aboutsummaryrefslogtreecommitdiff
path: root/libjava/classpath/lib/java/util/Collections$UnmodifiableMap$UnmodifiableEntrySet$UnmodifiableMapEntry.class
diff options
context:
space:
mode:
authorAndrew Haley <aph@redhat.com>2007-02-06 17:17:03 +0000
committerAndrew Haley <aph@gcc.gnu.org>2007-02-06 17:17:03 +0000
commit44fb8659c3c8e9309dff1b58a651ab47be827871 (patch)
treed02c5fe418c1ba987c47cd6a7e80e6c30485610c /libjava/classpath/lib/java/util/Collections$UnmodifiableMap$UnmodifiableEntrySet$UnmodifiableMapEntry.class
parent15814ba0a2c2bbbd8b4617447edb6065da14cb6a (diff)
downloadgcc-44fb8659c3c8e9309dff1b58a651ab47be827871.zip
gcc-44fb8659c3c8e9309dff1b58a651ab47be827871.tar.gz
gcc-44fb8659c3c8e9309dff1b58a651ab47be827871.tar.bz2
makemake.tcl (emit_bc_rule): Set the source filename.
2007-02-06 Andrew Haley <aph@redhat.com> * scripts/makemake.tcl (emit_bc_rule): Set the source filename. * sources.am: Rebuild. From-SVN: r121651
Diffstat (limited to 'libjava/classpath/lib/java/util/Collections$UnmodifiableMap$UnmodifiableEntrySet$UnmodifiableMapEntry.class')
0 files changed, 0 insertions, 0 deletions
' href='#n180'>180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291
/* Loop unrolling and peeling.
   Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc.

This file is part of GCC.

GCC is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 2, or (at your option) any later
version.

GCC is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
for more details.

You should have received a copy of the GNU General Public License
along with GCC; see the file COPYING.  If not, write to the Free
Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA
02110-1301, USA.  */

#include "config.h"
#include "system.h"
#include "coretypes.h"
#include "tm.h"
#include "rtl.h"
#include "hard-reg-set.h"
#include "obstack.h"
#include "basic-block.h"
#include "cfgloop.h"
#include "cfglayout.h"
#include "params.h"
#include "output.h"
#include "expr.h"
#include "hashtab.h"
#include "recog.h"    

/* This pass performs loop unrolling and peeling.  We only perform these
   optimizations on innermost loops (with single exception) because
   the impact on performance is greatest here, and we want to avoid
   unnecessary code size growth.  The gain is caused by greater sequentiality
   of code, better code to optimize for further passes and in some cases
   by fewer testings of exit conditions.  The main problem is code growth,
   that impacts performance negatively due to effect of caches.

   What we do:

   -- complete peeling of once-rolling loops; this is the above mentioned
      exception, as this causes loop to be cancelled completely and
      does not cause code growth
   -- complete peeling of loops that roll (small) constant times.
   -- simple peeling of first iterations of loops that do not roll much
      (according to profile feedback)
   -- unrolling of loops that roll constant times; this is almost always
      win, as we get rid of exit condition tests.
   -- unrolling of loops that roll number of times that we can compute
      in runtime; we also get rid of exit condition tests here, but there
      is the extra expense for calculating the number of iterations
   -- simple unrolling of remaining loops; this is performed only if we
      are asked to, as the gain is questionable in this case and often
      it may even slow down the code
   For more detailed descriptions of each of those, see comments at
   appropriate function below.

   There is a lot of parameters (defined and described in params.def) that
   control how much we unroll/peel.

   ??? A great problem is that we don't have a good way how to determine
   how many times we should unroll the loop; the experiments I have made
   showed that this choice may affect performance in order of several %.
   */

/* Information about induction variables to split.  */

struct iv_to_split
{
  rtx insn;		/* The insn in that the induction variable occurs.  */
  rtx base_var;		/* The variable on that the values in the further
			   iterations are based.  */
  rtx step;		/* Step of the induction variable.  */
  unsigned n_loc;
  unsigned loc[3];	/* Location where the definition of the induction
			   variable occurs in the insn.  For example if
			   N_LOC is 2, the expression is located at
			   XEXP (XEXP (single_set, loc[0]), loc[1]).  */ 
};

/* Information about accumulators to expand.  */

struct var_to_expand
{
  rtx insn;		           /* The insn in that the variable expansion occurs.  */
  rtx reg;                         /* The accumulator which is expanded.  */
  VEC(rtx,heap) *var_expansions;   /* The copies of the accumulator which is expanded.  */ 
  enum rtx_code op;                /* The type of the accumulation - addition, subtraction 
                                      or multiplication.  */
  int expansion_count;             /* Count the number of expansions generated so far.  */
  int reuse_expansion;             /* The expansion we intend to reuse to expand
                                      the accumulator.  If REUSE_EXPANSION is 0 reuse 
                                      the original accumulator.  Else use 
                                      var_expansions[REUSE_EXPANSION - 1].  */
  unsigned accum_pos;              /* The position in which the accumulator is placed in
                                      the insn src.  For example in x = x + something
                                      accum_pos is 0 while in x = something + x accum_pos
                                      is 1.  */
};

/* Information about optimization applied in
   the unrolled loop.  */

struct opt_info
{
  htab_t insns_to_split;           /* A hashtable of insns to split.  */
  htab_t insns_with_var_to_expand; /* A hashtable of insns with accumulators
                                      to expand.  */
  unsigned first_new_block;        /* The first basic block that was
                                      duplicated.  */
  basic_block loop_exit;           /* The loop exit basic block.  */
  basic_block loop_preheader;      /* The loop preheader basic block.  */
};

static void decide_unrolling_and_peeling (int);
static void peel_loops_completely (int);
static void decide_peel_simple (struct loop *, int);
static void decide_peel_once_rolling (struct loop *, int);
static void decide_peel_completely (struct loop *, int);
static void decide_unroll_stupid (struct loop *, int);
static void decide_unroll_constant_iterations (struct loop *, int);
static void decide_unroll_runtime_iterations (struct loop *, int);
static void peel_loop_simple (struct loop *);
static void peel_loop_completely (struct loop *);
static void unroll_loop_stupid (struct loop *);
static void unroll_loop_constant_iterations (struct loop *);
static void unroll_loop_runtime_iterations (struct loop *);
static struct opt_info *analyze_insns_in_loop (struct loop *);
static void opt_info_start_duplication (struct opt_info *);
static void apply_opt_in_copies (struct opt_info *, unsigned, bool, bool);
static void free_opt_info (struct opt_info *);
static struct var_to_expand *analyze_insn_to_expand_var (struct loop*, rtx);
static bool referenced_in_one_insn_in_loop_p (struct loop *, rtx);
static struct iv_to_split *analyze_iv_to_split_insn (rtx);
static void expand_var_during_unrolling (struct var_to_expand *, rtx);
static int insert_var_expansion_initialization (void **, void *);
static int combine_var_copies_in_loop_exit (void **, void *);
static int release_var_copies (void **, void *);
static rtx get_expansion (struct var_to_expand *);

/* Unroll and/or peel (depending on FLAGS) LOOPS.  */
void
unroll_and_peel_loops (int flags)
{
  struct loop *loop;
  bool check;
  loop_iterator li;

  /* First perform complete loop peeling (it is almost surely a win,
     and affects parameters for further decision a lot).  */
  peel_loops_completely (flags);

  /* Now decide rest of unrolling and peeling.  */
  decide_unrolling_and_peeling (flags);

  /* Scan the loops, inner ones first.  */
  FOR_EACH_LOOP (li, loop, LI_FROM_INNERMOST)
    {
      check = true;
      /* And perform the appropriate transformations.  */
      switch (loop->lpt_decision.decision)
	{
	case LPT_PEEL_COMPLETELY:
	  /* Already done.  */
	  gcc_unreachable ();
	case LPT_PEEL_SIMPLE:
	  peel_loop_simple (loop);
	  break;
	case LPT_UNROLL_CONSTANT:
	  unroll_loop_constant_iterations (loop);
	  break;
	case LPT_UNROLL_RUNTIME:
	  unroll_loop_runtime_iterations (loop);
	  break;
	case LPT_UNROLL_STUPID:
	  unroll_loop_stupid (loop);
	  break;
	case LPT_NONE:
	  check = false;
	  break;
	default:
	  gcc_unreachable ();
	}
      if (check)
	{
#ifdef ENABLE_CHECKING
	  verify_dominators (CDI_DOMINATORS);
	  verify_loop_structure ();
#endif
	}
    }

  iv_analysis_done ();
}

/* Check whether exit of the LOOP is at the end of loop body.  */

static bool
loop_exit_at_end_p (struct loop *loop)
{
  struct niter_desc *desc = get_simple_loop_desc (loop);
  rtx insn;

  if (desc->in_edge->dest != loop->latch)
    return false;

