llvm.org GIT mirror llvm / release_27 lib / Target / README.txt
release_27

Tree @release_27 (Download .tar.gz)

README.txt @release_27raw · history · blame

   1
   2
   3
   4
   5
   6
   7
   8
   9
  10
  11
  12
  13
  14
  15
  16
  17
  18
  19
  20
  21
  22
  23
  24
  25
  26
  27
  28
  29
  30
  31
  32
  33
  34
  35
  36
  37
  38
  39
  40
  41
  42
  43
  44
  45
  46
  47
  48
  49
  50
  51
  52
  53
  54
  55
  56
  57
  58
  59
  60
  61
  62
  63
  64
  65
  66
  67
  68
  69
  70
  71
  72
  73
  74
  75
  76
  77
  78
  79
  80
  81
  82
  83
  84
  85
  86
  87
  88
  89
  90
  91
  92
  93
  94
  95
  96
  97
  98
  99
 100
 101
 102
 103
 104
 105
 106
 107
 108
 109
 110
 111
 112
 113
 114
 115
 116
 117
 118
 119
 120
 121
 122
 123
 124
 125
 126
 127
 128
 129
 130
 131
 132
 133
 134
 135
 136
 137
 138
 139
 140
 141
 142
 143
 144
 145
 146
 147
 148
 149
 150
 151
 152
 153
 154
 155
 156
 157
 158
 159
 160
 161
 162
 163
 164
 165
 166
 167
 168
 169
 170
 171
 172
 173
 174
 175
 176
 177
 178
 179
 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
Target Independent Opportunities:

//===---------------------------------------------------------------------===//

Dead argument elimination should be enhanced to handle cases when an argument is
dead to an externally visible function.  Though the argument can't be removed
from the externally visible function, the caller doesn't need to pass it in.
For example in this testcase:

  void foo(int X) __attribute__((noinline));
  void foo(int X) { sideeffect(); }
  void bar(int A) { foo(A+1); }

We compile bar to:

define void @bar(i32 %A) nounwind ssp {
  %0 = add nsw i32 %A, 1                          ; <i32> [#uses=1]
  tail call void @foo(i32 %0) nounwind noinline ssp
  ret void
}

The add is dead, we could pass in 'i32 undef' instead.  This occurs for C++
templates etc, which usually have linkonce_odr/weak_odr linkage, not internal
linkage.

//===---------------------------------------------------------------------===//

With the recent changes to make the implicit def/use set explicit in
machineinstrs, we should change the target descriptions for 'call' instructions
so that the .td files don't list all the call-clobbered registers as implicit
defs.  Instead, these should be added by the code generator (e.g. on the dag).

This has a number of uses:

1. PPC32/64 and X86 32/64 can avoid having multiple copies of call instructions
   for their different impdef sets.
2. Targets with multiple calling convs (e.g. x86) which have different clobber
   sets don't need copies of call instructions.
3. 'Interprocedural register allocation' can be done to reduce the clobber sets
   of calls.

//===---------------------------------------------------------------------===//

Make the PPC branch selector target independant

//===---------------------------------------------------------------------===//

Get the C front-end to expand hypot(x,y) -> llvm.sqrt(x*x+y*y) when errno and
precision don't matter (ffastmath).  Misc/mandel will like this. :)  This isn't
safe in general, even on darwin.  See the libm implementation of hypot for
examples (which special case when x/y are exactly zero to get signed zeros etc
right).

//===---------------------------------------------------------------------===//

Solve this DAG isel folding deficiency:

int X, Y;

void fn1(void)
{
  X = X | (Y << 3);
}

compiles to

fn1:
	movl Y, %eax
	shll $3, %eax
	orl X, %eax
	movl %eax, X
	ret

The problem is the store's chain operand is not the load X but rather
a TokenFactor of the load X and load Y, which prevents the folding.

There are two ways to fix this:

1. The dag combiner can start using alias analysis to realize that y/x
   don't alias, making the store to X not dependent on the load from Y.
2. The generated isel could be made smarter in the case it can't
   disambiguate the pointers.

Number 1 is the preferred solution.

This has been "fixed" by a TableGen hack. But that is a short term workaround
which will be removed once the proper fix is made.

//===---------------------------------------------------------------------===//

On targets with expensive 64-bit multiply, we could LSR this:

for (i = ...; ++i) {
   x = 1ULL << i;

into:
 long long tmp = 1;
 for (i = ...; ++i, tmp+=tmp)
   x = tmp;

This would be a win on ppc32, but not x86 or ppc64.

//===---------------------------------------------------------------------===//

Shrink: (setlt (loadi32 P), 0) -> (setlt (loadi8 Phi), 0)

//===---------------------------------------------------------------------===//

Reassociate should turn things like:

int factorial(int X) {
 return X*X*X*X*X*X*X*X;
}

into llvm.powi calls, allowing the code generator to produce balanced
multiplication trees.

First, the intrinsic needs to be extended to support integers, and second the
code generator needs to be enhanced to lower these to multiplication trees.

//===---------------------------------------------------------------------===//

Interesting? testcase for add/shift/mul reassoc:

int bar(int x, int y) {
  return x*x*x+y+x*x*x*x*x*y*y*y*y;
}
int foo(int z, int n) {
  return bar(z, n) + bar(2*z, 2*n);
}

This is blocked on not handling X*X*X -> powi(X, 3) (see note above).  The issue
is that we end up getting t = 2*X  s = t*t   and don't turn this into 4*X*X,
which is the same number of multiplies and is canonical, because the 2*X has
multiple uses.  Here's a simple example:

define i32 @test15(i32 %X1) {
  %B = mul i32 %X1, 47   ; X1*47
  %C = mul i32 %B, %B
  ret i32 %C
}


//===---------------------------------------------------------------------===//

Reassociate should handle the example in GCC PR16157:

extern int a0, a1, a2, a3, a4; extern int b0, b1, b2, b3, b4; 
void f () {  /* this can be optimized to four additions... */ 
        b4 = a4 + a3 + a2 + a1 + a0; 
        b3 = a3 + a2 + a1 + a0; 
        b2 = a2 + a1 + a0; 
        b1 = a1 + a0; 
} 

This requires reassociating to forms of expressions that are already available,
something that reassoc doesn't think about yet.


//===---------------------------------------------------------------------===//

This function: (derived from GCC PR19988)
double foo(double x, double y) {
  return ((x + 0.1234 * y) * (x + -0.1234 * y));
}

compiles to:
_foo:
	movapd	%xmm1, %xmm2
	mulsd	LCPI1_1(%rip), %xmm1
	mulsd	LCPI1_0(%rip), %xmm2
	addsd	%xmm0, %xmm1
	addsd	%xmm0, %xmm2
	movapd	%xmm1, %xmm0
	mulsd	%xmm2, %xmm0
	ret

Reassociate should be able to turn it into:

double foo(double x, double y) {
  return ((x + 0.1234 * y) * (x - 0.1234 * y));
}

Which allows the multiply by constant to be CSE'd, producing:

_foo:
	mulsd	LCPI1_0(%rip), %xmm1
	movapd	%xmm1, %xmm2
	addsd	%xmm0, %xmm2
	subsd	%xmm1, %xmm0
	mulsd	%xmm2, %xmm0
	ret

This doesn't need -ffast-math support at all.  This is particularly bad because
the llvm-gcc frontend is canonicalizing the later into the former, but clang
doesn't have this problem.

