2012-02-18 12:03:15 +00:00
|
|
|
//===-- X86InstrInfo.cpp - X86 Instruction Information --------------------===//
|
2005-04-21 23:38:14 +00:00
|
|
|
//
|
2003-10-20 19:43:21 +00:00
|
|
|
// The LLVM Compiler Infrastructure
|
|
|
|
//
|
2007-12-29 20:36:04 +00:00
|
|
|
// This file is distributed under the University of Illinois Open Source
|
|
|
|
// License. See LICENSE.TXT for details.
|
2005-04-21 23:38:14 +00:00
|
|
|
//
|
2003-10-20 19:43:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
2002-10-25 22:55:53 +00:00
|
|
|
//
|
2003-01-14 22:00:31 +00:00
|
|
|
// This file contains the X86 implementation of the TargetInstrInfo class.
|
2002-10-25 22:55:53 +00:00
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2002-10-29 21:05:24 +00:00
|
|
|
#include "X86InstrInfo.h"
|
2002-12-03 05:42:53 +00:00
|
|
|
#include "X86.h"
|
2005-01-02 02:37:07 +00:00
|
|
|
#include "X86InstrBuilder.h"
|
2008-01-04 23:57:37 +00:00
|
|
|
#include "X86MachineFunctionInfo.h"
|
2006-05-30 21:45:53 +00:00
|
|
|
#include "X86Subtarget.h"
|
|
|
|
#include "X86TargetMachine.h"
|
2007-09-07 04:06:50 +00:00
|
|
|
#include "llvm/ADT/STLExtras.h"
|
2012-12-03 16:50:05 +00:00
|
|
|
#include "llvm/CodeGen/LiveVariables.h"
|
2008-12-03 05:21:24 +00:00
|
|
|
#include "llvm/CodeGen/MachineConstantPool.h"
|
2012-06-01 16:27:21 +00:00
|
|
|
#include "llvm/CodeGen/MachineDominators.h"
|
2008-01-04 23:57:37 +00:00
|
|
|
#include "llvm/CodeGen/MachineFrameInfo.h"
|
2003-05-24 00:09:50 +00:00
|
|
|
#include "llvm/CodeGen/MachineInstrBuilder.h"
|
2007-12-31 04:13:23 +00:00
|
|
|
#include "llvm/CodeGen/MachineRegisterInfo.h"
|
2013-01-02 11:36:10 +00:00
|
|
|
#include "llvm/IR/DerivedTypes.h"
|
|
|
|
#include "llvm/IR/LLVMContext.h"
|
2012-03-17 18:46:09 +00:00
|
|
|
#include "llvm/MC/MCAsmInfo.h"
|
2010-04-26 23:37:21 +00:00
|
|
|
#include "llvm/MC/MCInst.h"
|
2008-01-07 01:35:02 +00:00
|
|
|
#include "llvm/Support/CommandLine.h"
|
2010-01-05 01:29:29 +00:00
|
|
|
#include "llvm/Support/Debug.h"
|
2009-07-08 18:01:40 +00:00
|
|
|
#include "llvm/Support/ErrorHandling.h"
|
|
|
|
#include "llvm/Support/raw_ostream.h"
|
2007-09-25 01:57:46 +00:00
|
|
|
#include "llvm/Target/TargetOptions.h"
|
2009-11-12 20:55:29 +00:00
|
|
|
#include <limits>
|
|
|
|
|
2011-07-01 17:57:27 +00:00
|
|
|
#define GET_INSTRINFO_CTOR
|
2011-06-28 20:07:07 +00:00
|
|
|
#include "X86GenInstrInfo.inc"
|
|
|
|
|
2003-11-11 22:41:34 +00:00
|
|
|
using namespace llvm;
|
|
|
|
|
2009-08-23 03:41:05 +00:00
|
|
|
static cl::opt<bool>
|
|
|
|
NoFusing("disable-spill-fusing",
|
|
|
|
cl::desc("Disable fusing of spill code into instructions"));
|
|
|
|
static cl::opt<bool>
|
|
|
|
PrintFailedFusing("print-failed-fuse-candidates",
|
|
|
|
cl::desc("Print instructions that the allocator wants to"
|
|
|
|
" fuse, but the X86 backend currently can't"),
|
|
|
|
cl::Hidden);
|
|
|
|
static cl::opt<bool>
|
|
|
|
ReMatPICStubLoad("remat-pic-stub-load",
|
|
|
|
cl::desc("Re-materialize load from stub in PIC mode"),
|
|
|
|
cl::init(false), cl::Hidden);
|
2008-01-07 01:35:02 +00:00
|
|
|
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
enum {
|
|
|
|
// Select which memory operand is being unfolded.
|
2012-06-23 08:01:18 +00:00
|
|
|
// (stored in bits 0 - 3)
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
TB_INDEX_0 = 0,
|
|
|
|
TB_INDEX_1 = 1,
|
|
|
|
TB_INDEX_2 = 2,
|
2012-05-31 09:20:20 +00:00
|
|
|
TB_INDEX_3 = 3,
|
2012-06-23 08:01:18 +00:00
|
|
|
TB_INDEX_MASK = 0xf,
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
|
|
|
|
// Do not insert the reverse map (MemOp -> RegOp) into the table.
|
|
|
|
// This may be needed because there is a many -> one mapping.
|
2012-06-23 08:01:18 +00:00
|
|
|
TB_NO_REVERSE = 1 << 4,
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
|
|
|
|
// Do not insert the forward map (RegOp -> MemOp) into the table.
|
|
|
|
// This is needed for Native Client, which prohibits branch
|
|
|
|
// instructions from using a memory operand.
|
2012-06-23 08:01:18 +00:00
|
|
|
TB_NO_FORWARD = 1 << 5,
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
|
2012-06-23 08:01:18 +00:00
|
|
|
TB_FOLDED_LOAD = 1 << 6,
|
|
|
|
TB_FOLDED_STORE = 1 << 7,
|
|
|
|
|
|
|
|
// Minimum alignment required for load/store.
|
|
|
|
// Used for RegOp->MemOp conversion.
|
|
|
|
// (stored in bits 8 - 15)
|
|
|
|
TB_ALIGN_SHIFT = 8,
|
|
|
|
TB_ALIGN_NONE = 0 << TB_ALIGN_SHIFT,
|
|
|
|
TB_ALIGN_16 = 16 << TB_ALIGN_SHIFT,
|
|
|
|
TB_ALIGN_32 = 32 << TB_ALIGN_SHIFT,
|
2013-08-11 07:55:09 +00:00
|
|
|
TB_ALIGN_64 = 64 << TB_ALIGN_SHIFT,
|
2012-06-23 08:01:18 +00:00
|
|
|
TB_ALIGN_MASK = 0xff << TB_ALIGN_SHIFT
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
};
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
struct X86OpTblEntry {
|
|
|
|
uint16_t RegOp;
|
|
|
|
uint16_t MemOp;
|
2012-06-23 08:01:18 +00:00
|
|
|
uint16_t Flags;
|
2012-03-09 07:45:21 +00:00
|
|
|
};
|
|
|
|
|
2006-05-30 21:45:53 +00:00
|
|
|
X86InstrInfo::X86InstrInfo(X86TargetMachine &tm)
|
2011-07-01 17:57:27 +00:00
|
|
|
: X86GenInstrInfo((tm.getSubtarget<X86Subtarget>().is64Bit()
|
|
|
|
? X86::ADJCALLSTACKDOWN64
|
|
|
|
: X86::ADJCALLSTACKDOWN32),
|
|
|
|
(tm.getSubtarget<X86Subtarget>().is64Bit()
|
|
|
|
? X86::ADJCALLSTACKUP64
|
|
|
|
: X86::ADJCALLSTACKUP32)),
|
2013-06-07 21:00:34 +00:00
|
|
|
TM(tm), RI(tm) {
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const X86OpTblEntry OpTbl2Addr[] = {
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::ADC32ri, X86::ADC32mi, 0 },
|
|
|
|
{ X86::ADC32ri8, X86::ADC32mi8, 0 },
|
|
|
|
{ X86::ADC32rr, X86::ADC32mr, 0 },
|
|
|
|
{ X86::ADC64ri32, X86::ADC64mi32, 0 },
|
|
|
|
{ X86::ADC64ri8, X86::ADC64mi8, 0 },
|
|
|
|
{ X86::ADC64rr, X86::ADC64mr, 0 },
|
|
|
|
{ X86::ADD16ri, X86::ADD16mi, 0 },
|
|
|
|
{ X86::ADD16ri8, X86::ADD16mi8, 0 },
|
|
|
|
{ X86::ADD16ri_DB, X86::ADD16mi, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD16ri8_DB, X86::ADD16mi8, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD16rr, X86::ADD16mr, 0 },
|
|
|
|
{ X86::ADD16rr_DB, X86::ADD16mr, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD32ri, X86::ADD32mi, 0 },
|
|
|
|
{ X86::ADD32ri8, X86::ADD32mi8, 0 },
|
|
|
|
{ X86::ADD32ri_DB, X86::ADD32mi, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD32ri8_DB, X86::ADD32mi8, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD32rr, X86::ADD32mr, 0 },
|
|
|
|
{ X86::ADD32rr_DB, X86::ADD32mr, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD64ri32, X86::ADD64mi32, 0 },
|
|
|
|
{ X86::ADD64ri8, X86::ADD64mi8, 0 },
|
|
|
|
{ X86::ADD64ri32_DB,X86::ADD64mi32, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD64ri8_DB, X86::ADD64mi8, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD64rr, X86::ADD64mr, 0 },
|
|
|
|
{ X86::ADD64rr_DB, X86::ADD64mr, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD8ri, X86::ADD8mi, 0 },
|
|
|
|
{ X86::ADD8rr, X86::ADD8mr, 0 },
|
|
|
|
{ X86::AND16ri, X86::AND16mi, 0 },
|
|
|
|
{ X86::AND16ri8, X86::AND16mi8, 0 },
|
|
|
|
{ X86::AND16rr, X86::AND16mr, 0 },
|
|
|
|
{ X86::AND32ri, X86::AND32mi, 0 },
|
|
|
|
{ X86::AND32ri8, X86::AND32mi8, 0 },
|
|
|
|
{ X86::AND32rr, X86::AND32mr, 0 },
|
|
|
|
{ X86::AND64ri32, X86::AND64mi32, 0 },
|
|
|
|
{ X86::AND64ri8, X86::AND64mi8, 0 },
|
|
|
|
{ X86::AND64rr, X86::AND64mr, 0 },
|
|
|
|
{ X86::AND8ri, X86::AND8mi, 0 },
|
|
|
|
{ X86::AND8rr, X86::AND8mr, 0 },
|
|
|
|
{ X86::DEC16r, X86::DEC16m, 0 },
|
|
|
|
{ X86::DEC32r, X86::DEC32m, 0 },
|
|
|
|
{ X86::DEC64_16r, X86::DEC64_16m, 0 },
|
|
|
|
{ X86::DEC64_32r, X86::DEC64_32m, 0 },
|
|
|
|
{ X86::DEC64r, X86::DEC64m, 0 },
|
|
|
|
{ X86::DEC8r, X86::DEC8m, 0 },
|
|
|
|
{ X86::INC16r, X86::INC16m, 0 },
|
|
|
|
{ X86::INC32r, X86::INC32m, 0 },
|
|
|
|
{ X86::INC64_16r, X86::INC64_16m, 0 },
|
|
|
|
{ X86::INC64_32r, X86::INC64_32m, 0 },
|
|
|
|
{ X86::INC64r, X86::INC64m, 0 },
|
|
|
|
{ X86::INC8r, X86::INC8m, 0 },
|
|
|
|
{ X86::NEG16r, X86::NEG16m, 0 },
|
|
|
|
{ X86::NEG32r, X86::NEG32m, 0 },
|
|
|
|
{ X86::NEG64r, X86::NEG64m, 0 },
|
|
|
|
{ X86::NEG8r, X86::NEG8m, 0 },
|
|
|
|
{ X86::NOT16r, X86::NOT16m, 0 },
|
|
|
|
{ X86::NOT32r, X86::NOT32m, 0 },
|
|
|
|
{ X86::NOT64r, X86::NOT64m, 0 },
|
|
|
|
{ X86::NOT8r, X86::NOT8m, 0 },
|
|
|
|
{ X86::OR16ri, X86::OR16mi, 0 },
|
|
|
|
{ X86::OR16ri8, X86::OR16mi8, 0 },
|
|
|
|
{ X86::OR16rr, X86::OR16mr, 0 },
|
|
|
|
{ X86::OR32ri, X86::OR32mi, 0 },
|
|
|
|
{ X86::OR32ri8, X86::OR32mi8, 0 },
|
|
|
|
{ X86::OR32rr, X86::OR32mr, 0 },
|
|
|
|
{ X86::OR64ri32, X86::OR64mi32, 0 },
|
|
|
|
{ X86::OR64ri8, X86::OR64mi8, 0 },
|
|
|
|
{ X86::OR64rr, X86::OR64mr, 0 },
|
|
|
|
{ X86::OR8ri, X86::OR8mi, 0 },
|
|
|
|
{ X86::OR8rr, X86::OR8mr, 0 },
|
|
|
|
{ X86::ROL16r1, X86::ROL16m1, 0 },
|
|
|
|
{ X86::ROL16rCL, X86::ROL16mCL, 0 },
|
|
|
|
{ X86::ROL16ri, X86::ROL16mi, 0 },
|
|
|
|
{ X86::ROL32r1, X86::ROL32m1, 0 },
|
|
|
|
{ X86::ROL32rCL, X86::ROL32mCL, 0 },
|
|
|
|
{ X86::ROL32ri, X86::ROL32mi, 0 },
|
|
|
|
{ X86::ROL64r1, X86::ROL64m1, 0 },
|
|
|
|
{ X86::ROL64rCL, X86::ROL64mCL, 0 },
|
|
|
|
{ X86::ROL64ri, X86::ROL64mi, 0 },
|
|
|
|
{ X86::ROL8r1, X86::ROL8m1, 0 },
|
|
|
|
{ X86::ROL8rCL, X86::ROL8mCL, 0 },
|
|
|
|
{ X86::ROL8ri, X86::ROL8mi, 0 },
|
|
|
|
{ X86::ROR16r1, X86::ROR16m1, 0 },
|
|
|
|
{ X86::ROR16rCL, X86::ROR16mCL, 0 },
|
|
|
|
{ X86::ROR16ri, X86::ROR16mi, 0 },
|
|
|
|
{ X86::ROR32r1, X86::ROR32m1, 0 },
|
|
|
|
{ X86::ROR32rCL, X86::ROR32mCL, 0 },
|
|
|
|
{ X86::ROR32ri, X86::ROR32mi, 0 },
|
|
|
|
{ X86::ROR64r1, X86::ROR64m1, 0 },
|
|
|
|
{ X86::ROR64rCL, X86::ROR64mCL, 0 },
|
|
|
|
{ X86::ROR64ri, X86::ROR64mi, 0 },
|
|
|
|
{ X86::ROR8r1, X86::ROR8m1, 0 },
|
|
|
|
{ X86::ROR8rCL, X86::ROR8mCL, 0 },
|
|
|
|
{ X86::ROR8ri, X86::ROR8mi, 0 },
|
|
|
|
{ X86::SAR16r1, X86::SAR16m1, 0 },
|
|
|
|
{ X86::SAR16rCL, X86::SAR16mCL, 0 },
|
|
|
|
{ X86::SAR16ri, X86::SAR16mi, 0 },
|
|
|
|
{ X86::SAR32r1, X86::SAR32m1, 0 },
|
|
|
|
{ X86::SAR32rCL, X86::SAR32mCL, 0 },
|
|
|
|
{ X86::SAR32ri, X86::SAR32mi, 0 },
|
|
|
|
{ X86::SAR64r1, X86::SAR64m1, 0 },
|
|
|
|
{ X86::SAR64rCL, X86::SAR64mCL, 0 },
|
|
|
|
{ X86::SAR64ri, X86::SAR64mi, 0 },
|
|
|
|
{ X86::SAR8r1, X86::SAR8m1, 0 },
|
|
|
|
{ X86::SAR8rCL, X86::SAR8mCL, 0 },
|
|
|
|
{ X86::SAR8ri, X86::SAR8mi, 0 },
|
|
|
|
{ X86::SBB32ri, X86::SBB32mi, 0 },
|
|
|
|
{ X86::SBB32ri8, X86::SBB32mi8, 0 },
|
|
|
|
{ X86::SBB32rr, X86::SBB32mr, 0 },
|
|
|
|
{ X86::SBB64ri32, X86::SBB64mi32, 0 },
|
|
|
|
{ X86::SBB64ri8, X86::SBB64mi8, 0 },
|
|
|
|
{ X86::SBB64rr, X86::SBB64mr, 0 },
|
|
|
|
{ X86::SHL16rCL, X86::SHL16mCL, 0 },
|
|
|
|
{ X86::SHL16ri, X86::SHL16mi, 0 },
|
|
|
|
{ X86::SHL32rCL, X86::SHL32mCL, 0 },
|
|
|
|
{ X86::SHL32ri, X86::SHL32mi, 0 },
|
|
|
|
{ X86::SHL64rCL, X86::SHL64mCL, 0 },
|
|
|
|
{ X86::SHL64ri, X86::SHL64mi, 0 },
|
|
|
|
{ X86::SHL8rCL, X86::SHL8mCL, 0 },
|
|
|
|
{ X86::SHL8ri, X86::SHL8mi, 0 },
|
|
|
|
{ X86::SHLD16rrCL, X86::SHLD16mrCL, 0 },
|
|
|
|
{ X86::SHLD16rri8, X86::SHLD16mri8, 0 },
|
|
|
|
{ X86::SHLD32rrCL, X86::SHLD32mrCL, 0 },
|
|
|
|
{ X86::SHLD32rri8, X86::SHLD32mri8, 0 },
|
|
|
|
{ X86::SHLD64rrCL, X86::SHLD64mrCL, 0 },
|
|
|
|
{ X86::SHLD64rri8, X86::SHLD64mri8, 0 },
|
|
|
|
{ X86::SHR16r1, X86::SHR16m1, 0 },
|
|
|
|
{ X86::SHR16rCL, X86::SHR16mCL, 0 },
|
|
|
|
{ X86::SHR16ri, X86::SHR16mi, 0 },
|
|
|
|
{ X86::SHR32r1, X86::SHR32m1, 0 },
|
|
|
|
{ X86::SHR32rCL, X86::SHR32mCL, 0 },
|
|
|
|
{ X86::SHR32ri, X86::SHR32mi, 0 },
|
|
|
|
{ X86::SHR64r1, X86::SHR64m1, 0 },
|
|
|
|
{ X86::SHR64rCL, X86::SHR64mCL, 0 },
|
|
|
|
{ X86::SHR64ri, X86::SHR64mi, 0 },
|
|
|
|
{ X86::SHR8r1, X86::SHR8m1, 0 },
|
|
|
|
{ X86::SHR8rCL, X86::SHR8mCL, 0 },
|
|
|
|
{ X86::SHR8ri, X86::SHR8mi, 0 },
|
|
|
|
{ X86::SHRD16rrCL, X86::SHRD16mrCL, 0 },
|
|
|
|
{ X86::SHRD16rri8, X86::SHRD16mri8, 0 },
|
|
|
|
{ X86::SHRD32rrCL, X86::SHRD32mrCL, 0 },
|
|
|
|
{ X86::SHRD32rri8, X86::SHRD32mri8, 0 },
|
|
|
|
{ X86::SHRD64rrCL, X86::SHRD64mrCL, 0 },
|
|
|
|
{ X86::SHRD64rri8, X86::SHRD64mri8, 0 },
|
|
|
|
{ X86::SUB16ri, X86::SUB16mi, 0 },
|
|
|
|
{ X86::SUB16ri8, X86::SUB16mi8, 0 },
|
|
|
|
{ X86::SUB16rr, X86::SUB16mr, 0 },
|
|
|
|
{ X86::SUB32ri, X86::SUB32mi, 0 },
|
|
|
|
{ X86::SUB32ri8, X86::SUB32mi8, 0 },
|
|
|
|
{ X86::SUB32rr, X86::SUB32mr, 0 },
|
|
|
|
{ X86::SUB64ri32, X86::SUB64mi32, 0 },
|
|
|
|
{ X86::SUB64ri8, X86::SUB64mi8, 0 },
|
|
|
|
{ X86::SUB64rr, X86::SUB64mr, 0 },
|
|
|
|
{ X86::SUB8ri, X86::SUB8mi, 0 },
|
|
|
|
{ X86::SUB8rr, X86::SUB8mr, 0 },
|
|
|
|
{ X86::XOR16ri, X86::XOR16mi, 0 },
|
|
|
|
{ X86::XOR16ri8, X86::XOR16mi8, 0 },
|
|
|
|
{ X86::XOR16rr, X86::XOR16mr, 0 },
|
|
|
|
{ X86::XOR32ri, X86::XOR32mi, 0 },
|
|
|
|
{ X86::XOR32ri8, X86::XOR32mi8, 0 },
|
|
|
|
{ X86::XOR32rr, X86::XOR32mr, 0 },
|
|
|
|
{ X86::XOR64ri32, X86::XOR64mi32, 0 },
|
|
|
|
{ X86::XOR64ri8, X86::XOR64mi8, 0 },
|
|
|
|
{ X86::XOR64rr, X86::XOR64mr, 0 },
|
|
|
|
{ X86::XOR8ri, X86::XOR8mi, 0 },
|
|
|
|
{ X86::XOR8rr, X86::XOR8mr, 0 }
|
2008-01-07 01:35:02 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (unsigned i = 0, e = array_lengthof(OpTbl2Addr); i != e; ++i) {
|
2012-03-09 07:45:21 +00:00
|
|
|
unsigned RegOp = OpTbl2Addr[i].RegOp;
|
|
|
|
unsigned MemOp = OpTbl2Addr[i].MemOp;
|
|
|
|
unsigned Flags = OpTbl2Addr[i].Flags;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
AddTableEntry(RegOp2MemOpTable2Addr, MemOp2RegOpTable,
|
|
|
|
RegOp, MemOp,
|
|
|
|
// Index 0, folded load and store, no alignment requirement.
|
|
|
|
Flags | TB_INDEX_0 | TB_FOLDED_LOAD | TB_FOLDED_STORE);
|
|
|
|
}
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const X86OpTblEntry OpTbl0[] = {
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::BT16ri8, X86::BT16mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::BT32ri8, X86::BT32mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::BT64ri8, X86::BT64mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CALL32r, X86::CALL32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CALL64r, X86::CALL64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP16ri, X86::CMP16mi, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP16ri8, X86::CMP16mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP16rr, X86::CMP16mr, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP32ri, X86::CMP32mi, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP32ri8, X86::CMP32mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP32rr, X86::CMP32mr, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP64ri32, X86::CMP64mi32, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP64ri8, X86::CMP64mi8, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP64rr, X86::CMP64mr, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP8ri, X86::CMP8mi, TB_FOLDED_LOAD },
|
|
|
|
{ X86::CMP8rr, X86::CMP8mr, TB_FOLDED_LOAD },
|
|
|
|
{ X86::DIV16r, X86::DIV16m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::DIV32r, X86::DIV32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::DIV64r, X86::DIV64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::DIV8r, X86::DIV8m, TB_FOLDED_LOAD },
|
2012-12-26 01:47:12 +00:00
|
|
|
{ X86::EXTRACTPSrr, X86::EXTRACTPSmr, TB_FOLDED_STORE },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::IDIV16r, X86::IDIV16m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IDIV32r, X86::IDIV32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IDIV64r, X86::IDIV64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IDIV8r, X86::IDIV8m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IMUL16r, X86::IMUL16m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IMUL32r, X86::IMUL32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IMUL64r, X86::IMUL64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::IMUL8r, X86::IMUL8m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::JMP32r, X86::JMP32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::JMP64r, X86::JMP64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::MOV16ri, X86::MOV16mi, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV16rr, X86::MOV16mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV32ri, X86::MOV32mi, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV32rr, X86::MOV32mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV64ri32, X86::MOV64mi32, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV64rr, X86::MOV64mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV8ri, X86::MOV8mi, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV8rr, X86::MOV8mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOV8rr_NOREX, X86::MOV8mr_NOREX, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVAPDrr, X86::MOVAPDmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::MOVAPSrr, X86::MOVAPSmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::MOVDQArr, X86::MOVDQAmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::MOVPDI2DIrr, X86::MOVPDI2DImr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVPQIto64rr,X86::MOVPQI2QImr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVSDto64rr, X86::MOVSDto64mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVSS2DIrr, X86::MOVSS2DImr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVUPDrr, X86::MOVUPDmr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MOVUPSrr, X86::MOVUPSmr, TB_FOLDED_STORE },
|
|
|
|
{ X86::MUL16r, X86::MUL16m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::MUL32r, X86::MUL32m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::MUL64r, X86::MUL64m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::MUL8r, X86::MUL8m, TB_FOLDED_LOAD },
|
|
|
|
{ X86::SETAEr, X86::SETAEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETAr, X86::SETAm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETBEr, X86::SETBEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETBr, X86::SETBm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETEr, X86::SETEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETGEr, X86::SETGEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETGr, X86::SETGm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETLEr, X86::SETLEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETLr, X86::SETLm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETNEr, X86::SETNEm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETNOr, X86::SETNOm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETNPr, X86::SETNPm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETNSr, X86::SETNSm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETOr, X86::SETOm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETPr, X86::SETPm, TB_FOLDED_STORE },
|
|
|
|
{ X86::SETSr, X86::SETSm, TB_FOLDED_STORE },
|
|
|
|
{ X86::TAILJMPr, X86::TAILJMPm, TB_FOLDED_LOAD },
|
|
|
|
{ X86::TAILJMPr64, X86::TAILJMPm64, TB_FOLDED_LOAD },
|
|
|
|
{ X86::TEST16ri, X86::TEST16mi, TB_FOLDED_LOAD },
|
|
|
|
{ X86::TEST32ri, X86::TEST32mi, TB_FOLDED_LOAD },
|
|
|
|
{ X86::TEST64ri32, X86::TEST64mi32, TB_FOLDED_LOAD },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::TEST8ri, X86::TEST8mi, TB_FOLDED_LOAD },
|
|
|
|
// AVX 128-bit versions of foldable instructions
|
2012-12-26 01:47:12 +00:00
|
|
|
{ X86::VEXTRACTPSrr,X86::VEXTRACTPSmr, TB_FOLDED_STORE },
|
2012-01-14 18:14:53 +00:00
|
|
|
{ X86::VEXTRACTF128rr, X86::VEXTRACTF128mr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMOVAPDrr, X86::VMOVAPDmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVAPSrr, X86::VMOVAPSmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVDQArr, X86::VMOVDQAmr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVPDI2DIrr,X86::VMOVPDI2DImr, TB_FOLDED_STORE },
|
|
|
|
{ X86::VMOVPQIto64rr, X86::VMOVPQI2QImr,TB_FOLDED_STORE },
|
|
|
|
{ X86::VMOVSDto64rr,X86::VMOVSDto64mr, TB_FOLDED_STORE },
|
|
|
|
{ X86::VMOVSS2DIrr, X86::VMOVSS2DImr, TB_FOLDED_STORE },
|
|
|
|
{ X86::VMOVUPDrr, X86::VMOVUPDmr, TB_FOLDED_STORE },
|
|
|
|
{ X86::VMOVUPSrr, X86::VMOVUPSmr, TB_FOLDED_STORE },
|
|
|
|
// AVX 256-bit foldable instructions
|
2012-01-14 18:14:53 +00:00
|
|
|
{ X86::VEXTRACTI128rr, X86::VEXTRACTI128mr, TB_FOLDED_STORE | TB_ALIGN_16 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMOVAPDYrr, X86::VMOVAPDYmr, TB_FOLDED_STORE | TB_ALIGN_32 },
|
|
|
|
{ X86::VMOVAPSYrr, X86::VMOVAPSYmr, TB_FOLDED_STORE | TB_ALIGN_32 },
|
|
|
|
{ X86::VMOVDQAYrr, X86::VMOVDQAYmr, TB_FOLDED_STORE | TB_ALIGN_32 },
|
|
|
|
{ X86::VMOVUPDYrr, X86::VMOVUPDYmr, TB_FOLDED_STORE },
|
2013-09-02 07:12:29 +00:00
|
|
|
{ X86::VMOVUPSYrr, X86::VMOVUPSYmr, TB_FOLDED_STORE },
|
|
|
|
// AVX-512 foldable instructions
|
|
|
|
{ X86::VMOVPDI2DIZrr,X86::VMOVPDI2DIZmr, TB_FOLDED_STORE }
|
2008-01-07 01:35:02 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (unsigned i = 0, e = array_lengthof(OpTbl0); i != e; ++i) {
|
2012-03-09 07:45:21 +00:00
|
|
|
unsigned RegOp = OpTbl0[i].RegOp;
|
|
|
|
unsigned MemOp = OpTbl0[i].MemOp;
|
|
|
|
unsigned Flags = OpTbl0[i].Flags;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
AddTableEntry(RegOp2MemOpTable0, MemOp2RegOpTable,
|
|
|
|
RegOp, MemOp, TB_INDEX_0 | Flags);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const X86OpTblEntry OpTbl1[] = {
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::CMP16rr, X86::CMP16rm, 0 },
|
|
|
|
{ X86::CMP32rr, X86::CMP32rm, 0 },
|
|
|
|
{ X86::CMP64rr, X86::CMP64rm, 0 },
|
|
|
|
{ X86::CMP8rr, X86::CMP8rm, 0 },
|
|
|
|
{ X86::CVTSD2SSrr, X86::CVTSD2SSrm, 0 },
|
|
|
|
{ X86::CVTSI2SD64rr, X86::CVTSI2SD64rm, 0 },
|
|
|
|
{ X86::CVTSI2SDrr, X86::CVTSI2SDrm, 0 },
|
|
|
|
{ X86::CVTSI2SS64rr, X86::CVTSI2SS64rm, 0 },
|
|
|
|
{ X86::CVTSI2SSrr, X86::CVTSI2SSrm, 0 },
|
|
|
|
{ X86::CVTSS2SDrr, X86::CVTSS2SDrm, 0 },
|
|
|
|
{ X86::CVTTSD2SI64rr, X86::CVTTSD2SI64rm, 0 },
|
|
|
|
{ X86::CVTTSD2SIrr, X86::CVTTSD2SIrm, 0 },
|
|
|
|
{ X86::CVTTSS2SI64rr, X86::CVTTSS2SI64rm, 0 },
|
|
|
|
{ X86::CVTTSS2SIrr, X86::CVTTSS2SIrm, 0 },
|
|
|
|
{ X86::IMUL16rri, X86::IMUL16rmi, 0 },
|
|
|
|
{ X86::IMUL16rri8, X86::IMUL16rmi8, 0 },
|
|
|
|
{ X86::IMUL32rri, X86::IMUL32rmi, 0 },
|
|
|
|
{ X86::IMUL32rri8, X86::IMUL32rmi8, 0 },
|
|
|
|
{ X86::IMUL64rri32, X86::IMUL64rmi32, 0 },
|
|
|
|
{ X86::IMUL64rri8, X86::IMUL64rmi8, 0 },
|
|
|
|
{ X86::Int_COMISDrr, X86::Int_COMISDrm, 0 },
|
|
|
|
{ X86::Int_COMISSrr, X86::Int_COMISSrm, 0 },
|
|
|
|
{ X86::CVTSD2SI64rr, X86::CVTSD2SI64rm, 0 },
|
|
|
|
{ X86::CVTSD2SIrr, X86::CVTSD2SIrm, 0 },
|
2012-06-15 07:02:58 +00:00
|
|
|
{ X86::CVTSS2SI64rr, X86::CVTSS2SI64rm, 0 },
|
|
|
|
{ X86::CVTSS2SIrr, X86::CVTSS2SIrm, 0 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::CVTTPD2DQrr, X86::CVTTPD2DQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::CVTTPS2DQrr, X86::CVTTPS2DQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::Int_CVTTSD2SI64rr,X86::Int_CVTTSD2SI64rm, 0 },
|
|
|
|
{ X86::Int_CVTTSD2SIrr, X86::Int_CVTTSD2SIrm, 0 },
|
|
|
|
{ X86::Int_CVTTSS2SI64rr,X86::Int_CVTTSS2SI64rm, 0 },
|
|
|
|
{ X86::Int_CVTTSS2SIrr, X86::Int_CVTTSS2SIrm, 0 },
|
|
|
|
{ X86::Int_UCOMISDrr, X86::Int_UCOMISDrm, 0 },
|
|
|
|
{ X86::Int_UCOMISSrr, X86::Int_UCOMISSrm, 0 },
|
|
|
|
{ X86::MOV16rr, X86::MOV16rm, 0 },
|
|
|
|
{ X86::MOV32rr, X86::MOV32rm, 0 },
|
|
|
|
{ X86::MOV64rr, X86::MOV64rm, 0 },
|
|
|
|
{ X86::MOV64toPQIrr, X86::MOVQI2PQIrm, 0 },
|
|
|
|
{ X86::MOV64toSDrr, X86::MOV64toSDrm, 0 },
|
|
|
|
{ X86::MOV8rr, X86::MOV8rm, 0 },
|
|
|
|
{ X86::MOVAPDrr, X86::MOVAPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVAPSrr, X86::MOVAPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVDDUPrr, X86::MOVDDUPrm, 0 },
|
|
|
|
{ X86::MOVDI2PDIrr, X86::MOVDI2PDIrm, 0 },
|
|
|
|
{ X86::MOVDI2SSrr, X86::MOVDI2SSrm, 0 },
|
|
|
|
{ X86::MOVDQArr, X86::MOVDQArm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVSHDUPrr, X86::MOVSHDUPrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVSLDUPrr, X86::MOVSLDUPrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVSX16rr8, X86::MOVSX16rm8, 0 },
|
|
|
|
{ X86::MOVSX32rr16, X86::MOVSX32rm16, 0 },
|
|
|
|
{ X86::MOVSX32rr8, X86::MOVSX32rm8, 0 },
|
|
|
|
{ X86::MOVSX64rr16, X86::MOVSX64rm16, 0 },
|
|
|
|
{ X86::MOVSX64rr32, X86::MOVSX64rm32, 0 },
|
|
|
|
{ X86::MOVSX64rr8, X86::MOVSX64rm8, 0 },
|
|
|
|
{ X86::MOVUPDrr, X86::MOVUPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVUPSrr, X86::MOVUPSrm, 0 },
|
|
|
|
{ X86::MOVZDI2PDIrr, X86::MOVZDI2PDIrm, 0 },
|
|
|
|