  /* Check that the latch is empty.  */
  FOR_BB_INSNS (loop->latch, insn)
    {
      if (INSN_P (insn))
	return false;
    }

  return true;
}

/* Depending on FLAGS, check whether to peel loops completely and do so.  */
static void
peel_loops_completely (int flags)
{
  struct loop *loop;
  loop_iterator li;

  /* Scan the loops, the inner ones first.  */
  FOR_EACH_LOOP (li, loop, LI_FROM_INNERMOST)
    {
      loop->lpt_decision.decision = LPT_NONE;

      if (dump_file)
	fprintf (dump_file,
		 "\n;; *** Considering loop %d for complete peeling ***\n",
		 loop->num);

      loop->ninsns = num_loop_insns (loop);

      decide_peel_once_rolling (loop, flags);
      if (loop->lpt_decision.decision == LPT_NONE)
	decide_peel_completely (loop, flags);

      if (loop->lpt_decision.decision == LPT_PEEL_COMPLETELY)
	{
	  peel_loop_completely (loop);
#ifdef ENABLE_CHECKING
	  verify_dominators (CDI_DOMINATORS);
	  verify_loop_structure ();
#endif
	}
    }
}

/* Decide whether unroll or peel loops (depending on FLAGS) and how much.  */
static void
decide_unrolling_and_peeling (int flags)
{
  struct loop *loop;
  loop_iterator li;

  /* Scan the loops, inner ones first.  */
  FOR_EACH_LOOP (li, loop, LI_FROM_INNERMOST)
    {
      loop->lpt_decision.decision = LPT_NONE;

      if (dump_file)
	fprintf (dump_file, "\n;; *** Considering loop %d ***\n", loop->num);

      /* Do not peel cold areas.  */
      if (!maybe_hot_bb_p (loop->header))
	{
	  if (dump_file)
	    fprintf (dump_file, ";; Not considering loop, cold area\n");
	  continue;
	}

      /* Can the loop be manipulated?  */
      if (!can_duplicate_loop_p (loop))
	{
	  if (dump_file)
	    fprintf (dump_file,
		     ";; Not considering loop, cannot duplicate\n");
	  continue;
	}

      /* Skip non-innermost loops.  */
      if (loop->inner)
	{
	  if (dump_file)
	    fprintf (dump_file, ";; Not considering loop, is not innermost\n");
	  continue;
	}

      loop->ninsns = num_loop_insns (loop);
      loop->av_ninsns = average_num_loop_insns (loop);

      /* Try transformations one by one in decreasing order of
	 priority.  */

      decide_unroll_constant_iterations (loop, flags);
      if (loop->lpt_decision.decision == LPT_NONE)
	decide_unroll_runtime_iterations (loop, flags);
      if (loop->lpt_decision.decision == LPT_NONE)
	decide_unroll_stupid (loop, flags);
      if (loop->lpt_decision.decision == LPT_NONE)
	decide_peel_simple (loop, flags);
    }
}

/* Decide whether the LOOP is once rolling and suitable for complete
   peeling.  */
static void
decide_peel_once_rolling (struct loop *loop, int flags ATTRIBUTE_UNUSED)
{
  struct niter_desc *desc;

  if (dump_file)
    fprintf (dump_file, "\n;; Considering peeling once rolling loop\n");

  /* Is the loop small enough?  */
  if ((unsigned) PARAM_VALUE (PARAM_MAX_ONCE_PEELED_INSNS) < loop->ninsns)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check number of iterations.  */
  if (!desc->simple_p
      || desc->assumptions
      || desc->infinite
      || !desc->const_iter
      || desc->niter != 0)
    {
      if (dump_file)
	fprintf (dump_file,
		 ";; Unable to prove that the loop rolls exactly once\n");
      return;
    }

  /* Success.  */
  if (dump_file)
    fprintf (dump_file, ";; Decided to peel exactly once rolling loop\n");
  loop->lpt_decision.decision = LPT_PEEL_COMPLETELY;
}

/* Decide whether the LOOP is suitable for complete peeling.  */
static void
decide_peel_completely (struct loop *loop, int flags ATTRIBUTE_UNUSED)
{
  unsigned npeel;
  struct niter_desc *desc;

  if (dump_file)
    fprintf (dump_file, "\n;; Considering peeling completely\n");

  /* Skip non-innermost loops.  */
  if (loop->inner)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is not innermost\n");
      return;
    }

  /* Do not peel cold areas.  */
  if (!maybe_hot_bb_p (loop->header))
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, cold area\n");
      return;
    }

  /* Can the loop be manipulated?  */
  if (!can_duplicate_loop_p (loop))
    {
      if (dump_file)
	fprintf (dump_file,
		 ";; Not considering loop, cannot duplicate\n");
      return;
    }

  /* npeel = number of iterations to peel.  */
  npeel = PARAM_VALUE (PARAM_MAX_COMPLETELY_PEELED_INSNS) / loop->ninsns;
  if (npeel > (unsigned) PARAM_VALUE (PARAM_MAX_COMPLETELY_PEEL_TIMES))
    npeel = PARAM_VALUE (PARAM_MAX_COMPLETELY_PEEL_TIMES);

  /* Is the loop small enough?  */
  if (!npeel)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check number of iterations.  */
  if (!desc->simple_p
      || desc->assumptions
      || !desc->const_iter
      || desc->infinite)
    {
      if (dump_file)
	fprintf (dump_file,
		 ";; Unable to prove that the loop iterates constant times\n");
      return;
    }

  if (desc->niter > npeel - 1)
    {
      if (dump_file)
	{
	  fprintf (dump_file,
		   ";; Not peeling loop completely, rolls too much (");
	  fprintf (dump_file, HOST_WIDEST_INT_PRINT_DEC, desc->niter);
	  fprintf (dump_file, " iterations > %d [maximum peelings])\n", npeel);
	}
      return;
    }

  /* Success.  */
  if (dump_file)
    fprintf (dump_file, ";; Decided to peel loop completely\n");
  loop->lpt_decision.decision = LPT_PEEL_COMPLETELY;
}

/* Peel all iterations of LOOP, remove exit edges and cancel the loop
   completely.  The transformation done:

   for (i = 0; i < 4; i++)
     body;

   ==>

   i = 0;
   body; i++;
   body; i++;
   body; i++;
   body; i++;
   */
static void
peel_loop_completely (struct loop *loop)
{
  sbitmap wont_exit;
  unsigned HOST_WIDE_INT npeel;
  unsigned i;
  VEC (edge, heap) *remove_edges;
  edge ein;
  struct niter_desc *desc = get_simple_loop_desc (loop);
  struct opt_info *opt_info = NULL;
  
  npeel = desc->niter;

  if (npeel)
    {
      bool ok;
      
      wont_exit = sbitmap_alloc (npeel + 1);
      sbitmap_ones (wont_exit);
      RESET_BIT (wont_exit, 0);
      if (desc->noloop_assumptions)
	RESET_BIT (wont_exit, 1);

      remove_edges = NULL;

      if (flag_split_ivs_in_unroller)
        opt_info = analyze_insns_in_loop (loop);
      
      opt_info_start_duplication (opt_info);
      ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
					  npeel,
					  wont_exit, desc->out_edge,
					  &remove_edges,
					  DLTHE_FLAG_UPDATE_FREQ
					  | DLTHE_FLAG_COMPLETTE_PEEL
					  | (opt_info
					     ? DLTHE_RECORD_COPY_NUMBER : 0));
      gcc_assert (ok);

      free (wont_exit);
      
      if (opt_info)
 	{
 	  apply_opt_in_copies (opt_info, npeel, false, true);
 	  free_opt_info (opt_info);
 	}

      /* Remove the exit edges.  */
      for (i = 0; VEC_iterate (edge, remove_edges, i, ein); i++)
	remove_path (ein);
      VEC_free (edge, heap, remove_edges);
    }

  ein = desc->in_edge;
  free_simple_loop_desc (loop);

  /* Now remove the unreachable part of the last iteration and cancel
     the loop.  */
  remove_path (ein);

  if (dump_file)
    fprintf (dump_file, ";; Peeled loop completely, %d times\n", (int) npeel);
}

/* Decide whether to unroll LOOP iterating constant number of times
   and how much.  */

static void
decide_unroll_constant_iterations (struct loop *loop, int flags)
{
  unsigned nunroll, nunroll_by_av, best_copies, best_unroll = 0, n_copies, i;
  struct niter_desc *desc;

  if (!(flags & UAP_UNROLL))
    {
      /* We were not asked to, just return back silently.  */
      return;
    }

  if (dump_file)
    fprintf (dump_file,
	     "\n;; Considering unrolling loop with constant "
	     "number of iterations\n");

  /* nunroll = total number of copies of the original loop body in
     unrolled loop (i.e. if it is 2, we have to duplicate loop body once.  */
  nunroll = PARAM_VALUE (PARAM_MAX_UNROLLED_INSNS) / loop->ninsns;
  nunroll_by_av
    = PARAM_VALUE (PARAM_MAX_AVERAGE_UNROLLED_INSNS) / loop->av_ninsns;
  if (nunroll > nunroll_by_av)
    nunroll = nunroll_by_av;
  if (nunroll > (unsigned) PARAM_VALUE (PARAM_MAX_UNROLL_TIMES))
    nunroll = PARAM_VALUE (PARAM_MAX_UNROLL_TIMES);