//===---------------------------------------------------------------------===//

These two functions should generate the same code on big-endian systems:

int g(int *j,int *l)  {  return memcmp(j,l,4);  }
int h(int *j, int *l) {  return *j - *l; }

this could be done in SelectionDAGISel.cpp, along with other special cases,
for 1,2,4,8 bytes.

//===---------------------------------------------------------------------===//

It would be nice to revert this patch:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20060213/031986.html

And teach the dag combiner enough to simplify the code expanded before 
legalize.  It seems plausible that this knowledge would let it simplify other
stuff too.

//===---------------------------------------------------------------------===//

For vector types, TargetData.cpp::getTypeInfo() returns alignment that is equal
to the type size. It works but can be overly conservative as the alignment of
specific vector types are target dependent.

//===---------------------------------------------------------------------===//

We should produce an unaligned load from code like this:

v4sf example(float *P) {
  return (v4sf){P[0], P[1], P[2], P[3] };
}

//===---------------------------------------------------------------------===//

Add support for conditional increments, and other related patterns.  Instead
of:

	movl 136(%esp), %eax
	cmpl $0, %eax
	je LBB16_2	#cond_next
LBB16_1:	#cond_true
	incl _foo
LBB16_2:	#cond_next

emit:
	movl	_foo, %eax
	cmpl	$1, %edi
	sbbl	$-1, %eax
	movl	%eax, _foo

//===---------------------------------------------------------------------===//

Combine: a = sin(x), b = cos(x) into a,b = sincos(x).

Expand these to calls of sin/cos and stores:
      double sincos(double x, double *sin, double *cos);
      float sincosf(float x, float *sin, float *cos);
      long double sincosl(long double x, long double *sin, long double *cos);

Doing so could allow SROA of the destination pointers.  See also:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17687

This is now easily doable with MRVs.  We could even make an intrinsic for this
if anyone cared enough about sincos.

//===---------------------------------------------------------------------===//

Turn this into a single byte store with no load (the other 3 bytes are
unmodified):

define void @test(i32* %P) {
	%tmp = load i32* %P
        %tmp14 = or i32 %tmp, 3305111552
        %tmp15 = and i32 %tmp14, 3321888767
        store i32 %tmp15, i32* %P
        ret void
}

//===---------------------------------------------------------------------===//

quantum_sigma_x in 462.libquantum contains the following loop:

      for(i=0; i<reg->size; i++)
	{
	  /* Flip the target bit of each basis state */
	  reg->node[i].state ^= ((MAX_UNSIGNED) 1 << target);
	} 

Where MAX_UNSIGNED/state is a 64-bit int.  On a 32-bit platform it would be just
so cool to turn it into something like:

   long long Res = ((MAX_UNSIGNED) 1 << target);
   if (target < 32) {
     for(i=0; i<reg->size; i++)
       reg->node[i].state ^= Res & 0xFFFFFFFFULL;
   } else {
     for(i=0; i<reg->size; i++)
       reg->node[i].state ^= Res & 0xFFFFFFFF00000000ULL
   }
   
... which would only do one 32-bit XOR per loop iteration instead of two.

It would also be nice to recognize the reg->size doesn't alias reg->node[i], but
this requires TBAA.

//===---------------------------------------------------------------------===//

This isn't recognized as bswap by instcombine (yes, it really is bswap):

unsigned long reverse(unsigned v) {
    unsigned t;
    t = v ^ ((v << 16) | (v >> 16));
    t &= ~0xff0000;
    v = (v << 24) | (v >> 8);
    return v ^ (t >> 8);
}

//===---------------------------------------------------------------------===//

[LOOP RECOGNITION]

These idioms should be recognized as popcount (see PR1488):

unsigned countbits_slow(unsigned v) {
  unsigned c;
  for (c = 0; v; v >>= 1)
    c += v & 1;
  return c;
}
unsigned countbits_fast(unsigned v){
  unsigned c;
  for (c = 0; v; c++)
    v &= v - 1; // clear the least significant bit set
  return c;
}

BITBOARD = unsigned long long
int PopCnt(register BITBOARD a) {
  register int c=0;
  while(a) {
    c++;
    a &= a - 1;
  }
  return c;
}
unsigned int popcount(unsigned int input) {
  unsigned int count = 0;
  for (unsigned int i =  0; i < 4 * 8; i++)
    count += (input >> i) & i;
  return count;
}

This is a form of idiom recognition for loops, the same thing that could be
useful for recognizing memset/memcpy.

//===---------------------------------------------------------------------===//

These should turn into single 16-bit (unaligned?) loads on little/big endian
processors.

unsigned short read_16_le(const unsigned char *adr) {
  return adr[0] | (adr[1] << 8);
}
unsigned short read_16_be(const unsigned char *adr) {
  return (adr[0] << 8) | adr[1];
}

//===---------------------------------------------------------------------===//

-instcombine should handle this transform:
   icmp pred (sdiv X / C1 ), C2
when X, C1, and C2 are unsigned.  Similarly for udiv and signed operands. 

Currently InstCombine avoids this transform but will do it when the signs of
the operands and the sign of the divide match. See the FIXME in 
InstructionCombining.cpp in the visitSetCondInst method after the switch case 
for Instruction::UDiv (around line 4447) for more details.

The SingleSource/Benchmarks/Shootout-C++/hash and hash2 tests have examples of
this construct. 

//===---------------------------------------------------------------------===//

[LOOP RECOGNITION]

viterbi speeds up *significantly* if the various "history" related copy loops
are turned into memcpy calls at the source level.  We need a "loops to memcpy"
pass.

//===---------------------------------------------------------------------===//

[LOOP OPTIMIZATION]

SingleSource/Benchmarks/Misc/dt.c shows several interesting optimization
opportunities in its double_array_divs_variable function: it needs loop
interchange, memory promotion (which LICM already does), vectorization and
variable trip count loop unrolling (since it has a constant trip count). ICC
apparently produces this very nice code with -ffast-math:

..B1.70:                        # Preds ..B1.70 ..B1.69
       mulpd     %xmm0, %xmm1                                  #108.2
       mulpd     %xmm0, %xmm1                                  #108.2
       mulpd     %xmm0, %xmm1                                  #108.2
       mulpd     %xmm0, %xmm1                                  #108.2
       addl      $8, %edx                                      #
       cmpl      $131072, %edx                                 #108.2
       jb        ..B1.70       # Prob 99%                      #108.2

It would be better to count down to zero, but this is a lot better than what we
do.

//===---------------------------------------------------------------------===//

Consider:

typedef unsigned U32;
typedef unsigned long long U64;
int test (U32 *inst, U64 *regs) {
    U64 effective_addr2;
    U32 temp = *inst;
    int r1 = (temp >> 20) & 0xf;
    int b2 = (temp >> 16) & 0xf;
    effective_addr2 = temp & 0xfff;
    if (b2) effective_addr2 += regs[b2];
    b2 = (temp >> 12) & 0xf;
    if (b2) effective_addr2 += regs[b2];
    effective_addr2 &= regs[4];
     if ((effective_addr2 & 3) == 0)
        return 1;
    return 0;
}

Note that only the low 2 bits of effective_addr2 are used.  On 32-bit systems,
we don't eliminate the computation of the top half of effective_addr2 because
we don't have whole-function selection dags.  On x86, this means we use one
extra register for the function when effective_addr2 is declared as U64 than
when it is declared U32.