{ X86::MOVZQI2PQIrr, X86::MOVZQI2PQIrm, 0 },
|
|
|
|
{ X86::MOVZPQILo2PQIrr, X86::MOVZPQILo2PQIrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MOVZX16rr8, X86::MOVZX16rm8, 0 },
|
|
|
|
{ X86::MOVZX32rr16, X86::MOVZX32rm16, 0 },
|
|
|
|
{ X86::MOVZX32_NOREXrr8, X86::MOVZX32_NOREXrm8, 0 },
|
|
|
|
{ X86::MOVZX32rr8, X86::MOVZX32rm8, 0 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PABSBrr128, X86::PABSBrm128, TB_ALIGN_16 },
|
|
|
|
{ X86::PABSDrr128, X86::PABSDrm128, TB_ALIGN_16 },
|
|
|
|
{ X86::PABSWrr128, X86::PABSWrm128, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PSHUFDri, X86::PSHUFDmi, TB_ALIGN_16 },
|
|
|
|
{ X86::PSHUFHWri, X86::PSHUFHWmi, TB_ALIGN_16 },
|
|
|
|
{ X86::PSHUFLWri, X86::PSHUFLWmi, TB_ALIGN_16 },
|
|
|
|
{ X86::RCPPSr, X86::RCPPSm, TB_ALIGN_16 },
|
|
|
|
{ X86::RCPPSr_Int, X86::RCPPSm_Int, TB_ALIGN_16 },
|
|
|
|
{ X86::RSQRTPSr, X86::RSQRTPSm, TB_ALIGN_16 },
|
|
|
|
{ X86::RSQRTPSr_Int, X86::RSQRTPSm_Int, TB_ALIGN_16 },
|
|
|
|
{ X86::RSQRTSSr, X86::RSQRTSSm, 0 },
|
|
|
|
{ X86::RSQRTSSr_Int, X86::RSQRTSSm_Int, 0 },
|
|
|
|
{ X86::SQRTPDr, X86::SQRTPDm, TB_ALIGN_16 },
|
|
|
|
{ X86::SQRTPSr, X86::SQRTPSm, TB_ALIGN_16 },
|
|
|
|
{ X86::SQRTSDr, X86::SQRTSDm, 0 },
|
|
|
|
{ X86::SQRTSDr_Int, X86::SQRTSDm_Int, 0 },
|
|
|
|
{ X86::SQRTSSr, X86::SQRTSSm, 0 },
|
|
|
|
{ X86::SQRTSSr_Int, X86::SQRTSSm_Int, 0 },
|
|
|
|
{ X86::TEST16rr, X86::TEST16rm, 0 },
|
|
|
|
{ X86::TEST32rr, X86::TEST32rm, 0 },
|
|
|
|
{ X86::TEST64rr, X86::TEST64rm, 0 },
|
|
|
|
{ X86::TEST8rr, X86::TEST8rm, 0 },
|
2008-01-07 01:35:02 +00:00
|
|
|
// FIXME: TEST*rr EAX,EAX ---> CMP [mem], 0
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::UCOMISDrr, X86::UCOMISDrm, 0 },
|
|
|
|
{ X86::UCOMISSrr, X86::UCOMISSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
// AVX 128-bit versions of foldable instructions
|
|
|
|
{ X86::Int_VCOMISDrr, X86::Int_VCOMISDrm, 0 },
|
|
|
|
{ X86::Int_VCOMISSrr, X86::Int_VCOMISSrm, 0 },
|
|
|
|
{ X86::Int_VUCOMISDrr, X86::Int_VUCOMISDrm, 0 },
|
|
|
|
{ X86::Int_VUCOMISSrr, X86::Int_VUCOMISSrm, 0 },
|
2012-06-15 07:02:58 +00:00
|
|
|
{ X86::VCVTTSD2SI64rr, X86::VCVTTSD2SI64rm, 0 },
|
|
|
|
{ X86::Int_VCVTTSD2SI64rr,X86::Int_VCVTTSD2SI64rm,0 },
|
2012-06-14 22:12:58 +00:00
|
|
|
{ X86::VCVTTSD2SIrr, X86::VCVTTSD2SIrm, 0 },
|
2012-06-15 07:02:58 +00:00
|
|
|
{ X86::Int_VCVTTSD2SIrr,X86::Int_VCVTTSD2SIrm, 0 },
|
|
|
|
{ X86::VCVTTSS2SI64rr, X86::VCVTTSS2SI64rm, 0 },
|
|
|
|
{ X86::Int_VCVTTSS2SI64rr,X86::Int_VCVTTSS2SI64rm,0 },
|
|
|
|
{ X86::VCVTTSS2SIrr, X86::VCVTTSS2SIrm, 0 },
|
|
|
|
{ X86::Int_VCVTTSS2SIrr,X86::Int_VCVTTSS2SIrm, 0 },
|
|
|
|
{ X86::VCVTSD2SI64rr, X86::VCVTSD2SI64rm, 0 },
|
|
|
|
{ X86::VCVTSD2SIrr, X86::VCVTSD2SIrm, 0 },
|
|
|
|
{ X86::VCVTSS2SI64rr, X86::VCVTSS2SI64rm, 0 },
|
|
|
|
{ X86::VCVTSS2SIrr, X86::VCVTSS2SIrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMOV64toPQIrr, X86::VMOVQI2PQIrm, 0 },
|
|
|
|
{ X86::VMOV64toSDrr, X86::VMOV64toSDrm, 0 },
|
|
|
|
{ X86::VMOVAPDrr, X86::VMOVAPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVAPSrr, X86::VMOVAPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVDDUPrr, X86::VMOVDDUPrm, 0 },
|
|
|
|
{ X86::VMOVDI2PDIrr, X86::VMOVDI2PDIrm, 0 },
|
|
|
|
{ X86::VMOVDI2SSrr, X86::VMOVDI2SSrm, 0 },
|
|
|
|
{ X86::VMOVDQArr, X86::VMOVDQArm, TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVSLDUPrr, X86::VMOVSLDUPrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VMOVSHDUPrr, X86::VMOVSHDUPrm, TB_ALIGN_16 },
|
2012-12-26 02:14:19 +00:00
|
|
|
{ X86::VMOVUPDrr, X86::VMOVUPDrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMOVUPSrr, X86::VMOVUPSrm, 0 },
|
|
|
|
{ X86::VMOVZDI2PDIrr, X86::VMOVZDI2PDIrm, 0 },
|
|
|
|
{ X86::VMOVZQI2PQIrr, X86::VMOVZQI2PQIrm, 0 },
|
|
|
|
{ X86::VMOVZPQILo2PQIrr,X86::VMOVZPQILo2PQIrm, TB_ALIGN_16 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VPABSBrr128, X86::VPABSBrm128, 0 },
|
|
|
|
{ X86::VPABSDrr128, X86::VPABSDrm128, 0 },
|
|
|
|
{ X86::VPABSWrr128, X86::VPABSWrm128, 0 },
|
|
|
|
{ X86::VPERMILPDri, X86::VPERMILPDmi, 0 },
|
|
|
|
{ X86::VPERMILPSri, X86::VPERMILPSmi, 0 },
|
|
|
|
{ X86::VPSHUFDri, X86::VPSHUFDmi, 0 },
|
|
|
|
{ X86::VPSHUFHWri, X86::VPSHUFHWmi, 0 },
|
|
|
|
{ X86::VPSHUFLWri, X86::VPSHUFLWmi, 0 },
|
|
|
|
{ X86::VRCPPSr, X86::VRCPPSm, 0 },
|
|
|
|
{ X86::VRCPPSr_Int, X86::VRCPPSm_Int, 0 },
|
|
|
|
{ X86::VRSQRTPSr, X86::VRSQRTPSm, 0 },
|
|
|
|
{ X86::VRSQRTPSr_Int, X86::VRSQRTPSm_Int, 0 },
|
|
|
|
{ X86::VSQRTPDr, X86::VSQRTPDm, 0 },
|
|
|
|
{ X86::VSQRTPSr, X86::VSQRTPSm, 0 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::VUCOMISDrr, X86::VUCOMISDrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VUCOMISSrr, X86::VUCOMISSrm, 0 },
|
2012-07-15 12:26:30 +00:00
|
|
|
{ X86::VBROADCASTSSrr, X86::VBROADCASTSSrm, TB_NO_REVERSE },
|
|
|
|
|
2011-09-14 02:36:58 +00:00
|
|
|
// AVX 256-bit foldable instructions
|
|
|
|
{ X86::VMOVAPDYrr, X86::VMOVAPDYrm, TB_ALIGN_32 },
|
|
|
|
{ X86::VMOVAPSYrr, X86::VMOVAPSYrm, TB_ALIGN_32 },
|
2012-01-19 08:50:38 +00:00
|
|
|
{ X86::VMOVDQAYrr, X86::VMOVDQAYrm, TB_ALIGN_32 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMOVUPDYrr, X86::VMOVUPDYrm, 0 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::VMOVUPSYrr, X86::VMOVUPSYrm, 0 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VPERMILPDYri, X86::VPERMILPDYmi, 0 },
|
|
|
|
{ X86::VPERMILPSYri, X86::VPERMILPSYmi, 0 },
|
2012-07-15 12:26:30 +00:00
|
|
|
|
2011-11-14 08:07:55 +00:00
|
|
|
// AVX2 foldable instructions
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VPABSBrr256, X86::VPABSBrm256, 0 },
|
|
|
|
{ X86::VPABSDrr256, X86::VPABSDrm256, 0 },
|
|
|
|
{ X86::VPABSWrr256, X86::VPABSWrm256, 0 },
|
|
|
|
{ X86::VPSHUFDYri, X86::VPSHUFDYmi, 0 },
|
|
|
|
{ X86::VPSHUFHWYri, X86::VPSHUFHWYmi, 0 },
|
|
|
|
{ X86::VPSHUFLWYri, X86::VPSHUFLWYmi, 0 },
|
|
|
|
{ X86::VRCPPSYr, X86::VRCPPSYm, 0 },
|
|
|
|
{ X86::VRCPPSYr_Int, X86::VRCPPSYm_Int, 0 },
|
|
|
|
{ X86::VRSQRTPSYr, X86::VRSQRTPSYm, 0 },
|
|
|
|
{ X86::VSQRTPDYr, X86::VSQRTPDYm, 0 },
|
|
|
|
{ X86::VSQRTPSYr, X86::VSQRTPSYm, 0 },
|
2012-07-15 12:26:30 +00:00
|
|
|
{ X86::VBROADCASTSSYrr, X86::VBROADCASTSSYrm, TB_NO_REVERSE },
|
|
|
|
{ X86::VBROADCASTSDYrr, X86::VBROADCASTSDYrm, TB_NO_REVERSE },
|
2012-09-26 08:24:51 +00:00
|
|
|
|
2013-10-05 20:20:51 +00:00
|
|
|
// BMI/BMI2/LZCNT/POPCNT/TBM foldable instructions
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::BEXTR32rr, X86::BEXTR32rm, 0 },
|
|
|
|
{ X86::BEXTR64rr, X86::BEXTR64rm, 0 },
|
2013-10-05 20:20:51 +00:00
|
|
|
{ X86::BEXTRI32ri, X86::BEXTRI32mi, 0 },
|
|
|
|
{ X86::BEXTRI64ri, X86::BEXTRI64mi, 0 },
|
|
|
|
{ X86::BLCFILL32rr, X86::BLCFILL32rm, 0 },
|
|
|
|
{ X86::BLCFILL64rr, X86::BLCFILL64rm, 0 },
|
|
|
|
{ X86::BLCI32rr, X86::BLCI32rm, 0 },
|
|
|
|
{ X86::BLCI64rr, X86::BLCI64rm, 0 },
|
|
|
|
{ X86::BLCIC32rr, X86::BLCIC32rm, 0 },
|
|
|
|
{ X86::BLCIC64rr, X86::BLCIC64rm, 0 },
|
|
|
|
{ X86::BLCMSK32rr, X86::BLCMSK32rm, 0 },
|
|
|
|
{ X86::BLCMSK64rr, X86::BLCMSK64rm, 0 },
|
|
|
|
{ X86::BLCS32rr, X86::BLCS32rm, 0 },
|
|
|
|
{ X86::BLCS64rr, X86::BLCS64rm, 0 },
|
|
|
|
{ X86::BLSFILL32rr, X86::BLSFILL32rm, 0 },
|
|
|
|
{ X86::BLSFILL64rr, X86::BLSFILL64rm, 0 },
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::BLSI32rr, X86::BLSI32rm, 0 },
|
|
|
|
{ X86::BLSI64rr, X86::BLSI64rm, 0 },
|
2013-10-05 20:20:51 +00:00
|
|
|
{ X86::BLSIC32rr, X86::BLSIC32rm, 0 },
|
|
|
|
{ X86::BLSIC64rr, X86::BLSIC64rm, 0 },
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::BLSMSK32rr, X86::BLSMSK32rm, 0 },
|
|
|
|
{ X86::BLSMSK64rr, X86::BLSMSK64rm, 0 },
|
|
|
|
{ X86::BLSR32rr, X86::BLSR32rm, 0 },
|
|
|
|
{ X86::BLSR64rr, X86::BLSR64rm, 0 },
|
|
|
|
{ X86::BZHI32rr, X86::BZHI32rm, 0 },
|
|
|
|
{ X86::BZHI64rr, X86::BZHI64rm, 0 },
|
|
|
|
{ X86::LZCNT16rr, X86::LZCNT16rm, 0 },
|
|
|
|
{ X86::LZCNT32rr, X86::LZCNT32rm, 0 },
|
|
|
|
{ X86::LZCNT64rr, X86::LZCNT64rm, 0 },
|
|
|
|
{ X86::POPCNT16rr, X86::POPCNT16rm, 0 },
|
|
|
|
{ X86::POPCNT32rr, X86::POPCNT32rm, 0 },
|
|
|
|
{ X86::POPCNT64rr, X86::POPCNT64rm, 0 },
|
2012-09-26 08:24:51 +00:00
|
|
|
{ X86::RORX32ri, X86::RORX32mi, 0 },
|
|
|
|
{ X86::RORX64ri, X86::RORX64mi, 0 },
|
2012-09-26 08:26:25 +00:00
|
|
|
{ X86::SARX32rr, X86::SARX32rm, 0 },
|
|
|
|
{ X86::SARX64rr, X86::SARX64rm, 0 },
|
|
|
|
{ X86::SHRX32rr, X86::SHRX32rm, 0 },
|
|
|
|
{ X86::SHRX64rr, X86::SHRX64rm, 0 },
|
|
|
|
{ X86::SHLX32rr, X86::SHLX32rm, 0 },
|
|
|
|
{ X86::SHLX64rr, X86::SHLX64rm, 0 },
|
2013-10-05 20:20:51 +00:00
|
|
|
{ X86::T1MSKC32rr, X86::T1MSKC32rm, 0 },
|
|
|
|
{ X86::T1MSKC64rr, X86::T1MSKC64rm, 0 },
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::TZCNT16rr, X86::TZCNT16rm, 0 },
|
|
|
|
{ X86::TZCNT32rr, X86::TZCNT32rm, 0 },
|
|
|
|
{ X86::TZCNT64rr, X86::TZCNT64rm, 0 },
|
2013-10-05 20:20:51 +00:00
|
|
|
{ X86::TZMSK32rr, X86::TZMSK32rm, 0 },
|
|
|
|
{ X86::TZMSK64rr, X86::TZMSK64rm, 0 },
|
2013-09-02 07:12:29 +00:00
|
|
|
|
|
|
|
// AVX-512 foldable instructions
|
|
|
|
{ X86::VMOV64toPQIZrr, X86::VMOVQI2PQIZrm, 0 },
|
|
|
|
{ X86::VMOVDI2SSZrr, X86::VMOVDI2SSZrm, 0 },
|
2013-09-17 06:05:17 +00:00
|
|
|
{ X86::VMOVDQA32rr, X86::VMOVDQA32rm, TB_ALIGN_64 },
|
|
|
|
{ X86::VMOVDQA64rr, X86::VMOVDQA64rm, TB_ALIGN_64 },
|
|
|
|
{ X86::VMOVDQU32rr, X86::VMOVDQU32rm, 0 },
|
|
|
|
{ X86::VMOVDQU64rr, X86::VMOVDQU64rm, 0 },
|
2013-09-17 06:50:11 +00:00
|
|
|
|
|
|
|
// AES foldable instructions
|
|
|
|
{ X86::AESIMCrr, X86::AESIMCrm, TB_ALIGN_16 },
|
|
|
|
{ X86::AESKEYGENASSIST128rr, X86::AESKEYGENASSIST128rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESIMCrr, X86::VAESIMCrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESKEYGENASSIST128rr, X86::VAESKEYGENASSIST128rm, TB_ALIGN_16 },
|
2008-01-07 01:35:02 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (unsigned i = 0, e = array_lengthof(OpTbl1); i != e; ++i) {
|
2012-03-09 07:45:21 +00:00
|
|
|
unsigned RegOp = OpTbl1[i].RegOp;
|
|
|
|
unsigned MemOp = OpTbl1[i].MemOp;
|
|
|
|
unsigned Flags = OpTbl1[i].Flags;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
AddTableEntry(RegOp2MemOpTable1, MemOp2RegOpTable,
|
|
|
|
RegOp, MemOp,
|
|
|
|
// Index 1, folded load
|
|
|
|
Flags | TB_INDEX_1 | TB_FOLDED_LOAD);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const X86OpTblEntry OpTbl2[] = {
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::ADC32rr, X86::ADC32rm, 0 },
|
|
|
|
{ X86::ADC64rr, X86::ADC64rm, 0 },
|
|
|
|
{ X86::ADD16rr, X86::ADD16rm, 0 },
|
|
|
|
{ X86::ADD16rr_DB, X86::ADD16rm, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD32rr, X86::ADD32rm, 0 },
|
|
|
|
{ X86::ADD32rr_DB, X86::ADD32rm, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD64rr, X86::ADD64rm, 0 },
|
|
|
|
{ X86::ADD64rr_DB, X86::ADD64rm, TB_NO_REVERSE },
|
|
|
|
{ X86::ADD8rr, X86::ADD8rm, 0 },
|
|
|
|
{ X86::ADDPDrr, X86::ADDPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ADDPSrr, X86::ADDPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ADDSDrr, X86::ADDSDrm, 0 },
|
|
|
|
{ X86::ADDSSrr, X86::ADDSSrm, 0 },
|
|
|
|
{ X86::ADDSUBPDrr, X86::ADDSUBPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ADDSUBPSrr, X86::ADDSUBPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::AND16rr, X86::AND16rm, 0 },
|
|
|
|
{ X86::AND32rr, X86::AND32rm, 0 },
|
|
|
|
{ X86::AND64rr, X86::AND64rm, 0 },
|
|
|
|
{ X86::AND8rr, X86::AND8rm, 0 },
|
|
|
|
{ X86::ANDNPDrr, X86::ANDNPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ANDNPSrr, X86::ANDNPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ANDPDrr, X86::ANDPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ANDPSrr, X86::ANDPSrm, TB_ALIGN_16 },
|
2012-01-14 18:14:53 +00:00
|
|
|
{ X86::BLENDPDrri, X86::BLENDPDrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::BLENDPSrri, X86::BLENDPSrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::BLENDVPDrr0, X86::BLENDVPDrm0, TB_ALIGN_16 },
|
|
|
|
{ X86::BLENDVPSrr0, X86::BLENDVPSrm0, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::CMOVA16rr, X86::CMOVA16rm, 0 },
|
|
|
|
{ X86::CMOVA32rr, X86::CMOVA32rm, 0 },
|
|
|
|
{ X86::CMOVA64rr, X86::CMOVA64rm, 0 },
|
|
|
|
{ X86::CMOVAE16rr, X86::CMOVAE16rm, 0 },
|
|
|
|
{ X86::CMOVAE32rr, X86::CMOVAE32rm, 0 },
|
|
|
|
{ X86::CMOVAE64rr, X86::CMOVAE64rm, 0 },
|
|
|
|
{ X86::CMOVB16rr, X86::CMOVB16rm, 0 },
|
|
|
|
{ X86::CMOVB32rr, X86::CMOVB32rm, 0 },
|
|
|
|
{ X86::CMOVB64rr, X86::CMOVB64rm, 0 },
|
|
|
|
{ X86::CMOVBE16rr, X86::CMOVBE16rm, 0 },
|
|
|
|
{ X86::CMOVBE32rr, X86::CMOVBE32rm, 0 },
|
|
|
|
{ X86::CMOVBE64rr, X86::CMOVBE64rm, 0 },
|
|
|
|
{ X86::CMOVE16rr, X86::CMOVE16rm, 0 },
|
|
|
|
{ X86::CMOVE32rr, X86::CMOVE32rm, 0 },
|
|
|
|
{ X86::CMOVE64rr, X86::CMOVE64rm, 0 },
|
|
|
|
{ X86::CMOVG16rr, X86::CMOVG16rm, 0 },
|
|
|
|
{ X86::CMOVG32rr, X86::CMOVG32rm, 0 },
|
|
|
|
{ X86::CMOVG64rr, X86::CMOVG64rm, 0 },
|
|
|
|
{ X86::CMOVGE16rr, X86::CMOVGE16rm, 0 },
|
|
|
|
{ X86::CMOVGE32rr, X86::CMOVGE32rm, 0 },
|
|
|
|
{ X86::CMOVGE64rr, X86::CMOVGE64rm, 0 },
|
|
|
|
{ X86::CMOVL16rr, X86::CMOVL16rm, 0 },
|
|
|
|
{ X86::CMOVL32rr, X86::CMOVL32rm, 0 },
|
|
|
|
{ X86::CMOVL64rr, X86::CMOVL64rm, 0 },
|
|
|
|
{ X86::CMOVLE16rr, X86::CMOVLE16rm, 0 },
|
|
|
|
{ X86::CMOVLE32rr, X86::CMOVLE32rm, 0 },
|
|
|
|
{ X86::CMOVLE64rr, X86::CMOVLE64rm, 0 },
|
|
|
|
{ X86::CMOVNE16rr, X86::CMOVNE16rm, 0 },
|
|
|
|
{ X86::CMOVNE32rr, X86::CMOVNE32rm, 0 },
|
|
|
|
{ X86::CMOVNE64rr, X86::CMOVNE64rm, 0 },
|
|
|
|
{ X86::CMOVNO16rr, X86::CMOVNO16rm, 0 },
|
|
|
|
{ X86::CMOVNO32rr, X86::CMOVNO32rm, 0 },
|
|
|
|
{ X86::CMOVNO64rr, X86::CMOVNO64rm, 0 },
|
|
|
|
{ X86::CMOVNP16rr, X86::CMOVNP16rm, 0 },
|
|
|
|
{ X86::CMOVNP32rr, X86::CMOVNP32rm, 0 },
|
|
|
|
{ X86::CMOVNP64rr, X86::CMOVNP64rm, 0 },
|
|
|
|
{ X86::CMOVNS16rr, X86::CMOVNS16rm, 0 },
|
|
|
|
{ X86::CMOVNS32rr, X86::CMOVNS32rm, 0 },
|
|
|
|
{ X86::CMOVNS64rr, X86::CMOVNS64rm, 0 },
|
|
|
|
{ X86::CMOVO16rr, X86::CMOVO16rm, 0 },
|
|
|
|
{ X86::CMOVO32rr, X86::CMOVO32rm, 0 },
|
|
|
|
{ X86::CMOVO64rr, X86::CMOVO64rm, 0 },
|
|
|
|
{ X86::CMOVP16rr, X86::CMOVP16rm, 0 },
|
|
|
|
{ X86::CMOVP32rr, X86::CMOVP32rm, 0 },
|
|
|
|
{ X86::CMOVP64rr, X86::CMOVP64rm, 0 },
|
|
|
|
{ X86::CMOVS16rr, X86::CMOVS16rm, 0 },
|
|
|
|
{ X86::CMOVS32rr, X86::CMOVS32rm, 0 },
|
|
|
|
{ X86::CMOVS64rr, X86::CMOVS64rm, 0 },
|
|
|
|
{ X86::CMPPDrri, X86::CMPPDrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::CMPPSrri, X86::CMPPSrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::CMPSDrr, X86::CMPSDrm, 0 },
|
|
|
|
{ X86::CMPSSrr, X86::CMPSSrm, 0 },
|
|
|
|
{ X86::DIVPDrr, X86::DIVPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::DIVPSrr, X86::DIVPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::DIVSDrr, X86::DIVSDrm, 0 },
|
|
|
|
{ X86::DIVSSrr, X86::DIVSSrm, 0 },
|
|
|
|
{ X86::FsANDNPDrr, X86::FsANDNPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsANDNPSrr, X86::FsANDNPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsANDPDrr, X86::FsANDPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsANDPSrr, X86::FsANDPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsORPDrr, X86::FsORPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsORPSrr, X86::FsORPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsXORPDrr, X86::FsXORPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::FsXORPSrr, X86::FsXORPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::HADDPDrr, X86::HADDPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::HADDPSrr, X86::HADDPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::HSUBPDrr, X86::HSUBPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::HSUBPSrr, X86::HSUBPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::IMUL16rr, X86::IMUL16rm, 0 },
|
|
|
|
{ X86::IMUL32rr, X86::IMUL32rm, 0 },
|
|
|
|
{ X86::IMUL64rr, X86::IMUL64rm, 0 },
|
|
|
|
{ X86::Int_CMPSDrr, X86::Int_CMPSDrm, 0 },
|
|
|
|
{ X86::Int_CMPSSrr, X86::Int_CMPSSrm, 0 },
|
2012-08-13 18:29:41 +00:00
|
|
|
{ X86::Int_CVTSD2SSrr, X86::Int_CVTSD2SSrm, 0 },
|
|
|
|
{ X86::Int_CVTSI2SD64rr,X86::Int_CVTSI2SD64rm, 0 },
|
|
|
|
{ X86::Int_CVTSI2SDrr, X86::Int_CVTSI2SDrm, 0 },
|
|
|
|
{ X86::Int_CVTSI2SS64rr,X86::Int_CVTSI2SS64rm, 0 },
|
|
|
|
{ X86::Int_CVTSI2SSrr, X86::Int_CVTSI2SSrm, 0 },
|
|
|
|
{ X86::Int_CVTSS2SDrr, X86::Int_CVTSS2SDrm, 0 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::MAXPDrr, X86::MAXPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MAXPSrr, X86::MAXPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MAXSDrr, X86::MAXSDrm, 0 },
|
|
|
|
{ X86::MAXSSrr, X86::MAXSSrm, 0 },
|
|
|
|
{ X86::MINPDrr, X86::MINPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MINPSrr, X86::MINPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MINSDrr, X86::MINSDrm, 0 },
|
|
|
|
{ X86::MINSSrr, X86::MINSSrm, 0 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::MPSADBWrri, X86::MPSADBWrmi, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::MULPDrr, X86::MULPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MULPSrr, X86::MULPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::MULSDrr, X86::MULSDrm, 0 },
|
|
|
|
{ X86::MULSSrr, X86::MULSSrm, 0 },
|
|
|
|
{ X86::OR16rr, X86::OR16rm, 0 },
|
|
|
|
{ X86::OR32rr, X86::OR32rm, 0 },
|
|
|
|
{ X86::OR64rr, X86::OR64rm, 0 },
|
|
|
|
{ X86::OR8rr, X86::OR8rm, 0 },
|
|
|
|
{ X86::ORPDrr, X86::ORPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::ORPSrr, X86::ORPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PACKSSDWrr, X86::PACKSSDWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PACKSSWBrr, X86::PACKSSWBrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PACKUSDWrr, X86::PACKUSDWrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PACKUSWBrr, X86::PACKUSWBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDBrr, X86::PADDBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDDrr, X86::PADDDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDQrr, X86::PADDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDSBrr, X86::PADDSBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDSWrr, X86::PADDSWrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PADDUSBrr, X86::PADDUSBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PADDUSWrr, X86::PADDUSWrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PADDWrr, X86::PADDWrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PALIGNR128rr, X86::PALIGNR128rm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PANDNrr, X86::PANDNrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PANDrr, X86::PANDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PAVGBrr, X86::PAVGBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PAVGWrr, X86::PAVGWrm, TB_ALIGN_16 },
|
2012-01-14 18:14:53 +00:00
|
|
|
{ X86::PBLENDWrri, X86::PBLENDWrmi, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PCMPEQBrr, X86::PCMPEQBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PCMPEQDrr, X86::PCMPEQDrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PCMPEQQrr, X86::PCMPEQQrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PCMPEQWrr, X86::PCMPEQWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PCMPGTBrr, X86::PCMPGTBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PCMPGTDrr, X86::PCMPGTDrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PCMPGTQrr, X86::PCMPGTQrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PCMPGTWrr, X86::PCMPGTWrm, TB_ALIGN_16 },
|
2012-01-25 05:37:32 +00:00
|
|
|
{ X86::PHADDDrr, X86::PHADDDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PHADDWrr, X86::PHADDWrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PHADDSWrr128, X86::PHADDSWrm128, TB_ALIGN_16 },
|
2012-01-25 05:37:32 +00:00
|
|
|
{ X86::PHSUBDrr, X86::PHSUBDrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PHSUBSWrr128, X86::PHSUBSWrm128, TB_ALIGN_16 },
|
2012-01-25 05:37:32 +00:00
|
|
|
{ X86::PHSUBWrr, X86::PHSUBWrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PINSRWrri, X86::PINSRWrmi, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PMADDUBSWrr128, X86::PMADDUBSWrm128, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PMADDWDrr, X86::PMADDWDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXSWrr, X86::PMAXSWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXUBrr, X86::PMAXUBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMINSWrr, X86::PMINSWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMINUBrr, X86::PMINUBrm, TB_ALIGN_16 },
|
2012-12-21 14:04:55 +00:00
|
|
|
{ X86::PMINSBrr, X86::PMINSBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMINSDrr, X86::PMINSDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMINUDrr, X86::PMINUDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMINUWrr, X86::PMINUWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXSBrr, X86::PMAXSBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXSDrr, X86::PMAXSDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXUDrr, X86::PMAXUDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMAXUWrr, X86::PMAXUWrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PMULDQrr, X86::PMULDQrm, TB_ALIGN_16 },
|
2011-11-14 08:07:55 +00:00
|
|
|
{ X86::PMULHRSWrr128, X86::PMULHRSWrm128, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PMULHUWrr, X86::PMULHUWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMULHWrr, X86::PMULHWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMULLDrr, X86::PMULLDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMULLWrr, X86::PMULLWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PMULUDQrr, X86::PMULUDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PORrr, X86::PORrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSADBWrr, X86::PSADBWrm, TB_ALIGN_16 },
|
2012-01-25 06:43:11 +00:00
|
|
|
{ X86::PSHUFBrr, X86::PSHUFBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSIGNBrr, X86::PSIGNBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSIGNWrr, X86::PSIGNWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSIGNDrr, X86::PSIGNDrm, TB_ALIGN_16 },
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::PSLLDrr, X86::PSLLDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSLLQrr, X86::PSLLQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSLLWrr, X86::PSLLWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSRADrr, X86::PSRADrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSRAWrr, X86::PSRAWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSRLDrr, X86::PSRLDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSRLQrr, X86::PSRLQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSRLWrr, X86::PSRLWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSUBBrr, X86::PSUBBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSUBDrr, X86::PSUBDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSUBSBrr, X86::PSUBSBrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSUBSWrr, X86::PSUBSWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PSUBWrr, X86::PSUBWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKHBWrr, X86::PUNPCKHBWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKHDQrr, X86::PUNPCKHDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKHQDQrr, X86::PUNPCKHQDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKHWDrr, X86::PUNPCKHWDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKLBWrr, X86::PUNPCKLBWrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKLDQrr, X86::PUNPCKLDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKLQDQrr, X86::PUNPCKLQDQrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PUNPCKLWDrr, X86::PUNPCKLWDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::PXORrr, X86::PXORrm, TB_ALIGN_16 },
|
|
|
|
{ X86::SBB32rr, X86::SBB32rm, 0 },
|
|
|
|
{ X86::SBB64rr, X86::SBB64rm, 0 },
|
|
|
|
{ X86::SHUFPDrri, X86::SHUFPDrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::SHUFPSrri, X86::SHUFPSrmi, TB_ALIGN_16 },
|
|
|
|
{ X86::SUB16rr, X86::SUB16rm, 0 },
|
|
|
|
{ X86::SUB32rr, X86::SUB32rm, 0 },
|
|
|
|
{ X86::SUB64rr, X86::SUB64rm, 0 },
|
|
|
|
{ X86::SUB8rr, X86::SUB8rm, 0 },
|
|
|
|
{ X86::SUBPDrr, X86::SUBPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::SUBPSrr, X86::SUBPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::SUBSDrr, X86::SUBSDrm, 0 },
|
|
|
|
{ X86::SUBSSrr, X86::SUBSSrm, 0 },
|
2008-01-07 01:35:02 +00:00
|
|
|
// FIXME: TEST*rr -> swapped operand of TEST*mr.