  /* Skip big loops.  */
  if (nunroll <= 1)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check number of iterations.  */
  if (!desc->simple_p || !desc->const_iter || desc->assumptions)
    {
      if (dump_file)
	fprintf (dump_file,
		 ";; Unable to prove that the loop iterates constant times\n");
      return;
    }

  /* Check whether the loop rolls enough to consider.  */
  if (desc->niter < 2 * nunroll)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not unrolling loop, doesn't roll\n");
      return;
    }

  /* Success; now compute number of iterations to unroll.  We alter
     nunroll so that as few as possible copies of loop body are
     necessary, while still not decreasing the number of unrollings
     too much (at most by 1).  */
  best_copies = 2 * nunroll + 10;

  i = 2 * nunroll + 2;
  if (i - 1 >= desc->niter)
    i = desc->niter - 2;

  for (; i >= nunroll - 1; i--)
    {
      unsigned exit_mod = desc->niter % (i + 1);

      if (!loop_exit_at_end_p (loop))
	n_copies = exit_mod + i + 1;
      else if (exit_mod != (unsigned) i
	       || desc->noloop_assumptions != NULL_RTX)
	n_copies = exit_mod + i + 2;
      else
	n_copies = i + 1;

      if (n_copies < best_copies)
	{
	  best_copies = n_copies;
	  best_unroll = i;
	}
    }

  if (dump_file)
    fprintf (dump_file, ";; max_unroll %d (%d copies, initial %d).\n",
	     best_unroll + 1, best_copies, nunroll);

  loop->lpt_decision.decision = LPT_UNROLL_CONSTANT;
  loop->lpt_decision.times = best_unroll;
  
  if (dump_file)
    fprintf (dump_file,
	     ";; Decided to unroll the constant times rolling loop, %d times.\n",
	     loop->lpt_decision.times);
}

/* Unroll LOOP with constant number of iterations LOOP->LPT_DECISION.TIMES + 1
   times.  The transformation does this:

   for (i = 0; i < 102; i++)
     body;

   ==>

   i = 0;
   body; i++;
   body; i++;
   while (i < 102)
     {
       body; i++;
       body; i++;
       body; i++;
       body; i++;
     }
  */
static void
unroll_loop_constant_iterations (struct loop *loop)
{
  unsigned HOST_WIDE_INT niter;
  unsigned exit_mod;
  sbitmap wont_exit;
  unsigned i;
  VEC (edge, heap) *remove_edges;
  edge e;
  unsigned max_unroll = loop->lpt_decision.times;
  struct niter_desc *desc = get_simple_loop_desc (loop);
  bool exit_at_end = loop_exit_at_end_p (loop);
  struct opt_info *opt_info = NULL;
  bool ok;
  
  niter = desc->niter;

  /* Should not get here (such loop should be peeled instead).  */
  gcc_assert (niter > max_unroll + 1);

  exit_mod = niter % (max_unroll + 1);

  wont_exit = sbitmap_alloc (max_unroll + 1);
  sbitmap_ones (wont_exit);

  remove_edges = NULL;
  if (flag_split_ivs_in_unroller 
      || flag_variable_expansion_in_unroller)
    opt_info = analyze_insns_in_loop (loop);
  
  if (!exit_at_end)
    {
      /* The exit is not at the end of the loop; leave exit test
	 in the first copy, so that the loops that start with test
	 of exit condition have continuous body after unrolling.  */

      if (dump_file)
	fprintf (dump_file, ";; Condition on beginning of loop.\n");

      /* Peel exit_mod iterations.  */
      RESET_BIT (wont_exit, 0);
      if (desc->noloop_assumptions)
	RESET_BIT (wont_exit, 1);

      if (exit_mod)
	{
	  opt_info_start_duplication (opt_info);
          ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
					      exit_mod,
					      wont_exit, desc->out_edge,
					      &remove_edges,
					      DLTHE_FLAG_UPDATE_FREQ
					      | (opt_info && exit_mod > 1
						 ? DLTHE_RECORD_COPY_NUMBER
						   : 0));
	  gcc_assert (ok);

          if (opt_info && exit_mod > 1)
 	    apply_opt_in_copies (opt_info, exit_mod, false, false); 
          
	  desc->noloop_assumptions = NULL_RTX;
	  desc->niter -= exit_mod;
	  desc->niter_max -= exit_mod;
	}

      SET_BIT (wont_exit, 1);
    }
  else
    {
      /* Leave exit test in last copy, for the same reason as above if
	 the loop tests the condition at the end of loop body.  */

      if (dump_file)
	fprintf (dump_file, ";; Condition on end of loop.\n");

      /* We know that niter >= max_unroll + 2; so we do not need to care of
	 case when we would exit before reaching the loop.  So just peel
	 exit_mod + 1 iterations.  */
      if (exit_mod != max_unroll
	  || desc->noloop_assumptions)
	{
	  RESET_BIT (wont_exit, 0);
	  if (desc->noloop_assumptions)
	    RESET_BIT (wont_exit, 1);
         
          opt_info_start_duplication (opt_info);
	  ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
					      exit_mod + 1,
					      wont_exit, desc->out_edge,
					      &remove_edges,
					      DLTHE_FLAG_UPDATE_FREQ
					      | (opt_info && exit_mod > 0
						 ? DLTHE_RECORD_COPY_NUMBER
						   : 0));
	  gcc_assert (ok);
 
          if (opt_info && exit_mod > 0)
  	    apply_opt_in_copies (opt_info, exit_mod + 1, false, false);

	  desc->niter -= exit_mod + 1;
	  desc->niter_max -= exit_mod + 1;
	  desc->noloop_assumptions = NULL_RTX;

	  SET_BIT (wont_exit, 0);
	  SET_BIT (wont_exit, 1);
	}

      RESET_BIT (wont_exit, max_unroll);
    }

  /* Now unroll the loop.  */
  
  opt_info_start_duplication (opt_info);
  ok = duplicate_loop_to_header_edge (loop, loop_latch_edge (loop),
				      max_unroll,
				      wont_exit, desc->out_edge,
				      &remove_edges,
				      DLTHE_FLAG_UPDATE_FREQ
				      | (opt_info
					 ? DLTHE_RECORD_COPY_NUMBER
					   : 0));
  gcc_assert (ok);

  if (opt_info)
    {
      apply_opt_in_copies (opt_info, max_unroll, true, true);
      free_opt_info (opt_info);
    }

  free (wont_exit);

  if (exit_at_end)
    {
      basic_block exit_block = get_bb_copy (desc->in_edge->src);
      /* Find a new in and out edge; they are in the last copy we have made.  */
      
      if (EDGE_SUCC (exit_block, 0)->dest == desc->out_edge->dest)
	{
	  desc->out_edge = EDGE_SUCC (exit_block, 0);
	  desc->in_edge = EDGE_SUCC (exit_block, 1);
	}
      else
	{
	  desc->out_edge = EDGE_SUCC (exit_block, 1);
	  desc->in_edge = EDGE_SUCC (exit_block, 0);
	}
    }

  desc->niter /= max_unroll + 1;
  desc->niter_max /= max_unroll + 1;
  desc->niter_expr = GEN_INT (desc->niter);

  /* Remove the edges.  */
  for (i = 0; VEC_iterate (edge, remove_edges, i, e); i++)
    remove_path (e);
  VEC_free (edge, heap, remove_edges);

  if (dump_file)
    fprintf (dump_file,
	     ";; Unrolled loop %d times, constant # of iterations %i insns\n",
	     max_unroll, num_loop_insns (loop));
}

/* Decide whether to unroll LOOP iterating runtime computable number of times
   and how much.  */
static void
decide_unroll_runtime_iterations (struct loop *loop, int flags)
{
  unsigned nunroll, nunroll_by_av, i;
  struct niter_desc *desc;

  if (!(flags & UAP_UNROLL))
    {
      /* We were not asked to, just return back silently.  */
      return;
    }

  if (dump_file)
    fprintf (dump_file,
	     "\n;; Considering unrolling loop with runtime "
	     "computable number of iterations\n");

  /* nunroll = total number of copies of the original loop body in
     unrolled loop (i.e. if it is 2, we have to duplicate loop body once.  */
  nunroll = PARAM_VALUE (PARAM_MAX_UNROLLED_INSNS) / loop->ninsns;
  nunroll_by_av = PARAM_VALUE (PARAM_MAX_AVERAGE_UNROLLED_INSNS) / loop->av_ninsns;
  if (nunroll > nunroll_by_av)
    nunroll = nunroll_by_av;
  if (nunroll > (unsigned) PARAM_VALUE (PARAM_MAX_UNROLL_TIMES))
    nunroll = PARAM_VALUE (PARAM_MAX_UNROLL_TIMES);

