd = open(out.name, O_RDWR | OFLAGS, DEFFILEMODE);
 139                 /*
 140                  * May not have read access, so try again with write only.
 141                  * Without read we may have a problem if output also does
 142                  * not support seeks.
 143                  */
 144                 if (out.fd == -1) {
 145                         out.fd = open(out.name, O_WRONLY | OFLAGS, DEFFILEMODE);
 146                         out.flags |= NOREAD;
 147                 }
 148                 if (out.fd == -1)
 149                         err(1, "%s", out.name);
 150         }
 151
 152         getfdtype(&out);
 153
 154         /*
 155          * Allocate space for the input and output buffers.  If not doing
 156          * record oriented I/O, only need a single buffer.
 157          */
 158         if (!(ddflags & (C_BLOCK | C_UNBLOCK))) {
 159                 if ((in.db = malloc(out.dbsz + in.dbsz - 1)) == NULL)
 160                         err(1, "input buffer");
 161                 out.db = in.db;
 162         } else if ((in.db = malloc(MAX(in.dbsz, cbsz) + cbsz)) == NULL ||
 163             (out.db = malloc(out.dbsz + cbsz)) == NULL)
 164                 err(1, "output buffer");
 165         in.dbp = in.db;
 166         out.dbp = out.db;
 167
 168         /* Position the input/output streams. */
 169         if (in.offset)
 170                 pos_in();
 171         if (out.offset)
 172                 pos_out();
 173
 174         /*
 175          * Truncate the output file.  If it fails on a type of output file
 176          * that it should _not_ fail on, error out.
 177          */
 178         if ((ddflags & (C_OF | C_SEEK | C_NOTRUNC)) == (C_OF | C_SEEK) &&
 179             out.flags & ISTRUNC)
 180                 if (ftruncate(out.fd, out.offset * out.dbsz) == -1)
 181                         err(1, "truncating %s", out.name);
 182
 183         /*
 184          * If converting case at the same time as another conversion, build a
 185          * table that does both at once.  If just converting case, use the
 186          * built-in tables.
 187          */
 188         if (ddflags & (C_LCASE | C_UCASE)) {
 189                 if (ddflags & (C_ASCII | C_EBCDIC)) {
 190                         if (ddflags & C_LCASE) {
 191                                 for (cnt = 0; cnt <= 0377; ++cnt)
 192                                         casetab[cnt] = tolower(ctab[cnt]);
 193                         } else {
 194                                 for (cnt = 0; cnt <= 0377; ++cnt)
 195                                         casetab[cnt] = toupper(ctab[cnt]);
 196                         }
 197                 } else {
 198                         if (ddflags & C_LCASE) {
 199                                 for (cnt = 0; cnt <= 0377; ++cnt)
 200                                         casetab[cnt] = tolower((int)cnt);
 201                         } else {
 202                                 for (cnt = 0; cnt <= 0377; ++cnt)
 203                                         casetab[cnt] = toupper((int)cnt);
 204                         }
 205                 }
 206                 ctab = casetab;
 207         }
 208
 209         (void)gettimeofday(&tv, (struct timezone *)NULL);
 210         st.start = tv.tv_sec + tv.tv_usec * 1e-6;
 211 }
 212
 213 static void
 214 getfdtype(IO *io)
 215 {
 216         struct stat sb;
 217         int type;
 218
 219         if (fstat(io->fd, &sb) == -1)
 220                 err(1, "%s", io->name);
 221         if (S_ISREG(sb.st_mode))
 222                 io->flags |= ISTRUNC;
 223         if (S_ISCHR(sb.st_mode) || S_ISBLK(sb.st_mode)) {
 224                 if (ioctl(io->fd, FIODTYPE, &type) == -1) {
 225                         err(1, "%s", io->name);
 226                 } else {
 227 #ifdef __APPLE__                /* MacOSX uses enumeration for type not a bitmask */
 228                         if (type == D_TAPE)
 229                                 io->flags |= ISTAPE;
 230                         else if (type == D_DISK || type == D_TTY) {
 231 #else  /* !__APPLE__ */
 232                         if (type & D_TAPE)
 233                                 io->flags |= ISTAPE;
 234                         else if (type & (D_DISK | D_MEM)) {
 235                                 if (type & D_DISK) {
 236                                         const int one = 1;
 237
 238                                         (void)ioctl(io->fd, DIOCWLABEL, &one);
 239                                 }
 240 #endif /* __APPLE__ */
 241                                 io->flags |= ISSEEK;
 242                         }
 243 #ifdef __APPLE__
 244                         if (S_ISCHR(sb.st_mode) && (type != D_TAPE))
 245 #else  /* !__APPLE__ */