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
{ X86::UNPCKHPDrr, X86::UNPCKHPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::UNPCKHPSrr, X86::UNPCKHPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::UNPCKLPDrr, X86::UNPCKLPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::UNPCKLPSrr, X86::UNPCKLPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::XOR16rr, X86::XOR16rm, 0 },
|
|
|
|
{ X86::XOR32rr, X86::XOR32rm, 0 },
|
|
|
|
{ X86::XOR64rr, X86::XOR64rm, 0 },
|
|
|
|
{ X86::XOR8rr, X86::XOR8rm, 0 },
|
|
|
|
{ X86::XORPDrr, X86::XORPDrm, TB_ALIGN_16 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::XORPSrr, X86::XORPSrm, TB_ALIGN_16 },
|
|
|
|
// AVX 128-bit versions of foldable instructions
|
|
|
|
{ X86::VCVTSD2SSrr, X86::VCVTSD2SSrm, 0 },
|
|
|
|
{ X86::Int_VCVTSD2SSrr, X86::Int_VCVTSD2SSrm, 0 },
|
|
|
|
{ X86::VCVTSI2SD64rr, X86::VCVTSI2SD64rm, 0 },
|
|
|
|
{ X86::Int_VCVTSI2SD64rr, X86::Int_VCVTSI2SD64rm, 0 },
|
|
|
|
{ X86::VCVTSI2SDrr, X86::VCVTSI2SDrm, 0 },
|
|
|
|
{ X86::Int_VCVTSI2SDrr, X86::Int_VCVTSI2SDrm, 0 },
|
|
|
|
{ X86::VCVTSI2SS64rr, X86::VCVTSI2SS64rm, 0 },
|
|
|
|
{ X86::Int_VCVTSI2SS64rr, X86::Int_VCVTSI2SS64rm, 0 },
|
|
|
|
{ X86::VCVTSI2SSrr, X86::VCVTSI2SSrm, 0 },
|
|
|
|
{ X86::Int_VCVTSI2SSrr, X86::Int_VCVTSI2SSrm, 0 },
|
2012-12-26 00:35:47 +00:00
|
|
|
{ X86::VCVTSS2SDrr, X86::VCVTSS2SDrm, 0 },
|
|
|
|
{ X86::Int_VCVTSS2SDrr, X86::Int_VCVTSS2SDrm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VCVTTPD2DQrr, X86::VCVTTPD2DQXrm, 0 },
|
|
|
|
{ X86::VCVTTPS2DQrr, X86::VCVTTPS2DQrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VRSQRTSSr, X86::VRSQRTSSm, 0 },
|
|
|
|
{ X86::VSQRTSDr, X86::VSQRTSDm, 0 },
|
|
|
|
{ X86::VSQRTSSr, X86::VSQRTSSm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VADDPDrr, X86::VADDPDrm, 0 },
|
|
|
|
{ X86::VADDPSrr, X86::VADDPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VADDSDrr, X86::VADDSDrm, 0 },
|
|
|
|
{ X86::VADDSSrr, X86::VADDSSrm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VADDSUBPDrr, X86::VADDSUBPDrm, 0 },
|
|
|
|
{ X86::VADDSUBPSrr, X86::VADDSUBPSrm, 0 },
|
|
|
|
{ X86::VANDNPDrr, X86::VANDNPDrm, 0 },
|
|
|
|
{ X86::VANDNPSrr, X86::VANDNPSrm, 0 },
|
|
|
|
{ X86::VANDPDrr, X86::VANDPDrm, 0 },
|
|
|
|
{ X86::VANDPSrr, X86::VANDPSrm, 0 },
|
|
|
|
{ X86::VBLENDPDrri, X86::VBLENDPDrmi, 0 },
|
|
|
|
{ X86::VBLENDPSrri, X86::VBLENDPSrmi, 0 },
|
|
|
|
{ X86::VBLENDVPDrr, X86::VBLENDVPDrm, 0 },
|
|
|
|
{ X86::VBLENDVPSrr, X86::VBLENDVPSrm, 0 },
|
|
|
|
{ X86::VCMPPDrri, X86::VCMPPDrmi, 0 },
|
|
|
|
{ X86::VCMPPSrri, X86::VCMPPSrmi, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VCMPSDrr, X86::VCMPSDrm, 0 },
|
|
|
|
{ X86::VCMPSSrr, X86::VCMPSSrm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VDIVPDrr, X86::VDIVPDrm, 0 },
|
|
|
|
{ X86::VDIVPSrr, X86::VDIVPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VDIVSDrr, X86::VDIVSDrm, 0 },
|
|
|
|
{ X86::VDIVSSrr, X86::VDIVSSrm, 0 },
|
|
|
|
{ X86::VFsANDNPDrr, X86::VFsANDNPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsANDNPSrr, X86::VFsANDNPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsANDPDrr, X86::VFsANDPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsANDPSrr, X86::VFsANDPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsORPDrr, X86::VFsORPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsORPSrr, X86::VFsORPSrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsXORPDrr, X86::VFsXORPDrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFsXORPSrr, X86::VFsXORPSrm, TB_ALIGN_16 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VHADDPDrr, X86::VHADDPDrm, 0 },
|
|
|
|
{ X86::VHADDPSrr, X86::VHADDPSrm, 0 },
|
|
|
|
{ X86::VHSUBPDrr, X86::VHSUBPDrm, 0 },
|
|
|
|
{ X86::VHSUBPSrr, X86::VHSUBPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::Int_VCMPSDrr, X86::Int_VCMPSDrm, 0 },
|
|
|
|
{ X86::Int_VCMPSSrr, X86::Int_VCMPSSrm, 0 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VMAXPDrr, X86::VMAXPDrm, 0 },
|
|
|
|
{ X86::VMAXPSrr, X86::VMAXPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMAXSDrr, X86::VMAXSDrm, 0 },
|
|
|
|
{ X86::VMAXSSrr, X86::VMAXSSrm, 0 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VMINPDrr, X86::VMINPDrm, 0 },
|
|
|
|
{ X86::VMINPSrr, X86::VMINPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMINSDrr, X86::VMINSDrm, 0 },
|
|
|
|
{ X86::VMINSSrr, X86::VMINSSrm, 0 },
|
2012-12-26 02:44:47 +00:00
|
|
|
{ X86::VMPSADBWrri, X86::VMPSADBWrmi, 0 },
|
|
|
|
{ X86::VMULPDrr, X86::VMULPDrm, 0 },
|
|
|
|
{ X86::VMULPSrr, X86::VMULPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VMULSDrr, X86::VMULSDrm, 0 },
|
|
|
|
{ X86::VMULSSrr, X86::VMULSSrm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VORPDrr, X86::VORPDrm, 0 },
|
|
|
|
{ X86::VORPSrr, X86::VORPSrm, 0 },
|
|
|
|
{ X86::VPACKSSDWrr, X86::VPACKSSDWrm, 0 },
|
|
|
|
{ X86::VPACKSSWBrr, X86::VPACKSSWBrm, 0 },
|
|
|
|
{ X86::VPACKUSDWrr, X86::VPACKUSDWrm, 0 },
|
|
|
|
{ X86::VPACKUSWBrr, X86::VPACKUSWBrm, 0 },
|
|
|
|
{ X86::VPADDBrr, X86::VPADDBrm, 0 },
|
|
|
|
{ X86::VPADDDrr, X86::VPADDDrm, 0 },
|
|
|
|
{ X86::VPADDQrr, X86::VPADDQrm, 0 },
|
|
|
|
{ X86::VPADDSBrr, X86::VPADDSBrm, 0 },
|
|
|
|
{ X86::VPADDSWrr, X86::VPADDSWrm, 0 },
|
|
|
|
{ X86::VPADDUSBrr, X86::VPADDUSBrm, 0 },
|
|
|
|
{ X86::VPADDUSWrr, X86::VPADDUSWrm, 0 },
|
|
|
|
{ X86::VPADDWrr, X86::VPADDWrm, 0 },
|
|
|
|
{ X86::VPALIGNR128rr, X86::VPALIGNR128rm, 0 },
|
|
|
|
{ X86::VPANDNrr, X86::VPANDNrm, 0 },
|
|
|
|
{ X86::VPANDrr, X86::VPANDrm, 0 },
|
|
|
|
{ X86::VPAVGBrr, X86::VPAVGBrm, 0 },
|
|
|
|
{ X86::VPAVGWrr, X86::VPAVGWrm, 0 },
|
|
|
|
{ X86::VPBLENDWrri, X86::VPBLENDWrmi, 0 },
|
|
|
|
{ X86::VPCMPEQBrr, X86::VPCMPEQBrm, 0 },
|
|
|
|
{ X86::VPCMPEQDrr, X86::VPCMPEQDrm, 0 },
|
|
|
|
{ X86::VPCMPEQQrr, X86::VPCMPEQQrm, 0 },
|
|
|
|
{ X86::VPCMPEQWrr, X86::VPCMPEQWrm, 0 },
|
|
|
|
{ X86::VPCMPGTBrr, X86::VPCMPGTBrm, 0 },
|
|
|
|
{ X86::VPCMPGTDrr, X86::VPCMPGTDrm, 0 },
|
|
|
|
{ X86::VPCMPGTQrr, X86::VPCMPGTQrm, 0 },
|
|
|
|
{ X86::VPCMPGTWrr, X86::VPCMPGTWrm, 0 },
|
|
|
|
{ X86::VPHADDDrr, X86::VPHADDDrm, 0 },
|
|
|
|
{ X86::VPHADDSWrr128, X86::VPHADDSWrm128, 0 },
|
|
|
|
{ X86::VPHADDWrr, X86::VPHADDWrm, 0 },
|
|
|
|
{ X86::VPHSUBDrr, X86::VPHSUBDrm, 0 },
|
|
|
|
{ X86::VPHSUBSWrr128, X86::VPHSUBSWrm128, 0 },
|
|
|
|
{ X86::VPHSUBWrr, X86::VPHSUBWrm, 0 },
|
|
|
|
{ X86::VPERMILPDrr, X86::VPERMILPDrm, 0 },
|
|
|
|
{ X86::VPERMILPSrr, X86::VPERMILPSrm, 0 },
|
|
|
|
{ X86::VPINSRWrri, X86::VPINSRWrmi, 0 },
|
|
|
|
{ X86::VPMADDUBSWrr128, X86::VPMADDUBSWrm128, 0 },
|
|
|
|
{ X86::VPMADDWDrr, X86::VPMADDWDrm, 0 },
|
|
|
|
{ X86::VPMAXSWrr, X86::VPMAXSWrm, 0 },
|
|
|
|
{ X86::VPMAXUBrr, X86::VPMAXUBrm, 0 },
|
|
|
|
{ X86::VPMINSWrr, X86::VPMINSWrm, 0 },
|
|
|
|
{ X86::VPMINUBrr, X86::VPMINUBrm, 0 },
|
|
|
|
{ X86::VPMINSBrr, X86::VPMINSBrm, 0 },
|
|
|
|
{ X86::VPMINSDrr, X86::VPMINSDrm, 0 },
|
|
|
|
{ X86::VPMINUDrr, X86::VPMINUDrm, 0 },
|
|
|
|
{ X86::VPMINUWrr, X86::VPMINUWrm, 0 },
|
|
|
|
{ X86::VPMAXSBrr, X86::VPMAXSBrm, 0 },
|
|
|
|
{ X86::VPMAXSDrr, X86::VPMAXSDrm, 0 },
|
|
|
|
{ X86::VPMAXUDrr, X86::VPMAXUDrm, 0 },
|
|
|
|
{ X86::VPMAXUWrr, X86::VPMAXUWrm, 0 },
|
|
|
|
{ X86::VPMULDQrr, X86::VPMULDQrm, 0 },
|
|
|
|
{ X86::VPMULHRSWrr128, X86::VPMULHRSWrm128, 0 },
|
|
|
|
{ X86::VPMULHUWrr, X86::VPMULHUWrm, 0 },
|
|
|
|
{ X86::VPMULHWrr, X86::VPMULHWrm, 0 },
|
|
|
|
{ X86::VPMULLDrr, X86::VPMULLDrm, 0 },
|
|
|
|
{ X86::VPMULLWrr, X86::VPMULLWrm, 0 },
|
|
|
|
{ X86::VPMULUDQrr, X86::VPMULUDQrm, 0 },
|
|
|
|
{ X86::VPORrr, X86::VPORrm, 0 },
|
|
|
|
{ X86::VPSADBWrr, X86::VPSADBWrm, 0 },
|
|
|
|
{ X86::VPSHUFBrr, X86::VPSHUFBrm, 0 },
|
|
|
|
{ X86::VPSIGNBrr, X86::VPSIGNBrm, 0 },
|
|
|
|
{ X86::VPSIGNWrr, X86::VPSIGNWrm, 0 },
|
|
|
|
{ X86::VPSIGNDrr, X86::VPSIGNDrm, 0 },
|
|
|
|
{ X86::VPSLLDrr, X86::VPSLLDrm, 0 },
|
|
|
|
{ X86::VPSLLQrr, X86::VPSLLQrm, 0 },
|
|
|
|
{ X86::VPSLLWrr, X86::VPSLLWrm, 0 },
|
|
|
|
{ X86::VPSRADrr, X86::VPSRADrm, 0 },
|
|
|
|
{ X86::VPSRAWrr, X86::VPSRAWrm, 0 },
|
|
|
|
{ X86::VPSRLDrr, X86::VPSRLDrm, 0 },
|
|
|
|
{ X86::VPSRLQrr, X86::VPSRLQrm, 0 },
|
|
|
|
{ X86::VPSRLWrr, X86::VPSRLWrm, 0 },
|
|
|
|
{ X86::VPSUBBrr, X86::VPSUBBrm, 0 },
|
|
|
|
{ X86::VPSUBDrr, X86::VPSUBDrm, 0 },
|
|
|
|
{ X86::VPSUBSBrr, X86::VPSUBSBrm, 0 },
|
|
|
|
{ X86::VPSUBSWrr, X86::VPSUBSWrm, 0 },
|
|
|
|
{ X86::VPSUBWrr, X86::VPSUBWrm, 0 },
|
|
|
|
{ X86::VPUNPCKHBWrr, X86::VPUNPCKHBWrm, 0 },
|
|
|
|
{ X86::VPUNPCKHDQrr, X86::VPUNPCKHDQrm, 0 },
|
|
|
|
{ X86::VPUNPCKHQDQrr, X86::VPUNPCKHQDQrm, 0 },
|
|
|
|
{ X86::VPUNPCKHWDrr, X86::VPUNPCKHWDrm, 0 },
|
|
|
|
{ X86::VPUNPCKLBWrr, X86::VPUNPCKLBWrm, 0 },
|
|
|
|
{ X86::VPUNPCKLDQrr, X86::VPUNPCKLDQrm, 0 },
|
|
|
|
{ X86::VPUNPCKLQDQrr, X86::VPUNPCKLQDQrm, 0 },
|
|
|
|
{ X86::VPUNPCKLWDrr, X86::VPUNPCKLWDrm, 0 },
|
|
|
|
{ X86::VPXORrr, X86::VPXORrm, 0 },
|
|
|
|
{ X86::VSHUFPDrri, X86::VSHUFPDrmi, 0 },
|
|
|
|
{ X86::VSHUFPSrri, X86::VSHUFPSrmi, 0 },
|
|
|
|
{ X86::VSUBPDrr, X86::VSUBPDrm, 0 },
|
|
|
|
{ X86::VSUBPSrr, X86::VSUBPSrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
{ X86::VSUBSDrr, X86::VSUBSDrm, 0 },
|
|
|
|
{ X86::VSUBSSrr, X86::VSUBSSrm, 0 },
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VUNPCKHPDrr, X86::VUNPCKHPDrm, 0 },
|
|
|
|
{ X86::VUNPCKHPSrr, X86::VUNPCKHPSrm, 0 },
|
|
|
|
{ X86::VUNPCKLPDrr, X86::VUNPCKLPDrm, 0 },
|
|
|
|
{ X86::VUNPCKLPSrr, X86::VUNPCKLPSrm, 0 },
|
|
|
|
{ X86::VXORPDrr, X86::VXORPDrm, 0 },
|
|
|
|
{ X86::VXORPSrr, X86::VXORPSrm, 0 },
|
2012-01-14 18:14:53 +00:00
|
|
|
// AVX 256-bit foldable instructions
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VADDPDYrr, X86::VADDPDYrm, 0 },
|
|
|
|
{ X86::VADDPSYrr, X86::VADDPSYrm, 0 },
|
|
|
|
{ X86::VADDSUBPDYrr, X86::VADDSUBPDYrm, 0 },
|
|
|
|
{ X86::VADDSUBPSYrr, X86::VADDSUBPSYrm, 0 },
|
|
|
|
{ X86::VANDNPDYrr, X86::VANDNPDYrm, 0 },
|
|
|
|
{ X86::VANDNPSYrr, X86::VANDNPSYrm, 0 },
|
|
|
|
{ X86::VANDPDYrr, X86::VANDPDYrm, 0 },
|
|
|
|
{ X86::VANDPSYrr, X86::VANDPSYrm, 0 },
|
|
|
|
{ X86::VBLENDPDYrri, X86::VBLENDPDYrmi, 0 },
|
|
|
|
{ X86::VBLENDPSYrri, X86::VBLENDPSYrmi, 0 },
|
|
|
|
{ X86::VBLENDVPDYrr, X86::VBLENDVPDYrm, 0 },
|
|
|
|
{ X86::VBLENDVPSYrr, X86::VBLENDVPSYrm, 0 },
|
|
|
|
{ X86::VCMPPDYrri, X86::VCMPPDYrmi, 0 },
|
|
|
|
{ X86::VCMPPSYrri, X86::VCMPPSYrmi, 0 },
|
|
|
|
{ X86::VDIVPDYrr, X86::VDIVPDYrm, 0 },
|
|
|
|
{ X86::VDIVPSYrr, X86::VDIVPSYrm, 0 },
|
|
|
|
{ X86::VHADDPDYrr, X86::VHADDPDYrm, 0 },
|
|
|
|
{ X86::VHADDPSYrr, X86::VHADDPSYrm, 0 },
|
|
|
|
{ X86::VHSUBPDYrr, X86::VHSUBPDYrm, 0 },
|
|
|
|
{ X86::VHSUBPSYrr, X86::VHSUBPSYrm, 0 },
|
|
|
|
{ X86::VINSERTF128rr, X86::VINSERTF128rm, 0 },
|
|
|
|
{ X86::VMAXPDYrr, X86::VMAXPDYrm, 0 },
|
|
|
|
{ X86::VMAXPSYrr, X86::VMAXPSYrm, 0 },
|
|
|
|
{ X86::VMINPDYrr, X86::VMINPDYrm, 0 },
|
|
|
|
{ X86::VMINPSYrr, X86::VMINPSYrm, 0 },
|
|
|
|
{ X86::VMULPDYrr, X86::VMULPDYrm, 0 },
|
|
|
|
{ X86::VMULPSYrr, X86::VMULPSYrm, 0 },
|
|
|
|
{ X86::VORPDYrr, X86::VORPDYrm, 0 },
|
|
|
|
{ X86::VORPSYrr, X86::VORPSYrm, 0 },
|
|
|
|
{ X86::VPERM2F128rr, X86::VPERM2F128rm, 0 },
|
|
|
|
{ X86::VPERMILPDYrr, X86::VPERMILPDYrm, 0 },
|
|
|
|
{ X86::VPERMILPSYrr, X86::VPERMILPSYrm, 0 },
|
|
|
|
{ X86::VSHUFPDYrri, X86::VSHUFPDYrmi, 0 },
|
|
|
|
{ X86::VSHUFPSYrri, X86::VSHUFPSYrmi, 0 },
|
|
|
|
{ X86::VSUBPDYrr, X86::VSUBPDYrm, 0 },
|
|
|
|
{ X86::VSUBPSYrr, X86::VSUBPSYrm, 0 },
|
|
|
|
{ X86::VUNPCKHPDYrr, X86::VUNPCKHPDYrm, 0 },
|
|
|
|
{ X86::VUNPCKHPSYrr, X86::VUNPCKHPSYrm, 0 },
|
|
|
|
{ X86::VUNPCKLPDYrr, X86::VUNPCKLPDYrm, 0 },
|
|
|
|
{ X86::VUNPCKLPSYrr, X86::VUNPCKLPSYrm, 0 },
|
|
|
|
{ X86::VXORPDYrr, X86::VXORPDYrm, 0 },
|
|
|
|
{ X86::VXORPSYrr, X86::VXORPSYrm, 0 },
|
2011-11-14 08:07:55 +00:00
|
|
|
// AVX2 foldable instructions
|
2012-12-24 09:40:33 +00:00
|
|
|
{ X86::VINSERTI128rr, X86::VINSERTI128rm, 0 },
|
|
|
|
{ X86::VPACKSSDWYrr, X86::VPACKSSDWYrm, 0 },
|
|
|
|
{ X86::VPACKSSWBYrr, X86::VPACKSSWBYrm, 0 },
|
|
|
|
{ X86::VPACKUSDWYrr, X86::VPACKUSDWYrm, 0 },
|
|
|
|
{ X86::VPACKUSWBYrr, X86::VPACKUSWBYrm, 0 },
|
|
|
|
{ X86::VPADDBYrr, X86::VPADDBYrm, 0 },
|
|
|
|
{ X86::VPADDDYrr, X86::VPADDDYrm, 0 },
|
|
|
|
{ X86::VPADDQYrr, X86::VPADDQYrm, 0 },
|
|
|
|
{ X86::VPADDSBYrr, X86::VPADDSBYrm, 0 },
|
|
|
|
{ X86::VPADDSWYrr, X86::VPADDSWYrm, 0 },
|
|
|
|
{ X86::VPADDUSBYrr, X86::VPADDUSBYrm, 0 },
|
|
|
|
{ X86::VPADDUSWYrr, X86::VPADDUSWYrm, 0 },
|
|
|
|
{ X86::VPADDWYrr, X86::VPADDWYrm, 0 },
|
|
|
|
{ X86::VPALIGNR256rr, X86::VPALIGNR256rm, 0 },
|
|
|
|
{ X86::VPANDNYrr, X86::VPANDNYrm, 0 },
|
|
|
|
{ X86::VPANDYrr, X86::VPANDYrm, 0 },
|
|
|
|
{ X86::VPAVGBYrr, X86::VPAVGBYrm, 0 },
|
|
|
|
{ X86::VPAVGWYrr, X86::VPAVGWYrm, 0 },
|
|
|
|
{ X86::VPBLENDDrri, X86::VPBLENDDrmi, 0 },
|
|
|
|
{ X86::VPBLENDDYrri, X86::VPBLENDDYrmi, 0 },
|
|
|
|
{ X86::VPBLENDWYrri, X86::VPBLENDWYrmi, 0 },
|
|
|
|
{ X86::VPCMPEQBYrr, X86::VPCMPEQBYrm, 0 },
|
|
|
|
{ X86::VPCMPEQDYrr, X86::VPCMPEQDYrm, 0 },
|
|
|
|
{ X86::VPCMPEQQYrr, X86::VPCMPEQQYrm, 0 },
|
|
|
|
{ X86::VPCMPEQWYrr, X86::VPCMPEQWYrm, 0 },
|
|
|
|
{ X86::VPCMPGTBYrr, X86::VPCMPGTBYrm, 0 },
|
|
|
|
{ X86::VPCMPGTDYrr, X86::VPCMPGTDYrm, 0 },
|
|
|
|
{ X86::VPCMPGTQYrr, X86::VPCMPGTQYrm, 0 },
|
|
|
|
{ X86::VPCMPGTWYrr, X86::VPCMPGTWYrm, 0 },
|
|
|
|
{ X86::VPERM2I128rr, X86::VPERM2I128rm, 0 },
|
|
|
|
{ X86::VPERMDYrr, X86::VPERMDYrm, 0 },
|
|
|
|
{ X86::VPERMPDYri, X86::VPERMPDYmi, 0 },
|
|
|
|
{ X86::VPERMPSYrr, X86::VPERMPSYrm, 0 },
|
|
|
|
{ X86::VPERMQYri, X86::VPERMQYmi, 0 },
|
|
|
|
{ X86::VPHADDDYrr, X86::VPHADDDYrm, 0 },
|
|
|
|
{ X86::VPHADDSWrr256, X86::VPHADDSWrm256, 0 },
|
|
|
|
{ X86::VPHADDWYrr, X86::VPHADDWYrm, 0 },
|
|
|
|
{ X86::VPHSUBDYrr, X86::VPHSUBDYrm, 0 },
|
|
|
|
{ X86::VPHSUBSWrr256, X86::VPHSUBSWrm256, 0 },
|
|
|
|
{ X86::VPHSUBWYrr, X86::VPHSUBWYrm, 0 },
|
|
|
|
{ X86::VPMADDUBSWrr256, X86::VPMADDUBSWrm256, 0 },
|
|
|
|
{ X86::VPMADDWDYrr, X86::VPMADDWDYrm, 0 },
|
|
|
|
{ X86::VPMAXSWYrr, X86::VPMAXSWYrm, 0 },
|
|
|
|
{ X86::VPMAXUBYrr, X86::VPMAXUBYrm, 0 },
|
|
|
|
{ X86::VPMINSWYrr, X86::VPMINSWYrm, 0 },
|
|
|
|
{ X86::VPMINUBYrr, X86::VPMINUBYrm, 0 },
|
|
|
|
{ X86::VPMINSBYrr, X86::VPMINSBYrm, 0 },
|
|
|
|
{ X86::VPMINSDYrr, X86::VPMINSDYrm, 0 },
|
|
|
|
{ X86::VPMINUDYrr, X86::VPMINUDYrm, 0 },
|
|
|
|
{ X86::VPMINUWYrr, X86::VPMINUWYrm, 0 },
|
|
|
|
{ X86::VPMAXSBYrr, X86::VPMAXSBYrm, 0 },
|
|
|
|
{ X86::VPMAXSDYrr, X86::VPMAXSDYrm, 0 },
|
|
|
|
{ X86::VPMAXUDYrr, X86::VPMAXUDYrm, 0 },
|
|
|
|
{ X86::VPMAXUWYrr, X86::VPMAXUWYrm, 0 },
|
|
|
|
{ X86::VMPSADBWYrri, X86::VMPSADBWYrmi, 0 },
|
|
|
|
{ X86::VPMULDQYrr, X86::VPMULDQYrm, 0 },
|
|
|
|
{ X86::VPMULHRSWrr256, X86::VPMULHRSWrm256, 0 },
|
|
|
|
{ X86::VPMULHUWYrr, X86::VPMULHUWYrm, 0 },
|
|
|
|
{ X86::VPMULHWYrr, X86::VPMULHWYrm, 0 },
|
|
|
|
{ X86::VPMULLDYrr, X86::VPMULLDYrm, 0 },
|
|
|
|
{ X86::VPMULLWYrr, X86::VPMULLWYrm, 0 },
|
|
|
|
{ X86::VPMULUDQYrr, X86::VPMULUDQYrm, 0 },
|
|
|
|
{ X86::VPORYrr, X86::VPORYrm, 0 },
|
|
|
|
{ X86::VPSADBWYrr, X86::VPSADBWYrm, 0 },
|
|
|
|
{ X86::VPSHUFBYrr, X86::VPSHUFBYrm, 0 },
|
|
|
|
{ X86::VPSIGNBYrr, X86::VPSIGNBYrm, 0 },
|
|
|
|
{ X86::VPSIGNWYrr, X86::VPSIGNWYrm, 0 },
|
|
|
|
{ X86::VPSIGNDYrr, X86::VPSIGNDYrm, 0 },
|
|
|
|
{ X86::VPSLLDYrr, X86::VPSLLDYrm, 0 },
|
|
|
|
{ X86::VPSLLQYrr, X86::VPSLLQYrm, 0 },
|
|
|
|
{ X86::VPSLLWYrr, X86::VPSLLWYrm, 0 },
|
|
|
|
{ X86::VPSLLVDrr, X86::VPSLLVDrm, 0 },
|
|
|
|
{ X86::VPSLLVDYrr, X86::VPSLLVDYrm, 0 },
|
|
|
|
{ X86::VPSLLVQrr, X86::VPSLLVQrm, 0 },
|
|
|
|
{ X86::VPSLLVQYrr, X86::VPSLLVQYrm, 0 },
|
|
|
|
{ X86::VPSRADYrr, X86::VPSRADYrm, 0 },
|
|
|
|
{ X86::VPSRAWYrr, X86::VPSRAWYrm, 0 },
|
|
|
|
{ X86::VPSRAVDrr, X86::VPSRAVDrm, 0 },
|
|
|
|
{ X86::VPSRAVDYrr, X86::VPSRAVDYrm, 0 },
|
|
|
|
{ X86::VPSRLDYrr, X86::VPSRLDYrm, 0 },
|
|
|
|
{ X86::VPSRLQYrr, X86::VPSRLQYrm, 0 },
|
|
|
|
{ X86::VPSRLWYrr, X86::VPSRLWYrm, 0 },
|
|
|
|
{ X86::VPSRLVDrr, X86::VPSRLVDrm, 0 },
|
|
|
|
{ X86::VPSRLVDYrr, X86::VPSRLVDYrm, 0 },
|
|
|
|
{ X86::VPSRLVQrr, X86::VPSRLVQrm, 0 },
|
|
|
|
{ X86::VPSRLVQYrr, X86::VPSRLVQYrm, 0 },
|
|
|
|
{ X86::VPSUBBYrr, X86::VPSUBBYrm, 0 },
|
|
|
|
{ X86::VPSUBDYrr, X86::VPSUBDYrm, 0 },
|
|
|
|
{ X86::VPSUBSBYrr, X86::VPSUBSBYrm, 0 },
|
|
|
|
{ X86::VPSUBSWYrr, X86::VPSUBSWYrm, 0 },
|
|
|
|
{ X86::VPSUBWYrr, X86::VPSUBWYrm, 0 },
|
|
|
|
{ X86::VPUNPCKHBWYrr, X86::VPUNPCKHBWYrm, 0 },
|
|
|
|
{ X86::VPUNPCKHDQYrr, X86::VPUNPCKHDQYrm, 0 },
|
|
|
|
{ X86::VPUNPCKHQDQYrr, X86::VPUNPCKHQDQYrm, 0 },
|
|
|
|
{ X86::VPUNPCKHWDYrr, X86::VPUNPCKHWDYrm, 0 },
|
|
|
|
{ X86::VPUNPCKLBWYrr, X86::VPUNPCKLBWYrm, 0 },
|
|
|
|
{ X86::VPUNPCKLDQYrr, X86::VPUNPCKLDQYrm, 0 },
|
|
|
|
{ X86::VPUNPCKLQDQYrr, X86::VPUNPCKLQDQYrm, 0 },
|
|
|
|
{ X86::VPUNPCKLWDYrr, X86::VPUNPCKLWDYrm, 0 },
|
|
|
|
{ X86::VPXORYrr, X86::VPXORYrm, 0 },
|
2011-09-14 02:36:58 +00:00
|
|
|
// FIXME: add AVX 256-bit foldable instructions
|
2012-08-31 23:10:34 +00:00
|
|
|
|
|
|
|
// FMA4 foldable patterns
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFMADDSS4rr, X86::VFMADDSS4mr, 0 },
|
|
|
|
{ X86::VFMADDSD4rr, X86::VFMADDSD4mr, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFMADDPS4rr, X86::VFMADDPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPD4rr, X86::VFMADDPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPS4rrY, X86::VFMADDPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPD4rrY, X86::VFMADDPD4mrY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFNMADDSS4rr, X86::VFNMADDSS4mr, 0 },
|
|
|
|
{ X86::VFNMADDSD4rr, X86::VFNMADDSD4mr, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFNMADDPS4rr, X86::VFNMADDPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPD4rr, X86::VFNMADDPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPS4rrY, X86::VFNMADDPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPD4rrY, X86::VFNMADDPD4mrY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFMSUBSS4rr, X86::VFMSUBSS4mr, 0 },
|
|
|
|
{ X86::VFMSUBSD4rr, X86::VFMSUBSD4mr, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFMSUBPS4rr, X86::VFMSUBPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPD4rr, X86::VFMSUBPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPS4rrY, X86::VFMSUBPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPD4rrY, X86::VFMSUBPD4mrY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFNMSUBSS4rr, X86::VFNMSUBSS4mr, 0 },
|
|
|
|
{ X86::VFNMSUBSD4rr, X86::VFNMSUBSD4mr, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFNMSUBPS4rr, X86::VFNMSUBPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPD4rr, X86::VFNMSUBPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPS4rrY, X86::VFNMSUBPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPD4rrY, X86::VFNMSUBPD4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPS4rr, X86::VFMADDSUBPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPD4rr, X86::VFMADDSUBPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPS4rrY, X86::VFMADDSUBPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPD4rrY, X86::VFMADDSUBPD4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPS4rr, X86::VFMSUBADDPS4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPD4rr, X86::VFMSUBADDPD4mr, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPS4rrY, X86::VFMSUBADDPS4mrY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPD4rrY, X86::VFMSUBADDPD4mrY, TB_ALIGN_32 },
|
2012-09-26 08:22:37 +00:00
|
|
|
|
|
|
|
// BMI/BMI2 foldable instructions
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::ANDN32rr, X86::ANDN32rm, 0 },
|
|
|
|
{ X86::ANDN64rr, X86::ANDN64rm, 0 },
|
2012-09-26 08:22:37 +00:00
|
|
|
{ X86::MULX32rr, X86::MULX32rm, 0 },
|
|
|
|
{ X86::MULX64rr, X86::MULX64rm, 0 },
|
2012-12-17 05:02:29 +00:00
|
|
|
{ X86::PDEP32rr, X86::PDEP32rm, 0 },
|
|
|
|
{ X86::PDEP64rr, X86::PDEP64rm, 0 },
|
|
|
|
{ X86::PEXT32rr, X86::PEXT32rm, 0 },
|
|
|
|
{ X86::PEXT64rr, X86::PEXT64rm, 0 },
|
2013-08-11 07:55:09 +00:00
|
|
|
|
|
|
|
// AVX-512 foldable instructions
|
2013-09-02 07:12:29 +00:00
|
|
|
{ X86::VPADDDZrr, X86::VPADDDZrm, 0 },
|
|
|
|
{ X86::VPADDQZrr, X86::VPADDQZrm, 0 },
|
|
|
|
{ X86::VADDPSZrr, X86::VADDPSZrm, 0 },
|
|
|
|
{ X86::VADDPDZrr, X86::VADDPDZrm, 0 },
|
|
|
|
{ X86::VSUBPSZrr, X86::VSUBPSZrm, 0 },
|
|
|
|
{ X86::VSUBPDZrr, X86::VSUBPDZrm, 0 },
|
|
|
|
{ X86::VMULPSZrr, X86::VMULPSZrm, 0 },
|
|
|
|
{ X86::VMULPDZrr, X86::VMULPDZrm, 0 },
|
|
|
|
{ X86::VDIVPSZrr, X86::VDIVPSZrm, 0 },
|
|
|
|
{ X86::VDIVPDZrr, X86::VDIVPDZrm, 0 },
|
|
|
|
{ X86::VMINPSZrr, X86::VMINPSZrm, 0 },
|
|
|
|
{ X86::VMINPDZrr, X86::VMINPDZrm, 0 },
|
|
|
|
{ X86::VMAXPSZrr, X86::VMAXPSZrm, 0 },
|
|
|
|
{ X86::VMAXPDZrr, X86::VMAXPDZrm, 0 },
|
2013-08-11 07:55:09 +00:00
|
|
|
{ X86::VPERMPDZri, X86::VPERMPDZmi, 0 },
|
|
|
|
{ X86::VPERMPSZrr, X86::VPERMPSZrm, 0 },
|
2013-09-02 07:12:29 +00:00
|
|
|
{ X86::VPSLLVDZrr, X86::VPSLLVDZrm, 0 },
|
|
|
|
{ X86::VPSLLVQZrr, X86::VPSLLVQZrm, 0 },
|
|
|
|
{ X86::VPSRAVDZrr, X86::VPSRAVDZrm, 0 },
|
|
|
|
{ X86::VPSRLVDZrr, X86::VPSRLVDZrm, 0 },
|
|
|
|
{ X86::VPSRLVQZrr, X86::VPSRLVQZrm, 0 },
|
|
|
|
{ X86::VSHUFPDZrri, X86::VSHUFPDZrmi, 0 },
|
|
|
|
{ X86::VSHUFPSZrri, X86::VSHUFPSZrmi, 0 },
|
|
|
|
{ X86::VALIGNQrri, X86::VALIGNQrmi, 0 },
|
|
|
|
{ X86::VALIGNDrri, X86::VALIGNDrmi, 0 },
|
2013-09-17 06:50:11 +00:00
|
|
|
|
|
|
|
// AES foldable instructions
|
|
|
|
{ X86::AESDECLASTrr, X86::AESDECLASTrm, TB_ALIGN_16 },
|
|
|
|
{ X86::AESDECrr, X86::AESDECrm, TB_ALIGN_16 },
|
|
|
|
{ X86::AESENCLASTrr, X86::AESENCLASTrm, TB_ALIGN_16 },
|
|
|
|
{ X86::AESENCrr, X86::AESENCrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESDECLASTrr, X86::VAESDECLASTrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESDECrr, X86::VAESDECrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESENCLASTrr, X86::VAESENCLASTrm, TB_ALIGN_16 },
|
|
|
|
{ X86::VAESENCrr, X86::VAESENCrm, TB_ALIGN_16 },
|
|
|
|
|
|
|
|
// SHA foldable instructions
|
|
|
|
{ X86::SHA1MSG1rr, X86::SHA1MSG1rm, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA1MSG2rr, X86::SHA1MSG2rm, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA1NEXTErr, X86::SHA1NEXTErm, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA1RNDS4rri, X86::SHA1RNDS4rmi, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA256MSG1rr, X86::SHA256MSG1rm, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA256MSG2rr, X86::SHA256MSG2rm, TB_ALIGN_16 },
|
|
|
|
{ X86::SHA256RNDS2rr, X86::SHA256RNDS2rm, TB_ALIGN_16 },
|
2008-01-07 01:35:02 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (unsigned i = 0, e = array_lengthof(OpTbl2); i != e; ++i) {
|
2012-03-09 07:45:21 +00:00
|
|
|
unsigned RegOp = OpTbl2[i].RegOp;
|
|
|
|
unsigned MemOp = OpTbl2[i].MemOp;
|
|
|
|
unsigned Flags = OpTbl2[i].Flags;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
AddTableEntry(RegOp2MemOpTable2, MemOp2RegOpTable,
|
|
|
|
RegOp, MemOp,
|
|
|
|
// Index 2, folded load
|
|
|
|
Flags | TB_INDEX_2 | TB_FOLDED_LOAD);
|
|
|
|
}
|
2012-05-31 09:20:20 +00:00
|
|
|
|
|
|
|
static const X86OpTblEntry OpTbl3[] = {
|
|
|
|
// FMA foldable instructions
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMADDSSr231r, X86::VFMADDSSr231m, 0 },
|
|
|
|
{ X86::VFMADDSDr231r, X86::VFMADDSDr231m, 0 },
|
|
|
|
{ X86::VFMADDSSr132r, X86::VFMADDSSr132m, 0 },
|
|
|
|
{ X86::VFMADDSDr132r, X86::VFMADDSDr132m, 0 },
|
|
|
|
{ X86::VFMADDSSr213r, X86::VFMADDSSr213m, 0 },
|
|
|
|
{ X86::VFMADDSDr213r, X86::VFMADDSDr213m, 0 },
|
2012-08-01 12:06:00 +00:00
|
|
|
{ X86::VFMADDSSr213r_Int, X86::VFMADDSSr213m_Int, 0 },
|
|
|
|
{ X86::VFMADDSDr213r_Int, X86::VFMADDSDr213m_Int, 0 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMADDPSr231r, X86::VFMADDPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPDr231r, X86::VFMADDPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPSr132r, X86::VFMADDPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPDr132r, X86::VFMADDPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPSr213r, X86::VFMADDPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPDr213r, X86::VFMADDPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPSr231rY, X86::VFMADDPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPDr231rY, X86::VFMADDPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPSr132rY, X86::VFMADDPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPDr132rY, X86::VFMADDPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPSr213rY, X86::VFMADDPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPDr213rY, X86::VFMADDPDr213mY, TB_ALIGN_32 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFNMADDSSr231r, X86::VFNMADDSSr231m, 0 },
|
|
|
|
{ X86::VFNMADDSDr231r, X86::VFNMADDSDr231m, 0 },
|
|
|
|
{ X86::VFNMADDSSr132r, X86::VFNMADDSSr132m, 0 },
|
|
|
|
{ X86::VFNMADDSDr132r, X86::VFNMADDSDr132m, 0 },
|
|
|
|
{ X86::VFNMADDSSr213r, X86::VFNMADDSSr213m, 0 },
|
|
|
|
{ X86::VFNMADDSDr213r, X86::VFNMADDSDr213m, 0 },
|
2012-08-01 12:06:00 +00:00
|
|
|
{ X86::VFNMADDSSr213r_Int, X86::VFNMADDSSr213m_Int, 0 },
|
|
|
|
{ X86::VFNMADDSDr213r_Int, X86::VFNMADDSDr213m_Int, 0 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFNMADDPSr231r, X86::VFNMADDPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPDr231r, X86::VFNMADDPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPSr132r, X86::VFNMADDPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPDr132r, X86::VFNMADDPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPSr213r, X86::VFNMADDPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPDr213r, X86::VFNMADDPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPSr231rY, X86::VFNMADDPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPDr231rY, X86::VFNMADDPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPSr132rY, X86::VFNMADDPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPDr132rY, X86::VFNMADDPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPSr213rY, X86::VFNMADDPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPDr213rY, X86::VFNMADDPDr213mY, TB_ALIGN_32 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMSUBSSr231r, X86::VFMSUBSSr231m, 0 },
|
|
|
|
{ X86::VFMSUBSDr231r, X86::VFMSUBSDr231m, 0 },
|
|
|
|
{ X86::VFMSUBSSr132r, X86::VFMSUBSSr132m, 0 },
|
|
|
|
{ X86::VFMSUBSDr132r, X86::VFMSUBSDr132m, 0 },
|
|
|
|
{ X86::VFMSUBSSr213r, X86::VFMSUBSSr213m, 0 },
|
|
|
|
{ X86::VFMSUBSDr213r, X86::VFMSUBSDr213m, 0 },
|
2012-08-01 12:06:00 +00:00
|
|
|
{ X86::VFMSUBSSr213r_Int, X86::VFMSUBSSr213m_Int, 0 },
|
|
|
|
{ X86::VFMSUBSDr213r_Int, X86::VFMSUBSDr213m_Int, 0 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMSUBPSr231r, X86::VFMSUBPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPDr231r, X86::VFMSUBPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPSr132r, X86::VFMSUBPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPDr132r, X86::VFMSUBPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPSr213r, X86::VFMSUBPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPDr213r, X86::VFMSUBPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPSr231rY, X86::VFMSUBPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPDr231rY, X86::VFMSUBPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPSr132rY, X86::VFMSUBPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPDr132rY, X86::VFMSUBPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPSr213rY, X86::VFMSUBPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPDr213rY, X86::VFMSUBPDr213mY, TB_ALIGN_32 },
|
2012-05-31 09:20:20 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFNMSUBSSr231r, X86::VFNMSUBSSr231m, 0 },
|
|
|
|
{ X86::VFNMSUBSDr231r, X86::VFNMSUBSDr231m, 0 },
|
|
|
|
{ X86::VFNMSUBSSr132r, X86::VFNMSUBSSr132m, 0 },
|
|
|
|
{ X86::VFNMSUBSDr132r, X86::VFNMSUBSDr132m, 0 },
|
|
|
|
{ X86::VFNMSUBSSr213r, X86::VFNMSUBSSr213m, 0 },
|
|
|
|
{ X86::VFNMSUBSDr213r, X86::VFNMSUBSDr213m, 0 },
|
2012-08-01 12:06:00 +00:00
|
|
|
{ X86::VFNMSUBSSr213r_Int, X86::VFNMSUBSSr213m_Int, 0 },
|
|
|
|
{ X86::VFNMSUBSDr213r_Int, X86::VFNMSUBSDr213m_Int, 0 },
|
2012-06-01 05:48:39 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFNMSUBPSr231r, X86::VFNMSUBPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPDr231r, X86::VFNMSUBPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPSr132r, X86::VFNMSUBPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPDr132r, X86::VFNMSUBPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPSr213r, X86::VFNMSUBPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPDr213r, X86::VFNMSUBPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPSr231rY, X86::VFNMSUBPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPDr231rY, X86::VFNMSUBPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPSr132rY, X86::VFNMSUBPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPDr132rY, X86::VFNMSUBPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPSr213rY, X86::VFNMSUBPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPDr213rY, X86::VFNMSUBPDr213mY, TB_ALIGN_32 },
|
2012-06-04 07:08:21 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMADDSUBPSr231r, X86::VFMADDSUBPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPDr231r, X86::VFMADDSUBPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPSr132r, X86::VFMADDSUBPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPDr132r, X86::VFMADDSUBPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPSr213r, X86::VFMADDSUBPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPDr213r, X86::VFMADDSUBPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPSr231rY, X86::VFMADDSUBPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPDr231rY, X86::VFMADDSUBPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPSr132rY, X86::VFMADDSUBPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPDr132rY, X86::VFMADDSUBPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPSr213rY, X86::VFMADDSUBPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPDr213rY, X86::VFMADDSUBPDr213mY, TB_ALIGN_32 },
|
2012-06-04 07:08:21 +00:00
|
|
|
|
2012-06-04 07:46:16 +00:00
|
|
|
{ X86::VFMSUBADDPSr231r, X86::VFMSUBADDPSr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPDr231r, X86::VFMSUBADDPDr231m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPSr132r, X86::VFMSUBADDPSr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPDr132r, X86::VFMSUBADDPDr132m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPSr213r, X86::VFMSUBADDPSr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPDr213r, X86::VFMSUBADDPDr213m, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPSr231rY, X86::VFMSUBADDPSr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPDr231rY, X86::VFMSUBADDPDr231mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPSr132rY, X86::VFMSUBADDPSr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPDr132rY, X86::VFMSUBADDPDr132mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPSr213rY, X86::VFMSUBADDPSr213mY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPDr213rY, X86::VFMSUBADDPDr213mY, TB_ALIGN_32 },
|
2012-08-31 23:10:34 +00:00
|
|
|
|
|
|
|
// FMA4 foldable patterns
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFMADDSS4rr, X86::VFMADDSS4rm, 0 },
|
|
|
|
{ X86::VFMADDSD4rr, X86::VFMADDSD4rm, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFMADDPS4rr, X86::VFMADDPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPD4rr, X86::VFMADDPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDPS4rrY, X86::VFMADDPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDPD4rrY, X86::VFMADDPD4rmY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFNMADDSS4rr, X86::VFNMADDSS4rm, 0 },
|
|
|
|
{ X86::VFNMADDSD4rr, X86::VFNMADDSD4rm, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFNMADDPS4rr, X86::VFNMADDPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPD4rr, X86::VFNMADDPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMADDPS4rrY, X86::VFNMADDPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMADDPD4rrY, X86::VFNMADDPD4rmY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFMSUBSS4rr, X86::VFMSUBSS4rm, 0 },
|
|
|
|
{ X86::VFMSUBSD4rr, X86::VFMSUBSD4rm, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFMSUBPS4rr, X86::VFMSUBPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPD4rr, X86::VFMSUBPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBPS4rrY, X86::VFMSUBPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBPD4rrY, X86::VFMSUBPD4rmY, TB_ALIGN_32 },
|
2012-11-04 04:40:08 +00:00
|
|
|
{ X86::VFNMSUBSS4rr, X86::VFNMSUBSS4rm, 0 },
|
|
|
|
{ X86::VFNMSUBSD4rr, X86::VFNMSUBSD4rm, 0 },
|
2012-08-31 23:10:34 +00:00
|
|
|
{ X86::VFNMSUBPS4rr, X86::VFNMSUBPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPD4rr, X86::VFNMSUBPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFNMSUBPS4rrY, X86::VFNMSUBPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFNMSUBPD4rrY, X86::VFNMSUBPD4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPS4rr, X86::VFMADDSUBPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPD4rr, X86::VFMADDSUBPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMADDSUBPS4rrY, X86::VFMADDSUBPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMADDSUBPD4rrY, X86::VFMADDSUBPD4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPS4rr, X86::VFMSUBADDPS4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPD4rr, X86::VFMSUBADDPD4rm, TB_ALIGN_16 },
|
|
|
|
{ X86::VFMSUBADDPS4rrY, X86::VFMSUBADDPS4rmY, TB_ALIGN_32 },
|
|
|
|
{ X86::VFMSUBADDPD4rrY, X86::VFMSUBADDPD4rmY, TB_ALIGN_32 },
|
2013-10-06 13:11:09 +00:00
|
|
|
// AVX-512 VPERMI instructions with 3 source operands.