  /* Skip big loops.  */
  if (nunroll <= 1)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check simpleness.  */
  if (!desc->simple_p || desc->assumptions)
    {
      if (dump_file)
	fprintf (dump_file,
		 ";; Unable to prove that the number of iterations "
		 "can be counted in runtime\n");
      return;
    }

  if (desc->const_iter)
    {
      if (dump_file)
	fprintf (dump_file, ";; Loop iterates constant times\n");
      return;
    }

  /* If we have profile feedback, check whether the loop rolls.  */
  if (loop->header->count && expected_loop_iterations (loop) < 2 * nunroll)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not unrolling loop, doesn't roll\n");
      return;
    }

  /* Success; now force nunroll to be power of 2, as we are unable to
     cope with overflows in computation of number of iterations.  */
  for (i = 1; 2 * i <= nunroll; i *= 2)
    continue;

  loop->lpt_decision.decision = LPT_UNROLL_RUNTIME;
  loop->lpt_decision.times = i - 1;
  
  if (dump_file)
    fprintf (dump_file,
	     ";; Decided to unroll the runtime computable "
	     "times rolling loop, %d times.\n",
	     loop->lpt_decision.times);
}

/* Splits edge E and inserts the sequence of instructions INSNS on it, and
   returns the newly created block.  If INSNS is NULL_RTX, nothing is changed
   and NULL is returned instead.  */

basic_block
split_edge_and_insert (edge e, rtx insns)
{
  basic_block bb;

  if (!insns)
    return NULL;
  bb = split_edge (e); 
  emit_insn_after (insns, BB_END (bb));

  /* ??? We used to assume that INSNS can contain control flow insns, and
     that we had to try to find sub basic blocks in BB to maintain a valid
     CFG.  For this purpose we used to set the BB_SUPERBLOCK flag on BB
     and call break_superblocks when going out of cfglayout mode.  But it
     turns out that this never happens; and that if it does ever happen,
     the verify_flow_info call in loop_optimizer_finalize would fail.

     There are two reasons why we expected we could have control flow insns
     in INSNS.  The first is when a comparison has to be done in parts, and
     the second is when the number of iterations is computed for loops with
     the number of iterations known at runtime.  In both cases, test cases
     to get control flow in INSNS appear to be impossible to construct:

      * If do_compare_rtx_and_jump needs several branches to do comparison
	in a mode that needs comparison by parts, we cannot analyze the
	number of iterations of the loop, and we never get to unrolling it.

      * The code in expand_divmod that was suspected to cause creation of
	branching code seems to be only accessed for signed division.  The
	divisions used by # of iterations analysis are always unsigned.
	Problems might arise on architectures that emits branching code
	for some operations that may appear in the unroller (especially
	for division), but we have no such architectures.

     Considering all this, it was decided that we should for now assume
     that INSNS can in theory contain control flow insns, but in practice
     it never does.  So we don't handle the theoretical case, and should
     a real failure ever show up, we have a pretty good clue for how to
     fix it.  */

  return bb;
}

/* Unroll LOOP for that we are able to count number of iterations in runtime
   LOOP->LPT_DECISION.TIMES + 1 times.  The transformation does this (with some
   extra care for case n < 0):

   for (i = 0; i < n; i++)
     body;

   ==>

   i = 0;
   mod = n % 4;

   switch (mod)
     {
       case 3:
         body; i++;
       case 2:
         body; i++;
       case 1:
         body; i++;
       case 0: ;
     }

   while (i < n)
     {
       body; i++;
       body; i++;
       body; i++;
       body; i++;
     }
   */
static void
unroll_loop_runtime_iterations (struct loop *loop)
{
  rtx old_niter, niter, init_code, branch_code, tmp;
  unsigned i, j, p;
  basic_block preheader, *body, swtch, ezc_swtch;
  VEC (basic_block, heap) *dom_bbs;
  sbitmap wont_exit;
  int may_exit_copy;
  unsigned n_peel;
  VEC (edge, heap) *remove_edges;
  edge e;
  bool extra_zero_check, last_may_exit;
  unsigned max_unroll = loop->lpt_decision.times;
  struct niter_desc *desc = get_simple_loop_desc (loop);
  bool exit_at_end = loop_exit_at_end_p (loop);
  struct opt_info *opt_info = NULL;
  bool ok;
  
  if (flag_split_ivs_in_unroller
      || flag_variable_expansion_in_unroller)
    opt_info = analyze_insns_in_loop (loop);
  
  /* Remember blocks whose dominators will have to be updated.  */
  dom_bbs = NULL;

  body = get_loop_body (loop);
  for (i = 0; i < loop->num_nodes; i++)
    {
      VEC (basic_block, heap) *ldom;
      basic_block bb;

      ldom = get_dominated_by (CDI_DOMINATORS, body[i]);
      for (j = 0; VEC_iterate (basic_block, ldom, j, bb); j++)
	if (!flow_bb_inside_loop_p (loop, bb))
	  VEC_safe_push (basic_block, heap, dom_bbs, bb);

      VEC_free (basic_block, heap, ldom);
    }
  free (body);

  if (!exit_at_end)
    {
      /* Leave exit in first copy (for explanation why see comment in
	 unroll_loop_constant_iterations).  */
      may_exit_copy = 0;
      n_peel = max_unroll - 1;
      extra_zero_check = true;
      last_may_exit = false;
    }
  else
    {
      /* Leave exit in last copy (for explanation why see comment in
	 unroll_loop_constant_iterations).  */
      may_exit_copy = max_unroll;
      n_peel = max_unroll;
      extra_zero_check = false;
      last_may_exit = true;
    }

  /* Get expression for number of iterations.  */
  start_sequence ();
  old_niter = niter = gen_reg_rtx (desc->mode);
  tmp = force_operand (copy_rtx (desc->niter_expr), niter);
  if (tmp != niter)
    emit_move_insn (niter, tmp);

  /* Count modulo by ANDing it with max_unroll; we use the fact that
     the number of unrollings is a power of two, and thus this is correct
     even if there is overflow in the computation.  */
  niter = expand_simple_binop (desc->mode, AND,
			       niter,
			       GEN_INT (max_unroll),
			       NULL_RTX, 0, OPTAB_LIB_WIDEN);

  init_code = get_insns ();
  end_sequence ();

  /* Precondition the loop.  */
  split_edge_and_insert (loop_preheader_edge (loop), init_code);

  remove_edges = NULL;

  wont_exit = sbitmap_alloc (max_unroll + 2);

  /* Peel the first copy of loop body (almost always we must leave exit test
     here; the only exception is when we have extra zero check and the number
     of iterations is reliable.  Also record the place of (possible) extra
     zero check.  */
  sbitmap_zero (wont_exit);
  if (extra_zero_check
      && !desc->noloop_assumptions)
    SET_BIT (wont_exit, 1);
  ezc_swtch = loop_preheader_edge (loop)->src;
  ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
				      1, wont_exit, desc->out_edge,
				      &remove_edges,
				      DLTHE_FLAG_UPDATE_FREQ);
  gcc_assert (ok);

  /* Record the place where switch will be built for preconditioning.  */
  swtch = split_edge (loop_preheader_edge (loop));

  for (i = 0; i < n_peel; i++)
    {
      /* Peel the copy.  */
      sbitmap_zero (wont_exit);
      if (i != n_peel - 1 || !last_may_exit)
	SET_BIT (wont_exit, 1);
      ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
					  1, wont_exit, desc->out_edge,
					  &remove_edges,
					  DLTHE_FLAG_UPDATE_FREQ);
      gcc_assert (ok);

      /* Create item for switch.  */
      j = n_peel - i - (extra_zero_check ? 0 : 1);
      p = REG_BR_PROB_BASE / (i + 2);

      preheader = split_edge (loop_preheader_edge (loop));
      branch_code = compare_and_jump_seq (copy_rtx (niter), GEN_INT (j), EQ,
					  block_label (preheader), p,
					  NULL_RTX);

      /* We rely on the fact that the compare and jump cannot be optimized out,
	 and hence the cfg we create is correct.  */
      gcc_assert (branch_code != NULL_RTX);

      swtch = split_edge_and_insert (single_pred_edge (swtch), branch_code);
      set_immediate_dominator (CDI_DOMINATORS, preheader, swtch);
      single_pred_edge (swtch)->probability = REG_BR_PROB_BASE - p;
      e = make_edge (swtch, preheader,
		     single_succ_edge (swtch)->flags & EDGE_IRREDUCIBLE_LOOP);
      e->probability = p;
    }

  if (extra_zero_check)
    {
      /* Add branch for zero iterations.  */
      p = REG_BR_PROB_BASE / (max_unroll + 1);
      swtch = ezc_swtch;
      preheader = split_edge (loop_preheader_edge (loop));
      branch_code = compare_and_jump_seq (copy_rtx (niter), const0_rtx, EQ,
					  block_label (preheader), p,
					  NULL_RTX);
      gcc_assert (branch_code != NULL_RTX);

      swtch = split_edge_and_insert (single_succ_edge (swtch), branch_code);
      set_immediate_dominator (CDI_DOMINATORS, preheader, swtch);
      single_succ_edge (swtch)->probability = REG_BR_PROB_BASE - p;
      e = make_edge (swtch, preheader,
		     single_succ_edge (swtch)->flags & EDGE_IRREDUCIBLE_LOOP);
      e->probability = p;
    }