 246                         if (S_ISCHR(sb.st_mode) && (type & D_TAPE) == 0)
 247 #endif /* __APPLE__ */
 248                                 io->flags |= ISCHR;
 249                 }
 250                 return;
 251         }
 252         errno = 0;
 253         if (lseek(io->fd, (off_t)0, SEEK_CUR) == -1 && errno == ESPIPE)
 254                 io->flags |= ISPIPE;
 255         else
 256                 io->flags |= ISSEEK;
 257 }
 258
 259 static void
 260 dd_in(void)
 261 {
 262         ssize_t n;
 263
 264         for (;;) {
 265                 switch (cpy_cnt) {
 266                 case -1:                        /* count=0 was specified */
 267                         return;
 268                 case 0:
 269                         break;
 270                 default:
 271                         if (st.in_full + st.in_part >= (u_quad_t)cpy_cnt)
 272                                 return;
 273                         break;
 274                 }
 275
 276                 /*
 277                  * Zero the buffer first if sync; if doing block operations,
 278                  * use spaces.
 279                  */
 280                 if (ddflags & C_SYNC) {
 281                         if (ddflags & (C_BLOCK | C_UNBLOCK))
 282                                 memset(in.dbp, ' ', in.dbsz);
 283                         else
 284                                 memset(in.dbp, 0, in.dbsz);
 285                 }
 286
 287                 n = read(in.fd, in.dbp, in.dbsz);
 288                 if (n == 0) {
 289                         in.dbrcnt = 0;
 290                         return;
 291                 }
 292
 293                 /* Read error. */
 294                 if (n == -1) {
 295                         /*
 296                          * If noerror not specified, die.  POSIX requires that
 297                          * the warning message be followed by an I/O display.
 298                          */
 299                         if (!(ddflags & C_NOERROR))
 300                                 err(1, "%s", in.name);
 301                         warn("%s", in.name);
 302                         summary();
 303
 304                         /*
 305                          * If it's a seekable file descriptor, seek past the
 306                          * error.  If your OS doesn't do the right thing for
 307                          * raw disks this section should be modified to re-read
 308                          * in sector size chunks.
 309                          */
 310                         if (in.flags & ISSEEK &&
 311                             lseek(in.fd, (off_t)in.dbsz, SEEK_CUR))
 312                                 warn("%s", in.name);
 313
 314                         /* If sync not specified, omit block and continue. */
 315                         if (!(ddflags & C_SYNC))
 316                                 continue;
 317
 318                         /* Read errors count as full blocks. */
 319                         in.dbcnt += in.dbrcnt = in.dbsz;
 320                         ++st.in_full;
 321
 322                 /* Handle full input blocks. */
 323                 } else if ((size_t)n == in.dbsz) {
 324                         in.dbcnt += in.dbrcnt = n;
 325                         ++st.in_full;
 326
 327                 /* Handle partial input blocks. */
 328                 } else {
 329                         /* If sync, use the entire block. */
 330                         if (ddflags & C_SYNC)
 331                                 in.dbcnt += in.dbrcnt = in.dbsz;
 332                         else
 333                                 in.dbcnt += in.dbrcnt = n;
 334                         ++st.in_part;
 335                 }
 336
 337                 /*
 338                  * POSIX states that if bs is set and no other conversions
 339                  * than noerror, notrunc or sync are specified, the block
 340                  * is output without buffering as it is read.
 341                  */
 342                 if (ddflags & C_BS) {
 343                         out.dbcnt = in.dbcnt;
 344                         dd_out(1);
 345                         in.dbcnt = 0;
 346                         continue;
 347                 }
 348
 349                 if (ddflags & C_SWAB) {
 350                         if ((n = in.dbrcnt) & 1) {
 351                                 ++st.swab;
 352                                 --n;
 353                         }
 354                         swab(in.dbp, in.dbp, (size_t)n);
 355                 }
 356
 357                 in.dbp += in.dbrcnt;
 358                 (*cfunc)();
 359         }
 360 }
 361
 362 /*
 363  * Clean up any remaining I/O and flush output.  If necessary, the output file
 364  * is truncated.