|
|
|
|
{ X86::VPERMI2Drr, X86::VPERMI2Drm, 0 },
|
|
|
|
{ X86::VPERMI2Qrr, X86::VPERMI2Qrm, 0 },
|
|
|
|
{ X86::VPERMI2PSrr, X86::VPERMI2PSrm, 0 },
|
|
|
|
{ X86::VPERMI2PDrr, X86::VPERMI2PDrm, 0 },
|
2012-05-31 09:20:20 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
for (unsigned i = 0, e = array_lengthof(OpTbl3); i != e; ++i) {
|
|
|
|
unsigned RegOp = OpTbl3[i].RegOp;
|
|
|
|
unsigned MemOp = OpTbl3[i].MemOp;
|
|
|
|
unsigned Flags = OpTbl3[i].Flags;
|
|
|
|
AddTableEntry(RegOp2MemOpTable3, MemOp2RegOpTable,
|
|
|
|
RegOp, MemOp,
|
|
|
|
// Index 3, folded load
|
|
|
|
Flags | TB_INDEX_3 | TB_FOLDED_LOAD);
|
|
|
|
}
|
|
|
|
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
void
|
|
|
|
X86InstrInfo::AddTableEntry(RegOp2MemOpTableType &R2MTable,
|
|
|
|
MemOp2RegOpTableType &M2RTable,
|
|
|
|
unsigned RegOp, unsigned MemOp, unsigned Flags) {
|
|
|
|
if ((Flags & TB_NO_FORWARD) == 0) {
|
|
|
|
assert(!R2MTable.count(RegOp) && "Duplicate entry!");
|
|
|
|
R2MTable[RegOp] = std::make_pair(MemOp, Flags);
|
|
|
|
}
|
|
|
|
if ((Flags & TB_NO_REVERSE) == 0) {
|
|
|
|
assert(!M2RTable.count(MemOp) &&
|
2010-10-08 03:54:52 +00:00
|
|
|
"Duplicated entries in unfolding maps?");
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
M2RTable[MemOp] = std::make_pair(RegOp, Flags);
|
|
|
|
}
|
2002-10-25 22:55:53 +00:00
|
|
|
}
|
|
|
|
|
2010-01-12 00:09:37 +00:00
|
|
|
bool
|
2010-01-13 00:30:23 +00:00
|
|
|
X86InstrInfo::isCoalescableExtInstr(const MachineInstr &MI,
|
|
|
|
unsigned &SrcReg, unsigned &DstReg,
|
|
|
|
unsigned &SubIdx) const {
|
2010-01-12 00:09:37 +00:00
|
|
|
switch (MI.getOpcode()) {
|
|
|
|
default: break;
|
|
|
|
case X86::MOVSX16rr8:
|
|
|
|
case X86::MOVZX16rr8:
|
|
|
|
case X86::MOVSX32rr8:
|
|
|
|
case X86::MOVZX32rr8:
|
|
|
|
case X86::MOVSX64rr8:
|
2010-01-13 08:01:32 +00:00
|
|
|
if (!TM.getSubtarget<X86Subtarget>().is64Bit())
|
|
|
|
// It's not always legal to reference the low 8-bit of the larger
|
|
|
|
// register in 32-bit mode.
|
|
|
|
return false;
|
2010-01-12 00:09:37 +00:00
|
|
|
case X86::MOVSX32rr16:
|
|
|
|
case X86::MOVZX32rr16:
|
|
|
|
case X86::MOVSX64rr16:
|
2013-05-30 10:43:18 +00:00
|
|
|
case X86::MOVSX64rr32: {
|
2010-01-12 00:09:37 +00:00
|
|
|
if (MI.getOperand(0).getSubReg() || MI.getOperand(1).getSubReg())
|
|
|
|
// Be conservative.
|
|
|
|
return false;
|
|
|
|
SrcReg = MI.getOperand(1).getReg();
|
|
|
|
DstReg = MI.getOperand(0).getReg();
|
|
|
|
switch (MI.getOpcode()) {
|
2012-08-21 08:16:16 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2010-01-12 00:09:37 +00:00
|
|
|
case X86::MOVSX16rr8:
|
|
|
|
case X86::MOVZX16rr8:
|
|
|
|
case X86::MOVSX32rr8:
|
|
|
|
case X86::MOVZX32rr8:
|
|
|
|
case X86::MOVSX64rr8:
|
2010-05-25 17:04:16 +00:00
|
|
|
SubIdx = X86::sub_8bit;
|
2010-01-12 00:09:37 +00:00
|
|
|
break;
|
|
|
|
case X86::MOVSX32rr16:
|
|
|
|
case X86::MOVZX32rr16:
|
|
|
|
case X86::MOVSX64rr16:
|
2010-05-25 17:04:16 +00:00
|
|
|
SubIdx = X86::sub_16bit;
|
2010-01-12 00:09:37 +00:00
|
|
|
break;
|
|
|
|
case X86::MOVSX64rr32:
|
2010-05-25 17:04:16 +00:00
|
|
|
SubIdx = X86::sub_32bit;
|
2010-01-12 00:09:37 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-01-13 00:30:23 +00:00
|
|
|
return true;
|
2010-01-12 00:09:37 +00:00
|
|
|
}
|
|
|
|
}
|
2010-01-13 00:30:23 +00:00
|
|
|
return false;
|
2010-01-12 00:09:37 +00:00
|
|
|
}
|
|
|
|
|
2009-11-12 20:55:29 +00:00
|
|
|
/// isFrameOperand - Return true and the FrameIndex if the specified
|
|
|
|
/// operand and follow operands form a reference to the stack frame.
|
|
|
|
bool X86InstrInfo::isFrameOperand(const MachineInstr *MI, unsigned int Op,
|
|
|
|
int &FrameIndex) const {
|
|
|
|
if (MI->getOperand(Op).isFI() && MI->getOperand(Op+1).isImm() &&
|
|
|
|
MI->getOperand(Op+2).isReg() && MI->getOperand(Op+3).isImm() &&
|
|
|
|
MI->getOperand(Op+1).getImm() == 1 &&
|
|
|
|
MI->getOperand(Op+2).getReg() == 0 &&
|
|
|
|
MI->getOperand(Op+3).getImm() == 0) {
|
|
|
|
FrameIndex = MI->getOperand(Op).getIndex();
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2009-11-13 00:29:53 +00:00
|
|
|
static bool isFrameLoadOpcode(int Opcode) {
|
|
|
|
switch (Opcode) {
|
2012-01-20 21:51:11 +00:00
|
|
|
default:
|
|
|
|
return false;
|
2006-02-02 20:12:32 +00:00
|
|
|
case X86::MOV8rm:
|
|
|
|
case X86::MOV16rm:
|
|
|
|
case X86::MOV32rm:
|
2006-09-08 06:48:29 +00:00
|
|
|
case X86::MOV64rm:
|
2007-07-04 21:07:47 +00:00
|
|
|
case X86::LD_Fp64m:
|
2006-02-02 20:12:32 +00:00
|
|
|
case X86::MOVSSrm:
|
|
|
|
case X86::MOVSDrm:
|
2006-04-18 16:44:51 +00:00
|
|
|
case X86::MOVAPSrm:
|
|
|
|
case X86::MOVAPDrm:
|
2009-01-09 02:40:34 +00:00
|
|
|
case X86::MOVDQArm:
|
2011-09-14 02:36:58 +00:00
|
|
|
case X86::VMOVSSrm:
|
|
|
|
case X86::VMOVSDrm:
|
|
|
|
case X86::VMOVAPSrm:
|
|
|
|
case X86::VMOVAPDrm:
|
|
|
|
case X86::VMOVDQArm:
|
2011-07-14 18:50:58 +00:00
|
|
|
case X86::VMOVAPSYrm:
|
|
|
|
case X86::VMOVAPDYrm:
|
|
|
|
case X86::VMOVDQAYrm:
|
2007-04-03 06:00:37 +00:00
|
|
|
case X86::MMX_MOVD64rm:
|
|
|
|
case X86::MMX_MOVQ64rm:
|
2013-08-11 07:55:09 +00:00
|
|
|
case X86::VMOVDQA32rm:
|
|
|
|
case X86::VMOVDQA64rm:
|
2009-11-13 00:29:53 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool isFrameStoreOpcode(int Opcode) {
|
|
|
|
switch (Opcode) {
|
|
|
|
default: break;
|
|
|
|
case X86::MOV8mr:
|
|
|
|
case X86::MOV16mr:
|
|
|
|
case X86::MOV32mr:
|
|
|
|
case X86::MOV64mr:
|
|
|
|
case X86::ST_FpP64m:
|
|
|
|
case X86::MOVSSmr:
|
|
|
|
case X86::MOVSDmr:
|
|
|
|
case X86::MOVAPSmr:
|
|
|
|
case X86::MOVAPDmr:
|
|
|
|
case X86::MOVDQAmr:
|
2011-09-14 02:36:58 +00:00
|
|
|
case X86::VMOVSSmr:
|
|
|
|
case X86::VMOVSDmr:
|
|
|
|
case X86::VMOVAPSmr:
|
|
|
|
case X86::VMOVAPDmr:
|
|
|
|
case X86::VMOVDQAmr:
|
2011-07-14 18:50:58 +00:00
|
|
|
case X86::VMOVAPSYmr:
|
|
|
|
case X86::VMOVAPDYmr:
|
|
|
|
case X86::VMOVDQAYmr:
|
2009-11-13 00:29:53 +00:00
|
|
|
case X86::MMX_MOVD64mr:
|
|
|
|
case X86::MMX_MOVQ64mr:
|
|
|
|
case X86::MMX_MOVNTQmr:
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2011-01-26 02:03:37 +00:00
|
|
|
unsigned X86InstrInfo::isLoadFromStackSlot(const MachineInstr *MI,
|
2009-11-13 00:29:53 +00:00
|
|
|
int &FrameIndex) const {
|
|
|
|
if (isFrameLoadOpcode(MI->getOpcode()))
|
2010-07-27 04:17:01 +00:00
|
|
|
if (MI->getOperand(0).getSubReg() == 0 && isFrameOperand(MI, 1, FrameIndex))
|
2006-02-02 20:12:32 +00:00
|
|
|
return MI->getOperand(0).getReg();
|
2009-11-13 00:29:53 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-01-26 02:03:37 +00:00
|
|
|
unsigned X86InstrInfo::isLoadFromStackSlotPostFE(const MachineInstr *MI,
|
2009-11-13 00:29:53 +00:00
|
|
|
int &FrameIndex) const {
|
|
|
|
if (isFrameLoadOpcode(MI->getOpcode())) {
|
|
|
|
unsigned Reg;
|
|
|
|
if ((Reg = isLoadFromStackSlot(MI, FrameIndex)))
|
|
|
|
return Reg;
|
2009-11-12 20:55:29 +00:00
|
|
|
// Check for post-frame index elimination operations
|
2009-12-04 22:38:46 +00:00
|
|
|
const MachineMemOperand *Dummy;
|
|
|
|
return hasLoadFromStackSlot(MI, Dummy, FrameIndex);
|
2006-02-02 20:12:32 +00:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-11-18 19:49:32 +00:00
|
|
|
unsigned X86InstrInfo::isStoreToStackSlot(const MachineInstr *MI,
|
2006-02-02 20:12:32 +00:00
|
|
|
int &FrameIndex) const {
|
2009-11-13 00:29:53 +00:00
|
|
|
if (isFrameStoreOpcode(MI->getOpcode()))
|
2010-07-27 04:17:01 +00:00
|
|
|
if (MI->getOperand(X86::AddrNumOperands).getSubReg() == 0 &&
|
|
|
|
isFrameOperand(MI, 0, FrameIndex))
|
2010-07-08 22:41:28 +00:00
|
|
|
return MI->getOperand(X86::AddrNumOperands).getReg();
|
2009-11-13 00:29:53 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned X86InstrInfo::isStoreToStackSlotPostFE(const MachineInstr *MI,
|
|
|
|
int &FrameIndex) const {
|
|
|
|
if (isFrameStoreOpcode(MI->getOpcode())) {
|
|
|
|
unsigned Reg;
|
|
|
|
if ((Reg = isStoreToStackSlot(MI, FrameIndex)))
|
|
|
|
return Reg;
|
2009-11-12 20:55:29 +00:00
|
|
|
// Check for post-frame index elimination operations
|
2009-12-04 22:38:46 +00:00
|
|
|
const MachineMemOperand *Dummy;
|
|
|
|
return hasStoreToStackSlot(MI, Dummy, FrameIndex);
|
2006-02-02 20:12:32 +00:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-03-27 01:45:11 +00:00
|
|
|
/// regIsPICBase - Return true if register is PIC base (i.e.g defined by
|
|
|
|
/// X86::MOVPC32r.
|
2008-07-07 23:14:23 +00:00
|
|
|
static bool regIsPICBase(unsigned BaseReg, const MachineRegisterInfo &MRI) {
|
2012-08-08 00:40:47 +00:00
|
|
|
// Don't waste compile time scanning use-def chains of physregs.
|
|
|
|
if (!TargetRegisterInfo::isVirtualRegister(BaseReg))
|
|
|
|
return false;
|
2008-03-27 01:45:11 +00:00
|
|
|
bool isPICBase = false;
|
|
|
|
for (MachineRegisterInfo::def_iterator I = MRI.def_begin(BaseReg),
|
|
|
|
E = MRI.def_end(); I != E; ++I) {
|
|
|
|
MachineInstr *DefMI = I.getOperand().getParent();
|
|
|
|
if (DefMI->getOpcode() != X86::MOVPC32r)
|
|
|
|
return false;
|
|
|
|
assert(!isPICBase && "More than one PIC base?");
|
|
|
|
isPICBase = true;
|
|
|
|
}
|
|
|
|
return isPICBase;
|
|
|
|
}
|
2008-03-31 07:54:19 +00:00
|
|
|
|
2008-05-12 20:54:26 +00:00
|
|
|
bool
|
2009-10-10 00:34:18 +00:00
|
|
|
X86InstrInfo::isReallyTriviallyReMaterializable(const MachineInstr *MI,
|
|
|
|
AliasAnalysis *AA) const {
|
2007-06-14 20:50:44 +00:00
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: break;
|
2012-08-21 08:17:07 +00:00
|
|
|
case X86::MOV8rm:
|
|
|
|
case X86::MOV16rm:
|
|
|
|
case X86::MOV32rm:
|
|
|
|
case X86::MOV64rm:
|
|
|
|
case X86::LD_Fp64m:
|
|
|
|
case X86::MOVSSrm:
|
|
|
|
case X86::MOVSDrm:
|
|
|
|
case X86::MOVAPSrm:
|
|
|
|
case X86::MOVUPSrm:
|
|
|
|
case X86::MOVAPDrm:
|
|
|
|
case X86::MOVDQArm:
|
2012-12-06 06:49:16 +00:00
|
|
|
case X86::MOVDQUrm:
|
2012-08-21 08:17:07 +00:00
|
|
|
case X86::VMOVSSrm:
|
|
|
|
case X86::VMOVSDrm:
|
|
|
|
case X86::VMOVAPSrm:
|
|
|
|
case X86::VMOVUPSrm:
|
|
|
|
case X86::VMOVAPDrm:
|
|
|
|
case X86::VMOVDQArm:
|
2012-12-06 06:49:16 +00:00
|
|
|
case X86::VMOVDQUrm:
|
2012-08-21 08:17:07 +00:00
|
|
|
case X86::VMOVAPSYrm:
|
|
|
|
case X86::VMOVUPSYrm:
|
|
|
|
case X86::VMOVAPDYrm:
|
|
|
|
case X86::VMOVDQAYrm:
|
2012-12-06 06:49:16 +00:00
|
|
|
case X86::VMOVDQUYrm:
|
2012-08-21 08:17:07 +00:00
|
|
|
case X86::MMX_MOVD64rm:
|
|
|
|
case X86::MMX_MOVQ64rm:
|
|
|
|
case X86::FsVMOVAPSrm:
|
|
|
|
case X86::FsVMOVAPDrm:
|
|
|
|
case X86::FsMOVAPSrm:
|
|
|
|
case X86::FsMOVAPDrm: {
|
|
|
|
// Loads from constant pools are trivially rematerializable.
|
|
|
|
if (MI->getOperand(1).isReg() &&
|
|
|
|
MI->getOperand(2).isImm() &&
|
|
|
|
MI->getOperand(3).isReg() && MI->getOperand(3).getReg() == 0 &&
|
|
|
|
MI->isInvariantLoad(AA)) {
|
|
|
|
unsigned BaseReg = MI->getOperand(1).getReg();
|
|
|
|
if (BaseReg == 0 || BaseReg == X86::RIP)
|
|
|
|
return true;
|
|
|
|
// Allow re-materialization of PIC load.
|
|
|
|
if (!ReMatPICStubLoad && MI->getOperand(4).isGlobal())
|
|
|
|
return false;
|
|
|
|
const MachineFunction &MF = *MI->getParent()->getParent();
|
|
|
|
const MachineRegisterInfo &MRI = MF.getRegInfo();
|
|
|
|
return regIsPICBase(BaseReg, MRI);
|
2008-02-22 09:25:47 +00:00
|
|
|
}
|
2012-08-21 08:17:07 +00:00
|
|
|
return false;
|
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2012-08-21 08:17:07 +00:00
|
|
|
case X86::LEA32r:
|
|
|
|
case X86::LEA64r: {
|
|
|
|
if (MI->getOperand(2).isImm() &&
|
|
|
|
MI->getOperand(3).isReg() && MI->getOperand(3).getReg() == 0 &&
|
|
|
|
!MI->getOperand(4).isReg()) {
|
|
|
|
// lea fi#, lea GV, etc. are all rematerializable.
|
|
|
|
if (!MI->getOperand(1).isReg())
|
|
|
|
return true;
|
|
|
|
unsigned BaseReg = MI->getOperand(1).getReg();
|
|
|
|
if (BaseReg == 0)
|
|
|
|
return true;
|
|
|
|
// Allow re-materialization of lea PICBase + x.
|
|
|
|
const MachineFunction &MF = *MI->getParent()->getParent();
|
|
|
|
const MachineRegisterInfo &MRI = MF.getRegInfo();
|
|
|
|
return regIsPICBase(BaseReg, MRI);
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
2007-06-14 20:50:44 +00:00
|
|
|
}
|
2008-03-27 01:41:09 +00:00
|
|
|
|
2007-06-26 00:48:07 +00:00
|
|
|
// All other instructions marked M_REMATERIALIZABLE are always trivially
|
|
|
|
// rematerializable.
|
|
|
|
return true;
|
2007-06-14 20:50:44 +00:00
|
|
|
}
|
|
|
|
|
2008-06-24 07:10:51 +00:00
|
|
|
/// isSafeToClobberEFLAGS - Return true if it's safe insert an instruction that
|
|
|
|
/// would clobber the EFLAGS condition register. Note the result may be
|
|
|
|
/// conservative. If it cannot definitely determine the safety after visiting
|
2009-10-14 00:08:59 +00:00
|
|
|
/// a few instructions in each direction it assumes it's not safe.
|
2008-06-24 07:10:51 +00:00
|
|
|
static bool isSafeToClobberEFLAGS(MachineBasicBlock &MBB,
|
|
|
|
MachineBasicBlock::iterator I) {
|
2010-03-23 20:35:45 +00:00
|
|
|
MachineBasicBlock::iterator E = MBB.end();
|
|
|
|
|
2008-06-24 07:10:51 +00:00
|
|
|
// For compile time consideration, if we are not able to determine the
|
2009-10-14 00:08:59 +00:00
|
|
|
// safety after visiting 4 instructions in each direction, we will assume
|
|
|
|
// it's not safe.
|
|
|
|
MachineBasicBlock::iterator Iter = I;
|
2011-09-02 23:52:52 +00:00
|
|
|
for (unsigned i = 0; Iter != E && i < 4; ++i) {
|
2008-06-24 07:10:51 +00:00
|
|
|
bool SeenDef = false;
|
2009-10-14 00:08:59 +00:00
|
|
|
for (unsigned j = 0, e = Iter->getNumOperands(); j != e; ++j) {
|
|
|
|
MachineOperand &MO = Iter->getOperand(j);
|
2012-02-09 00:17:22 +00:00
|
|
|
if (MO.isRegMask() && MO.clobbersPhysReg(X86::EFLAGS))
|
|
|
|
SeenDef = true;
|
2008-10-03 15:45:36 +00:00
|
|
|
if (!MO.isReg())
|
2008-06-24 07:10:51 +00:00
|
|
|
continue;
|
|
|
|
if (MO.getReg() == X86::EFLAGS) {
|
|
|
|
if (MO.isUse())
|
|
|
|
return false;
|
|
|
|
SeenDef = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (SeenDef)
|
|
|
|
// This instruction defines EFLAGS, no need to look any further.
|
|
|
|
return true;
|
2009-10-14 00:08:59 +00:00
|
|
|
++Iter;
|
2010-03-23 20:35:45 +00:00
|
|
|
// Skip over DBG_VALUE.
|
|
|
|
while (Iter != E && Iter->isDebugValue())
|
|
|
|
++Iter;
|
2011-09-02 23:52:52 +00:00
|
|
|
}
|
2008-10-21 03:24:31 +00:00
|
|
|
|
2011-09-02 23:52:52 +00:00
|
|
|
// It is safe to clobber EFLAGS at the end of a block of no successor has it
|
|
|
|
// live in.
|
|
|
|
if (Iter == E) {
|
|
|
|
for (MachineBasicBlock::succ_iterator SI = MBB.succ_begin(),
|
|
|
|
SE = MBB.succ_end(); SI != SE; ++SI)
|
|
|
|
if ((*SI)->isLiveIn(X86::EFLAGS))
|
|
|
|
return false;
|
|
|
|
return true;
|
2009-10-14 00:08:59 +00:00
|
|
|
}
|
|
|
|
|
2010-03-23 20:35:45 +00:00
|
|
|
MachineBasicBlock::iterator B = MBB.begin();
|
2009-10-14 00:08:59 +00:00
|
|
|
Iter = I;
|
|
|
|
for (unsigned i = 0; i < 4; ++i) {
|
|
|
|
// If we make it to the beginning of the block, it's safe to clobber
|
2012-09-27 10:14:43 +00:00
|
|
|
// EFLAGS iff EFLAGS is not live-in.
|
2010-03-23 20:35:45 +00:00
|
|
|
if (Iter == B)
|
2009-10-14 00:08:59 +00:00
|
|
|
return !MBB.isLiveIn(X86::EFLAGS);
|
|
|
|
|
|
|
|
--Iter;
|
2010-03-23 20:35:45 +00:00
|
|
|
// Skip over DBG_VALUE.
|
|
|
|
while (Iter != B && Iter->isDebugValue())
|
|
|
|
--Iter;
|
|
|
|
|
2009-10-14 00:08:59 +00:00
|
|
|
bool SawKill = false;
|
|
|
|
for (unsigned j = 0, e = Iter->getNumOperands(); j != e; ++j) {
|
|
|
|
MachineOperand &MO = Iter->getOperand(j);
|
2012-02-09 00:17:22 +00:00
|
|
|
// A register mask may clobber EFLAGS, but we should still look for a
|
|
|
|
// live EFLAGS def.
|
|
|
|
if (MO.isRegMask() && MO.clobbersPhysReg(X86::EFLAGS))
|
|
|
|
SawKill = true;
|
2009-10-14 00:08:59 +00:00
|
|
|
if (MO.isReg() && MO.getReg() == X86::EFLAGS) {
|
|
|
|
if (MO.isDef()) return MO.isDead();
|
|
|
|
if (MO.isKill()) SawKill = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (SawKill)
|
|
|
|
// This instruction kills EFLAGS and doesn't redefine it, so
|
|
|
|
// there's no need to look further.
|
2008-10-21 03:24:31 +00:00
|
|
|
return true;
|
2008-06-24 07:10:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Conservative answer.
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2008-03-31 20:40:39 +00:00
|
|
|
void X86InstrInfo::reMaterialize(MachineBasicBlock &MBB,
|
|
|
|
MachineBasicBlock::iterator I,
|
2009-07-16 09:20:10 +00:00
|
|
|
unsigned DestReg, unsigned SubIdx,
|
2009-11-14 02:55:43 +00:00
|
|
|
const MachineInstr *Orig,
|
2010-06-02 22:47:25 +00:00
|
|
|
const TargetRegisterInfo &TRI) const {
|
2013-05-30 13:19:42 +00:00
|
|
|
// MOV32r0 is implemented with a xor which clobbers condition code.
|
|
|
|
// Re-materialize it as movri instructions to avoid side effects.
|
2009-07-16 09:20:10 +00:00
|
|
|
unsigned Opc = Orig->getOpcode();
|
2013-05-30 13:19:42 +00:00
|
|
|
if (Opc == X86::MOV32r0 && !isSafeToClobberEFLAGS(MBB, I)) {
|
|
|
|
DebugLoc DL = Orig->getDebugLoc();
|
|
|
|
BuildMI(MBB, I, DL, get(X86::MOV32ri)).addOperand(Orig->getOperand(0))
|
|
|
|
.addImm(0);
|
|
|
|
} else {
|
2008-07-07 23:14:23 +00:00
|
|
|
MachineInstr *MI = MBB.getParent()->CloneMachineInstr(Orig);
|
2008-03-31 20:40:39 +00:00
|
|
|
MBB.insert(I, MI);
|
|
|
|
}
|
2008-04-16 23:44:44 +00:00
|
|
|
|
2009-07-16 09:20:10 +00:00
|
|
|
MachineInstr *NewMI = prior(I);
|
2010-06-02 22:47:25 +00:00
|
|
|
NewMI->substituteRegister(Orig->getOperand(0).getReg(), DestReg, SubIdx, TRI);
|
2008-03-31 20:40:39 +00:00
|
|
|
}
|
|
|
|
|
2007-10-05 08:04:01 +00:00
|
|
|
/// hasLiveCondCodeDef - True if MI has a condition code def, e.g. EFLAGS, that
|
|
|
|
/// is not marked dead.
|
|
|
|
static bool hasLiveCondCodeDef(MachineInstr *MI) {
|
|
|
|
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
|
|
|
|
MachineOperand &MO = MI->getOperand(i);
|
2008-10-03 15:45:36 +00:00
|
|
|
if (MO.isReg() && MO.isDef() &&
|
2007-10-05 08:04:01 +00:00
|
|
|
MO.getReg() == X86::EFLAGS && !MO.isDead()) {
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2013-05-22 08:13:02 +00:00
|
|
|
/// getTruncatedShiftCount - check whether the shift count for a machine operand
|
|
|
|
/// is non-zero.
|
|
|
|
inline static unsigned getTruncatedShiftCount(MachineInstr *MI,
|
|
|
|
unsigned ShiftAmtOperandIdx) {
|
|
|
|
// The shift count is six bits with the REX.W prefix and five bits without.
|
|
|
|
unsigned ShiftCountMask = (MI->getDesc().TSFlags & X86II::REX_W) ? 63 : 31;
|
|
|
|
unsigned Imm = MI->getOperand(ShiftAmtOperandIdx).getImm();
|
|
|
|
return Imm & ShiftCountMask;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// isTruncatedShiftCountForLEA - check whether the given shift count is appropriate
|
|
|
|
/// can be represented by a LEA instruction.
|
|
|
|
inline static bool isTruncatedShiftCountForLEA(unsigned ShAmt) {
|
|
|
|
// Left shift instructions can be transformed into load-effective-address
|
|
|
|
// instructions if we can encode them appropriately.
|
|
|
|
// A LEA instruction utilizes a SIB byte to encode it's scale factor.
|
|
|
|
// The SIB.scale field is two bits wide which means that we can encode any
|
|
|
|
// shift amount less than 4.
|
|
|
|
return ShAmt < 4 && ShAmt > 0;
|
|
|
|
}
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
bool X86InstrInfo::classifyLEAReg(MachineInstr *MI, const MachineOperand &Src,
|
|
|
|
unsigned Opc, bool AllowSP,
|
|
|
|
unsigned &NewSrc, bool &isKill, bool &isUndef,
|
|
|
|
MachineOperand &ImplicitOp) const {
|
|
|
|
MachineFunction &MF = *MI->getParent()->getParent();
|
|
|
|
const TargetRegisterClass *RC;
|
|
|
|
if (AllowSP) {
|
|
|
|
RC = Opc != X86::LEA32r ? &X86::GR64RegClass : &X86::GR32RegClass;
|
|
|
|
} else {
|
|
|
|
RC = Opc != X86::LEA32r ?
|
|
|
|
&X86::GR64_NOSPRegClass : &X86::GR32_NOSPRegClass;
|
|
|
|
}
|
|
|
|
unsigned SrcReg = Src.getReg();
|
|
|
|
|
|
|
|
// For both LEA64 and LEA32 the register already has essentially the right
|
|
|
|
// type (32-bit or 64-bit) we may just need to forbid SP.
|
|
|
|
if (Opc != X86::LEA64_32r) {
|
|
|
|
NewSrc = SrcReg;
|
|
|
|
isKill = Src.isKill();
|
|
|
|
isUndef = Src.isUndef();
|
|
|
|
|
|
|
|
if (TargetRegisterInfo::isVirtualRegister(NewSrc) &&
|
|
|
|
!MF.getRegInfo().constrainRegClass(NewSrc, RC))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
// This is for an LEA64_32r and incoming registers are 32-bit. One way or
|
|
|
|
// another we need to add 64-bit registers to the final MI.
|
|
|
|
if (TargetRegisterInfo::isPhysicalRegister(SrcReg)) {
|
|
|
|
ImplicitOp = Src;
|
|
|
|
ImplicitOp.setImplicit();
|
|
|
|
|
|
|
|
NewSrc = getX86SubSuperRegister(Src.getReg(), MVT::i64);
|
|
|
|
MachineBasicBlock::LivenessQueryResult LQR =
|
|
|
|
MI->getParent()->computeRegisterLiveness(&getRegisterInfo(), NewSrc, MI);
|
|
|
|
|
|
|
|
switch (LQR) {
|
|
|
|
case MachineBasicBlock::LQR_Unknown:
|
|
|
|
// We can't give sane liveness flags to the instruction, abandon LEA
|
|
|
|
// formation.
|
|
|
|
return false;
|
|
|
|
case MachineBasicBlock::LQR_Live:
|
|
|
|
isKill = MI->killsRegister(SrcReg);
|
|
|
|
isUndef = false;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
// The physreg itself is dead, so we have to use it as an <undef>.
|
|
|
|
isKill = false;
|
|
|
|
isUndef = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// Virtual register of the wrong class, we have to create a temporary 64-bit
|
|
|
|
// vreg to feed into the LEA.
|
|
|
|
NewSrc = MF.getRegInfo().createVirtualRegister(RC);
|
|
|
|
BuildMI(*MI->getParent(), MI, MI->getDebugLoc(),
|
|
|
|
get(TargetOpcode::COPY))
|
|
|
|
.addReg(NewSrc, RegState::Define | RegState::Undef, X86::sub_32bit)
|
|
|
|
.addOperand(Src);
|
|
|
|
|
|
|
|
// Which is obviously going to be dead after we're done with it.
|
|
|
|
isKill = true;
|
|
|
|
isUndef = false;
|
|
|
|
}
|
|
|
|
|
|
|
|
// We've set all the parameters without issue.