  /* Recount dominators for outer blocks.  */
  iterate_fix_dominators (CDI_DOMINATORS, dom_bbs, false);

  /* And unroll loop.  */

  sbitmap_ones (wont_exit);
  RESET_BIT (wont_exit, may_exit_copy);
  opt_info_start_duplication (opt_info);
  
  ok = duplicate_loop_to_header_edge (loop, loop_latch_edge (loop),
				      max_unroll,
				      wont_exit, desc->out_edge,
				      &remove_edges,
				      DLTHE_FLAG_UPDATE_FREQ
				      | (opt_info
					 ? DLTHE_RECORD_COPY_NUMBER
					   : 0));
  gcc_assert (ok);
  
  if (opt_info)
    {
      apply_opt_in_copies (opt_info, max_unroll, true, true);
      free_opt_info (opt_info);
    }

  free (wont_exit);

  if (exit_at_end)
    {
      basic_block exit_block = get_bb_copy (desc->in_edge->src);
      /* Find a new in and out edge; they are in the last copy we have
	 made.  */
      
      if (EDGE_SUCC (exit_block, 0)->dest == desc->out_edge->dest)
	{
	  desc->out_edge = EDGE_SUCC (exit_block, 0);
	  desc->in_edge = EDGE_SUCC (exit_block, 1);
	}
      else
	{
	  desc->out_edge = EDGE_SUCC (exit_block, 1);
	  desc->in_edge = EDGE_SUCC (exit_block, 0);
	}
    }

  /* Remove the edges.  */
  for (i = 0; VEC_iterate (edge, remove_edges, i, e); i++)
    remove_path (e);
  VEC_free (edge, heap, remove_edges);

  /* We must be careful when updating the number of iterations due to
     preconditioning and the fact that the value must be valid at entry
     of the loop.  After passing through the above code, we see that
     the correct new number of iterations is this:  */
  gcc_assert (!desc->const_iter);
  desc->niter_expr =
    simplify_gen_binary (UDIV, desc->mode, old_niter,
			 GEN_INT (max_unroll + 1));
  desc->niter_max /= max_unroll + 1;
  if (exit_at_end)
    {
      desc->niter_expr =
	simplify_gen_binary (MINUS, desc->mode, desc->niter_expr, const1_rtx);
      desc->noloop_assumptions = NULL_RTX;
      desc->niter_max--;
    }

  if (dump_file)
    fprintf (dump_file,
	     ";; Unrolled loop %d times, counting # of iterations "
	     "in runtime, %i insns\n",
	     max_unroll, num_loop_insns (loop));

  VEC_free (basic_block, heap, dom_bbs);
}

/* Decide whether to simply peel LOOP and how much.  */
static void
decide_peel_simple (struct loop *loop, int flags)
{
  unsigned npeel;
  struct niter_desc *desc;

  if (!(flags & UAP_PEEL))
    {
      /* We were not asked to, just return back silently.  */
      return;
    }

  if (dump_file)
    fprintf (dump_file, "\n;; Considering simply peeling loop\n");

  /* npeel = number of iterations to peel.  */
  npeel = PARAM_VALUE (PARAM_MAX_PEELED_INSNS) / loop->ninsns;
  if (npeel > (unsigned) PARAM_VALUE (PARAM_MAX_PEEL_TIMES))
    npeel = PARAM_VALUE (PARAM_MAX_PEEL_TIMES);

  /* Skip big loops.  */
  if (!npeel)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check number of iterations.  */
  if (desc->simple_p && !desc->assumptions && desc->const_iter)
    {
      if (dump_file)
	fprintf (dump_file, ";; Loop iterates constant times\n");
      return;
    }

  /* Do not simply peel loops with branches inside -- it increases number
     of mispredicts.  */
  if (num_loop_branches (loop) > 1)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not peeling, contains branches\n");
      return;
    }

  if (loop->header->count)
    {
      unsigned niter = expected_loop_iterations (loop);
      if (niter + 1 > npeel)
	{
	  if (dump_file)
	    {
	      fprintf (dump_file, ";; Not peeling loop, rolls too much (");
	      fprintf (dump_file, HOST_WIDEST_INT_PRINT_DEC,
		       (HOST_WIDEST_INT) (niter + 1));
	      fprintf (dump_file, " iterations > %d [maximum peelings])\n",
		       npeel);
	    }
	  return;
	}
      npeel = niter + 1;
    }
  else
    {
      /* For now we have no good heuristics to decide whether loop peeling
         will be effective, so disable it.  */
      if (dump_file)
	fprintf (dump_file,
		 ";; Not peeling loop, no evidence it will be profitable\n");
      return;
    }

  /* Success.  */
  loop->lpt_decision.decision = LPT_PEEL_SIMPLE;
  loop->lpt_decision.times = npeel;
      
  if (dump_file)
    fprintf (dump_file, ";; Decided to simply peel the loop, %d times.\n",
	     loop->lpt_decision.times);
}

/* Peel a LOOP LOOP->LPT_DECISION.TIMES times.  The transformation:
   while (cond)
     body;

   ==>

   if (!cond) goto end;
   body;
   if (!cond) goto end;
   body;
   while (cond)
     body;
   end: ;
   */
static void
peel_loop_simple (struct loop *loop)
{
  sbitmap wont_exit;
  unsigned npeel = loop->lpt_decision.times;
  struct niter_desc *desc = get_simple_loop_desc (loop);
  struct opt_info *opt_info = NULL;
  bool ok;
  
  if (flag_split_ivs_in_unroller && npeel > 1)
    opt_info = analyze_insns_in_loop (loop);
  
  wont_exit = sbitmap_alloc (npeel + 1);
  sbitmap_zero (wont_exit);
  
  opt_info_start_duplication (opt_info);
  
  ok = duplicate_loop_to_header_edge (loop, loop_preheader_edge (loop),
				      npeel, wont_exit, NULL,
				      NULL, DLTHE_FLAG_UPDATE_FREQ
				      | (opt_info
					 ? DLTHE_RECORD_COPY_NUMBER
					   : 0));
  gcc_assert (ok);

  free (wont_exit);
  
  if (opt_info)
    {
      apply_opt_in_copies (opt_info, npeel, false, false);
      free_opt_info (opt_info);
    }

  if (desc->simple_p)
    {
      if (desc->const_iter)
	{
	  desc->niter -= npeel;
	  desc->niter_expr = GEN_INT (desc->niter);
	  desc->noloop_assumptions = NULL_RTX;
	}
      else
	{
	  /* We cannot just update niter_expr, as its value might be clobbered
	     inside loop.  We could handle this by counting the number into
	     temporary just like we do in runtime unrolling, but it does not
	     seem worthwhile.  */
	  free_simple_loop_desc (loop);
	}
    }
  if (dump_file)
    fprintf (dump_file, ";; Peeling loop %d times\n", npeel);
}

/* Decide whether to unroll LOOP stupidly and how much.  */
static void
decide_unroll_stupid (struct loop *loop, int flags)
{
  unsigned nunroll, nunroll_by_av, i;
  struct niter_desc *desc;

  if (!(flags & UAP_UNROLL_ALL))
    {
      /* We were not asked to, just return back silently.  */
      return;
    }

  if (dump_file)
    fprintf (dump_file, "\n;; Considering unrolling loop stupidly\n");

  /* nunroll = total number of copies of the original loop body in
     unrolled loop (i.e. if it is 2, we have to duplicate loop body once.  */
  nunroll = PARAM_VALUE (PARAM_MAX_UNROLLED_INSNS) / loop->ninsns;
  nunroll_by_av
    = PARAM_VALUE (PARAM_MAX_AVERAGE_UNROLLED_INSNS) / loop->av_ninsns;
  if (nunroll > nunroll_by_av)
    nunroll = nunroll_by_av;
  if (nunroll > (unsigned) PARAM_VALUE (PARAM_MAX_UNROLL_TIMES))
    nunroll = PARAM_VALUE (PARAM_MAX_UNROLL_TIMES);

  /* Skip big loops.  */
  if (nunroll <= 1)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not considering loop, is too big\n");
      return;
    }

  /* Check for simple loops.  */
  desc = get_simple_loop_desc (loop);

  /* Check simpleness.  */
  if (desc->simple_p && !desc->assumptions)
    {
      if (dump_file)
	fprintf (dump_file, ";; The loop is simple\n");
      return;
    }

  /* Do not unroll loops with branches inside -- it increases number
     of mispredicts.  */
  if (num_loop_branches (loop) > 1)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not unrolling, contains branches\n");
      return;
    }