PHI Slicing could be extended to do this.

//===---------------------------------------------------------------------===//

LSR should know what GPR types a target has from TargetData.  This code:

volatile short X, Y; // globals

void foo(int N) {
  int i;
  for (i = 0; i < N; i++) { X = i; Y = i*4; }
}

produces two near identical IV's (after promotion) on PPC/ARM:

LBB1_2:
	ldr r3, LCPI1_0
	ldr r3, [r3]
	strh r2, [r3]
	ldr r3, LCPI1_1
	ldr r3, [r3]
	strh r1, [r3]
	add r1, r1, #4
	add r2, r2, #1   <- [0,+,1]
	sub r0, r0, #1   <- [0,-,1]
	cmp r0, #0
	bne LBB1_2

LSR should reuse the "+" IV for the exit test.

//===---------------------------------------------------------------------===//

Tail call elim should be more aggressive, checking to see if the call is
followed by an uncond branch to an exit block.

; This testcase is due to tail-duplication not wanting to copy the return
; instruction into the terminating blocks because there was other code
; optimized out of the function after the taildup happened.
; RUN: llvm-as < %s | opt -tailcallelim | llvm-dis | not grep call

define i32 @t4(i32 %a) {
entry:
	%tmp.1 = and i32 %a, 1		; <i32> [#uses=1]
	%tmp.2 = icmp ne i32 %tmp.1, 0		; <i1> [#uses=1]
	br i1 %tmp.2, label %then.0, label %else.0

then.0:		; preds = %entry
	%tmp.5 = add i32 %a, -1		; <i32> [#uses=1]
	%tmp.3 = call i32 @t4( i32 %tmp.5 )		; <i32> [#uses=1]
	br label %return

else.0:		; preds = %entry
	%tmp.7 = icmp ne i32 %a, 0		; <i1> [#uses=1]
	br i1 %tmp.7, label %then.1, label %return

then.1:		; preds = %else.0
	%tmp.11 = add i32 %a, -2		; <i32> [#uses=1]
	%tmp.9 = call i32 @t4( i32 %tmp.11 )		; <i32> [#uses=1]
	br label %return

return:		; preds = %then.1, %else.0, %then.0
	%result.0 = phi i32 [ 0, %else.0 ], [ %tmp.3, %then.0 ],
                            [ %tmp.9, %then.1 ]
	ret i32 %result.0
}

//===---------------------------------------------------------------------===//

Tail recursion elimination should handle:

int pow2m1(int n) {
 if (n == 0)
   return 0;
 return 2 * pow2m1 (n - 1) + 1;
}

Also, multiplies can be turned into SHL's, so they should be handled as if
they were associative.  "return foo() << 1" can be tail recursion eliminated.

//===---------------------------------------------------------------------===//

Argument promotion should promote arguments for recursive functions, like 
this:

; RUN: llvm-as < %s | opt -argpromotion | llvm-dis | grep x.val

define internal i32 @foo(i32* %x) {
entry:
	%tmp = load i32* %x		; <i32> [#uses=0]
	%tmp.foo = call i32 @foo( i32* %x )		; <i32> [#uses=1]
	ret i32 %tmp.foo
}

define i32 @bar(i32* %x) {
entry:
	%tmp3 = call i32 @foo( i32* %x )		; <i32> [#uses=1]
	ret i32 %tmp3
}

//===---------------------------------------------------------------------===//

We should investigate an instruction sinking pass.  Consider this silly
example in pic mode:

#include <assert.h>
void foo(int x) {
  assert(x);
  //...
}

we compile this to:
_foo:
	subl	$28, %esp
	call	"L1$pb"
"L1$pb":
	popl	%eax
	cmpl	$0, 32(%esp)
	je	LBB1_2	# cond_true
LBB1_1:	# return
	# ...
	addl	$28, %esp
	ret
LBB1_2:	# cond_true
...

The PIC base computation (call+popl) is only used on one path through the 
code, but is currently always computed in the entry block.  It would be 
better to sink the picbase computation down into the block for the 
assertion, as it is the only one that uses it.  This happens for a lot of 
code with early outs.

Another example is loads of arguments, which are usually emitted into the 
entry block on targets like x86.  If not used in all paths through a 
function, they should be sunk into the ones that do.

In this case, whole-function-isel would also handle this.

//===---------------------------------------------------------------------===//

Investigate lowering of sparse switch statements into perfect hash tables:
http://burtleburtle.net/bob/hash/perfect.html

//===---------------------------------------------------------------------===//

We should turn things like "load+fabs+store" and "load+fneg+store" into the
corresponding integer operations.  On a yonah, this loop:

double a[256];
void foo() {
  int i, b;
  for (b = 0; b < 10000000; b++)
  for (i = 0; i < 256; i++)
    a[i] = -a[i];
}

is twice as slow as this loop:

long long a[256];
void foo() {
  int i, b;
  for (b = 0; b < 10000000; b++)
  for (i = 0; i < 256; i++)
    a[i] ^= (1ULL << 63);
}

and I suspect other processors are similar.  On X86 in particular this is a
big win because doing this with integers allows the use of read/modify/write
instructions.

//===---------------------------------------------------------------------===//

DAG Combiner should try to combine small loads into larger loads when 
profitable.  For example, we compile this C++ example:

struct THotKey { short Key; bool Control; bool Shift; bool Alt; };
extern THotKey m_HotKey;
THotKey GetHotKey () { return m_HotKey; }

into (-O3 -fno-exceptions -static -fomit-frame-pointer):

__Z9GetHotKeyv:
	pushl	%esi
	movl	8(%esp), %eax
	movb	_m_HotKey+3, %cl
	movb	_m_HotKey+4, %dl
	movb	_m_HotKey+2, %ch
	movw	_m_HotKey, %si
	movw	%si, (%eax)
	movb	%ch, 2(%eax)
	movb	%cl, 3(%eax)
	movb	%dl, 4(%eax)
	popl	%esi
	ret	$4

GCC produces:

__Z9GetHotKeyv:
	movl	_m_HotKey, %edx
	movl	4(%esp), %eax
	movl	%edx, (%eax)
	movzwl	_m_HotKey+4, %edx
	movw	%dx, 4(%eax)
	ret	$4

The LLVM IR contains the needed alignment info, so we should be able to 
merge the loads and stores into 4-byte loads:

	%struct.THotKey = type { i16, i8, i8, i8 }
define void @_Z9GetHotKeyv(%struct.THotKey* sret  %agg.result) nounwind  {
...
	%tmp2 = load i16* getelementptr (@m_HotKey, i32 0, i32 0), align 8
	%tmp5 = load i8* getelementptr (@m_HotKey, i32 0, i32 1), align 2
	%tmp8 = load i8* getelementptr (@m_HotKey, i32 0, i32 2), align 1
	%tmp11 = load i8* getelementptr (@m_HotKey, i32 0, i32 3), align 2

Alternatively, we should use a small amount of base-offset alias analysis
to make it so the scheduler doesn't need to hold all the loads in regs at
once.

