- Jan 17, 2005
-
-
Chris Lattner authored
llvm-svn: 19635
-
Chris Lattner authored
llvm-svn: 19634
-
Chris Lattner authored
int %foo(int %X) { %T = add int %X, 13 %S = mul int %T, 3 ret int %S } as this: mov %ECX, DWORD PTR [%ESP + 4] lea %EAX, DWORD PTR [%ECX + 2*%ECX + 39] ret instead of this: mov %ECX, DWORD PTR [%ESP + 4] mov %EAX, %ECX add %EAX, 13 imul %EAX, %EAX, 3 ret llvm-svn: 19633
-
Tanya Lattner authored
llvm-svn: 19632
-
Chris Lattner authored
Do not fold a load into an operation if it will induce a cycle in the DAG. Repeat after me: dAg. llvm-svn: 19631
-
Chris Lattner authored
llvm-svn: 19630
-
Chris Lattner authored
useness. llvm-svn: 19629
-
Chris Lattner authored
Disable the xform for < > cases. It turns out that the following is being miscompiled: bool %test(sbyte %S) { %T = cast sbyte %S to uint %V = setgt uint %T, 255 ret bool %V } llvm-svn: 19628
-
Chris Lattner authored
llvm-svn: 19627
-
Chris Lattner authored
The comparison will probably be folded, so this is not ok to do. This fixed 197.parser. llvm-svn: 19624
-
Reid Spencer authored
llvm-svn: 19623
-
Chris Lattner authored
of the bytereg. This fixes yacr2, 300.twolf and probably others. llvm-svn: 19622
-
Chris Lattner authored
llvm-svn: 19621
-
Chris Lattner authored
If we emit a load because we followed a token chain to get to it, try to fold it into its single user if possible. llvm-svn: 19620
-
Chris Lattner authored
llvm-svn: 19619
-
Chris Lattner authored
Add fields to hold the result type of setcc operations and shift amounts. llvm-svn: 19618
-
- Jan 16, 2005
-
-
Chris Lattner authored
llvm-svn: 19617
-
Alkis Evlogimenos authored
llvm-svn: 19616
-
Tanya Lattner authored
llvm-svn: 19615
-
Chris Lattner authored
* Remove custom promotion for bool and byte select ops. Legalize now promotes them for us. * Allow folding ConstantPoolIndexes into EXTLOAD's, useful for float immediates. * Declare which operations are not supported better. * Add some hacky code for TRUNCSTORE to pretend that we have truncstore for i16 types. This is useful for testing promotion code because I can just remove 16-bit registers all together and verify that programs work. llvm-svn: 19614
-
Chris Lattner authored
track of how to deal with it, and provide the target with a hook that they can use to legalize arbitrary operations in arbitrary ways. Implement custom lowering for a couple of ops, implement promotion for select operations (which x86 needs). llvm-svn: 19613
-
Chris Lattner authored
llvm-svn: 19612
-
Chris Lattner authored
llvm-svn: 19611
-
Chris Lattner authored
llvm-svn: 19610
-
Chris Lattner authored
track of how to deal with it, and provide the target with a hook that they can use to legalize arbitrary operations in arbitrary ways. llvm-svn: 19609
-
Reid Spencer authored
* the dist-clean target no longer implies a check llvm-svn: 19608
-
Reid Spencer authored
a symbolic link making the autoconf name for the directory (LLVM_OBJ_ROOT) and the "make" name for the directory (PROJ_OBJ_ROOT) different. llvm-svn: 19607
-
Chris Lattner authored
llvm-svn: 19606
-
Chris Lattner authored
llvm-svn: 19605
-
Chris Lattner authored
llvm-svn: 19604
-
Jeff Cohen authored
llvm-svn: 19603
-
Reid Spencer authored
(RogueWave). These are implemented in rw/stdex/hash_map.h and rw/stdex/hash_set.h on HP/UX. llvm-svn: 19600
-
Reid Spencer authored
llvm-svn: 19599
-
Reid Spencer authored
llvm-svn: 19598
-
Chris Lattner authored
llvm-svn: 19597
-
Chris Lattner authored
llvm-svn: 19596
-
Chris Lattner authored
llvm-svn: 19595
-
Reid Spencer authored
llvm-svn: 19594
-
Reid Spencer authored
llvm-svn: 19592
-
Reid Spencer authored
llvm-svn: 19591
-