  /* If we have profile feedback, check whether the loop rolls.  */
  if (loop->header->count
      && expected_loop_iterations (loop) < 2 * nunroll)
    {
      if (dump_file)
	fprintf (dump_file, ";; Not unrolling loop, doesn't roll\n");
      return;
    }

  /* Success.  Now force nunroll to be power of 2, as it seems that this
     improves results (partially because of better alignments, partially
     because of some dark magic).  */
  for (i = 1; 2 * i <= nunroll; i *= 2)
    continue;

  loop->lpt_decision.decision = LPT_UNROLL_STUPID;
  loop->lpt_decision.times = i - 1;
      
  if (dump_file)
    fprintf (dump_file,
	     ";; Decided to unroll the loop stupidly, %d times.\n",
	     loop->lpt_decision.times);
}

/* Unroll a LOOP LOOP->LPT_DECISION.TIMES times.  The transformation:
   while (cond)
     body;

   ==>

   while (cond)
     {
       body;
       if (!cond) break;
       body;
       if (!cond) break;
       body;
       if (!cond) break;
       body;
     }
   */
static void
unroll_loop_stupid (struct loop *loop)
{
  sbitmap wont_exit;
  unsigned nunroll = loop->lpt_decision.times;
  struct niter_desc *desc = get_simple_loop_desc (loop);
  struct opt_info *opt_info = NULL;
  bool ok;
  
  if (flag_split_ivs_in_unroller
      || flag_variable_expansion_in_unroller)
    opt_info = analyze_insns_in_loop (loop);
  
  
  wont_exit = sbitmap_alloc (nunroll + 1);
  sbitmap_zero (wont_exit);
  opt_info_start_duplication (opt_info);
  
  ok = duplicate_loop_to_header_edge (loop, loop_latch_edge (loop),
				      nunroll, wont_exit,
				      NULL, NULL,
				      DLTHE_FLAG_UPDATE_FREQ
				      | (opt_info
					 ? DLTHE_RECORD_COPY_NUMBER
					   : 0));
  gcc_assert (ok);
  
  if (opt_info)
    {
      apply_opt_in_copies (opt_info, nunroll, true, true);
      free_opt_info (opt_info);
    }

  free (wont_exit);

  if (desc->simple_p)
    {
      /* We indeed may get here provided that there are nontrivial assumptions
	 for a loop to be really simple.  We could update the counts, but the
	 problem is that we are unable to decide which exit will be taken
	 (not really true in case the number of iterations is constant,
	 but noone will do anything with this information, so we do not
	 worry about it).  */
      desc->simple_p = false;
    }

  if (dump_file)
    fprintf (dump_file, ";; Unrolled loop %d times, %i insns\n",
	     nunroll, num_loop_insns (loop));
}

/* A hash function for information about insns to split.  */

static hashval_t
si_info_hash (const void *ivts)
{
  return (hashval_t) INSN_UID (((struct iv_to_split *) ivts)->insn);
}

/* An equality functions for information about insns to split.  */

static int
si_info_eq (const void *ivts1, const void *ivts2)
{
  const struct iv_to_split *i1 = ivts1;
  const struct iv_to_split *i2 = ivts2;

  return i1->insn == i2->insn;
}

/* Return a hash for VES, which is really a "var_to_expand *".  */

static hashval_t
ve_info_hash (const void *ves)
{
  return (hashval_t) INSN_UID (((struct var_to_expand *) ves)->insn);
}

/* Return true if IVTS1 and IVTS2 (which are really both of type 
   "var_to_expand *") refer to the same instruction.  */

static int
ve_info_eq (const void *ivts1, const void *ivts2)
{
  const struct var_to_expand *i1 = ivts1;
  const struct var_to_expand *i2 = ivts2;
  
  return i1->insn == i2->insn;
}

/* Returns true if REG is referenced in one insn in LOOP.  */

bool
referenced_in_one_insn_in_loop_p (struct loop *loop, rtx reg)
{
  basic_block *body, bb;
  unsigned i;
  int count_ref = 0;
  rtx insn;
  
  body = get_loop_body (loop); 
  for (i = 0; i < loop->num_nodes; i++)
    {
      bb = body[i];
      
      FOR_BB_INSNS (bb, insn)
      {
        if (rtx_referenced_p (reg, insn))
          count_ref++;
      }
    }
  return (count_ref  == 1);
}

/* Determine whether INSN contains an accumulator
   which can be expanded into separate copies, 
   one for each copy of the LOOP body.
   
   for (i = 0 ; i < n; i++)
     sum += a[i];
   
   ==>
     
   sum += a[i]
   ....
   i = i+1;
   sum1 += a[i]
   ....
   i = i+1
   sum2 += a[i];
   ....

   Return NULL if INSN contains no opportunity for expansion of accumulator.  
   Otherwise, allocate a VAR_TO_EXPAND structure, fill it with the relevant 
   information and return a pointer to it.
*/

static struct var_to_expand *
analyze_insn_to_expand_var (struct loop *loop, rtx insn)
{
  rtx set, dest, src, op1, op2, something;
  struct var_to_expand *ves;
  enum machine_mode mode1, mode2;
  unsigned accum_pos;

  set = single_set (insn);
  if (!set)
    return NULL;
  
  dest = SET_DEST (set);
  src = SET_SRC (set);
  
  if (GET_CODE (src) != PLUS
      && GET_CODE (src) != MINUS
      && GET_CODE (src) != MULT)
    return NULL;

  /* Hmm, this is a bit paradoxical.  We know that INSN is a valid insn
     in MD.  But if there is no optab to generate the insn, we can not
     perform the variable expansion.  This can happen if an MD provides
     an insn but not a named pattern to generate it, for example to avoid
     producing code that needs additional mode switches like for x87/mmx.

     So we check have_insn_for which looks for an optab for the operation
     in SRC.  If it doesn't exist, we can't perform the expansion even
     though INSN is valid.  */
  if (!have_insn_for (GET_CODE (src), GET_MODE (src)))
    return NULL;

  op1 = XEXP (src, 0);
  op2 = XEXP (src, 1);
  
  if (!REG_P (dest)
      && !(GET_CODE (dest) == SUBREG
           && REG_P (SUBREG_REG (dest))))
    return NULL;
  
  if (rtx_equal_p (dest, op1))
    accum_pos = 0;
  else if (rtx_equal_p (dest, op2))
    accum_pos = 1;
  else
    return NULL;

  /* The method of expansion that we are using; which includes
     the initialization of the expansions with zero and the summation of
     the expansions at the end of the computation will yield wrong results
     for (x = something - x) thus avoid using it in that case.  */
  if (accum_pos == 1  
    && GET_CODE (src) == MINUS)
   return NULL;

  something = (accum_pos == 0)? op2 : op1;

  if (!referenced_in_one_insn_in_loop_p (loop, dest))
    return NULL;
  
  if (rtx_referenced_p (dest, something))
    return NULL;
  
  mode1 = GET_MODE (dest); 
  mode2 = GET_MODE (something);
  if ((FLOAT_MODE_P (mode1) 
       || FLOAT_MODE_P (mode2)) 
      && !flag_unsafe_math_optimizations) 
    return NULL;

  if (dump_file)
  {
    fprintf (dump_file,
    "\n;; Expanding Accumulator ");
    print_rtl (dump_file, dest);
    fprintf (dump_file, "\n");
  }

  /* Record the accumulator to expand.  */
  ves = XNEW (struct var_to_expand);
  ves->insn = insn;
  ves->var_expansions = VEC_alloc (rtx, heap, 1);
  ves->reg = copy_rtx (dest);
  ves->op = GET_CODE (src);
  ves->expansion_count = 0;
  ves->reuse_expansion = 0;
  ves->accum_pos = accum_pos;
  return ves; 
}

/* Determine whether there is an induction variable in INSN that
   we would like to split during unrolling.  

   I.e. replace

   i = i + 1;
   ...
   i = i + 1;
   ...
   i = i + 1;
   ...

   type chains by

   i0 = i + 1
   ...
   i = i0 + 1
   ...
   i = i0 + 2
   ...

   Return NULL if INSN contains no interesting IVs.  Otherwise, allocate 
   an IV_TO_SPLIT structure, fill it with the relevant information and return a
   pointer to it.  */

static struct iv_to_split *
analyze_iv_to_split_insn (rtx insn)
{
  rtx set, dest;
  struct rtx_iv iv;
  struct iv_to_split *ivts;
  bool ok;

  /* For now we just split the basic induction variables.  Later this may be
     extended for example by selecting also addresses of memory references.  */
  set = single_set (insn);
  if (!set)
    return NULL;

  dest = SET_DEST (set);
  if (!REG_P (dest))
    return NULL;

  if (!biv_p (insn, dest))
    return NULL;

  ok = iv_analyze_result (insn, dest, &iv);

  /* This used to be an assert under the assumption that if biv_p returns
     true that iv_analyze_result must also return true.  However, that
     assumption is not strictly correct as evidenced by pr25569.