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2009-12-12 20:03:14 +00:00
|
|
|
/// convertToThreeAddressWithLEA - Helper for convertToThreeAddress when
|
2009-12-11 06:01:48 +00:00
|
|
|
/// 16-bit LEA is disabled, use 32-bit LEA to form 3-address code by promoting
|
|
|
|
/// to a 32-bit superregister and then truncating back down to a 16-bit
|
|
|
|
/// subregister.
|
|
|
|
MachineInstr *
|
|
|
|
X86InstrInfo::convertToThreeAddressWithLEA(unsigned MIOpc,
|
|
|
|
MachineFunction::iterator &MFI,
|
|
|
|
MachineBasicBlock::iterator &MBBI,
|
|
|
|
LiveVariables *LV) const {
|
|
|
|
MachineInstr *MI = MBBI;
|
|
|
|
unsigned Dest = MI->getOperand(0).getReg();
|
|
|
|
unsigned Src = MI->getOperand(1).getReg();
|
|
|
|
bool isDead = MI->getOperand(0).isDead();
|
|
|
|
bool isKill = MI->getOperand(1).isKill();
|
|
|
|
|
|
|
|
MachineRegisterInfo &RegInfo = MFI->getParent()->getRegInfo();
|
|
|
|
unsigned leaOutReg = RegInfo.createVirtualRegister(&X86::GR32RegClass);
|
2013-06-10 20:43:49 +00:00
|
|
|
unsigned Opc, leaInReg;
|
|
|
|
if (TM.getSubtarget<X86Subtarget>().is64Bit()) {
|
|
|
|
Opc = X86::LEA64_32r;
|
|
|
|
leaInReg = RegInfo.createVirtualRegister(&X86::GR64_NOSPRegClass);
|
|
|
|
} else {
|
|
|
|
Opc = X86::LEA32r;
|
|
|
|
leaInReg = RegInfo.createVirtualRegister(&X86::GR32_NOSPRegClass);
|
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2009-12-11 06:01:48 +00:00
|
|
|
// Build and insert into an implicit UNDEF value. This is OK because
|
2011-01-26 02:03:37 +00:00
|
|
|
// well be shifting and then extracting the lower 16-bits.
|
2009-12-12 20:03:14 +00:00
|
|
|
// This has the potential to cause partial register stall. e.g.
|
2009-12-12 18:55:26 +00:00
|
|
|
// movw (%rbp,%rcx,2), %dx
|
|
|
|
// leal -65(%rdx), %esi
|
2009-12-12 20:03:14 +00:00
|
|
|
// But testing has shown this *does* help performance in 64-bit mode (at
|
|
|
|
// least on modern x86 machines).
|
2009-12-11 06:01:48 +00:00
|
|
|
BuildMI(*MFI, MBBI, MI->getDebugLoc(), get(X86::IMPLICIT_DEF), leaInReg);
|
|
|
|
MachineInstr *InsMI =
|
2010-07-08 16:40:15 +00:00
|
|
|
BuildMI(*MFI, MBBI, MI->getDebugLoc(), get(TargetOpcode::COPY))
|
|
|
|
.addReg(leaInReg, RegState::Define, X86::sub_16bit)
|
|
|
|
.addReg(Src, getKillRegState(isKill));
|
2009-12-11 06:01:48 +00:00
|
|
|
|
|
|
|
MachineInstrBuilder MIB = BuildMI(*MFI, MBBI, MI->getDebugLoc(),
|
|
|
|
get(Opc), leaOutReg);
|
|
|
|
switch (MIOpc) {
|
2012-08-21 08:16:16 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2009-12-11 06:01:48 +00:00
|
|
|
case X86::SHL16ri: {
|
|
|
|
unsigned ShAmt = MI->getOperand(2).getImm();
|
|
|
|
MIB.addReg(0).addImm(1 << ShAmt)
|
2010-07-08 23:46:44 +00:00
|
|
|
.addReg(leaInReg, RegState::Kill).addImm(0).addReg(0);
|
2009-12-11 06:01:48 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
case X86::INC16r:
|
|
|
|
case X86::INC64_16r:
|
2010-07-08 23:46:44 +00:00
|
|
|
addRegOffset(MIB, leaInReg, true, 1);
|
2009-12-11 06:01:48 +00:00
|
|
|
break;
|
|
|
|
case X86::DEC16r:
|
|
|
|
case X86::DEC64_16r:
|
2010-07-08 23:46:44 +00:00
|
|
|
addRegOffset(MIB, leaInReg, true, -1);
|
2009-12-11 06:01:48 +00:00
|
|
|
break;
|
|
|
|
case X86::ADD16ri:
|
|
|
|
case X86::ADD16ri8:
|
2010-10-08 03:57:25 +00:00
|
|
|
case X86::ADD16ri_DB:
|
|
|
|
case X86::ADD16ri8_DB:
|
2011-01-26 02:03:37 +00:00
|
|
|
addRegOffset(MIB, leaInReg, true, MI->getOperand(2).getImm());
|
2009-12-11 06:01:48 +00:00
|
|
|
break;
|
2010-10-08 03:54:52 +00:00
|
|
|
case X86::ADD16rr:
|
|
|
|
case X86::ADD16rr_DB: {
|
2009-12-11 06:01:48 +00:00
|
|
|
unsigned Src2 = MI->getOperand(2).getReg();
|
|
|
|
bool isKill2 = MI->getOperand(2).isKill();
|
|
|
|
unsigned leaInReg2 = 0;
|
|
|
|
MachineInstr *InsMI2 = 0;
|
|
|
|
if (Src == Src2) {
|
|
|
|
// ADD16rr %reg1028<kill>, %reg1028
|
|
|
|
// just a single insert_subreg.
|
|
|
|
addRegReg(MIB, leaInReg, true, leaInReg, false);
|
|
|
|
} else {
|
2013-06-10 20:43:49 +00:00
|
|
|
if (TM.getSubtarget<X86Subtarget>().is64Bit())
|
|
|
|
leaInReg2 = RegInfo.createVirtualRegister(&X86::GR64_NOSPRegClass);
|
|
|
|
else
|
|
|
|
leaInReg2 = RegInfo.createVirtualRegister(&X86::GR32_NOSPRegClass);
|
2009-12-11 06:01:48 +00:00
|
|
|
// Build and insert into an implicit UNDEF value. This is OK because
|
2011-01-26 02:03:37 +00:00
|
|
|
// well be shifting and then extracting the lower 16-bits.
|
2011-12-14 02:11:42 +00:00
|
|
|
BuildMI(*MFI, &*MIB, MI->getDebugLoc(), get(X86::IMPLICIT_DEF),leaInReg2);
|
2009-12-11 06:01:48 +00:00
|
|
|
InsMI2 =
|
2011-12-14 02:11:42 +00:00
|
|
|
BuildMI(*MFI, &*MIB, MI->getDebugLoc(), get(TargetOpcode::COPY))
|
2010-07-08 16:40:15 +00:00
|
|
|
.addReg(leaInReg2, RegState::Define, X86::sub_16bit)
|
|
|
|
.addReg(Src2, getKillRegState(isKill2));
|
2009-12-11 06:01:48 +00:00
|
|
|
addRegReg(MIB, leaInReg, true, leaInReg2, true);
|
|
|
|
}
|
|
|
|
if (LV && isKill2 && InsMI2)
|
|
|
|
LV->replaceKillInstruction(Src2, MI, InsMI2);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
MachineInstr *NewMI = MIB;
|
|
|
|
MachineInstr *ExtMI =
|
2010-07-08 16:40:22 +00:00
|
|
|
BuildMI(*MFI, MBBI, MI->getDebugLoc(), get(TargetOpcode::COPY))
|
2009-12-11 06:01:48 +00:00
|
|
|
.addReg(Dest, RegState::Define | getDeadRegState(isDead))
|
2010-07-08 16:40:22 +00:00
|
|
|
.addReg(leaOutReg, RegState::Kill, X86::sub_16bit);
|
2009-12-11 06:01:48 +00:00
|
|
|
|
|
|
|
if (LV) {
|
|
|
|
// Update live variables
|
|
|
|
LV->getVarInfo(leaInReg).Kills.push_back(NewMI);
|
|
|
|
LV->getVarInfo(leaOutReg).Kills.push_back(ExtMI);
|
|
|
|
if (isKill)
|
|
|
|
LV->replaceKillInstruction(Src, MI, InsMI);
|
|
|
|
if (isDead)
|
|
|
|
LV->replaceKillInstruction(Dest, MI, ExtMI);
|
|
|
|
}
|
|
|
|
|
|
|
|
return ExtMI;
|
|
|
|
}
|
|
|
|
|
2005-01-02 02:37:07 +00:00
|
|
|
/// convertToThreeAddress - This method must be implemented by targets that
|
|
|
|
/// set the M_CONVERTIBLE_TO_3_ADDR flag. When this flag is set, the target
|
|
|
|
/// may be able to convert a two-address instruction into a true
|
|
|
|
/// three-address instruction on demand. This allows the X86 target (for
|
|
|
|
/// example) to convert ADD and SHL instructions into LEA instructions if they
|
|
|
|
/// would require register copies due to two-addressness.
|
|
|
|
///
|
|
|
|
/// This method returns a null pointer if the transformation cannot be
|
|
|
|
/// performed, otherwise it returns the new instruction.
|
|
|
|
///
|
2006-12-01 21:52:41 +00:00
|
|
|
MachineInstr *
|
|
|
|
X86InstrInfo::convertToThreeAddress(MachineFunction::iterator &MFI,
|
|
|
|
MachineBasicBlock::iterator &MBBI,
|
2008-07-02 23:41:07 +00:00
|
|
|
LiveVariables *LV) const {
|
2006-12-01 21:52:41 +00:00
|
|
|
MachineInstr *MI = MBBI;
|
2013-05-22 08:13:02 +00:00
|
|
|
|
|
|
|
// The following opcodes also sets the condition code register(s). Only
|
|
|
|
// convert them to equivalent lea if the condition code register def's
|
|
|
|
// are dead!
|
|
|
|
if (hasLiveCondCodeDef(MI))
|
|
|
|
return 0;
|
|
|
|
|
2008-07-07 23:14:23 +00:00
|
|
|
MachineFunction &MF = *MI->getParent()->getParent();
|
2005-01-02 02:37:07 +00:00
|
|
|
// All instructions input are two-addr instructions. Get the known operands.
|
2012-08-23 22:36:31 +00:00
|
|
|
const MachineOperand &Dest = MI->getOperand(0);
|
|
|
|
const MachineOperand &Src = MI->getOperand(1);
|
2005-01-02 02:37:07 +00:00
|
|
|
|
2006-11-15 20:58:11 +00:00
|
|
|
MachineInstr *NewMI = NULL;
|
2006-12-01 21:52:41 +00:00
|
|
|
// FIXME: 16-bit LEA's are really slow on Athlons, but not bad on P4's. When
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
// we have better subtarget support, enable the 16-bit LEA generation here.
|
2009-12-12 20:03:14 +00:00
|
|
|
// 16-bit LEA is also slow on Core2.
|
2006-12-01 21:52:41 +00:00
|
|
|
bool DisableLEA16 = true;
|
2009-12-12 20:03:14 +00:00
|
|
|
bool is64Bit = TM.getSubtarget<X86Subtarget>().is64Bit();
|
2006-12-01 21:52:41 +00:00
|
|
|
|
2007-10-05 20:34:26 +00:00
|
|
|
unsigned MIOpc = MI->getOpcode();
|
|
|
|
switch (MIOpc) {
|
2006-05-30 20:26:50 +00:00
|
|
|
case X86::SHUFPSrri: {
|
|
|
|
assert(MI->getNumOperands() == 4 && "Unknown shufps instruction!");
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
if (!TM.getSubtarget<X86Subtarget>().hasSSE2()) return 0;
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2006-05-30 21:45:53 +00:00
|
|
|
unsigned B = MI->getOperand(1).getReg();
|
|
|
|
unsigned C = MI->getOperand(2).getReg();
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
if (B != C) return 0;
|
2008-07-03 09:09:37 +00:00
|
|
|
unsigned M = MI->getOperand(3).getImm();
|
2009-02-11 21:51:19 +00:00
|
|
|
NewMI = BuildMI(MF, MI->getDebugLoc(), get(X86::PSHUFDri))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest).addOperand(Src).addImm(M);
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
break;
|
|
|
|
}
|
2012-01-13 09:21:41 +00:00
|
|
|
case X86::SHUFPDrri: {
|
|
|
|
assert(MI->getNumOperands() == 4 && "Unknown shufpd instruction!");
|
|
|
|
if (!TM.getSubtarget<X86Subtarget>().hasSSE2()) return 0;
|
|
|
|
|
|
|
|
unsigned B = MI->getOperand(1).getReg();
|
|
|
|
unsigned C = MI->getOperand(2).getReg();
|
|
|
|
if (B != C) return 0;
|
|
|
|
unsigned M = MI->getOperand(3).getImm();
|
|
|
|
|
|
|
|
// Convert to PSHUFD mask.
|
|
|
|
M = ((M & 1) << 1) | ((M & 1) << 3) | ((M & 2) << 4) | ((M & 2) << 6)| 0x44;
|
|
|
|
|
|
|
|
NewMI = BuildMI(MF, MI->getDebugLoc(), get(X86::PSHUFDri))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest).addOperand(Src).addImm(M);
|
2012-01-13 09:21:41 +00:00
|
|
|
break;
|
|
|
|
}
|
2007-03-28 18:12:31 +00:00
|
|
|
case X86::SHL64ri: {
|
2007-09-14 21:48:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
|
2013-05-22 08:13:02 +00:00
|
|
|
unsigned ShAmt = getTruncatedShiftCount(MI, 2);
|
|
|
|
if (!isTruncatedShiftCountForLEA(ShAmt)) return 0;
|
2008-07-03 09:09:37 +00:00
|
|
|
|
2010-10-07 00:07:26 +00:00
|
|
|
// LEA can't handle RSP.
|
2012-08-23 22:36:31 +00:00
|
|
|
if (TargetRegisterInfo::isVirtualRegister(Src.getReg()) &&
|
|
|
|
!MF.getRegInfo().constrainRegClass(Src.getReg(),
|
|
|
|
&X86::GR64_NOSPRegClass))
|
2010-10-07 00:07:26 +00:00
|
|
|
return 0;
|
|
|
|
|
2009-02-11 21:51:19 +00:00
|
|
|
NewMI = BuildMI(MF, MI->getDebugLoc(), get(X86::LEA64r))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest)
|
|
|
|
.addReg(0).addImm(1 << ShAmt).addOperand(Src).addImm(0).addReg(0);
|
2007-03-28 18:12:31 +00:00
|
|
|
break;
|
|
|
|
}
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
case X86::SHL32ri: {
|
2007-09-14 21:48:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
|
2013-05-22 08:13:02 +00:00
|
|
|
unsigned ShAmt = getTruncatedShiftCount(MI, 2);
|
|
|
|
if (!isTruncatedShiftCountForLEA(ShAmt)) return 0;
|
2008-07-03 09:09:37 +00:00
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
unsigned Opc = is64Bit ? X86::LEA64_32r : X86::LEA32r;
|
|
|
|
|
2010-10-07 00:07:26 +00:00
|
|
|
// LEA can't handle ESP.
|
2013-06-10 20:43:49 +00:00
|
|
|
bool isKill, isUndef;
|
|
|
|
unsigned SrcReg;
|
|
|
|
MachineOperand ImplicitOp = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src, Opc, /*AllowSP=*/ false,
|
|
|
|
SrcReg, isKill, isUndef, ImplicitOp))
|
2010-10-07 00:07:26 +00:00
|
|
|
return 0;
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), get(Opc))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest)
|
2013-06-10 20:43:49 +00:00
|
|
|
.addReg(0).addImm(1 << ShAmt)
|
|
|
|
.addReg(SrcReg, getKillRegState(isKill) | getUndefRegState(isUndef))
|
|
|
|
.addImm(0).addReg(0);
|
|
|
|
if (ImplicitOp.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp);
|
|
|
|
NewMI = MIB;
|
|
|
|
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
case X86::SHL16ri: {
|
2007-09-14 21:48:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown shift instruction!");
|
2013-05-22 08:13:02 +00:00
|
|
|
unsigned ShAmt = getTruncatedShiftCount(MI, 2);
|
|
|
|
if (!isTruncatedShiftCountForLEA(ShAmt)) return 0;
|
2008-07-03 09:09:37 +00:00
|
|
|
|
2009-12-11 06:01:48 +00:00
|
|
|
if (DisableLEA16)
|
2009-12-12 20:03:14 +00:00
|
|
|
return is64Bit ? convertToThreeAddressWithLEA(MIOpc, MFI, MBBI, LV) : 0;
|
2009-12-11 06:01:48 +00:00
|
|
|
NewMI = BuildMI(MF, MI->getDebugLoc(), get(X86::LEA16r))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest)
|
|
|
|
.addReg(0).addImm(1 << ShAmt).addOperand(Src).addImm(0).addReg(0);
|
Two changes:
1) codegen a shift of a register as a shift, not an LEA.
2) teach the RA to convert a shift to an LEA instruction if it wants something
in three-address form.
This gives us asm diffs like:
- leal (,%eax,4), %eax
+ shll $2, %eax
which is faster on some processors and smaller on all of them.
and, more interestingly:
- movl 24(%esi), %eax
- leal (,%eax,4), %edi
+ movl 24(%esi), %edi
+ shll $2, %edi
Without #2, #1 was a significant pessimization in some cases.
This implements CodeGen/X86/shift-codegen.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@35204 91177308-0d34-0410-b5e6-96231b3b80d8
2007-03-20 06:08:29 +00:00
|
|
|
break;
|
2006-05-30 20:26:50 +00:00
|
|
|
}
|
2007-10-05 20:34:26 +00:00
|
|
|
default: {
|
2006-05-30 20:26:50 +00:00
|
|
|
|
2007-10-05 20:34:26 +00:00
|
|
|
switch (MIOpc) {
|
|
|
|
default: return 0;
|
|
|
|
case X86::INC64r:
|
2009-01-06 23:34:46 +00:00
|
|
|
case X86::INC32r:
|
|
|
|
case X86::INC64_32r: {
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 2 && "Unknown inc instruction!");
|
2007-10-09 07:14:53 +00:00
|
|
|
unsigned Opc = MIOpc == X86::INC64r ? X86::LEA64r
|
|
|
|
: (is64Bit ? X86::LEA64_32r : X86::LEA32r);
|
2013-06-10 20:43:49 +00:00
|
|
|
bool isKill, isUndef;
|
|
|
|
unsigned SrcReg;
|
|
|
|
MachineOperand ImplicitOp = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src, Opc, /*AllowSP=*/ false,
|
|
|
|
SrcReg, isKill, isUndef, ImplicitOp))
|
2010-10-07 00:07:26 +00:00
|
|
|
return 0;
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), get(Opc))
|
|
|
|
.addOperand(Dest)
|
|
|
|
.addReg(SrcReg, getKillRegState(isKill) | getUndefRegState(isUndef));
|
|
|
|
if (ImplicitOp.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp);
|
|
|
|
|
|
|
|
NewMI = addOffset(MIB, 1);
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
case X86::INC16r:
|
|
|
|
case X86::INC64_16r:
|
2009-12-11 06:01:48 +00:00
|
|
|
if (DisableLEA16)
|
2009-12-12 20:03:14 +00:00
|
|
|
return is64Bit ? convertToThreeAddressWithLEA(MIOpc, MFI, MBBI, LV) : 0;
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 2 && "Unknown inc instruction!");
|
2012-08-23 22:36:31 +00:00
|
|
|
NewMI = addOffset(BuildMI(MF, MI->getDebugLoc(), get(X86::LEA16r))
|
|
|
|
.addOperand(Dest).addOperand(Src), 1);
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
case X86::DEC64r:
|
2009-01-06 23:34:46 +00:00
|
|
|
case X86::DEC32r:
|
|
|
|
case X86::DEC64_32r: {
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 2 && "Unknown dec instruction!");
|
2007-10-09 07:14:53 +00:00
|
|
|
unsigned Opc = MIOpc == X86::DEC64r ? X86::LEA64r
|
|
|
|
: (is64Bit ? X86::LEA64_32r : X86::LEA32r);
|
2013-06-10 20:43:49 +00:00
|
|
|
|
|
|
|
bool isKill, isUndef;
|
|
|
|
unsigned SrcReg;
|
|
|
|
MachineOperand ImplicitOp = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src, Opc, /*AllowSP=*/ false,
|
|
|
|
SrcReg, isKill, isUndef, ImplicitOp))
|
2010-10-07 00:07:26 +00:00
|
|
|
return 0;
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), get(Opc))
|
|
|
|
.addOperand(Dest)
|
|
|
|
.addReg(SrcReg, getUndefRegState(isUndef) | getKillRegState(isKill));
|
|
|
|
if (ImplicitOp.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp);
|
|
|
|
|
|
|
|
NewMI = addOffset(MIB, -1);
|
|
|
|
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
case X86::DEC16r:
|
|
|
|
case X86::DEC64_16r:
|
2009-12-11 06:01:48 +00:00
|
|
|
if (DisableLEA16)
|
2009-12-12 20:03:14 +00:00
|
|
|
return is64Bit ? convertToThreeAddressWithLEA(MIOpc, MFI, MBBI, LV) : 0;
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 2 && "Unknown dec instruction!");
|
2012-08-23 22:36:31 +00:00
|
|
|
NewMI = addOffset(BuildMI(MF, MI->getDebugLoc(), get(X86::LEA16r))
|
|
|
|
.addOperand(Dest).addOperand(Src), -1);
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
case X86::ADD64rr:
|
2010-10-08 03:54:52 +00:00
|
|
|
case X86::ADD64rr_DB:
|
|
|
|
case X86::ADD32rr:
|
|
|
|
case X86::ADD32rr_DB: {
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown add instruction!");
|
2010-10-08 03:54:52 +00:00
|
|
|
unsigned Opc;
|
2013-06-10 20:43:49 +00:00
|
|
|
if (MIOpc == X86::ADD64rr || MIOpc == X86::ADD64rr_DB)
|
2010-10-08 03:54:52 +00:00
|
|
|
Opc = X86::LEA64r;
|
2013-06-10 20:43:49 +00:00
|
|
|
else
|
2010-10-08 03:54:52 +00:00
|
|
|
Opc = is64Bit ? X86::LEA64_32r : X86::LEA32r;
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
bool isKill, isUndef;
|
|
|
|
unsigned SrcReg;
|
|
|
|
MachineOperand ImplicitOp = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src, Opc, /*AllowSP=*/ true,
|
|
|
|
SrcReg, isKill, isUndef, ImplicitOp))
|
|
|
|
return 0;
|
2013-06-01 10:23:46 +00:00
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
const MachineOperand &Src2 = MI->getOperand(2);
|
|
|
|
bool isKill2, isUndef2;
|
|
|
|
unsigned SrcReg2;
|
|
|
|
MachineOperand ImplicitOp2 = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src2, Opc, /*AllowSP=*/ false,
|
|
|
|
SrcReg2, isKill2, isUndef2, ImplicitOp2))
|
2010-10-07 00:07:26 +00:00
|
|
|
return 0;
|
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), get(Opc))
|
|
|
|
.addOperand(Dest);
|
|
|
|
if (ImplicitOp.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp);
|
|
|
|
if (ImplicitOp2.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp2);
|
|
|
|
|
|
|
|
NewMI = addRegReg(MIB, SrcReg, isKill, SrcReg2, isKill2);
|
2012-07-16 10:52:25 +00:00
|
|
|
|
|
|
|
// Preserve undefness of the operands.
|
2013-06-01 10:23:46 +00:00
|
|
|
NewMI->getOperand(1).setIsUndef(isUndef);
|
|
|
|
NewMI->getOperand(3).setIsUndef(isUndef2);
|
2012-07-16 10:52:25 +00:00
|
|
|
|
2013-06-10 20:43:49 +00:00
|
|
|
if (LV && Src2.isKill())
|
|
|
|
LV->replaceKillInstruction(SrcReg2, MI, NewMI);
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-10-08 03:54:52 +00:00
|
|
|
case X86::ADD16rr:
|
|
|
|
case X86::ADD16rr_DB: {
|
2009-12-11 06:01:48 +00:00
|
|
|
if (DisableLEA16)
|
2009-12-12 20:03:14 +00:00
|
|
|
return is64Bit ? convertToThreeAddressWithLEA(MIOpc, MFI, MBBI, LV) : 0;
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown add instruction!");
|
2008-07-03 09:09:37 +00:00
|
|
|
unsigned Src2 = MI->getOperand(2).getReg();
|
|
|
|
bool isKill2 = MI->getOperand(2).isKill();
|
2009-02-11 21:51:19 +00:00
|
|
|
NewMI = addRegReg(BuildMI(MF, MI->getDebugLoc(), get(X86::LEA16r))
|
2012-08-23 22:36:31 +00:00
|
|
|
.addOperand(Dest),
|
|
|
|
Src.getReg(), Src.isKill(), Src2, isKill2);
|
|
|
|
|
|
|
|
// Preserve undefness of the operands.
|
|
|
|
bool isUndef = MI->getOperand(1).isUndef();
|
|
|
|
bool isUndef2 = MI->getOperand(2).isUndef();
|
|
|
|
NewMI->getOperand(1).setIsUndef(isUndef);
|
|
|
|
NewMI->getOperand(3).setIsUndef(isUndef2);
|
|
|
|
|
2008-07-03 09:09:37 +00:00
|
|
|
if (LV && isKill2)
|
|
|
|
LV->replaceKillInstruction(Src2, MI, NewMI);
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
2008-07-03 09:09:37 +00:00
|
|
|
}
|
2007-10-05 20:34:26 +00:00
|
|
|
case X86::ADD64ri32:
|
|
|
|
case X86::ADD64ri8:
|
2010-10-08 03:57:25 +00:00
|
|
|
case X86::ADD64ri32_DB:
|
|
|
|
case X86::ADD64ri8_DB:
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown add instruction!");
|
2012-08-23 22:36:31 +00:00
|
|
|
NewMI = addOffset(BuildMI(MF, MI->getDebugLoc(), get(X86::LEA64r))
|
|
|
|
.addOperand(Dest).addOperand(Src),
|
|
|
|
MI->getOperand(2).getImm());
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
case X86::ADD32ri:
|
2010-10-08 03:57:25 +00:00
|
|
|
case X86::ADD32ri8:
|
|
|
|
case X86::ADD32ri_DB:
|
|
|
|
case X86::ADD32ri8_DB: {
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown add instruction!");
|
2013-06-01 10:23:46 +00:00
|
|
|
unsigned Opc = is64Bit ? X86::LEA64_32r : X86::LEA32r;
|
2013-06-10 20:43:49 +00:00
|
|
|
|
|
|
|
bool isKill, isUndef;
|
|
|
|
unsigned SrcReg;
|
|
|
|
MachineOperand ImplicitOp = MachineOperand::CreateReg(0, false);
|
|
|
|
if (!classifyLEAReg(MI, Src, Opc, /*AllowSP=*/ true,
|
|
|
|
SrcReg, isKill, isUndef, ImplicitOp))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), get(Opc))
|
|
|
|
.addOperand(Dest)
|
|
|
|
.addReg(SrcReg, getUndefRegState(isUndef) | getKillRegState(isKill));
|
|
|
|
if (ImplicitOp.getReg() != 0)
|
|
|
|
MIB.addOperand(ImplicitOp);
|
|
|
|
|
|
|
|
NewMI = addOffset(MIB, MI->getOperand(2).getImm());
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
2009-12-11 06:01:48 +00:00
|
|
|
}
|
2007-10-05 20:34:26 +00:00
|
|
|
case X86::ADD16ri:
|
|
|
|
case X86::ADD16ri8:
|
2010-10-08 03:57:25 +00:00
|
|
|
case X86::ADD16ri_DB:
|
|
|
|
case X86::ADD16ri8_DB:
|
2009-12-11 06:01:48 +00:00
|
|
|
if (DisableLEA16)
|
2009-12-12 20:03:14 +00:00
|
|
|
return is64Bit ? convertToThreeAddressWithLEA(MIOpc, MFI, MBBI, LV) : 0;
|
2007-10-05 20:34:26 +00:00
|
|
|
assert(MI->getNumOperands() >= 3 && "Unknown add instruction!");
|
2012-08-23 22:36:31 +00:00
|
|
|
NewMI = addOffset(BuildMI(MF, MI->getDebugLoc(), get(X86::LEA16r))
|
|
|
|
.addOperand(Dest).addOperand(Src),
|
|
|
|
MI->getOperand(2).getImm());
|
2007-10-05 20:34:26 +00:00
|
|
|
break;
|
|
|
|
}
|
2005-01-02 02:37:07 +00:00
|
|
|
}
|
2006-12-01 21:52:41 +00:00
|
|
|
}
|
2007-10-05 20:34:26 +00:00
|
|
|
|
2008-02-07 08:29:53 +00:00
|
|
|
if (!NewMI) return 0;
|
|
|
|
|
2008-07-03 09:09:37 +00:00
|
|
|
if (LV) { // Update live variables
|
2012-08-23 22:36:31 +00:00
|
|
|
if (Src.isKill())
|
|
|
|
LV->replaceKillInstruction(Src.getReg(), MI, NewMI);
|
|
|
|
if (Dest.isDead())
|
|
|
|
LV->replaceKillInstruction(Dest.getReg(), MI, NewMI);
|
2008-07-03 09:09:37 +00:00
|
|
|
}
|
|
|
|
|
2011-01-26 02:03:37 +00:00
|
|
|
MFI->insert(MBBI, NewMI); // Insert the new inst
|
2006-11-15 20:58:11 +00:00
|
|
|
return NewMI;
|
2005-01-02 02:37:07 +00:00
|
|
|
}
|
|
|
|
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
/// commuteInstruction - We have a few instructions that must be hacked on to
|
|
|
|
/// commute them.
|
|
|
|
///
|
2008-06-16 07:33:11 +00:00
|
|
|
MachineInstr *
|
|
|
|
X86InstrInfo::commuteInstruction(MachineInstr *MI, bool NewMI) const {
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
switch (MI->getOpcode()) {
|
2005-01-19 07:31:24 +00:00
|
|
|
case X86::SHRD16rri8: // A = SHRD16rri8 B, C, I -> A = SHLD16rri8 C, B, (16-I)
|
|
|
|
case X86::SHLD16rri8: // A = SHLD16rri8 B, C, I -> A = SHRD16rri8 C, B, (16-I)
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
case X86::SHRD32rri8: // A = SHRD32rri8 B, C, I -> A = SHLD32rri8 C, B, (32-I)
|
2007-09-14 23:17:45 +00:00
|
|
|
case X86::SHLD32rri8: // A = SHLD32rri8 B, C, I -> A = SHRD32rri8 C, B, (32-I)
|
|
|
|
case X86::SHRD64rri8: // A = SHRD64rri8 B, C, I -> A = SHLD64rri8 C, B, (64-I)
|
|
|
|
case X86::SHLD64rri8:{// A = SHLD64rri8 B, C, I -> A = SHRD64rri8 C, B, (64-I)
|
2005-01-19 07:31:24 +00:00
|
|
|
unsigned Opc;
|
|
|
|
unsigned Size;
|
|
|
|
switch (MI->getOpcode()) {
|
2009-07-14 16:55:14 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2005-01-19 07:31:24 +00:00
|
|
|
case X86::SHRD16rri8: Size = 16; Opc = X86::SHLD16rri8; break;
|
|
|
|
case X86::SHLD16rri8: Size = 16; Opc = X86::SHRD16rri8; break;
|
|
|
|
case X86::SHRD32rri8: Size = 32; Opc = X86::SHLD32rri8; break;
|
|
|
|
case X86::SHLD32rri8: Size = 32; Opc = X86::SHRD32rri8; break;
|
2007-09-14 23:17:45 +00:00
|
|
|
case X86::SHRD64rri8: Size = 64; Opc = X86::SHLD64rri8; break;
|
|
|
|
case X86::SHLD64rri8: Size = 64; Opc = X86::SHRD64rri8; break;
|
2005-01-19 07:31:24 +00:00
|
|
|
}
|
2007-12-30 20:49:49 +00:00
|
|
|
unsigned Amt = MI->getOperand(3).getImm();
|
2008-10-17 01:23:35 +00:00
|
|
|
if (NewMI) {
|
|
|
|
MachineFunction &MF = *MI->getParent()->getParent();
|
|
|
|
MI = MF.CloneMachineInstr(MI);
|
|
|
|
NewMI = false;
|
2008-02-13 02:46:49 +00:00
|
|
|
}
|
2008-10-17 01:23:35 +00:00
|
|
|
MI->setDesc(get(Opc));
|
|
|
|
MI->getOperand(3).setImm(Size-Amt);
|
2012-11-28 02:35:17 +00:00
|
|
|
return TargetInstrInfo::commuteInstruction(MI, NewMI);
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
}
|
2012-08-21 07:32:16 +00:00
|
|
|
case X86::CMOVB16rr: case X86::CMOVB32rr: case X86::CMOVB64rr:
|
|
|
|
case X86::CMOVAE16rr: case X86::CMOVAE32rr: case X86::CMOVAE64rr:
|
|
|
|
case X86::CMOVE16rr: case X86::CMOVE32rr: case X86::CMOVE64rr:
|
|
|
|
case X86::CMOVNE16rr: case X86::CMOVNE32rr: case X86::CMOVNE64rr:
|
|
|
|
case X86::CMOVBE16rr: case X86::CMOVBE32rr: case X86::CMOVBE64rr:
|
|
|
|
case X86::CMOVA16rr: case X86::CMOVA32rr: case X86::CMOVA64rr:
|
|
|
|
case X86::CMOVL16rr: case X86::CMOVL32rr: case X86::CMOVL64rr:
|
|
|
|
case X86::CMOVGE16rr: case X86::CMOVGE32rr: case X86::CMOVGE64rr:
|
|
|
|
case X86::CMOVLE16rr: case X86::CMOVLE32rr: case X86::CMOVLE64rr:
|
|
|
|
case X86::CMOVG16rr: case X86::CMOVG32rr: case X86::CMOVG64rr:
|
|
|
|
case X86::CMOVS16rr: case X86::CMOVS32rr: case X86::CMOVS64rr:
|
|
|
|
case X86::CMOVNS16rr: case X86::CMOVNS32rr: case X86::CMOVNS64rr:
|
|
|
|
case X86::CMOVP16rr: case X86::CMOVP32rr: case X86::CMOVP64rr:
|
|
|
|
case X86::CMOVNP16rr: case X86::CMOVNP32rr: case X86::CMOVNP64rr:
|
|
|
|
case X86::CMOVO16rr: case X86::CMOVO32rr: case X86::CMOVO64rr:
|
|
|
|
case X86::CMOVNO16rr: case X86::CMOVNO32rr: case X86::CMOVNO64rr: {
|
|
|
|
unsigned Opc;
|
2007-10-05 23:13:21 +00:00
|
|
|
switch (MI->getOpcode()) {
|
2012-08-21 07:32:16 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2007-10-05 23:13:21 +00:00
|
|
|
case X86::CMOVB16rr: Opc = X86::CMOVAE16rr; break;
|
|
|
|
case X86::CMOVB32rr: Opc = X86::CMOVAE32rr; break;
|
|
|
|
case X86::CMOVB64rr: Opc = X86::CMOVAE64rr; break;
|
|
|
|
case X86::CMOVAE16rr: Opc = X86::CMOVB16rr; break;
|
|
|
|
case X86::CMOVAE32rr: Opc = X86::CMOVB32rr; break;
|
|
|
|
case X86::CMOVAE64rr: Opc = X86::CMOVB64rr; break;
|
|
|
|
case X86::CMOVE16rr: Opc = X86::CMOVNE16rr; break;
|
|
|
|
case X86::CMOVE32rr: Opc = X86::CMOVNE32rr; break;
|
|
|
|
case X86::CMOVE64rr: Opc = X86::CMOVNE64rr; break;
|
|
|
|
case X86::CMOVNE16rr: Opc = X86::CMOVE16rr; break;
|
|
|
|
case X86::CMOVNE32rr: Opc = X86::CMOVE32rr; break;
|
|
|
|
case X86::CMOVNE64rr: Opc = X86::CMOVE64rr; break;
|
2010-10-05 23:00:14 +00:00
|
|
|
case X86::CMOVBE16rr: Opc = X86::CMOVA16rr; break;
|
|
|
|
case X86::CMOVBE32rr: Opc = X86::CMOVA32rr; break;
|
|
|
|
case X86::CMOVBE64rr: Opc = X86::CMOVA64rr; break;
|
|
|
|
case X86::CMOVA16rr: Opc = X86::CMOVBE16rr; break;
|
|
|
|
case X86::CMOVA32rr: Opc = X86::CMOVBE32rr; break;
|
|
|
|
case X86::CMOVA64rr: Opc = X86::CMOVBE64rr; break;
|
2007-10-05 23:13:21 +00:00
|
|
|
case X86::CMOVL16rr: Opc = X86::CMOVGE16rr; break;
|
|
|
|
case X86::CMOVL32rr: Opc = X86::CMOVGE32rr; break;
|
|
|
|
case X86::CMOVL64rr: Opc = X86::CMOVGE64rr; break;
|
|
|
|
case X86::CMOVGE16rr: Opc = X86::CMOVL16rr; break;
|
|
|
|
case X86::CMOVGE32rr: Opc = X86::CMOVL32rr; break;
|
|
|
|
case X86::CMOVGE64rr: Opc = X86::CMOVL64rr; break;
|
|
|
|
case X86::CMOVLE16rr: Opc = X86::CMOVG16rr; break;
|
|
|
|
case X86::CMOVLE32rr: Opc = X86::CMOVG32rr; break;
|
|
|
|
case X86::CMOVLE64rr: Opc = X86::CMOVG64rr; break;
|
|
|
|
case X86::CMOVG16rr: Opc = X86::CMOVLE16rr; break;
|
|
|
|
case X86::CMOVG32rr: Opc = X86::CMOVLE32rr; break;
|
|
|
|
case X86::CMOVG64rr: Opc = X86::CMOVLE64rr; break;
|
|
|
|
case X86::CMOVS16rr: Opc = X86::CMOVNS16rr; break;
|
|
|
|
case X86::CMOVS32rr: Opc = X86::CMOVNS32rr; break;
|
2009-04-18 05:16:01 +00:00
|
|
|
case X86::CMOVS64rr: Opc = X86::CMOVNS64rr; break;
|
2007-10-05 23:13:21 +00:00
|
|
|
case X86::CMOVNS16rr: Opc = X86::CMOVS16rr; break;
|
|
|
|
case X86::CMOVNS32rr: Opc = X86::CMOVS32rr; break;
|
|
|
|
case X86::CMOVNS64rr: Opc = X86::CMOVS64rr; break;
|
|
|
|
case X86::CMOVP16rr: Opc = X86::CMOVNP16rr; break;
|
|
|
|
case X86::CMOVP32rr: Opc = X86::CMOVNP32rr; break;
|
2009-04-18 05:16:01 +00:00
|
|
|
case X86::CMOVP64rr: Opc = X86::CMOVNP64rr; break;
|
2007-10-05 23:13:21 +00:00
|
|
|
case X86::CMOVNP16rr: Opc = X86::CMOVP16rr; break;
|
|
|
|
case X86::CMOVNP32rr: Opc = X86::CMOVP32rr; break;
|
|
|
|
case X86::CMOVNP64rr: Opc = X86::CMOVP64rr; break;
|
2009-01-07 00:35:10 +00:00
|
|
|
case X86::CMOVO16rr: Opc = X86::CMOVNO16rr; break;
|
|
|
|
case X86::CMOVO32rr: Opc = X86::CMOVNO32rr; break;
|
2009-04-18 05:16:01 +00:00
|
|
|
case X86::CMOVO64rr: Opc = X86::CMOVNO64rr; break;
|
2009-01-07 00:35:10 +00:00
|
|
|
case X86::CMOVNO16rr: Opc = X86::CMOVO16rr; break;
|
|
|
|
case X86::CMOVNO32rr: Opc = X86::CMOVO32rr; break;
|
|
|
|
case X86::CMOVNO64rr: Opc = X86::CMOVO64rr; break;
|
2007-10-05 23:13:21 +00:00
|
|
|
}
|
2008-10-17 01:23:35 +00:00
|
|
|
if (NewMI) {
|
|
|
|
MachineFunction &MF = *MI->getParent()->getParent();
|
|
|
|
MI = MF.CloneMachineInstr(MI);
|
|
|
|
NewMI = false;
|
|
|
|
}
|
2008-01-11 18:10:50 +00:00
|
|
|
MI->setDesc(get(Opc));
|
2007-10-05 23:13:21 +00:00
|
|
|
// Fallthrough intended.