//===---------------------------------------------------------------------===//

We should add an FRINT node to the DAG to model targets that have legal
implementations of ceil/floor/rint.

//===---------------------------------------------------------------------===//

Consider:

int test() {
  long long input[8] = {1,1,1,1,1,1,1,1};
  foo(input);
}

We currently compile this into a memcpy from a global array since the 
initializer is fairly large and not memset'able.  This is good, but the memcpy
gets lowered to load/stores in the code generator.  This is also ok, except
that the codegen lowering for memcpy doesn't handle the case when the source
is a constant global.  This gives us atrocious code like this:

	call	"L1$pb"
"L1$pb":
	popl	%eax
	movl	_C.0.1444-"L1$pb"+32(%eax), %ecx
	movl	%ecx, 40(%esp)
	movl	_C.0.1444-"L1$pb"+20(%eax), %ecx
	movl	%ecx, 28(%esp)
	movl	_C.0.1444-"L1$pb"+36(%eax), %ecx
	movl	%ecx, 44(%esp)
	movl	_C.0.1444-"L1$pb"+44(%eax), %ecx
	movl	%ecx, 52(%esp)
	movl	_C.0.1444-"L1$pb"+40(%eax), %ecx
	movl	%ecx, 48(%esp)
	movl	_C.0.1444-"L1$pb"+12(%eax), %ecx
	movl	%ecx, 20(%esp)
	movl	_C.0.1444-"L1$pb"+4(%eax), %ecx
...

instead of:
	movl	$1, 16(%esp)
	movl	$0, 20(%esp)
	movl	$1, 24(%esp)
	movl	$0, 28(%esp)
	movl	$1, 32(%esp)
	movl	$0, 36(%esp)
	...

//===---------------------------------------------------------------------===//

http://llvm.org/PR717:

The following code should compile into "ret int undef". Instead, LLVM
produces "ret int 0":

int f() {
  int x = 4;
  int y;
  if (x == 3) y = 0;
  return y;
}

//===---------------------------------------------------------------------===//

The loop unroller should partially unroll loops (instead of peeling them)
when code growth isn't too bad and when an unroll count allows simplification
of some code within the loop.  One trivial example is:

#include <stdio.h>
int main() {
    int nRet = 17;
    int nLoop;
    for ( nLoop = 0; nLoop < 1000; nLoop++ ) {
        if ( nLoop & 1 )
            nRet += 2;
        else
            nRet -= 1;
    }
    return nRet;
}

Unrolling by 2 would eliminate the '&1' in both copies, leading to a net
reduction in code size.  The resultant code would then also be suitable for
exit value computation.

//===---------------------------------------------------------------------===//

We miss a bunch of rotate opportunities on various targets, including ppc, x86,
etc.  On X86, we miss a bunch of 'rotate by variable' cases because the rotate
matching code in dag combine doesn't look through truncates aggressively 
enough.  Here are some testcases reduces from GCC PR17886:

unsigned long long f(unsigned long long x, int y) {
  return (x << y) | (x >> 64-y); 
} 
unsigned f2(unsigned x, int y){
  return (x << y) | (x >> 32-y); 
} 
unsigned long long f3(unsigned long long x){
  int y = 9;
  return (x << y) | (x >> 64-y); 
} 
unsigned f4(unsigned x){
  int y = 10;
  return (x << y) | (x >> 32-y); 
}
unsigned long long f5(unsigned long long x, unsigned long long y) {
  return (x << 8) | ((y >> 48) & 0xffull);
}
unsigned long long f6(unsigned long long x, unsigned long long y, int z) {
  switch(z) {
  case 1:
    return (x << 8) | ((y >> 48) & 0xffull);
  case 2:
    return (x << 16) | ((y >> 40) & 0xffffull);
  case 3:
    return (x << 24) | ((y >> 32) & 0xffffffull);
  case 4:
    return (x << 32) | ((y >> 24) & 0xffffffffull);
  default:
    return (x << 40) | ((y >> 16) & 0xffffffffffull);
  }
}

On X86-64, we only handle f2/f3/f4 right.  On x86-32, a few of these 
generate truly horrible code, instead of using shld and friends.  On
ARM, we end up with calls to L___lshrdi3/L___ashldi3 in f, which is
badness.  PPC64 misses f, f5 and f6.  CellSPU aborts in isel.

//===---------------------------------------------------------------------===//

We do a number of simplifications in simplify libcalls to strength reduce
standard library functions, but we don't currently merge them together.  For
example, it is useful to merge memcpy(a,b,strlen(b)) -> strcpy.  This can only
be done safely if "b" isn't modified between the strlen and memcpy of course.

//===---------------------------------------------------------------------===//

We compile this program: (from GCC PR11680)
http://gcc.gnu.org/bugzilla/attachment.cgi?id=4487

Into code that runs the same speed in fast/slow modes, but both modes run 2x
slower than when compile with GCC (either 4.0 or 4.2):

$ llvm-g++ perf.cpp -O3 -fno-exceptions
$ time ./a.out fast
1.821u 0.003s 0:01.82 100.0%	0+0k 0+0io 0pf+0w

$ g++ perf.cpp -O3 -fno-exceptions
$ time ./a.out fast
0.821u 0.001s 0:00.82 100.0%	0+0k 0+0io 0pf+0w

It looks like we are making the same inlining decisions, so this may be raw
codegen badness or something else (haven't investigated).

//===---------------------------------------------------------------------===//

We miss some instcombines for stuff like this:
void bar (void);
void foo (unsigned int a) {
  /* This one is equivalent to a >= (3 << 2).  */
  if ((a >> 2) >= 3)
    bar ();
}

A few other related ones are in GCC PR14753.

//===---------------------------------------------------------------------===//

Divisibility by constant can be simplified (according to GCC PR12849) from
being a mulhi to being a mul lo (cheaper).  Testcase:

void bar(unsigned n) {
  if (n % 3 == 0)
    true();
}

This is equivalent to the following, where 2863311531 is the multiplicative
inverse of 3, and 1431655766 is ((2^32)-1)/3+1:
void bar(unsigned n) {
  if (n * 2863311531U < 1431655766U)
    true();
}

The same transformation can work with an even modulo with the addition of a
rotate: rotate the result of the multiply to the right by the number of bits
which need to be zero for the condition to be true, and shrink the compare RHS
by the same amount.  Unless the target supports rotates, though, that
transformation probably isn't worthwhile.

The transformation can also easily be made to work with non-zero equality
comparisons: just transform, for example, "n % 3 == 1" to "(n-1) % 3 == 0".