     Returning NULL when iv_analyze_result returns false is safe and
     avoids the problems in pr25569 until the iv_analyze_* routines
     can be fixed, which is apparently hard and time consuming
     according to their author.  */
  if (! ok)
    return NULL;

  if (iv.step == const0_rtx
      || iv.mode != iv.extend_mode)
    return NULL;

  /* Record the insn to split.  */
  ivts = XNEW (struct iv_to_split);
  ivts->insn = insn;
  ivts->base_var = NULL_RTX;
  ivts->step = iv.step;
  ivts->n_loc = 1;
  ivts->loc[0] = 1;
  
  return ivts;
}

/* Determines which of insns in LOOP can be optimized.
   Return a OPT_INFO struct with the relevant hash tables filled
   with all insns to be optimized.  The FIRST_NEW_BLOCK field
   is undefined for the return value.  */

static struct opt_info *
analyze_insns_in_loop (struct loop *loop)
{
  basic_block *body, bb;
  unsigned i;
  struct opt_info *opt_info = XCNEW (struct opt_info);
  rtx insn;
  struct iv_to_split *ivts = NULL;
  struct var_to_expand *ves = NULL;
  PTR *slot1;
  PTR *slot2;
  VEC (edge, heap) *edges = get_loop_exit_edges (loop);
  edge exit;
  bool can_apply = false;
  
  iv_analysis_loop_init (loop);

  body = get_loop_body (loop);

  if (flag_split_ivs_in_unroller)
    opt_info->insns_to_split = htab_create (5 * loop->num_nodes,
                                            si_info_hash, si_info_eq, free);
  
  /* Record the loop exit bb and loop preheader before the unrolling.  */
  opt_info->loop_preheader = loop_preheader_edge (loop)->src;
  
  if (VEC_length (edge, edges) == 1)
    {
      exit = VEC_index (edge, edges, 0);
      if (!(exit->flags & EDGE_COMPLEX))
	{
	  opt_info->loop_exit = split_edge (exit);
	  can_apply = true;
	}
    }
  
  if (flag_variable_expansion_in_unroller
      && can_apply)
    opt_info->insns_with_var_to_expand = htab_create (5 * loop->num_nodes,
						      ve_info_hash, ve_info_eq, free);
  
  for (i = 0; i < loop->num_nodes; i++)
    {
      bb = body[i];
      if (!dominated_by_p (CDI_DOMINATORS, loop->latch, bb))
	continue;

      FOR_BB_INSNS (bb, insn)
      {
        if (!INSN_P (insn))
          continue;
        
        if (opt_info->insns_to_split)
          ivts = analyze_iv_to_split_insn (insn);
        
        if (ivts)
          {
            slot1 = htab_find_slot (opt_info->insns_to_split, ivts, INSERT);
            *slot1 = ivts;
            continue;
          }
        
        if (opt_info->insns_with_var_to_expand)
          ves = analyze_insn_to_expand_var (loop, insn);
        
        if (ves)
          {
            slot2 = htab_find_slot (opt_info->insns_with_var_to_expand, ves, INSERT);
            *slot2 = ves;
          }
      }
    }
  
  VEC_free (edge, heap, edges);
  free (body);
  return opt_info;
}

/* Called just before loop duplication.  Records start of duplicated area
   to OPT_INFO.  */

static void 
opt_info_start_duplication (struct opt_info *opt_info)
{
  if (opt_info)
    opt_info->first_new_block = last_basic_block;
}

/* Determine the number of iterations between initialization of the base
   variable and the current copy (N_COPY).  N_COPIES is the total number
   of newly created copies.  UNROLLING is true if we are unrolling
   (not peeling) the loop.  */

static unsigned
determine_split_iv_delta (unsigned n_copy, unsigned n_copies, bool unrolling)
{
  if (unrolling)
    {
      /* If we are unrolling, initialization is done in the original loop
	 body (number 0).  */
      return n_copy;
    }
  else
    {
      /* If we are peeling, the copy in that the initialization occurs has
	 number 1.  The original loop (number 0) is the last.  */
      if (n_copy)
	return n_copy - 1;
      else
	return n_copies;
    }
}

/* Locate in EXPR the expression corresponding to the location recorded
   in IVTS, and return a pointer to the RTX for this location.  */

static rtx *
get_ivts_expr (rtx expr, struct iv_to_split *ivts)
{
  unsigned i;
  rtx *ret = &expr;

  for (i = 0; i < ivts->n_loc; i++)
    ret = &XEXP (*ret, ivts->loc[i]);

  return ret;
}

/* Allocate basic variable for the induction variable chain.  Callback for
   htab_traverse.  */

static int
allocate_basic_variable (void **slot, void *data ATTRIBUTE_UNUSED)
{
  struct iv_to_split *ivts = *slot;
  rtx expr = *get_ivts_expr (single_set (ivts->insn), ivts);

  ivts->base_var = gen_reg_rtx (GET_MODE (expr));

  return 1;
}

/* Insert initialization of basic variable of IVTS before INSN, taking
   the initial value from INSN.  */

static void
insert_base_initialization (struct iv_to_split *ivts, rtx insn)
{
  rtx expr = copy_rtx (*get_ivts_expr (single_set (insn), ivts));
  rtx seq;

  start_sequence ();
  expr = force_operand (expr, ivts->base_var);
  if (expr != ivts->base_var)
    emit_move_insn (ivts->base_var, expr);
  seq = get_insns ();
  end_sequence ();

  emit_insn_before (seq, insn);
}

/* Replace the use of induction variable described in IVTS in INSN
   by base variable + DELTA * step.  */

static void
split_iv (struct iv_to_split *ivts, rtx insn, unsigned delta)
{
  rtx expr, *loc, seq, incr, var;
  enum machine_mode mode = GET_MODE (ivts->base_var);
  rtx src, dest, set;

  /* Construct base + DELTA * step.  */
  if (!delta)
    expr = ivts->base_var;
  else
    {
      incr = simplify_gen_binary (MULT, mode,
				  ivts->step, gen_int_mode (delta, mode));
      expr = simplify_gen_binary (PLUS, GET_MODE (ivts->base_var),
				  ivts->base_var, incr);
    }

  /* Figure out where to do the replacement.  */
  loc = get_ivts_expr (single_set (insn), ivts);

  /* If we can make the replacement right away, we're done.  */
  if (validate_change (insn, loc, expr, 0))
    return;

  /* Otherwise, force EXPR into a register and try again.  */
  start_sequence ();
  var = gen_reg_rtx (mode);
  expr = force_operand (expr, var);
  if (expr != var)
    emit_move_insn (var, expr);
  seq = get_insns ();
  end_sequence ();
  emit_insn_before (seq, insn);
      
  if (validate_change (insn, loc, var, 0))
    return;

  /* The last chance.  Try recreating the assignment in insn
     completely from scratch.  */
  set = single_set (insn);
  gcc_assert (set);

  start_sequence ();
  *loc = var;
  src = copy_rtx (SET_SRC (set));
  dest = copy_rtx (SET_DEST (set));
  src = force_operand (src, dest);
  if (src != dest)
    emit_move_insn (dest, src);
  seq = get_insns ();
  end_sequence ();
     
  emit_insn_before (seq, insn);
  delete_insn (insn);
}


/* Return one expansion of the accumulator recorded in struct VE.  */

static rtx
get_expansion (struct var_to_expand *ve)
{
  rtx reg;
  
  if (ve->reuse_expansion == 0)
    reg = ve->reg;
  else
    reg = VEC_index (rtx, ve->var_expansions, ve->reuse_expansion - 1);
  
  if (VEC_length (rtx, ve->var_expansions) == (unsigned) ve->reuse_expansion)
    ve->reuse_expansion = 0;
  else 
    ve->reuse_expansion++;
  
  return reg;
}


/* Given INSN replace the uses of the accumulator recorded in VE 
   with a new register.  */

static void
expand_var_during_unrolling (struct var_to_expand *ve, rtx insn)
{
  rtx new_reg, set;
  bool really_new_expansion = false;
  
  set = single_set (insn);
  gcc_assert (set);
  
  /* Generate a new register only if the expansion limit has not been
     reached.  Else reuse an already existing expansion.  */
  if (PARAM_VALUE (PARAM_MAX_VARIABLE_EXPANSIONS) > ve->expansion_count)
    {
      really_new_expansion = true;
      new_reg = gen_reg_rtx (GET_MODE (ve->reg));
    }
  else
    new_reg = get_expansion (ve);

  validate_change (insn, &SET_DEST (set), new_reg, 1);
  validate_change (insn, &XEXP (SET_SRC (set), ve->accum_pos), new_reg, 1);
  
  if (apply_change_group ())
    if (really_new_expansion)
      {
        VEC_safe_push (rtx, heap, ve->var_expansions, new_reg);
        ve->expansion_count++;
      }
}

/* Initialize the variable expansions in loop preheader.  
   Callbacks for htab_traverse.  PLACE_P is the loop-preheader 
   basic block where the initialization of the expansions 
   should take place.  The expansions are initialized with (-0)
   when the operation is plus or minus to honor sign zero.
   This way we can prevent cases where the sign of the final result is
   effected by the sign of the expansion.
   Here is an example to demonstrate this:
   
   for (i = 0 ; i < n; i++)
     sum += something;

   ==>

   sum += something
   ....
   i = i+1;
   sum1 += something
   ....
   i = i+1
   sum2 += something;
   ....
   