|
|
|
|
}
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
default:
|
2012-11-28 02:35:17 +00:00
|
|
|
return TargetInstrInfo::commuteInstruction(MI, NewMI);
|
Teach the code generator that shrd/shld is commutable if it has an immediate.
This allows us to generate this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %EDX, DWORD PTR [%ESP + 8]
shld %EDX, %EDX, 2
shl %EAX, 2
ret
instead of this:
foo:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EDX, %EAX
shrd %EDX, %ECX, 30
shl %EAX, 2
ret
Note the magically transmogrifying immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@19686 91177308-0d34-0410-b5e6-96231b3b80d8
2005-01-19 07:11:01 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-09 18:57:12 +00:00
|
|
|
static X86::CondCode getCondFromBranchOpc(unsigned BrOpc) {
|
2006-10-20 17:42:20 +00:00
|
|
|
switch (BrOpc) {
|
|
|
|
default: return X86::COND_INVALID;
|
2010-02-11 19:25:55 +00:00
|
|
|
case X86::JE_4: return X86::COND_E;
|
|
|
|
case X86::JNE_4: return X86::COND_NE;
|
|
|
|
case X86::JL_4: return X86::COND_L;
|
|
|
|
case X86::JLE_4: return X86::COND_LE;
|
|
|
|
case X86::JG_4: return X86::COND_G;
|
|
|
|
case X86::JGE_4: return X86::COND_GE;
|
|
|
|
case X86::JB_4: return X86::COND_B;
|
|
|
|
case X86::JBE_4: return X86::COND_BE;
|
|
|
|
case X86::JA_4: return X86::COND_A;
|
|
|
|
case X86::JAE_4: return X86::COND_AE;
|
|
|
|
case X86::JS_4: return X86::COND_S;
|
|
|
|
case X86::JNS_4: return X86::COND_NS;
|
|
|
|
case X86::JP_4: return X86::COND_P;
|
|
|
|
case X86::JNP_4: return X86::COND_NP;
|
|
|
|
case X86::JO_4: return X86::COND_O;
|
|
|
|
case X86::JNO_4: return X86::COND_NO;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-09 18:57:12 +00:00
|
|
|
/// getCondFromSETOpc - return condition code of a SET opcode.
|
|
|
|
static X86::CondCode getCondFromSETOpc(unsigned Opc) {
|
|
|
|
switch (Opc) {
|
|
|
|
default: return X86::COND_INVALID;
|
|
|
|
case X86::SETAr: case X86::SETAm: return X86::COND_A;
|
|
|
|
case X86::SETAEr: case X86::SETAEm: return X86::COND_AE;
|
|
|
|
case X86::SETBr: case X86::SETBm: return X86::COND_B;
|
|
|
|
case X86::SETBEr: case X86::SETBEm: return X86::COND_BE;
|
|
|
|
case X86::SETEr: case X86::SETEm: return X86::COND_E;
|
|
|
|
case X86::SETGr: case X86::SETGm: return X86::COND_G;
|
|
|
|
case X86::SETGEr: case X86::SETGEm: return X86::COND_GE;
|
|
|
|
case X86::SETLr: case X86::SETLm: return X86::COND_L;
|
|
|
|
case X86::SETLEr: case X86::SETLEm: return X86::COND_LE;
|
|
|
|
case X86::SETNEr: case X86::SETNEm: return X86::COND_NE;
|
|
|
|
case X86::SETNOr: case X86::SETNOm: return X86::COND_NO;
|
|
|
|
case X86::SETNPr: case X86::SETNPm: return X86::COND_NP;
|
|
|
|
case X86::SETNSr: case X86::SETNSm: return X86::COND_NS;
|
|
|
|
case X86::SETOr: case X86::SETOm: return X86::COND_O;
|
|
|
|
case X86::SETPr: case X86::SETPm: return X86::COND_P;
|
|
|
|
case X86::SETSr: case X86::SETSm: return X86::COND_S;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/// getCondFromCmovOpc - return condition code of a CMov opcode.
|
2012-09-20 03:06:15 +00:00
|
|
|
X86::CondCode X86::getCondFromCMovOpc(unsigned Opc) {
|
2012-07-09 18:57:12 +00:00
|
|
|
switch (Opc) {
|
|
|
|
default: return X86::COND_INVALID;
|
|
|
|
case X86::CMOVA16rm: case X86::CMOVA16rr: case X86::CMOVA32rm:
|
|
|
|
case X86::CMOVA32rr: case X86::CMOVA64rm: case X86::CMOVA64rr:
|
|
|
|
return X86::COND_A;
|
|
|
|
case X86::CMOVAE16rm: case X86::CMOVAE16rr: case X86::CMOVAE32rm:
|
|
|
|
case X86::CMOVAE32rr: case X86::CMOVAE64rm: case X86::CMOVAE64rr:
|
|
|
|
return X86::COND_AE;
|
|
|
|
case X86::CMOVB16rm: case X86::CMOVB16rr: case X86::CMOVB32rm:
|
|
|
|
case X86::CMOVB32rr: case X86::CMOVB64rm: case X86::CMOVB64rr:
|
|
|
|
return X86::COND_B;
|
|
|
|
case X86::CMOVBE16rm: case X86::CMOVBE16rr: case X86::CMOVBE32rm:
|
|
|
|
case X86::CMOVBE32rr: case X86::CMOVBE64rm: case X86::CMOVBE64rr:
|
|
|
|
return X86::COND_BE;
|
|
|
|
case X86::CMOVE16rm: case X86::CMOVE16rr: case X86::CMOVE32rm:
|
|
|
|
case X86::CMOVE32rr: case X86::CMOVE64rm: case X86::CMOVE64rr:
|
|
|
|
return X86::COND_E;
|
|
|
|
case X86::CMOVG16rm: case X86::CMOVG16rr: case X86::CMOVG32rm:
|
|
|
|
case X86::CMOVG32rr: case X86::CMOVG64rm: case X86::CMOVG64rr:
|
|
|
|
return X86::COND_G;
|
|
|
|
case X86::CMOVGE16rm: case X86::CMOVGE16rr: case X86::CMOVGE32rm:
|
|
|
|
case X86::CMOVGE32rr: case X86::CMOVGE64rm: case X86::CMOVGE64rr:
|
|
|
|
return X86::COND_GE;
|
|
|
|
case X86::CMOVL16rm: case X86::CMOVL16rr: case X86::CMOVL32rm:
|
|
|
|
case X86::CMOVL32rr: case X86::CMOVL64rm: case X86::CMOVL64rr:
|
|
|
|
return X86::COND_L;
|
|
|
|
case X86::CMOVLE16rm: case X86::CMOVLE16rr: case X86::CMOVLE32rm:
|
|
|
|
case X86::CMOVLE32rr: case X86::CMOVLE64rm: case X86::CMOVLE64rr:
|
|
|
|
return X86::COND_LE;
|
|
|
|
case X86::CMOVNE16rm: case X86::CMOVNE16rr: case X86::CMOVNE32rm:
|
|
|
|
case X86::CMOVNE32rr: case X86::CMOVNE64rm: case X86::CMOVNE64rr:
|
|
|
|
return X86::COND_NE;
|
|
|
|
case X86::CMOVNO16rm: case X86::CMOVNO16rr: case X86::CMOVNO32rm:
|
|
|
|
case X86::CMOVNO32rr: case X86::CMOVNO64rm: case X86::CMOVNO64rr:
|
|
|
|
return X86::COND_NO;
|
|
|
|
case X86::CMOVNP16rm: case X86::CMOVNP16rr: case X86::CMOVNP32rm:
|
|
|
|
case X86::CMOVNP32rr: case X86::CMOVNP64rm: case X86::CMOVNP64rr:
|
|
|
|
return X86::COND_NP;
|
|
|
|
case X86::CMOVNS16rm: case X86::CMOVNS16rr: case X86::CMOVNS32rm:
|
|
|
|
case X86::CMOVNS32rr: case X86::CMOVNS64rm: case X86::CMOVNS64rr:
|
|
|
|
return X86::COND_NS;
|
|
|
|
case X86::CMOVO16rm: case X86::CMOVO16rr: case X86::CMOVO32rm:
|
|
|
|
case X86::CMOVO32rr: case X86::CMOVO64rm: case X86::CMOVO64rr:
|
|
|
|
return X86::COND_O;
|
|
|
|
case X86::CMOVP16rm: case X86::CMOVP16rr: case X86::CMOVP32rm:
|
|
|
|
case X86::CMOVP32rr: case X86::CMOVP64rm: case X86::CMOVP64rr:
|
|
|
|
return X86::COND_P;
|
|
|
|
case X86::CMOVS16rm: case X86::CMOVS16rr: case X86::CMOVS32rm:
|
|
|
|
case X86::CMOVS32rr: case X86::CMOVS64rm: case X86::CMOVS64rr:
|
|
|
|
return X86::COND_S;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-10-20 17:42:20 +00:00
|
|
|
unsigned X86::GetCondBranchFromCond(X86::CondCode CC) {
|
|
|
|
switch (CC) {
|
2009-07-14 16:55:14 +00:00
|
|
|
default: llvm_unreachable("Illegal condition code!");
|
2010-02-11 19:25:55 +00:00
|
|
|
case X86::COND_E: return X86::JE_4;
|
|
|
|
case X86::COND_NE: return X86::JNE_4;
|
|
|
|
case X86::COND_L: return X86::JL_4;
|
|
|
|
case X86::COND_LE: return X86::JLE_4;
|
|
|
|
case X86::COND_G: return X86::JG_4;
|
|
|
|
case X86::COND_GE: return X86::JGE_4;
|
|
|
|
case X86::COND_B: return X86::JB_4;
|
|
|
|
case X86::COND_BE: return X86::JBE_4;
|
|
|
|
case X86::COND_A: return X86::JA_4;
|
|
|
|
case X86::COND_AE: return X86::JAE_4;
|
|
|
|
case X86::COND_S: return X86::JS_4;
|
|
|
|
case X86::COND_NS: return X86::JNS_4;
|
|
|
|
case X86::COND_P: return X86::JP_4;
|
|
|
|
case X86::COND_NP: return X86::JNP_4;
|
|
|
|
case X86::COND_O: return X86::JO_4;
|
|
|
|
case X86::COND_NO: return X86::JNO_4;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-10-21 05:52:40 +00:00
|
|
|
/// GetOppositeBranchCondition - Return the inverse of the specified condition,
|
|
|
|
/// e.g. turning COND_E to COND_NE.
|
|
|
|
X86::CondCode X86::GetOppositeBranchCondition(X86::CondCode CC) {
|
|
|
|
switch (CC) {
|
2009-07-14 16:55:14 +00:00
|
|
|
default: llvm_unreachable("Illegal condition code!");
|
2006-10-21 05:52:40 +00:00
|
|
|
case X86::COND_E: return X86::COND_NE;
|
|
|
|
case X86::COND_NE: return X86::COND_E;
|
|
|
|
case X86::COND_L: return X86::COND_GE;
|
|
|
|
case X86::COND_LE: return X86::COND_G;
|
|
|
|
case X86::COND_G: return X86::COND_LE;
|
|
|
|
case X86::COND_GE: return X86::COND_L;
|
|
|
|
case X86::COND_B: return X86::COND_AE;
|
|
|
|
case X86::COND_BE: return X86::COND_A;
|
|
|
|
case X86::COND_A: return X86::COND_BE;
|
|
|
|
case X86::COND_AE: return X86::COND_B;
|
|
|
|
case X86::COND_S: return X86::COND_NS;
|
|
|
|
case X86::COND_NS: return X86::COND_S;
|
|
|
|
case X86::COND_P: return X86::COND_NP;
|
|
|
|
case X86::COND_NP: return X86::COND_P;
|
|
|
|
case X86::COND_O: return X86::COND_NO;
|
|
|
|
case X86::COND_NO: return X86::COND_O;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-09 18:57:12 +00:00
|
|
|
/// getSwappedCondition - assume the flags are set by MI(a,b), return
|
|
|
|
/// the condition code if we modify the instructions such that flags are
|
|
|
|
/// set by MI(b,a).
|
2012-07-13 13:25:15 +00:00
|
|
|
static X86::CondCode getSwappedCondition(X86::CondCode CC) {
|
2012-07-09 18:57:12 +00:00
|
|
|
switch (CC) {
|
|
|
|
default: return X86::COND_INVALID;
|
|
|
|
case X86::COND_E: return X86::COND_E;
|
|
|
|
case X86::COND_NE: return X86::COND_NE;
|
|
|
|
case X86::COND_L: return X86::COND_G;
|
|
|
|
case X86::COND_LE: return X86::COND_GE;
|
|
|
|
case X86::COND_G: return X86::COND_L;
|
|
|
|
case X86::COND_GE: return X86::COND_LE;
|
|
|
|
case X86::COND_B: return X86::COND_A;
|
|
|
|
case X86::COND_BE: return X86::COND_AE;
|
|
|
|
case X86::COND_A: return X86::COND_B;
|
|
|
|
case X86::COND_AE: return X86::COND_BE;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/// getSETFromCond - Return a set opcode for the given condition and
|
|
|
|
/// whether it has memory operand.
|
|
|
|
static unsigned getSETFromCond(X86::CondCode CC,
|
|
|
|
bool HasMemoryOperand) {
|
2012-08-21 08:23:21 +00:00
|
|
|
static const uint16_t Opc[16][2] = {
|
2012-07-09 18:57:12 +00:00
|
|
|
{ X86::SETAr, X86::SETAm },
|
|
|
|
{ X86::SETAEr, X86::SETAEm },
|
|
|
|
{ X86::SETBr, X86::SETBm },
|
|
|
|
{ X86::SETBEr, X86::SETBEm },
|
|
|
|
{ X86::SETEr, X86::SETEm },
|
|
|
|
{ X86::SETGr, X86::SETGm },
|
|
|
|
{ X86::SETGEr, X86::SETGEm },
|
|
|
|
{ X86::SETLr, X86::SETLm },
|
|
|
|
{ X86::SETLEr, X86::SETLEm },
|
|
|
|
{ X86::SETNEr, X86::SETNEm },
|
|
|
|
{ X86::SETNOr, X86::SETNOm },
|
|
|
|
{ X86::SETNPr, X86::SETNPm },
|
|
|
|
{ X86::SETNSr, X86::SETNSm },
|
|
|
|
{ X86::SETOr, X86::SETOm },
|
|
|
|
{ X86::SETPr, X86::SETPm },
|
|
|
|
{ X86::SETSr, X86::SETSm }
|
|
|
|
};
|
|
|
|
|
|
|
|
assert(CC < 16 && "Can only handle standard cond codes");
|
|
|
|
return Opc[CC][HasMemoryOperand ? 1 : 0];
|
|
|
|
}
|
|
|
|
|
|
|
|
/// getCMovFromCond - Return a cmov opcode for the given condition,
|
|
|
|
/// register size in bytes, and operand type.
|
|
|
|
static unsigned getCMovFromCond(X86::CondCode CC, unsigned RegBytes,
|
|
|
|
bool HasMemoryOperand) {
|
2012-08-21 08:23:21 +00:00
|
|
|
static const uint16_t Opc[32][3] = {
|
2012-07-04 00:09:58 +00:00
|
|
|
{ X86::CMOVA16rr, X86::CMOVA32rr, X86::CMOVA64rr },
|
|
|
|
{ X86::CMOVAE16rr, X86::CMOVAE32rr, X86::CMOVAE64rr },
|
|
|
|
{ X86::CMOVB16rr, X86::CMOVB32rr, X86::CMOVB64rr },
|
|
|
|
{ X86::CMOVBE16rr, X86::CMOVBE32rr, X86::CMOVBE64rr },
|
|
|
|
{ X86::CMOVE16rr, X86::CMOVE32rr, X86::CMOVE64rr },
|
|
|
|
{ X86::CMOVG16rr, X86::CMOVG32rr, X86::CMOVG64rr },
|
|
|
|
{ X86::CMOVGE16rr, X86::CMOVGE32rr, X86::CMOVGE64rr },
|
|
|
|
{ X86::CMOVL16rr, X86::CMOVL32rr, X86::CMOVL64rr },
|
|
|
|
{ X86::CMOVLE16rr, X86::CMOVLE32rr, X86::CMOVLE64rr },
|
|
|
|
{ X86::CMOVNE16rr, X86::CMOVNE32rr, X86::CMOVNE64rr },
|
|
|
|
{ X86::CMOVNO16rr, X86::CMOVNO32rr, X86::CMOVNO64rr },
|
|
|
|
{ X86::CMOVNP16rr, X86::CMOVNP32rr, X86::CMOVNP64rr },
|
|
|
|
{ X86::CMOVNS16rr, X86::CMOVNS32rr, X86::CMOVNS64rr },
|
|
|
|
{ X86::CMOVO16rr, X86::CMOVO32rr, X86::CMOVO64rr },
|
|
|
|
{ X86::CMOVP16rr, X86::CMOVP32rr, X86::CMOVP64rr },
|
2012-07-09 18:57:12 +00:00
|
|
|
{ X86::CMOVS16rr, X86::CMOVS32rr, X86::CMOVS64rr },
|
|
|
|
{ X86::CMOVA16rm, X86::CMOVA32rm, X86::CMOVA64rm },
|
|
|
|
{ X86::CMOVAE16rm, X86::CMOVAE32rm, X86::CMOVAE64rm },
|
|
|
|
{ X86::CMOVB16rm, X86::CMOVB32rm, X86::CMOVB64rm },
|
|
|
|
{ X86::CMOVBE16rm, X86::CMOVBE32rm, X86::CMOVBE64rm },
|
|
|
|
{ X86::CMOVE16rm, X86::CMOVE32rm, X86::CMOVE64rm },
|
|
|
|
{ X86::CMOVG16rm, X86::CMOVG32rm, X86::CMOVG64rm },
|
|
|
|
{ X86::CMOVGE16rm, X86::CMOVGE32rm, X86::CMOVGE64rm },
|
|
|
|
{ X86::CMOVL16rm, X86::CMOVL32rm, X86::CMOVL64rm },
|
|
|
|
{ X86::CMOVLE16rm, X86::CMOVLE32rm, X86::CMOVLE64rm },
|
|
|
|
{ X86::CMOVNE16rm, X86::CMOVNE32rm, X86::CMOVNE64rm },
|
|
|
|
{ X86::CMOVNO16rm, X86::CMOVNO32rm, X86::CMOVNO64rm },
|
|
|
|
{ X86::CMOVNP16rm, X86::CMOVNP32rm, X86::CMOVNP64rm },
|
|
|
|
{ X86::CMOVNS16rm, X86::CMOVNS32rm, X86::CMOVNS64rm },
|
|
|
|
{ X86::CMOVO16rm, X86::CMOVO32rm, X86::CMOVO64rm },
|
|
|
|
{ X86::CMOVP16rm, X86::CMOVP32rm, X86::CMOVP64rm },
|
|
|
|
{ X86::CMOVS16rm, X86::CMOVS32rm, X86::CMOVS64rm }
|
2012-07-04 00:09:58 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
assert(CC < 16 && "Can only handle standard cond codes");
|
2012-07-09 18:57:12 +00:00
|
|
|
unsigned Idx = HasMemoryOperand ? 16+CC : CC;
|
2012-07-04 00:09:58 +00:00
|
|
|
switch(RegBytes) {
|
|
|
|
default: llvm_unreachable("Illegal register size!");
|
2012-07-09 18:57:12 +00:00
|
|
|
case 2: return Opc[Idx][0];
|
|
|
|
case 4: return Opc[Idx][1];
|
|
|
|
case 8: return Opc[Idx][2];
|
2012-07-04 00:09:58 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-06-14 22:03:45 +00:00
|
|
|
bool X86InstrInfo::isUnpredicatedTerminator(const MachineInstr *MI) const {
|
2011-12-07 07:15:52 +00:00
|
|
|
if (!MI->isTerminator()) return false;
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:56:04 +00:00
|
|
|
// Conditional branch is a special case.
|
2011-12-07 07:15:52 +00:00
|
|
|
if (MI->isBranch() && !MI->isBarrier())
|
2008-01-07 01:56:04 +00:00
|
|
|
return true;
|
2011-12-07 07:15:52 +00:00
|
|
|
if (!MI->isPredicable())
|
2008-01-07 01:56:04 +00:00
|
|
|
return true;
|
|
|
|
return !isPredicated(MI);
|
2007-06-14 22:03:45 +00:00
|
|
|
}
|
2006-10-21 05:52:40 +00:00
|
|
|
|
2011-01-26 02:03:37 +00:00
|
|
|
bool X86InstrInfo::AnalyzeBranch(MachineBasicBlock &MBB,
|
2006-10-20 17:42:20 +00:00
|
|
|
MachineBasicBlock *&TBB,
|
|
|
|
MachineBasicBlock *&FBB,
|
2009-02-09 07:14:22 +00:00
|
|
|
SmallVectorImpl<MachineOperand> &Cond,
|
|
|
|
bool AllowModify) const {
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Start from the bottom of the block and work up, examining the
|
|
|
|
// terminator instructions.
|
2006-10-20 17:42:20 +00:00
|
|
|
MachineBasicBlock::iterator I = MBB.end();
|
2010-04-13 18:50:27 +00:00
|
|
|
MachineBasicBlock::iterator UnCondBrIter = MBB.end();
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
while (I != MBB.begin()) {
|
|
|
|
--I;
|
2010-04-02 01:38:09 +00:00
|
|
|
if (I->isDebugValue())
|
|
|
|
continue;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
|
|
|
// Working from the bottom, when we see a non-terminator instruction, we're
|
|
|
|
// done.
|
2010-07-16 17:41:44 +00:00
|
|
|
if (!isUnpredicatedTerminator(I))
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
break;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
|
|
|
// A terminator that isn't a branch can't easily be handled by this
|
|
|
|
// analysis.
|
2011-12-07 07:15:52 +00:00
|
|
|
if (!I->isBranch())
|
2006-10-20 17:42:20 +00:00
|
|
|
return true;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Handle unconditional branches.
|
2010-02-11 19:25:55 +00:00
|
|
|
if (I->getOpcode() == X86::JMP_4) {
|
2010-04-13 18:50:27 +00:00
|
|
|
UnCondBrIter = I;
|
|
|
|
|
2009-02-09 07:14:22 +00:00
|
|
|
if (!AllowModify) {
|
|
|
|
TBB = I->getOperand(0).getMBB();
|
2009-05-08 06:34:09 +00:00
|
|
|
continue;
|
2009-02-09 07:14:22 +00:00
|
|
|
}
|
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// If the block has any instructions after a JMP, delete them.
|
2009-12-03 00:50:42 +00:00
|
|
|
while (llvm::next(I) != MBB.end())
|
|
|
|
llvm::next(I)->eraseFromParent();
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
Cond.clear();
|
|
|
|
FBB = 0;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Delete the JMP if it's equivalent to a fall-through.
|
|
|
|
if (MBB.isLayoutSuccessor(I->getOperand(0).getMBB())) {
|
|
|
|
TBB = 0;
|
|
|
|
I->eraseFromParent();
|
|
|
|
I = MBB.end();
|
2010-04-13 18:50:27 +00:00
|
|
|
UnCondBrIter = MBB.end();
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
continue;
|
|
|
|
}
|
2009-12-14 06:51:19 +00:00
|
|
|
|
2010-04-13 18:50:27 +00:00
|
|
|
// TBB is used to indicate the unconditional destination.
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
TBB = I->getOperand(0).getMBB();
|
|
|
|
continue;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Handle conditional branches.
|
2012-07-09 18:57:12 +00:00
|
|
|
X86::CondCode BranchCode = getCondFromBranchOpc(I->getOpcode());
|
2006-10-20 17:42:20 +00:00
|
|
|
if (BranchCode == X86::COND_INVALID)
|
|
|
|
return true; // Can't handle indirect branch.
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Working from the bottom, handle the first conditional branch.
|
|
|
|
if (Cond.empty()) {
|
2010-04-13 18:50:27 +00:00
|
|
|
MachineBasicBlock *TargetBB = I->getOperand(0).getMBB();
|
|
|
|
if (AllowModify && UnCondBrIter != MBB.end() &&
|
|
|
|
MBB.isLayoutSuccessor(TargetBB)) {
|
|
|
|
// If we can modify the code and it ends in something like:
|
|
|
|
//
|
|
|
|
// jCC L1
|
|
|
|
// jmp L2
|
|
|
|
// L1:
|
|
|
|
// ...
|
|
|
|
// L2:
|
|
|
|
//
|
|
|
|
// Then we can change this to:
|
|
|
|
//
|
|
|
|
// jnCC L2
|
|
|
|
// L1:
|
|
|
|
// ...
|
|
|
|
// L2:
|
|
|
|
//
|
|
|
|
// Which is a bit more efficient.
|
|
|
|
// We conditionally jump to the fall-through block.
|
|
|
|
BranchCode = GetOppositeBranchCondition(BranchCode);
|
|
|
|
unsigned JNCC = GetCondBranchFromCond(BranchCode);
|
|
|
|
MachineBasicBlock::iterator OldInst = I;
|
|
|
|
|
|
|
|
BuildMI(MBB, UnCondBrIter, MBB.findDebugLoc(I), get(JNCC))
|
|
|
|
.addMBB(UnCondBrIter->getOperand(0).getMBB());
|
|
|
|
BuildMI(MBB, UnCondBrIter, MBB.findDebugLoc(I), get(X86::JMP_4))
|
|
|
|
.addMBB(TargetBB);
|
|
|
|
|
|
|
|
OldInst->eraseFromParent();
|
|
|
|
UnCondBrIter->eraseFromParent();
|
|
|
|
|
|
|
|
// Restart the analysis.
|
|
|
|
UnCondBrIter = MBB.end();
|
|
|
|
I = MBB.end();
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
FBB = TBB;
|
|
|
|
TBB = I->getOperand(0).getMBB();
|
|
|
|
Cond.push_back(MachineOperand::CreateImm(BranchCode));
|
|
|
|
continue;
|
|
|
|
}
|
2009-12-14 06:51:19 +00:00
|
|
|
|
|
|
|
// Handle subsequent conditional branches. Only handle the case where all
|
|
|
|
// conditional branches branch to the same destination and their condition
|
|
|
|
// opcodes fit one of the special multi-branch idioms.
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
assert(Cond.size() == 1);
|
|
|
|
assert(TBB);
|
2009-12-14 06:51:19 +00:00
|
|
|
|
|
|
|
// Only handle the case where all conditional branches branch to the same
|
|
|
|
// destination.
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
if (TBB != I->getOperand(0).getMBB())
|
|
|
|
return true;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// If the conditions are the same, we can leave them alone.
|
2009-12-14 06:51:19 +00:00
|
|
|
X86::CondCode OldBranchCode = (X86::CondCode)Cond[0].getImm();
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
if (OldBranchCode == BranchCode)
|
|
|
|
continue;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
|
|
|
// If they differ, see if they fit one of the known patterns. Theoretically,
|
|
|
|
// we could handle more patterns here, but we shouldn't expect to see them
|
|
|
|
// if instruction selection has done a reasonable job.
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
if ((OldBranchCode == X86::COND_NP &&
|
|
|
|
BranchCode == X86::COND_E) ||
|
|
|
|
(OldBranchCode == X86::COND_E &&
|
|
|
|
BranchCode == X86::COND_NP))
|
|
|
|
BranchCode = X86::COND_NP_OR_E;
|
|
|
|
else if ((OldBranchCode == X86::COND_P &&
|
|
|
|
BranchCode == X86::COND_NE) ||
|
|
|
|
(OldBranchCode == X86::COND_NE &&
|
|
|
|
BranchCode == X86::COND_P))
|
|
|
|
BranchCode = X86::COND_NE_OR_P;
|
|
|
|
else
|
|
|
|
return true;
|
2009-12-14 06:51:19 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
// Update the MachineOperand.
|
|
|
|
Cond[0].setImm(BranchCode);
|
2007-06-13 17:59:52 +00:00
|
|
|
}
|
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
return false;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
|
2007-05-18 00:18:17 +00:00
|
|
|
unsigned X86InstrInfo::RemoveBranch(MachineBasicBlock &MBB) const {
|
2006-10-20 17:42:20 +00:00
|
|
|
MachineBasicBlock::iterator I = MBB.end();
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
unsigned Count = 0;
|
|
|
|
|
|
|
|
while (I != MBB.begin()) {
|
|
|
|
--I;
|
2010-04-02 01:38:09 +00:00
|
|
|
if (I->isDebugValue())
|
|
|
|
continue;
|
2010-02-11 19:25:55 +00:00
|
|
|
if (I->getOpcode() != X86::JMP_4 &&
|
2012-07-09 18:57:12 +00:00
|
|
|
getCondFromBranchOpc(I->getOpcode()) == X86::COND_INVALID)
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
break;
|
|
|
|
// Remove the branch.
|
|
|
|
I->eraseFromParent();
|
|
|
|
I = MBB.end();
|
|
|
|
++Count;
|
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
return Count;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
|
2007-05-18 00:18:17 +00:00
|
|
|
unsigned
|
|
|
|
X86InstrInfo::InsertBranch(MachineBasicBlock &MBB, MachineBasicBlock *TBB,
|
|
|
|
MachineBasicBlock *FBB,
|
2010-06-17 22:43:56 +00:00
|
|
|
const SmallVectorImpl<MachineOperand> &Cond,
|
|
|
|
DebugLoc DL) const {
|
2006-10-20 17:42:20 +00:00
|
|
|
// Shouldn't be a fall through.
|
|
|
|
assert(TBB && "InsertBranch must not be told to insert a fallthrough");
|
2006-10-21 05:34:23 +00:00
|
|
|
assert((Cond.size() == 1 || Cond.size() == 0) &&
|
|
|
|
"X86 branch conditions have one component!");
|
|
|
|
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
if (Cond.empty()) {
|
|
|
|
// Unconditional branch?
|
|
|
|
assert(!FBB && "Unconditional branch with multiple successors!");
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JMP_4)).addMBB(TBB);
|
2007-05-18 00:18:17 +00:00
|
|
|
return 1;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
|
|
|
|
// Conditional branch.
|
|
|
|
unsigned Count = 0;
|
|
|
|
X86::CondCode CC = (X86::CondCode)Cond[0].getImm();
|
|
|
|
switch (CC) {
|
|
|
|
case X86::COND_NP_OR_E:
|
|
|
|
// Synthesize NP_OR_E with two branches.
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JNP_4)).addMBB(TBB);
|
2010-03-05 00:33:59 +00:00
|
|
|
++Count;
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JE_4)).addMBB(TBB);
|
2010-03-05 00:33:59 +00:00
|
|
|
++Count;
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
break;
|
|
|
|
case X86::COND_NE_OR_P:
|
|
|
|
// Synthesize NE_OR_P with two branches.
