- Mar 04, 2005
-
-
Jeff Cohen authored
power of two. This emphatically includes the zeroeth power of two. llvm-svn: 20429
-
- Mar 03, 2005
-
-
Andrew Lenharth authored
llvm-svn: 20425
-
Andrew Lenharth authored
llvm-svn: 20424
-
Chris Lattner authored
double test(double l1_X) { return (-l1_X); } instead of like this: double test(double l1_X) { return (-0x0p+0 - l1_X); } llvm-svn: 20423
-
Andrew Lenharth authored
llvm-svn: 20422
-
Chris Lattner authored
ltmp_0_7 = malloc(((unsigned )(&(((signed char (*)[784])/*NULL*/0)[1u])))); Instead, just emit the literal constant, like this: ltmp_0_7 = malloc(784u); This works around a bug in ICC 8.1 compiling the CBE generated code. :-( llvm-svn: 20415
-
Chris Lattner authored
to a "sizeof" expression. llvm-svn: 20414
-
Misha Brukman authored
llvm-svn: 20412
-
Chris Lattner authored
llvm-svn: 20411
-
- Mar 02, 2005
-
-
Chris Lattner authored
llvm-svn: 20410
-
Andrew Lenharth authored
llvm-svn: 20408
-
Andrew Lenharth authored
llvm-svn: 20407
-
Chris Lattner authored
Loop Strength Reduction. llvm-svn: 20399
-
Reid Spencer authored
llvm-svn: 20397
-
Chris Lattner authored
explanation from IRC: |sabre| I think it's an order of evaluation thing |sabre| for me, the RHS of the assignment is evaluated first |sabre| getTypeDescription checks to see if ConcreteTypeDescription[Ty] contains anything |sabre| since it doesn't, it computes and returns the value |sabre| this gets put into the map. |sabre| For you, the LHS is evaluated first. |sabre| Map[Ty] (aka ConcreteTypeDescriptions[Ty]) inserts an empty string into the map, returning a reference |sabre| getTypeDesc then sees the empty string in the map |sabre| and returns it |sabre| bork :) llvm-svn: 20394
-
- Mar 01, 2005
-
-
Jeff Cohen authored
* Loop invariant code does not dominate the loop header, but rather the end of the loop preheader. * The base for a reduced GEP isn't a constant unless all of its operands (preceding the induction variable) are constant. * Allow induction variable elimination for the simple case after all. Also made changes recommended by Chris for properly deleting instructions. llvm-svn: 20383
-
Alkis Evlogimenos authored
llvm-svn: 20382
-
- Feb 28, 2005
-
-
Chris Lattner authored
llvm-svn: 20380
-
Chris Lattner authored
Justin Wick! llvm-svn: 20378
-
Chris Lattner authored
llvm-svn: 20377
-
Chris Lattner authored
llvm-svn: 20376
-
Chris Lattner authored
llvm-svn: 20375
-
Andrew Lenharth authored
llvm-svn: 20372
-
Jeff Cohen authored
for reasons explained in the comments, I also deactivated this code as it needs more thought. llvm-svn: 20367
-
- Feb 27, 2005
-
-
Jeff Cohen authored
llvm-svn: 20360
-
Jeff Cohen authored
llvm-svn: 20358
-
Chris Lattner authored
llvm-svn: 20354
-
Chris Lattner authored
llvm-svn: 20352
-
Chris Lattner authored
llvm-svn: 20343
-
Chris Lattner authored
llvm-svn: 20342
-
Chris Lattner authored
llvm-svn: 20340
-
Chris Lattner authored
llvm-svn: 20339
-
- Feb 26, 2005
-
-
Reid Spencer authored
bytecode file members or not. Patch Contributed By Adam Treat llvm-svn: 20338
-
Chris Lattner authored
If we fold three constants together (c1+c2+c3), make sure to keep LHSC updated, instead of reusing (in this case), the 1 instead of the partial sum. llvm-svn: 20337
-
Chris Lattner authored
llvm-svn: 20334
-
- Feb 25, 2005
-
-
Andrew Lenharth authored
make BB labels be exported for debuging, add fp negation optimization, further pecimise the FP instructions llvm-svn: 20332
-
Chris Lattner authored
Actually teach dsa about select instructions. This doesn't affect the graph in any way other than not setting a spurious U marker on pointer nodes that are selected. llvm-svn: 20324
-
- Feb 24, 2005
-
-
Chris Lattner authored
X = gep null, ... Used to not create a scalar map entry for X, which caused clients to barf. This is bad. llvm-svn: 20316
-
Chris Lattner authored
llvm-svn: 20308
-
Andrew Lenharth authored
llvm-svn: 20306
-