llvm-6502/lib
Chris Lattner e2ed057562 Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allows
us to compile oh-so-realistic stuff like this:

 vec_vperm(A, B, (vector unsigned char){14});

to:
        vspltb v0, v0, 14

instead of:

        vspltisb v0, 14
        vperm v0, v2, v1, v0


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@27452 91177308-0d34-0410-b5e6-96231b3b80d8
2006-04-06 19:19:17 +00:00
..
Analysis revert this, this is safe, if conservative. leave a note to that effect 2006-04-05 02:42:36 +00:00
Archive
AsmParser Add new generated files. 2006-03-23 23:48:12 +00:00
Bytecode
CodeGen Make a vector live across blocks have the correct Vec type. This fixes 2006-04-05 06:54:42 +00:00
Debugger Handle new forms of llvm.dbg intrinsics. 2006-03-23 18:06:46 +00:00
ExecutionEngine Get JIT/Interpreter working on Windows again. 2006-03-24 02:53:49 +00:00
Linker
Support
System
Target Support pattern matching vsldoi(x,y) and vsldoi(x,x), which allows the f.e. to 2006-04-06 18:26:28 +00:00
Transforms Lower vperm(x,y, mask) -> shuffle(x,y,mask) if mask is constant. This allows 2006-04-06 19:19:17 +00:00
VMCore vector casts never reinterpret bits 2006-04-02 05:40:28 +00:00
Makefile