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JNE_4)).addMBB(TBB);
|
2010-03-05 00:33:59 +00:00
|
|
|
++Count;
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JP_4)).addMBB(TBB);
|
2010-03-05 00:33:59 +00:00
|
|
|
++Count;
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
break;
|
2010-03-05 00:33:59 +00:00
|
|
|
default: {
|
|
|
|
unsigned Opc = GetCondBranchFromCond(CC);
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(Opc)).addMBB(TBB);
|
2010-03-05 00:33:59 +00:00
|
|
|
++Count;
|
|
|
|
}
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
}
|
|
|
|
if (FBB) {
|
|
|
|
// Two-way Conditional branch. Insert the second branch.
|
2010-06-17 22:43:56 +00:00
|
|
|
BuildMI(&MBB, DL, get(X86::JMP_4)).addMBB(FBB);
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
++Count;
|
|
|
|
}
|
|
|
|
return Count;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
|
2012-07-04 00:09:58 +00:00
|
|
|
bool X86InstrInfo::
|
|
|
|
canInsertSelect(const MachineBasicBlock &MBB,
|
|
|
|
const SmallVectorImpl<MachineOperand> &Cond,
|
|
|
|
unsigned TrueReg, unsigned FalseReg,
|
|
|
|
int &CondCycles, int &TrueCycles, int &FalseCycles) const {
|
|
|
|
// Not all subtargets have cmov instructions.
|
|
|
|
if (!TM.getSubtarget<X86Subtarget>().hasCMov())
|
|
|
|
return false;
|
|
|
|
if (Cond.size() != 1)
|
|
|
|
return false;
|
|
|
|
// We cannot do the composite conditions, at least not in SSA form.
|
|
|
|
if ((X86::CondCode)Cond[0].getImm() > X86::COND_S)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
// Check register classes.
|
|
|
|
const MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
|
|
|
|
const TargetRegisterClass *RC =
|
|
|
|
RI.getCommonSubClass(MRI.getRegClass(TrueReg), MRI.getRegClass(FalseReg));
|
|
|
|
if (!RC)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
// We have cmov instructions for 16, 32, and 64 bit general purpose registers.
|
|
|
|
if (X86::GR16RegClass.hasSubClassEq(RC) ||
|
|
|
|
X86::GR32RegClass.hasSubClassEq(RC) ||
|
|
|
|
X86::GR64RegClass.hasSubClassEq(RC)) {
|
|
|
|
// This latency applies to Pentium M, Merom, Wolfdale, Nehalem, and Sandy
|
|
|
|
// Bridge. Probably Ivy Bridge as well.
|
|
|
|
CondCycles = 2;
|
|
|
|
TrueCycles = 2;
|
|
|
|
FalseCycles = 2;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Can't do vectors.
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
void X86InstrInfo::insertSelect(MachineBasicBlock &MBB,
|
|
|
|
MachineBasicBlock::iterator I, DebugLoc DL,
|
|
|
|
unsigned DstReg,
|
|
|
|
const SmallVectorImpl<MachineOperand> &Cond,
|
|
|
|
unsigned TrueReg, unsigned FalseReg) const {
|
|
|
|
MachineRegisterInfo &MRI = MBB.getParent()->getRegInfo();
|
|
|
|
assert(Cond.size() == 1 && "Invalid Cond array");
|
|
|
|
unsigned Opc = getCMovFromCond((X86::CondCode)Cond[0].getImm(),
|
2012-07-09 18:57:12 +00:00
|
|
|
MRI.getRegClass(DstReg)->getSize(),
|
|
|
|
false/*HasMemoryOperand*/);
|
2012-07-04 00:09:58 +00:00
|
|
|
BuildMI(MBB, I, DL, get(Opc), DstReg).addReg(FalseReg).addReg(TrueReg);
|
|
|
|
}
|
|
|
|
|
2009-04-15 00:04:23 +00:00
|
|
|
/// isHReg - Test if the given register is a physical h register.
|
|
|
|
static bool isHReg(unsigned Reg) {
|
2009-04-27 16:41:36 +00:00
|
|
|
return X86::GR8_ABCD_HRegClass.contains(Reg);
|
2009-04-15 00:04:23 +00:00
|
|
|
}
|
|
|
|
|
2010-08-27 14:43:06 +00:00
|
|
|
// Try and copy between VR128/VR64 and GR64 registers.
|
2011-09-14 02:36:58 +00:00
|
|
|
static unsigned CopyToFromAsymmetricReg(unsigned DestReg, unsigned SrcReg,
|
2013-08-11 07:55:09 +00:00
|
|
|
const X86Subtarget& Subtarget) {
|
|
|
|
|
|
|
|
|
2010-08-27 14:43:06 +00:00
|
|
|
// SrcReg(VR128) -> DestReg(GR64)
|
|
|
|
// SrcReg(VR64) -> DestReg(GR64)
|
|
|
|
// SrcReg(GR64) -> DestReg(VR128)
|
|
|
|
// SrcReg(GR64) -> DestReg(VR64)
|
|
|
|
|
2013-08-11 07:55:09 +00:00
|
|
|
bool HasAVX = Subtarget.hasAVX();
|
2013-08-18 13:08:57 +00:00
|
|
|
bool HasAVX512 = Subtarget.hasAVX512();
|
2010-08-27 14:43:06 +00:00
|
|
|
if (X86::GR64RegClass.contains(DestReg)) {
|
2013-08-18 13:08:57 +00:00
|
|
|
if (X86::VR128XRegClass.contains(SrcReg))
|
2010-08-27 14:43:06 +00:00
|
|
|
// Copy from a VR128 register to a GR64 register.
|
2013-08-18 13:08:57 +00:00
|
|
|
return HasAVX512 ? X86::VMOVPQIto64Zrr: (HasAVX ? X86::VMOVPQIto64rr :
|
|
|
|
X86::MOVPQIto64rr);
|
2012-08-21 08:29:51 +00:00
|
|
|
if (X86::VR64RegClass.contains(SrcReg))
|
2010-08-27 14:43:06 +00:00
|
|
|
// Copy from a VR64 register to a GR64 register.
|
|
|
|
return X86::MOVSDto64rr;
|
|
|
|
} else if (X86::GR64RegClass.contains(SrcReg)) {
|
|
|
|
// Copy from a GR64 register to a VR128 register.
|
2013-08-18 13:08:57 +00:00
|
|
|
if (X86::VR128XRegClass.contains(DestReg))
|
|
|
|
return HasAVX512 ? X86::VMOV64toPQIZrr: (HasAVX ? X86::VMOV64toPQIrr :
|
|
|
|
X86::MOV64toPQIrr);
|
2010-08-27 14:43:06 +00:00
|
|
|
// Copy from a GR64 register to a VR64 register.
|
2012-08-21 08:29:51 +00:00
|
|
|
if (X86::VR64RegClass.contains(DestReg))
|
2010-08-27 14:43:06 +00:00
|
|
|
return X86::MOV64toSDrr;
|
|
|
|
}
|
|
|
|
|
2011-09-22 22:45:24 +00:00
|
|
|
// SrcReg(FR32) -> DestReg(GR32)
|
|
|
|
// SrcReg(GR32) -> DestReg(FR32)
|
|
|
|
|
2013-08-18 13:08:57 +00:00
|
|
|
if (X86::GR32RegClass.contains(DestReg) && X86::FR32XRegClass.contains(SrcReg))
|
2012-08-21 08:29:51 +00:00
|
|
|
// Copy from a FR32 register to a GR32 register.
|
2013-08-18 13:08:57 +00:00
|
|
|
return HasAVX512 ? X86::VMOVSS2DIZrr : (HasAVX ? X86::VMOVSS2DIrr : X86::MOVSS2DIrr);
|
2011-09-22 22:45:24 +00:00
|
|
|
|
2013-08-18 13:08:57 +00:00
|
|
|
if (X86::FR32XRegClass.contains(DestReg) && X86::GR32RegClass.contains(SrcReg))
|
2012-08-21 08:29:51 +00:00
|
|
|
// Copy from a GR32 register to a FR32 register.
|
2013-08-18 13:08:57 +00:00
|
|
|
return HasAVX512 ? X86::VMOVDI2SSZrr : (HasAVX ? X86::VMOVDI2SSrr : X86::MOVDI2SSrr);
|
2010-08-27 14:43:06 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-08-11 07:55:09 +00:00
|
|
|
static
|
|
|
|
unsigned copyPhysRegOpcode_AVX512(unsigned& DestReg, unsigned& SrcReg) {
|
|
|
|
if (X86::VR128XRegClass.contains(DestReg, SrcReg) ||
|
|
|
|
X86::VR256XRegClass.contains(DestReg, SrcReg) ||
|
|
|
|
X86::VR512RegClass.contains(DestReg, SrcReg)) {
|
|
|
|
DestReg = get512BitSuperRegister(DestReg);
|
|
|
|
SrcReg = get512BitSuperRegister(SrcReg);
|
|
|
|
return X86::VMOVAPSZrr;
|
|
|
|
}
|
|
|
|
if ((X86::VK8RegClass.contains(DestReg) ||
|
|
|
|
X86::VK16RegClass.contains(DestReg)) &&
|
|
|
|
(X86::VK8RegClass.contains(SrcReg) ||
|
|
|
|
X86::VK16RegClass.contains(SrcReg)))
|
|
|
|
return X86::KMOVWkk;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2010-07-08 19:46:25 +00:00
|
|
|
void X86InstrInfo::copyPhysReg(MachineBasicBlock &MBB,
|
|
|
|
MachineBasicBlock::iterator MI, DebugLoc DL,
|
|
|
|
unsigned DestReg, unsigned SrcReg,
|
|
|
|
bool KillSrc) const {
|
|
|
|
// First deal with the normal symmetric copies.
|
2011-09-14 02:36:58 +00:00
|
|
|
bool HasAVX = TM.getSubtarget<X86Subtarget>().hasAVX();
|
2013-08-11 07:55:09 +00:00
|
|
|
bool HasAVX512 = TM.getSubtarget<X86Subtarget>().hasAVX512();
|
|
|
|
unsigned Opc = 0;
|
2010-07-08 19:46:25 +00:00
|
|
|
if (X86::GR64RegClass.contains(DestReg, SrcReg))
|
|
|
|
Opc = X86::MOV64rr;
|
|
|
|
else if (X86::GR32RegClass.contains(DestReg, SrcReg))
|
|
|
|
Opc = X86::MOV32rr;
|
|
|
|
else if (X86::GR16RegClass.contains(DestReg, SrcReg))
|
|
|
|
Opc = X86::MOV16rr;
|
|
|
|
else if (X86::GR8RegClass.contains(DestReg, SrcReg)) {
|
|
|
|
// Copying to or from a physical H register on x86-64 requires a NOREX
|
|
|
|
// move. Otherwise use a normal move.
|
|
|
|
if ((isHReg(DestReg) || isHReg(SrcReg)) &&
|
2011-10-07 20:15:54 +00:00
|
|
|
TM.getSubtarget<X86Subtarget>().is64Bit()) {
|
2010-07-08 19:46:25 +00:00
|
|
|
Opc = X86::MOV8rr_NOREX;
|
2011-10-07 20:15:54 +00:00
|
|
|
// Both operands must be encodable without an REX prefix.
|
|
|
|
assert(X86::GR8_NOREXRegClass.contains(SrcReg, DestReg) &&
|
|
|
|
"8-bit H register can not be copied outside GR8_NOREX");
|
|
|
|
} else
|
2010-07-08 19:46:25 +00:00
|
|
|
Opc = X86::MOV8rr;
|
2013-08-11 07:55:09 +00:00
|
|
|
}
|
|
|
|
else if (X86::VR64RegClass.contains(DestReg, SrcReg))
|
|
|
|
Opc = X86::MMX_MOVQ64rr;
|
|
|
|
else if (HasAVX512)
|
|
|
|
Opc = copyPhysRegOpcode_AVX512(DestReg, SrcReg);
|
|
|
|
else if (X86::VR128RegClass.contains(DestReg, SrcReg))
|
2011-09-14 02:36:58 +00:00
|
|
|
Opc = HasAVX ? X86::VMOVAPSrr : X86::MOVAPSrr;
|
2011-07-14 18:50:58 +00:00
|
|
|
else if (X86::VR256RegClass.contains(DestReg, SrcReg))
|
|
|
|
Opc = X86::VMOVAPSYrr;
|
2013-08-11 07:55:09 +00:00
|
|
|
if (!Opc)
|
|
|
|
Opc = CopyToFromAsymmetricReg(DestReg, SrcReg, TM.getSubtarget<X86Subtarget>());
|
2010-07-08 19:46:25 +00:00
|
|
|
|
|
|
|
if (Opc) {
|
|
|
|
BuildMI(MBB, MI, DL, get(Opc), DestReg)
|
|
|
|
.addReg(SrcReg, getKillRegState(KillSrc));
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Moving EFLAGS to / from another register requires a push and a pop.
|
2012-12-21 23:48:49 +00:00
|
|
|
// Notice that we have to adjust the stack if we don't want to clobber the
|
|
|
|
// first frame index. See X86FrameLowering.cpp - colobbersTheStack.
|
2010-07-08 19:46:25 +00:00
|
|
|
if (SrcReg == X86::EFLAGS) {
|
|
|
|
if (X86::GR64RegClass.contains(DestReg)) {
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::PUSHF64));
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::POP64r), DestReg);
|
|
|
|
return;
|
2012-08-21 08:29:51 +00:00
|
|
|
}
|
|
|
|
if (X86::GR32RegClass.contains(DestReg)) {
|
2010-07-08 19:46:25 +00:00
|
|
|
BuildMI(MBB, MI, DL, get(X86::PUSHF32));
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::POP32r), DestReg);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (DestReg == X86::EFLAGS) {
|
|
|
|
if (X86::GR64RegClass.contains(SrcReg)) {
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::PUSH64r))
|
|
|
|
.addReg(SrcReg, getKillRegState(KillSrc));
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::POPF64));
|
|
|
|
return;
|
2012-08-21 08:29:51 +00:00
|
|
|
}
|
|
|
|
if (X86::GR32RegClass.contains(SrcReg)) {
|
2010-07-08 19:46:25 +00:00
|
|
|
BuildMI(MBB, MI, DL, get(X86::PUSH32r))
|
|
|
|
.addReg(SrcReg, getKillRegState(KillSrc));
|
|
|
|
BuildMI(MBB, MI, DL, get(X86::POPF32));
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
DEBUG(dbgs() << "Cannot copy " << RI.getName(SrcReg)
|
|
|
|
<< " to " << RI.getName(DestReg) << '\n');
|
|
|
|
llvm_unreachable("Cannot emit physreg copy instruction");
|
|
|
|
}
|
|
|
|
|
2010-06-12 20:13:29 +00:00
|
|
|
static unsigned getLoadStoreRegOpcode(unsigned Reg,
|
|
|
|
const TargetRegisterClass *RC,
|
|
|
|
bool isStackAligned,
|
|
|
|
const TargetMachine &TM,
|
|
|
|
bool load) {
|
2013-08-18 13:08:57 +00:00
|
|
|
if (TM.getSubtarget<X86Subtarget>().hasAVX512()) {
|
2013-10-14 22:18:56 +00:00
|
|
|
if (X86::VK8RegClass.hasSubClassEq(RC) ||
|
2013-08-18 13:08:57 +00:00
|
|
|
X86::VK16RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::KMOVWkm : X86::KMOVWmk;
|
2013-10-02 12:20:42 +00:00
|
|
|
if (RC->getSize() == 4 && X86::FR32XRegClass.hasSubClassEq(RC))
|
2013-08-18 13:08:57 +00:00
|
|
|
return load ? X86::VMOVSSZrm : X86::VMOVSSZmr;
|
2013-10-02 12:20:42 +00:00
|
|
|
if (RC->getSize() == 8 && X86::FR64XRegClass.hasSubClassEq(RC))
|
2013-08-18 13:08:57 +00:00
|
|
|
return load ? X86::VMOVSDZrm : X86::VMOVSDZmr;
|
2013-10-02 12:20:42 +00:00
|
|
|
if (X86::VR512RegClass.hasSubClassEq(RC))
|
2013-08-18 13:08:57 +00:00
|
|
|
return load ? X86::VMOVUPSZrm : X86::VMOVUPSZmr;
|
|
|
|
}
|
|
|
|
|
2011-09-14 02:36:58 +00:00
|
|
|
bool HasAVX = TM.getSubtarget<X86Subtarget>().hasAVX();
|
2011-06-01 15:32:10 +00:00
|
|
|
switch (RC->getSize()) {
|
2010-07-12 03:43:04 +00:00
|
|
|
default:
|
2011-06-01 15:32:10 +00:00
|
|
|
llvm_unreachable("Unknown spill size");
|
|
|
|
case 1:
|
|
|
|
assert(X86::GR8RegClass.hasSubClassEq(RC) && "Unknown 1-byte regclass");
|
2009-04-27 16:41:36 +00:00
|
|
|
if (TM.getSubtarget<X86Subtarget>().is64Bit())
|
2011-06-01 15:32:10 +00:00
|
|
|
// Copying to or from a physical H register on x86-64 requires a NOREX
|
|
|
|
// move. Otherwise use a normal move.
|
|
|
|
if (isHReg(Reg) || X86::GR8_ABCD_HRegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::MOV8rm_NOREX : X86::MOV8mr_NOREX;
|
|
|
|
return load ? X86::MOV8rm : X86::MOV8mr;
|
|
|
|
case 2:
|
|
|
|
assert(X86::GR16RegClass.hasSubClassEq(RC) && "Unknown 2-byte regclass");
|
|
|
|
return load ? X86::MOV16rm : X86::MOV16mr;
|
|
|
|
case 4:
|
|
|
|
if (X86::GR32RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::MOV32rm : X86::MOV32mr;
|
|
|
|
if (X86::FR32RegClass.hasSubClassEq(RC))
|
2011-09-14 02:36:58 +00:00
|
|
|
return load ?
|
|
|
|
(HasAVX ? X86::VMOVSSrm : X86::MOVSSrm) :
|
|
|
|
(HasAVX ? X86::VMOVSSmr : X86::MOVSSmr);
|
2011-06-01 15:32:10 +00:00
|
|
|
if (X86::RFP32RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::LD_Fp32m : X86::ST_Fp32m;
|
|
|
|
llvm_unreachable("Unknown 4-byte regclass");
|
|
|
|
case 8:
|
|
|
|
if (X86::GR64RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::MOV64rm : X86::MOV64mr;
|
|
|
|
if (X86::FR64RegClass.hasSubClassEq(RC))
|
2011-09-14 02:36:58 +00:00
|
|
|
return load ?
|
|
|
|
(HasAVX ? X86::VMOVSDrm : X86::MOVSDrm) :
|
|
|
|
(HasAVX ? X86::VMOVSDmr : X86::MOVSDmr);
|
2011-06-01 15:32:10 +00:00
|
|
|
if (X86::VR64RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::MMX_MOVQ64rm : X86::MMX_MOVQ64mr;
|
|
|
|
if (X86::RFP64RegClass.hasSubClassEq(RC))
|
|
|
|
return load ? X86::LD_Fp64m : X86::ST_Fp64m;
|
|
|
|
llvm_unreachable("Unknown 8-byte regclass");
|
|
|
|
case 10:
|
|
|
|
assert(X86::RFP80RegClass.hasSubClassEq(RC) && "Unknown 10-byte regclass");
|
2010-06-12 20:13:29 +00:00
|
|
|
return load ? X86::LD_Fp80m : X86::ST_FpP80m;
|
2011-08-31 03:04:09 +00:00
|
|
|
case 16: {
|
2011-06-01 15:32:10 +00:00
|
|
|
assert(X86::VR128RegClass.hasSubClassEq(RC) && "Unknown 16-byte regclass");
|
2008-07-19 06:30:51 +00:00
|
|
|
// If stack is realigned we can use aligned stores.
|
2010-06-12 20:13:29 +00:00
|
|
|
if (isStackAligned)
|
2011-08-31 03:04:09 +00:00
|
|
|
return load ?
|
|
|
|
(HasAVX ? X86::VMOVAPSrm : X86::MOVAPSrm) :
|
|
|
|
(HasAVX ? X86::VMOVAPSmr : X86::MOVAPSmr);
|
2010-06-12 20:13:29 +00:00
|
|
|
else
|
2011-08-31 03:04:09 +00:00
|
|
|
return load ?
|
|
|
|
(HasAVX ? X86::VMOVUPSrm : X86::MOVUPSrm) :
|
|
|
|
(HasAVX ? X86::VMOVUPSmr : X86::MOVUPSmr);
|
|
|
|
}
|
2011-07-14 18:50:58 +00:00
|
|
|
case 32:
|
|
|
|
assert(X86::VR256RegClass.hasSubClassEq(RC) && "Unknown 32-byte regclass");
|
|
|
|
// If stack is realigned we can use aligned stores.
|
|
|
|
if (isStackAligned)
|
|
|
|
return load ? X86::VMOVAPSYrm : X86::VMOVAPSYmr;
|
|
|
|
else
|
|
|
|
return load ? X86::VMOVUPSYrm : X86::VMOVUPSYmr;
|
2013-08-18 13:08:57 +00:00
|
|
|
case 64:
|
|
|
|
assert(X86::VR512RegClass.hasSubClassEq(RC) && "Unknown 64-byte regclass");
|
|
|
|
if (isStackAligned)
|
|
|
|
return load ? X86::VMOVAPSZrm : X86::VMOVAPSZmr;
|
|
|
|
else
|
|
|
|
return load ? X86::VMOVUPSZrm : X86::VMOVUPSZmr;
|
2008-01-01 21:11:32 +00:00
|
|
|
}
|
2010-06-12 20:13:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static unsigned getStoreRegOpcode(unsigned SrcReg,
|
|
|
|
const TargetRegisterClass *RC,
|
|
|
|
bool isStackAligned,
|
|
|
|
TargetMachine &TM) {
|
|
|
|
return getLoadStoreRegOpcode(SrcReg, RC, isStackAligned, TM, false);
|
|
|
|
}
|
|
|
|
|
2008-01-01 21:11:32 +00:00
|
|
|
|
2010-06-12 20:13:29 +00:00
|
|
|
static unsigned getLoadRegOpcode(unsigned DestReg,
|
|
|
|
const TargetRegisterClass *RC,
|
|
|
|
bool isStackAligned,
|
|
|
|
const TargetMachine &TM) {
|
|
|
|
return getLoadStoreRegOpcode(DestReg, RC, isStackAligned, TM, true);
|
2008-01-01 21:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void X86InstrInfo::storeRegToStackSlot(MachineBasicBlock &MBB,
|
|
|
|
MachineBasicBlock::iterator MI,
|
|
|
|
unsigned SrcReg, bool isKill, int FrameIdx,
|
2010-05-06 19:06:44 +00:00
|
|
|
const TargetRegisterClass *RC,
|
|
|
|
const TargetRegisterInfo *TRI) const {
|
2008-07-19 06:30:51 +00:00
|
|
|
const MachineFunction &MF = *MBB.getParent();
|
2010-07-27 04:16:58 +00:00
|
|
|
assert(MF.getFrameInfo()->getObjectSize(FrameIdx) >= RC->getSize() &&
|
|
|
|
"Stack slot too small for store");
|
2013-08-18 13:08:57 +00:00
|
|
|
unsigned Alignment = std::max<uint32_t>(RC->getSize(), 16);
|
2011-09-14 02:36:58 +00:00
|
|
|
bool isAligned = (TM.getFrameLowering()->getStackAlignment() >= Alignment) ||
|
2011-06-23 01:53:43 +00:00
|
|
|
RI.canRealignStack(MF);
|
2009-04-27 16:41:36 +00:00
|
|
|
unsigned Opc = getStoreRegOpcode(SrcReg, RC, isAligned, TM);
|
2010-01-26 00:03:12 +00:00
|
|
|
DebugLoc DL = MBB.findDebugLoc(MI);
|
2009-02-11 21:51:19 +00:00
|
|
|
addFrameReference(BuildMI(MBB, MI, DL, get(Opc)), FrameIdx)
|
2009-05-13 21:33:08 +00:00
|
|
|
.addReg(SrcReg, getKillRegState(isKill));
|
2008-01-01 21:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void X86InstrInfo::storeRegToAddr(MachineFunction &MF, unsigned SrcReg,
|
|
|
|
bool isKill,
|
|
|
|
SmallVectorImpl<MachineOperand> &Addr,
|
|
|
|
const TargetRegisterClass *RC,
|
2009-10-09 18:10:05 +00:00
|
|
|
MachineInstr::mmo_iterator MMOBegin,
|
|
|
|
MachineInstr::mmo_iterator MMOEnd,
|
2008-01-01 21:11:32 +00:00
|
|
|
SmallVectorImpl<MachineInstr*> &NewMIs) const {
|
2013-08-18 13:08:57 +00:00
|
|
|
unsigned Alignment = std::max<uint32_t>(RC->getSize(), 16);
|
2011-09-14 02:36:58 +00:00
|
|
|
bool isAligned = MMOBegin != MMOEnd &&
|
|
|
|
(*MMOBegin)->getAlignment() >= Alignment;
|
2009-04-27 16:41:36 +00:00
|
|
|
unsigned Opc = getStoreRegOpcode(SrcReg, RC, isAligned, TM);
|
2010-04-02 20:16:16 +00:00
|
|
|
DebugLoc DL;
|
2009-02-12 23:08:38 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, DL, get(Opc));
|
2008-01-01 21:11:32 +00:00
|
|
|
for (unsigned i = 0, e = Addr.size(); i != e; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(Addr[i]);
|
2009-05-13 21:33:08 +00:00
|
|
|
MIB.addReg(SrcReg, getKillRegState(isKill));
|
2009-10-09 18:10:05 +00:00
|
|
|
(*MIB).setMemRefs(MMOBegin, MMOEnd);
|
2008-01-01 21:11:32 +00:00
|
|
|
NewMIs.push_back(MIB);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void X86InstrInfo::loadRegFromStackSlot(MachineBasicBlock &MBB,
|
2008-07-19 06:30:51 +00:00
|
|
|
MachineBasicBlock::iterator MI,
|
|
|
|
unsigned DestReg, int FrameIdx,
|
2010-05-06 19:06:44 +00:00
|
|
|
const TargetRegisterClass *RC,
|
|
|
|
const TargetRegisterInfo *TRI) const {
|
2008-07-19 06:30:51 +00:00
|
|
|
const MachineFunction &MF = *MBB.getParent();
|
2013-08-18 13:08:57 +00:00
|
|
|
unsigned Alignment = std::max<uint32_t>(RC->getSize(), 16);
|
2011-09-14 02:36:58 +00:00
|
|
|
bool isAligned = (TM.getFrameLowering()->getStackAlignment() >= Alignment) ||
|
2011-06-23 01:53:43 +00:00
|
|
|
RI.canRealignStack(MF);
|
2009-04-27 16:41:36 +00:00
|
|
|
unsigned Opc = getLoadRegOpcode(DestReg, RC, isAligned, TM);
|
2010-01-26 00:03:12 +00:00
|
|
|
DebugLoc DL = MBB.findDebugLoc(MI);
|
2009-02-11 21:51:19 +00:00
|
|
|
addFrameReference(BuildMI(MBB, MI, DL, get(Opc), DestReg), FrameIdx);
|
2008-01-01 21:11:32 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void X86InstrInfo::loadRegFromAddr(MachineFunction &MF, unsigned DestReg,
|
2008-07-03 09:09:37 +00:00
|
|
|
SmallVectorImpl<MachineOperand> &Addr,
|
|
|
|
const TargetRegisterClass *RC,
|
2009-10-09 18:10:05 +00:00
|
|
|
MachineInstr::mmo_iterator MMOBegin,
|
|
|
|
MachineInstr::mmo_iterator MMOEnd,
|
2008-01-01 21:11:32 +00:00
|
|
|
SmallVectorImpl<MachineInstr*> &NewMIs) const {
|
2013-08-18 13:08:57 +00:00
|
|
|
unsigned Alignment = std::max<uint32_t>(RC->getSize(), 16);
|
2011-09-14 02:36:58 +00:00
|
|
|
bool isAligned = MMOBegin != MMOEnd &&
|
|
|
|
(*MMOBegin)->getAlignment() >= Alignment;
|
2009-04-27 16:41:36 +00:00
|
|
|
unsigned Opc = getLoadRegOpcode(DestReg, RC, isAligned, TM);
|
2010-04-02 20:16:16 +00:00
|
|
|
DebugLoc DL;
|
2009-02-12 23:08:38 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, DL, get(Opc), DestReg);
|
2008-01-01 21:11:32 +00:00
|
|
|
for (unsigned i = 0, e = Addr.size(); i != e; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(Addr[i]);
|
2009-10-09 18:10:05 +00:00
|
|
|
(*MIB).setMemRefs(MMOBegin, MMOEnd);
|
2008-01-01 21:11:32 +00:00
|
|
|
NewMIs.push_back(MIB);
|
|
|
|
}
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
bool X86InstrInfo::
|
|
|
|
analyzeCompare(const MachineInstr *MI, unsigned &SrcReg, unsigned &SrcReg2,
|
|
|
|
int &CmpMask, int &CmpValue) const {
|
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: break;
|
|
|
|
case X86::CMP64ri32:
|
|
|
|
case X86::CMP64ri8:
|
|
|
|
case X86::CMP32ri:
|
|
|
|
case X86::CMP32ri8:
|
|
|
|
case X86::CMP16ri:
|
|
|
|
case X86::CMP16ri8:
|
|
|
|
case X86::CMP8ri:
|
|
|
|
SrcReg = MI->getOperand(0).getReg();
|
|
|
|
SrcReg2 = 0;
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = MI->getOperand(1).getImm();
|
|
|
|
return true;
|
2012-08-08 00:51:41 +00:00
|
|
|
// A SUB can be used to perform comparison.
|
|
|
|
case X86::SUB64rm:
|
|
|
|
case X86::SUB32rm:
|
|
|
|
case X86::SUB16rm:
|
|
|
|
case X86::SUB8rm:
|
|
|
|
SrcReg = MI->getOperand(1).getReg();
|
|
|
|
SrcReg2 = 0;
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = 0;
|
|
|
|
return true;
|
|
|
|
case X86::SUB64rr:
|
|
|
|
case X86::SUB32rr:
|
|
|
|
case X86::SUB16rr:
|
|
|
|
case X86::SUB8rr:
|
|
|
|
SrcReg = MI->getOperand(1).getReg();
|
|
|
|
SrcReg2 = MI->getOperand(2).getReg();
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = 0;
|
|
|
|
return true;
|
|
|
|
case X86::SUB64ri32:
|
|
|
|
case X86::SUB64ri8:
|
|
|
|
case X86::SUB32ri:
|
|
|
|
case X86::SUB32ri8:
|
|
|
|
case X86::SUB16ri:
|
|
|
|
case X86::SUB16ri8:
|
|
|
|
case X86::SUB8ri:
|
|
|
|
SrcReg = MI->getOperand(1).getReg();
|
|
|
|
SrcReg2 = 0;
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = MI->getOperand(2).getImm();
|
|
|
|
return true;
|
2012-07-06 17:36:20 +00:00
|
|
|
case X86::CMP64rr:
|
|
|
|
case X86::CMP32rr:
|
|
|
|
case X86::CMP16rr:
|
|
|
|
case X86::CMP8rr:
|
|
|
|
SrcReg = MI->getOperand(0).getReg();
|
|
|
|
SrcReg2 = MI->getOperand(1).getReg();
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = 0;
|
|
|
|
return true;
|
2012-07-18 21:40:01 +00:00
|
|
|
case X86::TEST8rr:
|
|
|
|
case X86::TEST16rr:
|
|
|
|
case X86::TEST32rr:
|
|
|
|
case X86::TEST64rr:
|
|
|
|
SrcReg = MI->getOperand(0).getReg();
|
|
|
|
if (MI->getOperand(1).getReg() != SrcReg) return false;
|
|
|
|
// Compare against zero.
|
|
|
|
SrcReg2 = 0;
|
|
|
|
CmpMask = ~0;
|
|
|
|
CmpValue = 0;
|
|
|
|
return true;
|
2012-07-06 17:36:20 +00:00
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// isRedundantFlagInstr - check whether the first instruction, whose only
|
|
|
|
/// purpose is to update flags, can be made redundant.
|
|
|
|
/// CMPrr can be made redundant by SUBrr if the operands are the same.
|
|
|
|
/// This function can be extended later on.
|
|
|
|
/// SrcReg, SrcRegs: register operands for FlagI.
|
|
|
|
/// ImmValue: immediate for FlagI if it takes an immediate.
|
|
|
|
inline static bool isRedundantFlagInstr(MachineInstr *FlagI, unsigned SrcReg,
|
|
|
|
unsigned SrcReg2, int ImmValue,
|
|
|
|
MachineInstr *OI) {
|
|
|
|
if (((FlagI->getOpcode() == X86::CMP64rr &&
|
|
|
|
OI->getOpcode() == X86::SUB64rr) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP32rr &&
|
|
|
|
OI->getOpcode() == X86::SUB32rr)||
|
|
|
|
(FlagI->getOpcode() == X86::CMP16rr &&
|
|
|
|
OI->getOpcode() == X86::SUB16rr)||
|
|
|
|
(FlagI->getOpcode() == X86::CMP8rr &&
|
|
|
|
OI->getOpcode() == X86::SUB8rr)) &&
|
|
|
|
((OI->getOperand(1).getReg() == SrcReg &&
|
|
|
|
OI->getOperand(2).getReg() == SrcReg2) ||
|
|
|
|
(OI->getOperand(1).getReg() == SrcReg2 &&
|
|
|
|
OI->getOperand(2).getReg() == SrcReg)))
|
|
|
|
return true;
|
|
|
|
|
|
|
|
if (((FlagI->getOpcode() == X86::CMP64ri32 &&
|
|
|
|
OI->getOpcode() == X86::SUB64ri32) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP64ri8 &&
|
|
|
|
OI->getOpcode() == X86::SUB64ri8) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP32ri &&
|
|
|
|
OI->getOpcode() == X86::SUB32ri) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP32ri8 &&
|
|
|
|
OI->getOpcode() == X86::SUB32ri8) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP16ri &&
|
|
|
|
OI->getOpcode() == X86::SUB16ri) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP16ri8 &&
|
|
|
|
OI->getOpcode() == X86::SUB16ri8) ||
|
|
|
|
(FlagI->getOpcode() == X86::CMP8ri &&
|
|
|
|
OI->getOpcode() == X86::SUB8ri)) &&
|
|
|
|
OI->getOperand(1).getReg() == SrcReg &&
|
|
|
|
OI->getOperand(2).getImm() == ImmValue)
|
|
|
|
return true;
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2012-07-18 21:40:01 +00:00
|
|
|
/// isDefConvertible - check whether the definition can be converted
|
|
|
|
/// to remove a comparison against zero.
|
|
|
|
inline static bool isDefConvertible(MachineInstr *MI) {
|
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: return false;
|
2013-05-22 08:13:02 +00:00
|
|
|
|
|
|
|
// The shift instructions only modify ZF if their shift count is non-zero.
|
|
|
|
// N.B.: The processor truncates the shift count depending on the encoding.
|
|
|
|
case X86::SAR8ri: case X86::SAR16ri: case X86::SAR32ri:case X86::SAR64ri:
|
|
|
|
case X86::SHR8ri: case X86::SHR16ri: case X86::SHR32ri:case X86::SHR64ri:
|
|
|
|
return getTruncatedShiftCount(MI, 2) != 0;
|
|
|
|
|
|
|
|
// Some left shift instructions can be turned into LEA instructions but only
|
|
|
|
// if their flags aren't used. Avoid transforming such instructions.
