mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-08-19 19:29:31 +00:00
XMM registers. There are many known deficiencies and fixmes, which will be addressed ASAP. The major benefit of this work is that it will allow the LLVM register allocator to allocate FP registers across basic blocks. The x86 backend will still default to x87 style FP. To enable this work, you must pass -enable-sse-scalar-fp and either -sse2 or -sse3 to llc. An example before and after would be for: double foo(double *P) { double Sum = 0; int i; for (i = 0; i < 1000; ++i) Sum += P[i]; return Sum; } The inner loop looks like the following: x87: .LBB_foo_1: # no_exit fldl (%esp) faddl (%eax,%ecx,8) fstpl (%esp) incl %ecx cmpl $1000, %ecx #FP_REG_KILL jne .LBB_foo_1 # no_exit SSE2: addsd (%eax,%ecx,8), %xmm0 incl %ecx cmpl $1000, %ecx #FP_REG_KILL jne .LBB_foo_1 # no_exit git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@22340 91177308-0d34-0410-b5e6-96231b3b80d8
97 lines
3.2 KiB
C++
97 lines
3.2 KiB
C++
//===-- X86.h - Top-level interface for X86 representation ------*- C++ -*-===//
|
|
//
|
|
// The LLVM Compiler Infrastructure
|
|
//
|
|
// This file was developed by the LLVM research group and is distributed under
|
|
// the University of Illinois Open Source License. See LICENSE.TXT for details.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
// This file contains the entry points for global functions defined in the x86
|
|
// target library, as used by the LLVM JIT.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#ifndef TARGET_X86_H
|
|
#define TARGET_X86_H
|
|
|
|
#include <iosfwd>
|
|
|
|
namespace llvm {
|
|
|
|
class TargetMachine;
|
|
class FunctionPass;
|
|
class IntrinsicLowering;
|
|
|
|
enum X86VectorEnum {
|
|
NoSSE, SSE, SSE2, SSE3
|
|
};
|
|
|
|
extern X86VectorEnum X86Vector;
|
|
extern bool X86ScalarSSE;
|
|
|
|
/// createX86SimpleInstructionSelector - This pass converts an LLVM function
|
|
/// into a machine code representation in a very simple peep-hole fashion. The
|
|
/// generated code sucks but the implementation is nice and simple.
|
|
///
|
|
FunctionPass *createX86SimpleInstructionSelector(TargetMachine &TM);
|
|
|
|
/// createX86PatternInstructionSelector - This pass converts an LLVM function
|
|
/// into a machine code representation in a more aggressive way.
|
|
///
|
|
FunctionPass *createX86PatternInstructionSelector(TargetMachine &TM);
|
|
|
|
/// createX86SSAPeepholeOptimizerPass - Create a pass to perform SSA-based X86
|
|
/// specific peephole optimizations.
|
|
///
|
|
FunctionPass *createX86SSAPeepholeOptimizerPass();
|
|
|
|
/// createX86PeepholeOptimizer - Create a pass to perform X86 specific peephole
|
|
/// optimizations.
|
|
///
|
|
FunctionPass *createX86PeepholeOptimizerPass();
|
|
|
|
/// createX86FloatingPointKiller - This function returns a pass which
|
|
/// kills every floating point register at the end of each basic block
|
|
/// because our FloatingPointStackifier cannot handle them.
|
|
///
|
|
FunctionPass *createX86FloatingPointKillerPass();
|
|
|
|
/// createX86FloatingPointStackifierPass - This function returns a pass which
|
|
/// converts floating point register references and pseudo instructions into
|
|
/// floating point stack references and physical instructions.
|
|
///
|
|
FunctionPass *createX86FloatingPointStackifierPass();
|
|
|
|
/// createX86CodePrinterPass - Returns a pass that prints the X86
|
|
/// assembly code for a MachineFunction to the given output stream,
|
|
/// using the given target machine description. This should work
|
|
/// regardless of whether the function is in SSA form.
|
|
///
|
|
FunctionPass *createX86CodePrinterPass(std::ostream &o,TargetMachine &tm);
|
|
|
|
/// createX86ELFObjectWriterPass - Returns a pass that outputs the generated
|
|
/// code as an ELF object file.
|
|
///
|
|
FunctionPass *createX86ELFObjectWriterPass(std::ostream &o, TargetMachine &tm);
|
|
|
|
|
|
/// createX86EmitCodeToMemory - Returns a pass that converts a register
|
|
/// allocated function into raw machine code in a dynamically
|
|
/// allocated chunk of memory.
|
|
///
|
|
FunctionPass *createEmitX86CodeToMemory();
|
|
|
|
} // End llvm namespace
|
|
|
|
// Defines symbolic names for X86 registers. This defines a mapping from
|
|
// register name to register number.
|
|
//
|
|
#include "X86GenRegisterNames.inc"
|
|
|
|
// Defines symbolic names for the X86 instructions.
|
|
//
|
|
#include "X86GenInstrNames.inc"
|
|
|
|
#endif
|