//===---------------------------------------------------------------------===//

Better mod/ref analysis for scanf would allow us to eliminate the vtable and a
bunch of other stuff from this example (see PR1604): 

#include <cstdio>
struct test {
    int val;
    virtual ~test() {}
};

int main() {
    test t;
    std::scanf("%d", &t.val);
    std::printf("%d\n", t.val);
}

//===---------------------------------------------------------------------===//

These functions perform the same computation, but produce different assembly.

define i8 @select(i8 %x) readnone nounwind {
  %A = icmp ult i8 %x, 250
  %B = select i1 %A, i8 0, i8 1
  ret i8 %B 
}

define i8 @addshr(i8 %x) readnone nounwind {
  %A = zext i8 %x to i9
  %B = add i9 %A, 6       ;; 256 - 250 == 6
  %C = lshr i9 %B, 8
  %D = trunc i9 %C to i8
  ret i8 %D
}

//===---------------------------------------------------------------------===//

From gcc bug 24696:
int
f (unsigned long a, unsigned long b, unsigned long c)
{
  return ((a & (c - 1)) != 0) || ((b & (c - 1)) != 0);
}
int
f (unsigned long a, unsigned long b, unsigned long c)
{
  return ((a & (c - 1)) != 0) | ((b & (c - 1)) != 0);
}
Both should combine to ((a|b) & (c-1)) != 0.  Currently not optimized with
"clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

From GCC Bug 20192:
#define PMD_MASK    (~((1UL << 23) - 1))
void clear_pmd_range(unsigned long start, unsigned long end)
{
   if (!(start & ~PMD_MASK) && !(end & ~PMD_MASK))
       f();
}
The expression should optimize to something like
"!((start|end)&~PMD_MASK). Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

From GCC Bug 3756:
int
pn (int n)
{
 return (n >= 0 ? 1 : -1);
}
Should combine to (n >> 31) | 1.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts | llc".

//===---------------------------------------------------------------------===//

void a(int variable)
{
 if (variable == 4 || variable == 6)
   bar();
}
This should optimize to "if ((variable | 2) == 6)".  Currently not
optimized with "clang -emit-llvm-bc | opt -std-compile-opts | llc".

//===---------------------------------------------------------------------===//

unsigned int f(unsigned int i, unsigned int n) {++i; if (i == n) ++i; return
i;}
unsigned int f2(unsigned int i, unsigned int n) {++i; i += i == n; return i;}
These should combine to the same thing.  Currently, the first function
produces better code on X86.

//===---------------------------------------------------------------------===//

From GCC Bug 15784:
#define abs(x) x>0?x:-x
int f(int x, int y)
{
 return (abs(x)) >= 0;
}
This should optimize to x == INT_MIN. (With -fwrapv.)  Currently not
optimized with "clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

From GCC Bug 14753:
void
rotate_cst (unsigned int a)
{
 a = (a << 10) | (a >> 22);
 if (a == 123)
   bar ();
}
void
minus_cst (unsigned int a)
{
 unsigned int tem;

 tem = 20 - a;
 if (tem == 5)
   bar ();
}
void
mask_gt (unsigned int a)
{
 /* This is equivalent to a > 15.  */
 if ((a & ~7) > 8)
   bar ();
}
void
rshift_gt (unsigned int a)
{
 /* This is equivalent to a > 23.  */
 if ((a >> 2) > 5)
   bar ();
}
All should simplify to a single comparison.  All of these are
currently not optimized with "clang -emit-llvm-bc | opt
-std-compile-opts".

//===---------------------------------------------------------------------===//

From GCC Bug 32605:
int c(int* x) {return (char*)x+2 == (char*)x;}
Should combine to 0.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts" (although llc can optimize it).

//===---------------------------------------------------------------------===//

int a(unsigned b) {return ((b << 31) | (b << 30)) >> 31;}
Should be combined to  "((b >> 1) | b) & 1".  Currently not optimized
with "clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

unsigned a(unsigned x, unsigned y) { return x | (y & 1) | (y & 2);}
Should combine to "x | (y & 3)".  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int a, int b, int c) {return (~a & c) | ((c|a) & b);}
Should fold to "(~a & c) | (a & b)".  Currently not optimized with
"clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int a,int b) {return (~(a|b))|a;}
Should fold to "a|~b".  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int a, int b) {return (a&&b) || (a&&!b);}
Should fold to "a".  Currently not optimized with "clang -emit-llvm-bc
| opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int a, int b, int c) {return (a&&b) || (!a&&c);}
Should fold to "a ? b : c", or at least something sane.  Currently not
optimized with "clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int a, int b, int c) {return (a&&b) || (a&&c) || (a&&b&&c);}
Should fold to a && (b || c).  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int x) {return x | ((x & 8) ^ 8);}
Should combine to x | 8.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int x) {return x ^ ((x & 8) ^ 8);}
Should also combine to x | 8.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int x) {return (x & 8) == 0 ? -1 : -9;}
Should combine to (x | -9) ^ 8.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int x) {return (x & 8) == 0 ? -9 : -1;}
Should combine to x | -9.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

int a(int x) {return ((x | -9) ^ 8) & x;}
Should combine to x & -9.  Currently not optimized with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

unsigned a(unsigned a) {return a * 0x11111111 >> 28 & 1;}
Should combine to "a * 0x88888888 >> 31".  Currently not optimized
with "clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

unsigned a(char* x) {if ((*x & 32) == 0) return b();}
There's an unnecessary zext in the generated code with "clang
-emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

unsigned a(unsigned long long x) {return 40 * (x >> 1);}
Should combine to "20 * (((unsigned)x) & -2)".  Currently not
optimized with "clang -emit-llvm-bc | opt -std-compile-opts".

//===---------------------------------------------------------------------===//

This was noticed in the entryblock for grokdeclarator in 403.gcc:

        %tmp = icmp eq i32 %decl_context, 4          
        %decl_context_addr.0 = select i1 %tmp, i32 3, i32 %decl_context 
        %tmp1 = icmp eq i32 %decl_context_addr.0, 1 
        %decl_context_addr.1 = select i1 %tmp1, i32 0, i32 %decl_context_addr.0

tmp1 should be simplified to something like:
  (!tmp || decl_context == 1)

This allows recursive simplifications, tmp1 is used all over the place in
the function, e.g. by:

        %tmp23 = icmp eq i32 %decl_context_addr.1, 0            ; <i1> [#uses=1]
        %tmp24 = xor i1 %tmp1, true             ; <i1> [#uses=1]
        %or.cond8 = and i1 %tmp23, %tmp24               ; <i1> [#uses=1]

later.

//===---------------------------------------------------------------------===//

[STORE SINKING]

Store sinking: This code:

void f (int n, int *cond, int *res) {
    int i;
    *res = 0;
    for (i = 0; i < n; i++)
        if (*cond)
            *res ^= 234; /* (*) */
}

On this function GVN hoists the fully redundant value of *res, but nothing
moves the store out.  This gives us this code:

bb:		; preds = %bb2, %entry
	%.rle = phi i32 [ 0, %entry ], [ %.rle6, %bb2 ]	
	%i.05 = phi i32 [ 0, %entry ], [ %indvar.next, %bb2 ]
	%1 = load i32* %cond, align 4
	%2 = icmp eq i32 %1, 0
	br i1 %2, label %bb2, label %bb1

bb1:		; preds = %bb
	%3 = xor i32 %.rle, 234	
	store i32 %3, i32* %res, align 4
	br label %bb2

bb2:		; preds = %bb, %bb1
	%.rle6 = phi i32 [ %3, %bb1 ], [ %.rle, %bb ]	
	%indvar.next = add i32 %i.05, 1	
	%exitcond = icmp eq i32 %indvar.next, %n
	br i1 %exitcond, label %return, label %bb

DSE should sink partially dead stores to get the store out of the loop.

Here's another partial dead case:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12395

//===---------------------------------------------------------------------===//

Scalar PRE hoists the mul in the common block up to the else:

int test (int a, int b, int c, int g) {
  int d, e;
  if (a)
    d = b * c;
  else
    d = b - c;
  e = b * c + g;
  return d + e;
}

It would be better to do the mul once to reduce codesize above the if.
This is GCC PR38204.