 365  */
 366 static void
 367 dd_close(void)
 368 {
 369         if (cfunc == def)
 370                 def_close();
 371         else if (cfunc == block)
 372                 block_close();
 373         else if (cfunc == unblock)
 374                 unblock_close();
 375         if (ddflags & C_OSYNC && out.dbcnt && out.dbcnt < out.dbsz) {
 376                 if (ddflags & (C_BLOCK | C_UNBLOCK))
 377                         memset(out.dbp, ' ', out.dbsz - out.dbcnt);
 378                 else
 379                         memset(out.dbp, 0, out.dbsz - out.dbcnt);
 380                 out.dbcnt = out.dbsz;
 381         }
 382         if (out.dbcnt || pending)
 383                 dd_out(1);
 384 }
 385
 386 void
 387 dd_out(int force)
 388 {
 389         u_char *outp;
 390         size_t cnt, i, n;
 391         ssize_t nw;
 392         static int warned;
 393         int sparse;
 394
 395         /*
 396          * Write one or more blocks out.  The common case is writing a full
 397          * output block in a single write; increment the full block stats.
 398          * Otherwise, we're into partial block writes.  If a partial write,
 399          * and it's a character device, just warn.  If a tape device, quit.
 400          *
 401          * The partial writes represent two cases.  1: Where the input block
 402          * was less than expected so the output block was less than expected.
 403          * 2: Where the input block was the right size but we were forced to
 404          * write the block in multiple chunks.  The original versions of dd(1)
 405          * never wrote a block in more than a single write, so the latter case
 406          * never happened.
 407          *
 408          * One special case is if we're forced to do the write -- in that case
 409          * we play games with the buffer size, and it's usually a partial write.
 410          */
 411         outp = out.db;
 412         for (n = force ? out.dbcnt : out.dbsz;; n = out.dbsz) {
 413                 for (cnt = n;; cnt -= nw) {
 414                         sparse = 0;
 415                         if (ddflags & C_SPARSE) {
 416                                 sparse = 1;     /* Is buffer sparse? */
 417                                 for (i = 0; i < cnt; i++)
 418                                         if (outp[i] != 0) {
 419                                                 sparse = 0;
 420                                                 break;
 421                                         }
 422                         }
 423                         if (sparse && !force) {
 424                                 pending += cnt;
 425                                 nw = cnt;
 426                         } else {
 427                                 if (pending != 0) {
 428                                         if (force)
 429                                                 pending--;
 430                                         if (lseek(out.fd, pending, SEEK_CUR) ==
 431                                             -1)
 432                                                 err(2, "%s: seek error creating sparse file",
 433                                                     out.name);
 434                                         if (force)
 435                                                 write(out.fd, outp, 1);
 436                                         pending = 0;
 437                                 }
 438                                 if (cnt)
 439                                         nw = write(out.fd, outp, cnt);
 440                                 else
 441                                         return;
 442                         }
 443
 444                         if (nw <= 0) {
 445                                 if (nw == 0)
 446                                         errx(1, "%s: end of device", out.name);
 447                                 if (errno != EINTR)
 448                                         err(1, "%s", out.name);
 449                                 nw = 0;
 450                         }
 451                         outp += nw;
 452                         st.bytes += nw;
 453                         if ((size_t)nw == n) {
 454                                 if (n != out.dbsz)
 455                                         ++st.out_part;
 456                                 else
 457                                         ++st.out_full;
 458                                 break;
 459                         }
 460                         ++st.out_part;
 461                         if ((size_t)nw == cnt)
 462                                 break;
 463                         if (out.flags & ISTAPE)
 464                                 errx(1, "%s: short write on tape device",
 465                                     out.name);
 466                         if (out.flags & ISCHR && !warned) {
 467                                 warned = 1;
 468                                 warnx("%s: short write on character device",
 469                                     out.name);
 470                         }
 471                 }
 472                 if ((out.dbcnt -= n) < out.dbsz)
 473                         break;
 474         }
 475
 476         /* Reassemble the output block. */
 477         if (out.dbcnt)
 478                 (void)memmove(out.db, out.dbp - out.dbcnt, out.dbcnt);
 479         out.dbp = out.db + out.dbcnt;
 480 }