   When SUM is initialized with -zero and SOMETHING is also -zero; the
   final result of sum should be -zero thus the expansions sum1 and sum2
   should be initialized with -zero as well (otherwise we will get +zero
   as the final result).  */

static int
insert_var_expansion_initialization (void **slot, void *place_p)
{
  struct var_to_expand *ve = *slot;
  basic_block place = (basic_block)place_p;
  rtx seq, var, zero_init, insn;
  unsigned i;
  enum machine_mode mode = GET_MODE (ve->reg);
  bool honor_signed_zero_p = HONOR_SIGNED_ZEROS (mode);

  if (VEC_length (rtx, ve->var_expansions) == 0)
    return 1;
  
  start_sequence ();
  if (ve->op == PLUS || ve->op == MINUS) 
    for (i = 0; VEC_iterate (rtx, ve->var_expansions, i, var); i++)
      {
	if (honor_signed_zero_p)
	  zero_init = simplify_gen_unary (NEG, mode, CONST0_RTX (mode), mode);
	else
	  zero_init = CONST0_RTX (mode);
       	
        emit_move_insn (var, zero_init);
      }
  else if (ve->op == MULT)
    for (i = 0; VEC_iterate (rtx, ve->var_expansions, i, var); i++)
      {
        zero_init =  CONST1_RTX (GET_MODE (var));
        emit_move_insn (var, zero_init);
      }
  
  seq = get_insns ();
  end_sequence ();
  
  insn = BB_HEAD (place);
  while (!NOTE_INSN_BASIC_BLOCK_P (insn))
    insn = NEXT_INSN (insn);
  
  emit_insn_after (seq, insn); 
  /* Continue traversing the hash table.  */
  return 1;   
}

/*  Combine the variable expansions at the loop exit.  
    Callbacks for htab_traverse.  PLACE_P is the loop exit
    basic block where the summation of the expansions should 
    take place.  */

static int
combine_var_copies_in_loop_exit (void **slot, void *place_p)
{
  struct var_to_expand *ve = *slot;
  basic_block place = (basic_block)place_p;
  rtx sum = ve->reg;
  rtx expr, seq, var, insn;
  unsigned i;

  if (VEC_length (rtx, ve->var_expansions) == 0)
    return 1;
  
  start_sequence ();
  if (ve->op == PLUS || ve->op == MINUS)
    for (i = 0; VEC_iterate (rtx, ve->var_expansions, i, var); i++)
      {
        sum = simplify_gen_binary (PLUS, GET_MODE (ve->reg),
                                   var, sum);
      }
  else if (ve->op == MULT)
    for (i = 0; VEC_iterate (rtx, ve->var_expansions, i, var); i++)
      {
        sum = simplify_gen_binary (MULT, GET_MODE (ve->reg),
                                   var, sum);
      }
  
  expr = force_operand (sum, ve->reg);
  if (expr != ve->reg)
    emit_move_insn (ve->reg, expr);
  seq = get_insns ();
  end_sequence ();
  
  insn = BB_HEAD (place);
  while (!NOTE_INSN_BASIC_BLOCK_P (insn))
    insn = NEXT_INSN (insn);

  emit_insn_after (seq, insn);
  
  /* Continue traversing the hash table.  */
  return 1;
}

/* Apply loop optimizations in loop copies using the 
   data which gathered during the unrolling.  Structure 
   OPT_INFO record that data.
   
   UNROLLING is true if we unrolled (not peeled) the loop.
   REWRITE_ORIGINAL_BODY is true if we should also rewrite the original body of
   the loop (as it should happen in complete unrolling, but not in ordinary
   peeling of the loop).  */

static void
apply_opt_in_copies (struct opt_info *opt_info, 
                     unsigned n_copies, bool unrolling, 
                     bool rewrite_original_loop)
{
  unsigned i, delta;
  basic_block bb, orig_bb;
  rtx insn, orig_insn, next;
  struct iv_to_split ivts_templ, *ivts;
  struct var_to_expand ve_templ, *ves;
  
  /* Sanity check -- we need to put initialization in the original loop
     body.  */
  gcc_assert (!unrolling || rewrite_original_loop);
  
  /* Allocate the basic variables (i0).  */
  if (opt_info->insns_to_split)
    htab_traverse (opt_info->insns_to_split, allocate_basic_variable, NULL);
  
  for (i = opt_info->first_new_block; i < (unsigned) last_basic_block; i++)
    {
      bb = BASIC_BLOCK (i);
      orig_bb = get_bb_original (bb);
      
      /* bb->aux holds position in copy sequence initialized by
	 duplicate_loop_to_header_edge.  */
      delta = determine_split_iv_delta ((size_t)bb->aux, n_copies,
					unrolling);
      bb->aux = 0;
      orig_insn = BB_HEAD (orig_bb);
      for (insn = BB_HEAD (bb); insn != NEXT_INSN (BB_END (bb)); insn = next)
        {
          next = NEXT_INSN (insn);
          if (!INSN_P (insn))
            continue;
          
          while (!INSN_P (orig_insn))
            orig_insn = NEXT_INSN (orig_insn);
          
          ivts_templ.insn = orig_insn;
          ve_templ.insn = orig_insn;
          
          /* Apply splitting iv optimization.  */
          if (opt_info->insns_to_split)
            {
              ivts = htab_find (opt_info->insns_to_split, &ivts_templ);
              
              if (ivts)
                {
		  gcc_assert (GET_CODE (PATTERN (insn))
			      == GET_CODE (PATTERN (orig_insn)));
                  
                  if (!delta)
                    insert_base_initialization (ivts, insn);
                  split_iv (ivts, insn, delta);
                }
            }
          /* Apply variable expansion optimization.  */
          if (unrolling && opt_info->insns_with_var_to_expand)
            {
              ves = htab_find (opt_info->insns_with_var_to_expand, &ve_templ);
              if (ves)
                { 
		  gcc_assert (GET_CODE (PATTERN (insn))
			      == GET_CODE (PATTERN (orig_insn)));
                  expand_var_during_unrolling (ves, insn);
                }
            }
          orig_insn = NEXT_INSN (orig_insn);
        }
    }

  if (!rewrite_original_loop)
    return;
  
  /* Initialize the variable expansions in the loop preheader
     and take care of combining them at the loop exit.  */ 
  if (opt_info->insns_with_var_to_expand)
    {
      htab_traverse (opt_info->insns_with_var_to_expand, 
                     insert_var_expansion_initialization, 
                     opt_info->loop_preheader);
      htab_traverse (opt_info->insns_with_var_to_expand, 
                     combine_var_copies_in_loop_exit, 
                     opt_info->loop_exit);
    }
  
  /* Rewrite also the original loop body.  Find them as originals of the blocks
     in the last copied iteration, i.e. those that have
     get_bb_copy (get_bb_original (bb)) == bb.  */
  for (i = opt_info->first_new_block; i < (unsigned) last_basic_block; i++)
    {
      bb = BASIC_BLOCK (i);
      orig_bb = get_bb_original (bb);
      if (get_bb_copy (orig_bb) != bb)
	continue;
      
      delta = determine_split_iv_delta (0, n_copies, unrolling);
      for (orig_insn = BB_HEAD (orig_bb);
           orig_insn != NEXT_INSN (BB_END (bb));
           orig_insn = next)
        {
          next = NEXT_INSN (orig_insn);
          
          if (!INSN_P (orig_insn))
 	    continue;
          
          ivts_templ.insn = orig_insn;
          if (opt_info->insns_to_split)
            {
              ivts = htab_find (opt_info->insns_to_split, &ivts_templ);
              if (ivts)
                {
                  if (!delta)
                    insert_base_initialization (ivts, orig_insn);
                  split_iv (ivts, orig_insn, delta);
                  continue;
                }
            }
          
        }
    }
}

/*  Release the data structures used for the variable expansion
    optimization.  Callbacks for htab_traverse.  */

static int
release_var_copies (void **slot, void *data ATTRIBUTE_UNUSED)
{
  struct var_to_expand *ve = *slot;
  
  VEC_free (rtx, heap, ve->var_expansions);
  
  /* Continue traversing the hash table.  */
  return 1;
}

/* Release OPT_INFO.  */

static void
free_opt_info (struct opt_info *opt_info)
{
  if (opt_info->insns_to_split)
    htab_delete (opt_info->insns_to_split);
  if (opt_info->insns_with_var_to_expand)
    {
      htab_traverse (opt_info->insns_with_var_to_expand, 
                     release_var_copies, NULL);
      htab_delete (opt_info->insns_with_var_to_expand);
    }
  free (opt_info);
}