//===---------------------------------------------------------------------===//

[STORE SINKING]

GCC PR37810 is an interesting case where we should sink load/store reload
into the if block and outside the loop, so we don't reload/store it on the
non-call path.

for () {
  *P += 1;
  if ()
    call();
  else
    ...
->
tmp = *P
for () {
  tmp += 1;
  if () {
    *P = tmp;
    call();
    tmp = *P;
  } else ...
}
*P = tmp;

We now hoist the reload after the call (Transforms/GVN/lpre-call-wrap.ll), but
we don't sink the store.  We need partially dead store sinking.

//===---------------------------------------------------------------------===//

[LOAD PRE CRIT EDGE SPLITTING]

GCC PR37166: Sinking of loads prevents SROA'ing the "g" struct on the stack
leading to excess stack traffic. This could be handled by GVN with some crazy
symbolic phi translation.  The code we get looks like (g is on the stack):

bb2:		; preds = %bb1
..
	%9 = getelementptr %struct.f* %g, i32 0, i32 0		
	store i32 %8, i32* %9, align  bel %bb3

bb3:		; preds = %bb1, %bb2, %bb
	%c_addr.0 = phi %struct.f* [ %g, %bb2 ], [ %c, %bb ], [ %c, %bb1 ]
	%b_addr.0 = phi %struct.f* [ %b, %bb2 ], [ %g, %bb ], [ %b, %bb1 ]
	%10 = getelementptr %struct.f* %c_addr.0, i32 0, i32 0
	%11 = load i32* %10, align 4

%11 is partially redundant, an in BB2 it should have the value %8.

GCC PR33344 and PR35287 are similar cases.


//===---------------------------------------------------------------------===//

[LOAD PRE]

There are many load PRE testcases in testsuite/gcc.dg/tree-ssa/loadpre* in the
GCC testsuite, ones we don't get yet are (checked through loadpre25):

[CRIT EDGE BREAKING]
loadpre3.c predcom-4.c

[PRE OF READONLY CALL]
loadpre5.c

[TURN SELECT INTO BRANCH]
loadpre14.c loadpre15.c 

actually a conditional increment: loadpre18.c loadpre19.c


//===---------------------------------------------------------------------===//

[SCALAR PRE]
There are many PRE testcases in testsuite/gcc.dg/tree-ssa/ssa-pre-*.c in the
GCC testsuite.

//===---------------------------------------------------------------------===//

There are some interesting cases in testsuite/gcc.dg/tree-ssa/pred-comm* in the
GCC testsuite.  For example, we get the first example in predcom-1.c, but 
miss the second one:

unsigned fib[1000];
unsigned avg[1000];

__attribute__ ((noinline))
void count_averages(int n) {
  int i;
  for (i = 1; i < n; i++)
    avg[i] = (((unsigned long) fib[i - 1] + fib[i] + fib[i + 1]) / 3) & 0xffff;
}

which compiles into two loads instead of one in the loop.

predcom-2.c is the same as predcom-1.c

predcom-3.c is very similar but needs loads feeding each other instead of
store->load.


//===---------------------------------------------------------------------===//

[ALIAS ANALYSIS]

Type based alias analysis:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14705

We should do better analysis of posix_memalign.  At the least it should
no-capture its pointer argument, at best, we should know that the out-value
result doesn't point to anything (like malloc).  One example of this is in
SingleSource/Benchmarks/Misc/dt.c

//===---------------------------------------------------------------------===//

A/B get pinned to the stack because we turn an if/then into a select instead
of PRE'ing the load/store.  This may be fixable in instcombine:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37892

struct X { int i; };
int foo (int x) {
  struct X a;
  struct X b;
  struct X *p;
  a.i = 1;
  b.i = 2;
  if (x)
    p = &a;
  else
    p = &b;
  return p->i;
}

//===---------------------------------------------------------------------===//

Interesting missed case because of control flow flattening (should be 2 loads):
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26629
With: llvm-gcc t2.c -S -o - -O0 -emit-llvm | llvm-as | 
             opt -mem2reg -gvn -instcombine | llvm-dis
we miss it because we need 1) CRIT EDGE 2) MULTIPLE DIFFERENT
VALS PRODUCED BY ONE BLOCK OVER DIFFERENT PATHS

//===---------------------------------------------------------------------===//

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19633
We could eliminate the branch condition here, loading from null is undefined:

struct S { int w, x, y, z; };
struct T { int r; struct S s; };
void bar (struct S, int);
void foo (int a, struct T b)
{
  struct S *c = 0;
  if (a)
    c = &b.s;
  bar (*c, a);
}

//===---------------------------------------------------------------------===//

simplifylibcalls should do several optimizations for strspn/strcspn:

strcspn(x, "") -> strlen(x)
strcspn("", x) -> 0
strspn("", x) -> 0
strspn(x, "") -> strlen(x)
strspn(x, "a") -> strchr(x, 'a')-x

strcspn(x, "a") -> inlined loop for up to 3 letters (similarly for strspn):

size_t __strcspn_c3 (__const char *__s, int __reject1, int __reject2,
                     int __reject3) {
  register size_t __result = 0;
  while (__s[__result] != '\0' && __s[__result] != __reject1 &&
         __s[__result] != __reject2 && __s[__result] != __reject3)
    ++__result;
  return __result;
}

This should turn into a switch on the character.  See PR3253 for some notes on
codegen.

456.hmmer apparently uses strcspn and strspn a lot.  471.omnetpp uses strspn.

//===---------------------------------------------------------------------===//

"gas" uses this idiom:
  else if (strchr ("+-/*%|&^:[]()~", *intel_parser.op_string))
..
  else if (strchr ("<>", *intel_parser.op_string)

Those should be turned into a switch.

//===---------------------------------------------------------------------===//

252.eon contains this interesting code:

        %3072 = getelementptr [100 x i8]* %tempString, i32 0, i32 0
        %3073 = call i8* @strcpy(i8* %3072, i8* %3071) nounwind
        %strlen = call i32 @strlen(i8* %3072)    ; uses = 1
        %endptr = getelementptr [100 x i8]* %tempString, i32 0, i32 %strlen
        call void @llvm.memcpy.i32(i8* %endptr, 
          i8* getelementptr ([5 x i8]* @"\01LC42", i32 0, i32 0), i32 5, i32 1)
        %3074 = call i32 @strlen(i8* %endptr) nounwind readonly 
        
This is interesting for a couple reasons.  First, in this:

        %3073 = call i8* @strcpy(i8* %3072, i8* %3071) nounwind
        %strlen = call i32 @strlen(i8* %3072)  

The strlen could be replaced with: %strlen = sub %3072, %3073, because the
strcpy call returns a pointer to the end of the string.  Based on that, the
endptr GEP just becomes equal to 3073, which eliminates a strlen call and GEP.

Second, the memcpy+strlen strlen can be replaced with:

        %3074 = call i32 @strlen([5 x i8]* @"\01LC42") nounwind readonly 

Because the destination was just copied into the specified memory buffer.  This,
in turn, can be constant folded to "4".

In other code, it contains:

        %endptr6978 = bitcast i8* %endptr69 to i32*            
        store i32 7107374, i32* %endptr6978, align 1
        %3167 = call i32 @strlen(i8* %endptr69) nounwind readonly    

Which could also be constant folded.  Whatever is producing this should probably
be fixed to leave this as a memcpy from a string.

Further, eon also has an interesting partially redundant strlen call:

bb8:            ; preds = %_ZN18eonImageCalculatorC1Ev.exit
        %682 = getelementptr i8** %argv, i32 6          ; <i8**> [#uses=2]
        %683 = load i8** %682, align 4          ; <i8*> [#uses=4]
        %684 = load i8* %683, align 1           ; <i8> [#uses=1]
        %685 = icmp eq i8 %684, 0               ; <i1> [#uses=1]
        br i1 %685, label %bb10, label %bb9

bb9:            ; preds = %bb8
        %686 = call i32 @strlen(i8* %683) nounwind readonly          
        %687 = icmp ugt i32 %686, 254           ; <i1> [#uses=1]
        br i1 %687, label %bb10, label %bb11

bb10:           ; preds = %bb9, %bb8
        %688 = call i32 @strlen(i8* %683) nounwind readonly          

This could be eliminated by doing the strlen once in bb8, saving code size and
improving perf on the bb8->9->10 path.

//===---------------------------------------------------------------------===//

I see an interesting fully redundant call to strlen left in 186.crafty:InputMove
which looks like:
       %movetext11 = getelementptr [128 x i8]* %movetext, i32 0, i32 0 
 

bb62:           ; preds = %bb55, %bb53
        %promote.0 = phi i32 [ %169, %bb55 ], [ 0, %bb53 ]             
        %171 = call i32 @strlen(i8* %movetext11) nounwind readonly align 1
        %172 = add i32 %171, -1         ; <i32> [#uses=1]
        %173 = getelementptr [128 x i8]* %movetext, i32 0, i32 %172       

...  no stores ...
       br i1 %or.cond, label %bb65, label %bb72

bb65:           ; preds = %bb62
        store i8 0, i8* %173, align 1
        br label %bb72

bb72:           ; preds = %bb65, %bb62
        %trank.1 = phi i32 [ %176, %bb65 ], [ -1, %bb62 ]            
        %177 = call i32 @strlen(i8* %movetext11) nounwind readonly align 1

Note that on the bb62->bb72 path, that the %177 strlen call is partially
redundant with the %171 call.  At worst, we could shove the %177 strlen call
up into the bb65 block moving it out of the bb62->bb72 path.   However, note
that bb65 stores to the string, zeroing out the last byte.  This means that on
that path the value of %177 is actually just %171-1.  A sub is cheaper than a
strlen!

This pattern repeats several times, basically doing:

  A = strlen(P);
  P[A-1] = 0;
  B = strlen(P);
  where it is "obvious" that B = A-1.

//===---------------------------------------------------------------------===//

186.crafty contains this interesting pattern:

%77 = call i8* @strstr(i8* getelementptr ([6 x i8]* @"\01LC5", i32 0, i32 0),
                       i8* %30)
%phitmp648 = icmp eq i8* %77, getelementptr ([6 x i8]* @"\01LC5", i32 0, i32 0)
br i1 %phitmp648, label %bb70, label %bb76

bb70:           ; preds = %OptionMatch.exit91, %bb69
        %78 = call i32 @strlen(i8* %30) nounwind readonly align 1               ; <i32> [#uses=1]

This is basically:
  cststr = "abcdef";
  if (strstr(cststr, P) == cststr) {
     x = strlen(P);
     ...

The strstr call would be significantly cheaper written as:

cststr = "abcdef";
if (memcmp(P, str, strlen(P)))
  x = strlen(P);

This is memcmp+strlen instead of strstr.  This also makes the strlen fully
redundant.

//===---------------------------------------------------------------------===//

186.crafty also contains this code:

%1906 = call i32 @strlen(i8* getelementptr ([32 x i8]* @pgn_event, i32 0,i32 0))
%1907 = getelementptr [32 x i8]* @pgn_event, i32 0, i32 %1906
%1908 = call i8* @strcpy(i8* %1907, i8* %1905) nounwind align 1
%1909 = call i32 @strlen(i8* getelementptr ([32 x i8]* @pgn_event, i32 0,i32 0))
%1910 = getelementptr [32 x i8]* @pgn_event, i32 0, i32 %1909         

The last strlen is computable as 1908-@pgn_event, which means 1910=1908.

//===---------------------------------------------------------------------===//

186.crafty has this interesting pattern with the "out.4543" variable:

call void @llvm.memcpy.i32(
        i8* getelementptr ([10 x i8]* @out.4543, i32 0, i32 0),
       i8* getelementptr ([7 x i8]* @"\01LC28700", i32 0, i32 0), i32 7, i32 1) 
%101 = call@printf(i8* ...   @out.4543, i32 0, i32 0)) nounwind 

It is basically doing:

  memcpy(globalarray, "string");
  printf(...,  globalarray);
  
Anyway, by knowing that printf just reads the memory and forward substituting
the string directly into the printf, this eliminates reads from globalarray.
Since this pattern occurs frequently in crafty (due to the "DisplayTime" and
other similar functions) there are many stores to "out".  Once all the printfs
stop using "out", all that is left is the memcpy's into it.  This should allow
globalopt to remove the "stored only" global.

//===---------------------------------------------------------------------===//

This code:

define inreg i32 @foo(i8* inreg %p) nounwind {
  %tmp0 = load i8* %p
  %tmp1 = ashr i8 %tmp0, 5
  %tmp2 = sext i8 %tmp1 to i32
  ret i32 %tmp2
}

could be dagcombine'd to a sign-extending load with a shift.
For example, on x86 this currently gets this:

	movb	(%eax), %al
	sarb	$5, %al
	movsbl	%al, %eax

while it could get this:

	movsbl	(%eax), %eax
	sarl	$5, %eax

//===---------------------------------------------------------------------===//

GCC PR31029:

int test(int x) { return 1-x == x; }     // --> return false
int test2(int x) { return 2-x == x; }    // --> return x == 1 ?

Always foldable for odd constants, what is the rule for even?

//===---------------------------------------------------------------------===//

PR 3381: GEP to field of size 0 inside a struct could be turned into GEP
for next field in struct (which is at same address).

For example: store of float into { {{}}, float } could be turned into a store to
the float directly.

//===---------------------------------------------------------------------===//

#include <math.h>
double foo(double a) {    return sin(a); }

This compiles into this on x86-64 Linux:
foo:
	subq	$8, %rsp
	call	sin
	addq	$8, %rsp
	ret
vs:

foo:
        jmp sin

//===---------------------------------------------------------------------===//

The arg promotion pass should make use of nocapture to make its alias analysis
stuff much more precise.

//===---------------------------------------------------------------------===//

The following functions should be optimized to use a select instead of a
branch (from gcc PR40072):

char char_int(int m) {if(m>7) return 0; return m;}
int int_char(char m) {if(m>7) return 0; return m;}

//===---------------------------------------------------------------------===//

int func(int a, int b) { if (a & 0x80) b |= 0x80; else b &= ~0x80; return b; }

Generates this:

define i32 @func(i32 %a, i32 %b) nounwind readnone ssp {
entry:
  %0 = and i32 %a, 128                            ; <i32> [#uses=1]
  %1 = icmp eq i32 %0, 0                          ; <i1> [#uses=1]
  %2 = or i32 %b, 128                             ; <i32> [#uses=1]
  %3 = and i32 %b, -129                           ; <i32> [#uses=1]
  %b_addr.0 = select i1 %1, i32 %3, i32 %2        ; <i32> [#uses=1]
  ret i32 %b_addr.0
}

However, it's functionally equivalent to:

         b = (b & ~0x80) | (a & 0x80);

Which generates this:

define i32 @func(i32 %a, i32 %b) nounwind readnone ssp {
entry:
  %0 = and i32 %b, -129                           ; <i32> [#uses=1]
  %1 = and i32 %a, 128                            ; <i32> [#uses=1]
  %2 = or i32 %0, %1                              ; <i32> [#uses=1]
  ret i32 %2
}

This can be generalized for other forms:

     b = (b & ~0x80) | (a & 0x40) << 1;

//===---------------------------------------------------------------------===//

These two functions produce different code. They shouldn't:

#include <stdint.h>
 
uint8_t p1(uint8_t b, uint8_t a) {
  b = (b & ~0xc0) | (a & 0xc0);
  return (b);
}
 
uint8_t p2(uint8_t b, uint8_t a) {
  b = (b & ~0x40) | (a & 0x40);
  b = (b & ~0x80) | (a & 0x80);
  return (b);
}

define zeroext i8 @p1(i8 zeroext %b, i8 zeroext %a) nounwind readnone ssp {
entry:
  %0 = and i8 %b, 63                              ; <i8> [#uses=1]
  %1 = and i8 %a, -64                             ; <i8> [#uses=1]
  %2 = or i8 %1, %0                               ; <i8> [#uses=1]
  ret i8 %2
}

define zeroext i8 @p2(i8 zeroext %b, i8 zeroext %a) nounwind readnone ssp {
entry:
  %0 = and i8 %b, 63                              ; <i8> [#uses=1]
  %.masked = and i8 %a, 64                        ; <i8> [#uses=1]
  %1 = and i8 %a, -128                            ; <i8> [#uses=1]
  %2 = or i8 %1, %0                               ; <i8> [#uses=1]
  %3 = or i8 %2, %.masked                         ; <i8> [#uses=1]
  ret i8 %3
}

//===---------------------------------------------------------------------===//

IPSCCP does not currently propagate argument dependent constants through
functions where it does not not all of the callers.  This includes functions
with normal external linkage as well as templates, C99 inline functions etc.
Specifically, it does nothing to:

define i32 @test(i32 %x, i32 %y, i32 %z) nounwind {
entry:
  %0 = add nsw i32 %y, %z                         
  %1 = mul i32 %0, %x                             
  %2 = mul i32 %y, %z                             
  %3 = add nsw i32 %1, %2                         
  ret i32 %3
}

define i32 @test2() nounwind {
entry:
  %0 = call i32 @test(i32 1, i32 2, i32 4) nounwind
  ret i32 %0
}

It would be interesting extend IPSCCP to be able to handle simple cases like
this, where all of the arguments to a call are constant.  Because IPSCCP runs
before inlining, trivial templates and inline functions are not yet inlined.
The results for a function + set of constant arguments should be memoized in a
map.

//===---------------------------------------------------------------------===//

The libcall constant folding stuff should be moved out of SimplifyLibcalls into
libanalysis' constantfolding logic.  This would allow IPSCCP to be able to
handle simple things like this:

static int foo(const char *X) { return strlen(X); }
int bar() { return foo("abcd"); }

//===---------------------------------------------------------------------===//

InstCombine should use SimplifyDemandedBits to remove the or instruction:

define i1 @test(i8 %x, i8 %y) {
  %A = or i8 %x, 1
  %B = icmp ugt i8 %A, 3
  ret i1 %B
}

Currently instcombine calls SimplifyDemandedBits with either all bits or just
the sign bit, if the comparison is obviously a sign test. In this case, we only
need all but the bottom two bits from %A, and if we gave that mask to SDB it
would delete the or instruction for us.

//===---------------------------------------------------------------------===//

functionattrs doesn't know much about memcpy/memset.  This function should be
marked readnone rather than readonly, since it only twiddles local memory, but
functionattrs doesn't handle memset/memcpy/memmove aggressively:

struct X { int *p; int *q; };
int foo() {
 int i = 0, j = 1;
 struct X x, y;
 int **p;
 y.p = &i;
 x.q = &j;
 p = __builtin_memcpy (&x, &y, sizeof (int *));
 return **p;
}

//===---------------------------------------------------------------------===//

Missed instcombine transformation:
define i1 @a(i32 %x) nounwind readnone {
entry:
  %cmp = icmp eq i32 %x, 30
  %sub = add i32 %x, -30
  %cmp2 = icmp ugt i32 %sub, 9
  %or = or i1 %cmp, %cmp2
  ret i1 %or
}
This should be optimized to a single compare.  Testcase derived from gcc.

//===---------------------------------------------------------------------===//

Missed instcombine transformation:
void b();
void a(int x) { if (((1<<x)&8)==0) b(); }

The shift should be optimized out.  Testcase derived from gcc.

//===---------------------------------------------------------------------===//

Missed instcombine or reassociate transformation:
int a(int a, int b) { return (a==12)&(b>47)&(b<58); }

The sgt and slt should be combined into a single comparison. Testcase derived
from gcc.

//===---------------------------------------------------------------------===//

Missed instcombine transformation:
define i32 @a(i32 %x) nounwind readnone {
entry:
  %rem = srem i32 %x, 32
  %shl = shl i32 1, %rem
  ret i32 %shl
}

The srem can be transformed to an and because if x is negative, the shift is
undefined. Testcase derived from gcc.

//===---------------------------------------------------------------------===//

Missed instcombine/dagcombine transformation:
define i32 @a(i32 %x, i32 %y) nounwind readnone {
entry:
  %mul = mul i32 %y, -8
  %sub = sub i32 %x, %mul
  ret i32 %sub
}

Should compile to something like x+y*8, but currently compiles to an
inefficient result.  Testcase derived from gcc.

//===---------------------------------------------------------------------===//

Missed instcombine/dagcombine transformation:
define void @lshift_lt(i8 zeroext %a) nounwind {
entry:
  %conv = zext i8 %a to i32
  %shl = shl i32 %conv, 3
  %cmp = icmp ult i32 %shl, 33
  br i1 %cmp, label %if.then, label %if.end

if.then:
  tail call void @bar() nounwind
  ret void

if.end:
  ret void
}
declare void @bar() nounwind

The shift should be eliminated.  Testcase derived from gcc.

//===---------------------------------------------------------------------===//

These compile into different code, one gets recognized as a switch and the
other doesn't due to phase ordering issues (PR6212):

int test1(int mainType, int subType) {
  if (mainType == 7)
    subType = 4;
  else if (mainType == 9)
    subType = 6;
  else if (mainType == 11)
    subType = 9;
  return subType;
}

int test2(int mainType, int subType) {
  if (mainType == 7)
    subType = 4;
  if (mainType == 9)
    subType = 6;
  if (mainType == 11)
    subType = 9;
  return subType;
}

//===---------------------------------------------------------------------===//