|
|
|
|
case X86::SHL8ri: case X86::SHL16ri: case X86::SHL32ri:case X86::SHL64ri:{
|
|
|
|
unsigned ShAmt = getTruncatedShiftCount(MI, 2);
|
|
|
|
if (isTruncatedShiftCountForLEA(ShAmt)) return false;
|
|
|
|
return ShAmt != 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
case X86::SHRD16rri8:case X86::SHRD32rri8:case X86::SHRD64rri8:
|
|
|
|
case X86::SHLD16rri8:case X86::SHLD32rri8:case X86::SHLD64rri8:
|
|
|
|
return getTruncatedShiftCount(MI, 3) != 0;
|
|
|
|
|
2012-07-18 21:40:01 +00:00
|
|
|
case X86::SUB64ri32: case X86::SUB64ri8: case X86::SUB32ri:
|
|
|
|
case X86::SUB32ri8: case X86::SUB16ri: case X86::SUB16ri8:
|
|
|
|
case X86::SUB8ri: case X86::SUB64rr: case X86::SUB32rr:
|
|
|
|
case X86::SUB16rr: case X86::SUB8rr: case X86::SUB64rm:
|
|
|
|
case X86::SUB32rm: case X86::SUB16rm: case X86::SUB8rm:
|
2012-12-17 04:55:07 +00:00
|
|
|
case X86::DEC64r: case X86::DEC32r: case X86::DEC16r: case X86::DEC8r:
|
2012-09-17 22:04:23 +00:00
|
|
|
case X86::DEC64_32r: case X86::DEC64_16r:
|
2012-07-18 21:40:01 +00:00
|
|
|
case X86::ADD64ri32: case X86::ADD64ri8: case X86::ADD32ri:
|
|
|
|
case X86::ADD32ri8: case X86::ADD16ri: case X86::ADD16ri8:
|
|
|
|
case X86::ADD8ri: case X86::ADD64rr: case X86::ADD32rr:
|
|
|
|
case X86::ADD16rr: case X86::ADD8rr: case X86::ADD64rm:
|
|
|
|
case X86::ADD32rm: case X86::ADD16rm: case X86::ADD8rm:
|
2012-12-17 04:55:07 +00:00
|
|
|
case X86::INC64r: case X86::INC32r: case X86::INC16r: case X86::INC8r:
|
2012-09-17 22:04:23 +00:00
|
|
|
case X86::INC64_32r: case X86::INC64_16r:
|
2012-07-18 21:40:01 +00:00
|
|
|
case X86::AND64ri32: case X86::AND64ri8: case X86::AND32ri:
|
|
|
|
case X86::AND32ri8: case X86::AND16ri: case X86::AND16ri8:
|
|
|
|
case X86::AND8ri: case X86::AND64rr: case X86::AND32rr:
|
|
|
|
case X86::AND16rr: case X86::AND8rr: case X86::AND64rm:
|
|
|
|
case X86::AND32rm: case X86::AND16rm: case X86::AND8rm:
|
|
|
|
case X86::XOR64ri32: case X86::XOR64ri8: case X86::XOR32ri:
|
|
|
|
case X86::XOR32ri8: case X86::XOR16ri: case X86::XOR16ri8:
|
|
|
|
case X86::XOR8ri: case X86::XOR64rr: case X86::XOR32rr:
|
|
|
|
case X86::XOR16rr: case X86::XOR8rr: case X86::XOR64rm:
|
|
|
|
case X86::XOR32rm: case X86::XOR16rm: case X86::XOR8rm:
|
|
|
|
case X86::OR64ri32: case X86::OR64ri8: case X86::OR32ri:
|
|
|
|
case X86::OR32ri8: case X86::OR16ri: case X86::OR16ri8:
|
|
|
|
case X86::OR8ri: case X86::OR64rr: case X86::OR32rr:
|
|
|
|
case X86::OR16rr: case X86::OR8rr: case X86::OR64rm:
|
|
|
|
case X86::OR32rm: case X86::OR16rm: case X86::OR8rm:
|
2013-05-15 22:03:08 +00:00
|
|
|
case X86::NEG8r: case X86::NEG16r: case X86::NEG32r: case X86::NEG64r:
|
|
|
|
case X86::SAR8r1: case X86::SAR16r1: case X86::SAR32r1:case X86::SAR64r1:
|
|
|
|
case X86::SHR8r1: case X86::SHR16r1: case X86::SHR32r1:case X86::SHR64r1:
|
|
|
|
case X86::SHL8r1: case X86::SHL16r1: case X86::SHL32r1:case X86::SHL64r1:
|
|
|
|
case X86::ADC32ri: case X86::ADC32ri8:
|
|
|
|
case X86::ADC32rr: case X86::ADC64ri32:
|
|
|
|
case X86::ADC64ri8: case X86::ADC64rr:
|
|
|
|
case X86::SBB32ri: case X86::SBB32ri8:
|
|
|
|
case X86::SBB32rr: case X86::SBB64ri32:
|
|
|
|
case X86::SBB64ri8: case X86::SBB64rr:
|
2012-12-17 05:12:30 +00:00
|
|
|
case X86::ANDN32rr: case X86::ANDN32rm:
|
|
|
|
case X86::ANDN64rr: case X86::ANDN64rm:
|
2013-05-15 22:03:08 +00:00
|
|
|
case X86::BEXTR32rr: case X86::BEXTR64rr:
|
|
|
|
case X86::BEXTR32rm: case X86::BEXTR64rm:
|
|
|
|
case X86::BLSI32rr: case X86::BLSI32rm:
|
|
|
|
case X86::BLSI64rr: case X86::BLSI64rm:
|
|
|
|
case X86::BLSMSK32rr:case X86::BLSMSK32rm:
|
|
|
|
case X86::BLSMSK64rr:case X86::BLSMSK64rm:
|
|
|
|
case X86::BLSR32rr: case X86::BLSR32rm:
|
|
|
|
case X86::BLSR64rr: case X86::BLSR64rm:
|
|
|
|
case X86::BZHI32rr: case X86::BZHI32rm:
|
|
|
|
case X86::BZHI64rr: case X86::BZHI64rm:
|
|
|
|
case X86::LZCNT16rr: case X86::LZCNT16rm:
|
|
|
|
case X86::LZCNT32rr: case X86::LZCNT32rm:
|
|
|
|
case X86::LZCNT64rr: case X86::LZCNT64rm:
|
|
|
|
case X86::POPCNT16rr:case X86::POPCNT16rm:
|
|
|
|
case X86::POPCNT32rr:case X86::POPCNT32rm:
|
|
|
|
case X86::POPCNT64rr:case X86::POPCNT64rm:
|
|
|
|
case X86::TZCNT16rr: case X86::TZCNT16rm:
|
|
|
|
case X86::TZCNT32rr: case X86::TZCNT32rm:
|
|
|
|
case X86::TZCNT64rr: case X86::TZCNT64rm:
|
2012-07-18 21:40:01 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
/// optimizeCompareInstr - Check if there exists an earlier instruction that
|
|
|
|
/// operates on the same source operands and sets flags in the same way as
|
|
|
|
/// Compare; remove Compare if possible.
|
|
|
|
bool X86InstrInfo::
|
|
|
|
optimizeCompareInstr(MachineInstr *CmpInstr, unsigned SrcReg, unsigned SrcReg2,
|
|
|
|
int CmpMask, int CmpValue,
|
|
|
|
const MachineRegisterInfo *MRI) const {
|
2012-08-08 00:51:41 +00:00
|
|
|
// Check whether we can replace SUB with CMP.
|
|
|
|
unsigned NewOpcode = 0;
|
|
|
|
switch (CmpInstr->getOpcode()) {
|
|
|
|
default: break;
|
|
|
|
case X86::SUB64ri32:
|
|
|
|
case X86::SUB64ri8:
|
|
|
|
case X86::SUB32ri:
|
|
|
|
case X86::SUB32ri8:
|
|
|
|
case X86::SUB16ri:
|
|
|
|
case X86::SUB16ri8:
|
|
|
|
case X86::SUB8ri:
|
|
|
|
case X86::SUB64rm:
|
|
|
|
case X86::SUB32rm:
|
|
|
|
case X86::SUB16rm:
|
|
|
|
case X86::SUB8rm:
|
|
|
|
case X86::SUB64rr:
|
|
|
|
case X86::SUB32rr:
|
|
|
|
case X86::SUB16rr:
|
|
|
|
case X86::SUB8rr: {
|
|
|
|
if (!MRI->use_nodbg_empty(CmpInstr->getOperand(0).getReg()))
|
|
|
|
return false;
|
|
|
|
// There is no use of the destination register, we can replace SUB with CMP.
|
|
|
|
switch (CmpInstr->getOpcode()) {
|
2012-08-21 08:16:16 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2012-08-08 00:51:41 +00:00
|
|
|
case X86::SUB64rm: NewOpcode = X86::CMP64rm; break;
|
|
|
|
case X86::SUB32rm: NewOpcode = X86::CMP32rm; break;
|
|
|
|
case X86::SUB16rm: NewOpcode = X86::CMP16rm; break;
|
|
|
|
case X86::SUB8rm: NewOpcode = X86::CMP8rm; break;
|
|
|
|
case X86::SUB64rr: NewOpcode = X86::CMP64rr; break;
|
|
|
|
case X86::SUB32rr: NewOpcode = X86::CMP32rr; break;
|
|
|
|
case X86::SUB16rr: NewOpcode = X86::CMP16rr; break;
|
|
|
|
case X86::SUB8rr: NewOpcode = X86::CMP8rr; break;
|
|
|
|
case X86::SUB64ri32: NewOpcode = X86::CMP64ri32; break;
|
|
|
|
case X86::SUB64ri8: NewOpcode = X86::CMP64ri8; break;
|
|
|
|
case X86::SUB32ri: NewOpcode = X86::CMP32ri; break;
|
|
|
|
case X86::SUB32ri8: NewOpcode = X86::CMP32ri8; break;
|
|
|
|
case X86::SUB16ri: NewOpcode = X86::CMP16ri; break;
|
|
|
|
case X86::SUB16ri8: NewOpcode = X86::CMP16ri8; break;
|
|
|
|
case X86::SUB8ri: NewOpcode = X86::CMP8ri; break;
|
|
|
|
}
|
|
|
|
CmpInstr->setDesc(get(NewOpcode));
|
|
|
|
CmpInstr->RemoveOperand(0);
|
|
|
|
// Fall through to optimize Cmp if Cmp is CMPrr or CMPri.
|
|
|
|
if (NewOpcode == X86::CMP64rm || NewOpcode == X86::CMP32rm ||
|
|
|
|
NewOpcode == X86::CMP16rm || NewOpcode == X86::CMP8rm)
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
// Get the unique definition of SrcReg.
|
|
|
|
MachineInstr *MI = MRI->getUniqueVRegDef(SrcReg);
|
|
|
|
if (!MI) return false;
|
|
|
|
|
|
|
|
// CmpInstr is the first instruction of the BB.
|
|
|
|
MachineBasicBlock::iterator I = CmpInstr, Def = MI;
|
|
|
|
|
2012-07-18 21:40:01 +00:00
|
|
|
// If we are comparing against zero, check whether we can use MI to update
|
|
|
|
// EFLAGS. If MI is not in the same BB as CmpInstr, do not optimize.
|
|
|
|
bool IsCmpZero = (SrcReg2 == 0 && CmpValue == 0);
|
|
|
|
if (IsCmpZero && (MI->getParent() != CmpInstr->getParent() ||
|
|
|
|
!isDefConvertible(MI)))
|
|
|
|
return false;
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
// We are searching for an earlier instruction that can make CmpInstr
|
|
|
|
// redundant and that instruction will be saved in Sub.
|
|
|
|
MachineInstr *Sub = NULL;
|
|
|
|
const TargetRegisterInfo *TRI = &getRegisterInfo();
|
2012-07-09 18:57:12 +00:00
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
// We iterate backward, starting from the instruction before CmpInstr and
|
|
|
|
// stop when reaching the definition of a source register or done with the BB.
|
|
|
|
// RI points to the instruction before CmpInstr.
|
|
|
|
// If the definition is in this basic block, RE points to the definition;
|
|
|
|
// otherwise, RE is the rend of the basic block.
|
|
|
|
MachineBasicBlock::reverse_iterator
|
|
|
|
RI = MachineBasicBlock::reverse_iterator(I),
|
|
|
|
RE = CmpInstr->getParent() == MI->getParent() ?
|
|
|
|
MachineBasicBlock::reverse_iterator(++Def) /* points to MI */ :
|
|
|
|
CmpInstr->getParent()->rend();
|
2012-07-11 19:35:12 +00:00
|
|
|
MachineInstr *Movr0Inst = 0;
|
2012-07-06 17:36:20 +00:00
|
|
|
for (; RI != RE; ++RI) {
|
|
|
|
MachineInstr *Instr = &*RI;
|
|
|
|
// Check whether CmpInstr can be made redundant by the current instruction.
|
2012-07-18 21:40:01 +00:00
|
|
|
if (!IsCmpZero &&
|
|
|
|
isRedundantFlagInstr(CmpInstr, SrcReg, SrcReg2, CmpValue, Instr)) {
|
2012-07-06 17:36:20 +00:00
|
|
|
Sub = Instr;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (Instr->modifiesRegister(X86::EFLAGS, TRI) ||
|
2012-07-11 19:35:12 +00:00
|
|
|
Instr->readsRegister(X86::EFLAGS, TRI)) {
|
2012-07-06 17:36:20 +00:00
|
|
|
// This instruction modifies or uses EFLAGS.
|
2012-07-11 19:35:12 +00:00
|
|
|
|
|
|
|
// MOV32r0 etc. are implemented with xor which clobbers condition code.
|
|
|
|
// They are safe to move up, if the definition to EFLAGS is dead and
|
|
|
|
// earlier instructions do not read or write EFLAGS.
|
2013-05-30 13:19:42 +00:00
|
|
|
if (!Movr0Inst && Instr->getOpcode() == X86::MOV32r0 &&
|
2012-07-11 19:35:12 +00:00
|
|
|
Instr->registerDefIsDead(X86::EFLAGS, TRI)) {
|
|
|
|
Movr0Inst = Instr;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
// We can't remove CmpInstr.
|
|
|
|
return false;
|
2012-07-11 19:35:12 +00:00
|
|
|
}
|
2012-07-06 17:36:20 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Return false if no candidates exist.
|
2012-07-18 21:40:01 +00:00
|
|
|
if (!IsCmpZero && !Sub)
|
2012-07-06 17:36:20 +00:00
|
|
|
return false;
|
|
|
|
|
2012-07-07 03:34:46 +00:00
|
|
|
bool IsSwapped = (SrcReg2 != 0 && Sub->getOperand(1).getReg() == SrcReg2 &&
|
|
|
|
Sub->getOperand(2).getReg() == SrcReg);
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
// Scan forward from the instruction after CmpInstr for uses of EFLAGS.
|
2012-07-07 03:34:46 +00:00
|
|
|
// It is safe to remove CmpInstr if EFLAGS is redefined or killed.
|
|
|
|
// If we are done with the basic block, we need to check whether EFLAGS is
|
|
|
|
// live-out.
|
|
|
|
bool IsSafe = false;
|
2012-07-06 17:36:20 +00:00
|
|
|
SmallVector<std::pair<MachineInstr*, unsigned /*NewOpc*/>, 4> OpsToUpdate;
|
|
|
|
MachineBasicBlock::iterator E = CmpInstr->getParent()->end();
|
|
|
|
for (++I; I != E; ++I) {
|
|
|
|
const MachineInstr &Instr = *I;
|
2012-07-28 03:15:46 +00:00
|
|
|
bool ModifyEFLAGS = Instr.modifiesRegister(X86::EFLAGS, TRI);
|
|
|
|
bool UseEFLAGS = Instr.readsRegister(X86::EFLAGS, TRI);
|
|
|
|
// We should check the usage if this instruction uses and updates EFLAGS.
|
|
|
|
if (!UseEFLAGS && ModifyEFLAGS) {
|
2012-07-06 17:36:20 +00:00
|
|
|
// It is safe to remove CmpInstr if EFLAGS is updated again.
|
2012-07-07 03:34:46 +00:00
|
|
|
IsSafe = true;
|
2012-07-06 17:36:20 +00:00
|
|
|
break;
|
2012-07-07 03:34:46 +00:00
|
|
|
}
|
2012-07-28 03:15:46 +00:00
|
|
|
if (!UseEFLAGS && !ModifyEFLAGS)
|
2012-07-06 17:36:20 +00:00
|
|
|
continue;
|
|
|
|
|
|
|
|
// EFLAGS is used by this instruction.
|
2012-07-18 21:40:01 +00:00
|
|
|
X86::CondCode OldCC;
|
|
|
|
bool OpcIsSET = false;
|
|
|
|
if (IsCmpZero || IsSwapped) {
|
|
|
|
// We decode the condition code from opcode.
|
2012-07-09 18:57:12 +00:00
|
|
|
if (Instr.isBranch())
|
|
|
|
OldCC = getCondFromBranchOpc(Instr.getOpcode());
|
|
|
|
else {
|
|
|
|
OldCC = getCondFromSETOpc(Instr.getOpcode());
|
|
|
|
if (OldCC != X86::COND_INVALID)
|
|
|
|
OpcIsSET = true;
|
|
|
|
else
|
2012-09-20 03:06:15 +00:00
|
|
|
OldCC = X86::getCondFromCMovOpc(Instr.getOpcode());
|
2012-07-09 18:57:12 +00:00
|
|
|
}
|
|
|
|
if (OldCC == X86::COND_INVALID) return false;
|
2012-07-18 21:40:01 +00:00
|
|
|
}
|
|
|
|
if (IsCmpZero) {
|
|
|
|
switch (OldCC) {
|
|
|
|
default: break;
|
|
|
|
case X86::COND_A: case X86::COND_AE:
|
|
|
|
case X86::COND_B: case X86::COND_BE:
|
|
|
|
case X86::COND_G: case X86::COND_GE:
|
|
|
|
case X86::COND_L: case X86::COND_LE:
|
|
|
|
case X86::COND_O: case X86::COND_NO:
|
|
|
|
// CF and OF are used, we can't perform this optimization.
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
} else if (IsSwapped) {
|
|
|
|
// If we have SUB(r1, r2) and CMP(r2, r1), the condition code needs
|
|
|
|
// to be changed from r2 > r1 to r1 < r2, from r2 < r1 to r1 > r2, etc.
|
|
|
|
// We swap the condition code and synthesize the new opcode.
|
2012-07-09 18:57:12 +00:00
|
|
|
X86::CondCode NewCC = getSwappedCondition(OldCC);
|
|
|
|
if (NewCC == X86::COND_INVALID) return false;
|
|
|
|
|
|
|
|
// Synthesize the new opcode.
|
|
|
|
bool HasMemoryOperand = Instr.hasOneMemOperand();
|
|
|
|
unsigned NewOpc;
|
|
|
|
if (Instr.isBranch())
|
|
|
|
NewOpc = GetCondBranchFromCond(NewCC);
|
|
|
|
else if(OpcIsSET)
|
|
|
|
NewOpc = getSETFromCond(NewCC, HasMemoryOperand);
|
|
|
|
else {
|
|
|
|
unsigned DstReg = Instr.getOperand(0).getReg();
|
|
|
|
NewOpc = getCMovFromCond(NewCC, MRI->getRegClass(DstReg)->getSize(),
|
|
|
|
HasMemoryOperand);
|
|
|
|
}
|
2012-07-06 17:36:20 +00:00
|
|
|
|
|
|
|
// Push the MachineInstr to OpsToUpdate.
|
|
|
|
// If it is safe to remove CmpInstr, the condition code of these
|
|
|
|
// instructions will be modified.
|
|
|
|
OpsToUpdate.push_back(std::make_pair(&*I, NewOpc));
|
|
|
|
}
|
2012-07-28 03:15:46 +00:00
|
|
|
if (ModifyEFLAGS || Instr.killsRegister(X86::EFLAGS, TRI)) {
|
|
|
|
// It is safe to remove CmpInstr if EFLAGS is updated again or killed.
|
2012-07-07 03:34:46 +00:00
|
|
|
IsSafe = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// If EFLAGS is not killed nor re-defined, we should check whether it is
|
|
|
|
// live-out. If it is live-out, do not optimize.
|
2012-07-18 21:40:01 +00:00
|
|
|
if ((IsCmpZero || IsSwapped) && !IsSafe) {
|
2012-07-07 03:34:46 +00:00
|
|
|
MachineBasicBlock *MBB = CmpInstr->getParent();
|
|
|
|
for (MachineBasicBlock::succ_iterator SI = MBB->succ_begin(),
|
|
|
|
SE = MBB->succ_end(); SI != SE; ++SI)
|
|
|
|
if ((*SI)->isLiveIn(X86::EFLAGS))
|
|
|
|
return false;
|
2012-07-06 17:36:20 +00:00
|
|
|
}
|
|
|
|
|
2012-07-18 21:40:01 +00:00
|
|
|
// The instruction to be updated is either Sub or MI.
|
|
|
|
Sub = IsCmpZero ? MI : Sub;
|
2013-05-18 01:02:03 +00:00
|
|
|
// Move Movr0Inst to the appropriate place before Sub.
|
2012-07-11 19:35:12 +00:00
|
|
|
if (Movr0Inst) {
|
2013-05-18 01:02:03 +00:00
|
|
|
// Look backwards until we find a def that doesn't use the current EFLAGS.
|
|
|
|
Def = Sub;
|
|
|
|
MachineBasicBlock::reverse_iterator
|
|
|
|
InsertI = MachineBasicBlock::reverse_iterator(++Def),
|
|
|
|
InsertE = Sub->getParent()->rend();
|
|
|
|
for (; InsertI != InsertE; ++InsertI) {
|
|
|
|
MachineInstr *Instr = &*InsertI;
|
|
|
|
if (!Instr->readsRegister(X86::EFLAGS, TRI) &&
|
|
|
|
Instr->modifiesRegister(X86::EFLAGS, TRI)) {
|
|
|
|
Sub->getParent()->remove(Movr0Inst);
|
|
|
|
Instr->getParent()->insert(MachineBasicBlock::iterator(Instr),
|
|
|
|
Movr0Inst);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (InsertI == InsertE)
|
|
|
|
return false;
|
2012-07-11 19:35:12 +00:00
|
|
|
}
|
|
|
|
|
2012-09-17 22:04:23 +00:00
|
|
|
// Make sure Sub instruction defines EFLAGS and mark the def live.
|
2013-05-15 22:03:08 +00:00
|
|
|
unsigned i = 0, e = Sub->getNumOperands();
|
|
|
|
for (; i != e; ++i) {
|
|
|
|
MachineOperand &MO = Sub->getOperand(i);
|
|
|
|
if (MO.isReg() && MO.isDef() && MO.getReg() == X86::EFLAGS) {
|
|
|
|
MO.setIsDead(false);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
assert(i != e && "Unable to locate a def EFLAGS operand");
|
|
|
|
|
2012-07-06 17:36:20 +00:00
|
|
|
CmpInstr->eraseFromParent();
|
|
|
|
|
|
|
|
// Modify the condition code of instructions in OpsToUpdate.
|
|
|
|
for (unsigned i = 0, e = OpsToUpdate.size(); i < e; i++)
|
|
|
|
OpsToUpdate[i].first->setDesc(get(OpsToUpdate[i].second));
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2012-08-02 00:56:42 +00:00
|
|
|
/// optimizeLoadInstr - Try to remove the load by folding it to a register
|
|
|
|
/// operand at the use. We fold the load instructions if load defines a virtual
|
|
|
|
/// register, the virtual register is used once in the same BB, and the
|
|
|
|
/// instructions in-between do not load or store, and have no side effects.
|
|
|
|
MachineInstr* X86InstrInfo::
|
|
|
|
optimizeLoadInstr(MachineInstr *MI, const MachineRegisterInfo *MRI,
|
|
|
|
unsigned &FoldAsLoadDefReg,
|
|
|
|
MachineInstr *&DefMI) const {
|
|
|
|
if (FoldAsLoadDefReg == 0)
|
|
|
|
return 0;
|
|
|
|
// To be conservative, if there exists another load, clear the load candidate.
|
|
|
|
if (MI->mayLoad()) {
|
|
|
|
FoldAsLoadDefReg = 0;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Check whether we can move DefMI here.
|
|
|
|
DefMI = MRI->getVRegDef(FoldAsLoadDefReg);
|
|
|
|
assert(DefMI);
|
|
|
|
bool SawStore = false;
|
|
|
|
if (!DefMI->isSafeToMove(this, 0, SawStore))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
// We try to commute MI if possible.
|
|
|
|
unsigned IdxEnd = (MI->isCommutable()) ? 2 : 1;
|
|
|
|
for (unsigned Idx = 0; Idx < IdxEnd; Idx++) {
|
|
|
|
// Collect information about virtual register operands of MI.
|
|
|
|
unsigned SrcOperandId = 0;
|
|
|
|
bool FoundSrcOperand = false;
|
|
|
|
for (unsigned i = 0, e = MI->getDesc().getNumOperands(); i != e; ++i) {
|
|
|
|
MachineOperand &MO = MI->getOperand(i);
|
|
|
|
if (!MO.isReg())
|
|
|
|
continue;
|
|
|
|
unsigned Reg = MO.getReg();
|
|
|
|
if (Reg != FoldAsLoadDefReg)
|
|
|
|
continue;
|
|
|
|
// Do not fold if we have a subreg use or a def or multiple uses.
|
|
|
|
if (MO.getSubReg() || MO.isDef() || FoundSrcOperand)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
SrcOperandId = i;
|
|
|
|
FoundSrcOperand = true;
|
|
|
|
}
|
|
|
|
if (!FoundSrcOperand) return 0;
|
|
|
|
|
|
|
|
// Check whether we can fold the def into SrcOperandId.
|
|
|
|
SmallVector<unsigned, 8> Ops;
|
|
|
|
Ops.push_back(SrcOperandId);
|
|
|
|
MachineInstr *FoldMI = foldMemoryOperand(MI, Ops, DefMI);
|
|
|
|
if (FoldMI) {
|
|
|
|
FoldAsLoadDefReg = 0;
|
|
|
|
return FoldMI;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (Idx == 1) {
|
|
|
|
// MI was changed but it didn't help, commute it back!
|
|
|
|
commuteInstruction(MI, false);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Check whether we can commute MI and enable folding.
|
|
|
|
if (MI->isCommutable()) {
|
|
|
|
MachineInstr *NewMI = commuteInstruction(MI, false);
|
|
|
|
// Unable to commute.
|
|
|
|
if (!NewMI) return 0;
|
|
|
|
if (NewMI != MI) {
|
|
|
|
// New instruction. It doesn't need to be kept.
|
|
|
|
NewMI->eraseFromParent();
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-09-29 05:10:54 +00:00
|
|
|
/// Expand2AddrUndef - Expand a single-def pseudo instruction to a two-addr
|
|
|
|
/// instruction with two undef reads of the register being defined. This is
|
|
|
|
/// used for mapping:
|
|
|
|
/// %xmm4 = V_SET0
|
|
|
|
/// to:
|
|
|
|
/// %xmm4 = PXORrr %xmm4<undef>, %xmm4<undef>
|
|
|
|
///
|
2012-12-19 21:31:56 +00:00
|
|
|
static bool Expand2AddrUndef(MachineInstrBuilder &MIB,
|
|
|
|
const MCInstrDesc &Desc) {
|
2011-09-29 05:10:54 +00:00
|
|
|
assert(Desc.getNumOperands() == 3 && "Expected two-addr instruction.");
|
2012-12-19 21:31:56 +00:00
|
|
|
unsigned Reg = MIB->getOperand(0).getReg();
|
|
|
|
MIB->setDesc(Desc);
|
2011-09-29 05:10:54 +00:00
|
|
|
|
|
|
|
// MachineInstr::addOperand() will insert explicit operands before any
|
|
|
|
// implicit operands.
|
2012-12-19 21:31:56 +00:00
|
|
|
MIB.addReg(Reg, RegState::Undef).addReg(Reg, RegState::Undef);
|
2011-09-29 05:10:54 +00:00
|
|
|
// But we don't trust that.
|
2012-12-19 21:31:56 +00:00
|
|
|
assert(MIB->getOperand(1).getReg() == Reg &&
|
|
|
|
MIB->getOperand(2).getReg() == Reg && "Misplaced operand");
|
2011-09-29 05:10:54 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool X86InstrInfo::expandPostRAPseudo(MachineBasicBlock::iterator MI) const {
|
|
|
|
bool HasAVX = TM.getSubtarget<X86Subtarget>().hasAVX();
|
2012-12-19 21:31:56 +00:00
|
|
|
MachineInstrBuilder MIB(*MI->getParent()->getParent(), MI);
|
2011-09-29 05:10:54 +00:00
|
|
|
switch (MI->getOpcode()) {
|
2012-10-05 06:05:15 +00:00
|
|
|
case X86::SETB_C8r:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::SBB8rr));
|
2012-10-05 06:05:15 +00:00
|
|
|
case X86::SETB_C16r:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::SBB16rr));
|
2012-10-05 06:05:15 +00:00
|
|
|
case X86::SETB_C32r:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::SBB32rr));
|
2012-10-05 06:05:15 +00:00
|
|
|
case X86::SETB_C64r:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::SBB64rr));
|
2011-09-29 05:10:54 +00:00
|
|
|
case X86::V_SET0:
|
2011-11-29 22:27:25 +00:00
|
|
|
case X86::FsFLD0SS:
|
|
|
|
case X86::FsFLD0SD:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(HasAVX ? X86::VXORPSrr : X86::XORPSrr));
|
2012-08-28 07:05:28 +00:00
|
|
|
case X86::AVX_SET0:
|
|
|
|
assert(HasAVX && "AVX not supported");
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::VXORPSYrr));
|
2013-08-25 12:54:30 +00:00
|
|
|
case X86::AVX512_512_SET0:
|
|
|
|
return Expand2AddrUndef(MIB, get(X86::VPXORDZrr));
|
2012-08-28 07:30:47 +00:00
|
|
|
case X86::V_SETALLONES:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(HasAVX ? X86::VPCMPEQDrr : X86::PCMPEQDrr));
|
2012-08-28 07:30:47 +00:00
|
|
|
case X86::AVX2_SETALLONES:
|
2012-12-19 21:31:56 +00:00
|
|
|
return Expand2AddrUndef(MIB, get(X86::VPCMPEQDYrr));
|
2011-10-08 18:28:28 +00:00
|
|
|
case X86::TEST8ri_NOREX:
|
|
|
|
MI->setDesc(get(X86::TEST8ri));
|
|
|
|
return true;
|
2013-08-25 12:54:30 +00:00
|
|
|
case X86::KSET0W: return Expand2AddrUndef(MIB, get(X86::KXORWrr));
|
|
|
|
case X86::KSET1B:
|
|
|
|
case X86::KSET1W: return Expand2AddrUndef(MIB, get(X86::KXNORWrr));
|
2011-09-29 05:10:54 +00:00
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2008-07-07 23:14:23 +00:00
|
|
|
static MachineInstr *FuseTwoAddrInst(MachineFunction &MF, unsigned Opcode,
|
2009-01-05 17:59:02 +00:00
|
|
|
const SmallVectorImpl<MachineOperand> &MOs,
|
2009-02-03 00:55:04 +00:00
|
|
|
MachineInstr *MI,
|
|
|
|
const TargetInstrInfo &TII) {
|
2008-01-07 01:35:02 +00:00
|
|
|
// Create the base instruction with the memory operand as the first part.
|
2012-12-19 21:31:56 +00:00
|
|
|
// Omit the implicit operands, something BuildMI can't do.
|
2009-02-03 00:55:04 +00:00
|
|
|
MachineInstr *NewMI = MF.CreateMachineInstr(TII.get(Opcode),
|
|
|
|
MI->getDebugLoc(), true);
|
2012-12-19 21:31:56 +00:00
|
|
|
MachineInstrBuilder MIB(MF, NewMI);
|
2008-01-07 01:35:02 +00:00
|
|
|
unsigned NumAddrOps = MOs.size();
|
|
|
|
for (unsigned i = 0; i != NumAddrOps; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MOs[i]);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NumAddrOps < 4) // FrameIndex only
|
2009-04-08 21:14:34 +00:00
|
|
|
addOffset(MIB, 0);
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
// Loop over the rest of the ri operands, converting them over.
|
2008-01-07 07:27:27 +00:00
|
|
|
unsigned NumOps = MI->getDesc().getNumOperands()-2;
|
2008-01-07 01:35:02 +00:00
|
|
|
for (unsigned i = 0; i != NumOps; ++i) {
|
|
|
|
MachineOperand &MO = MI->getOperand(i+2);
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MO);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
for (unsigned i = NumOps+2, e = MI->getNumOperands(); i != e; ++i) {
|
|
|
|
MachineOperand &MO = MI->getOperand(i);
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MO);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
return MIB;
|
|
|
|
}
|
|
|
|
|
2008-07-07 23:14:23 +00:00
|
|
|
static MachineInstr *FuseInst(MachineFunction &MF,
|
|
|
|
unsigned Opcode, unsigned OpNo,
|
2009-01-05 17:59:02 +00:00
|
|
|
const SmallVectorImpl<MachineOperand> &MOs,
|
2008-01-07 01:35:02 +00:00
|
|
|
MachineInstr *MI, const TargetInstrInfo &TII) {
|
2012-12-19 21:31:56 +00:00
|
|
|
// Omit the implicit operands, something BuildMI can't do.
|
2009-02-03 00:55:04 +00:00
|
|
|
MachineInstr *NewMI = MF.CreateMachineInstr(TII.get(Opcode),
|
|
|
|
MI->getDebugLoc(), true);
|
2012-12-19 21:31:56 +00:00
|
|
|
MachineInstrBuilder MIB(MF, NewMI);
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
|
|
|
|
MachineOperand &MO = MI->getOperand(i);
|
|
|
|
if (i == OpNo) {
|
2008-10-03 15:45:36 +00:00
|
|
|
assert(MO.isReg() && "Expected to fold into reg operand!");
|
2008-01-07 01:35:02 +00:00
|
|
|
unsigned NumAddrOps = MOs.size();
|
|
|
|
for (unsigned i = 0; i != NumAddrOps; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MOs[i]);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NumAddrOps < 4) // FrameIndex only
|
2009-04-08 21:14:34 +00:00
|
|
|
addOffset(MIB, 0);
|
2008-01-07 01:35:02 +00:00
|
|
|
} else {
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MO);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
return MIB;
|
|
|
|
}
|
|
|
|
|
|
|
|
static MachineInstr *MakeM0Inst(const TargetInstrInfo &TII, unsigned Opcode,
|
2009-01-05 17:59:02 +00:00
|
|
|
const SmallVectorImpl<MachineOperand> &MOs,
|
2008-01-07 01:35:02 +00:00
|
|
|
MachineInstr *MI) {
|
2008-07-07 23:14:23 +00:00
|
|
|
MachineFunction &MF = *MI->getParent()->getParent();
|
2009-02-11 21:51:19 +00:00
|
|
|
MachineInstrBuilder MIB = BuildMI(MF, MI->getDebugLoc(), TII.get(Opcode));
|
2008-01-07 01:35:02 +00:00
|
|
|
|
|
|
|
unsigned NumAddrOps = MOs.size();
|
|
|
|
for (unsigned i = 0; i != NumAddrOps; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(MOs[i]);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NumAddrOps < 4) // FrameIndex only
|
2009-04-08 21:14:34 +00:00
|
|
|
addOffset(MIB, 0);
|
2008-01-07 01:35:02 +00:00
|
|
|
return MIB.addImm(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
MachineInstr*
|
2008-12-03 18:43:12 +00:00
|
|
|
X86InstrInfo::foldMemoryOperandImpl(MachineFunction &MF,
|
|
|
|
MachineInstr *MI, unsigned i,
|
2009-07-15 06:10:07 +00:00
|
|
|
const SmallVectorImpl<MachineOperand> &MOs,
|
2009-09-11 00:39:26 +00:00
|
|
|
unsigned Size, unsigned Align) const {
|
2010-10-07 23:08:41 +00:00
|
|
|
const DenseMap<unsigned, std::pair<unsigned,unsigned> > *OpcodeTablePtr = 0;
|
2013-03-27 23:16:18 +00:00
|
|
|
bool isCallRegIndirect = TM.getSubtarget<X86Subtarget>().callRegIndirect();
|
2008-01-07 01:35:02 +00:00
|
|
|
bool isTwoAddrFold = false;
|
2013-03-27 23:16:18 +00:00
|
|
|
|
|
|
|
// Atom favors register form of call. So, we do not fold loads into calls
|
|
|
|
// when X86Subtarget is Atom.
|
|
|
|
if (isCallRegIndirect &&
|
|
|
|
(MI->getOpcode() == X86::CALL32r || MI->getOpcode() == X86::CALL64r)) {
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2008-01-07 07:27:27 +00:00
|
|
|
unsigned NumOps = MI->getDesc().getNumOperands();
|
2008-01-07 01:35:02 +00:00
|
|
|
bool isTwoAddr = NumOps > 1 &&
|
2011-06-28 19:10:37 +00:00
|
|
|
MI->getDesc().getOperandConstraint(1, MCOI::TIED_TO) != -1;
|
2008-01-07 01:35:02 +00:00
|
|
|
|
2011-04-30 23:00:05 +00:00
|
|
|
// FIXME: AsmPrinter doesn't know how to handle
|
|
|
|
// X86II::MO_GOT_ABSOLUTE_ADDRESS after folding.
|
|
|
|
if (MI->getOpcode() == X86::ADD32ri &&
|
|
|
|
MI->getOperand(2).getTargetFlags() == X86II::MO_GOT_ABSOLUTE_ADDRESS)
|
|
|
|
return NULL;
|
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
MachineInstr *NewMI = NULL;
|
|
|
|
// Folding a memory location into the two-address part of a two-address
|
|
|
|
// instruction is different than folding it other places. It requires
|
|
|
|
// replacing the *two* registers with the memory location.
|
|
|
|
if (isTwoAddr && NumOps >= 2 && i < 2 &&
|
2008-10-03 15:45:36 +00:00
|
|
|
MI->getOperand(0).isReg() &&
|
|
|
|
MI->getOperand(1).isReg() &&
|
2011-01-26 02:03:37 +00:00
|
|
|
MI->getOperand(0).getReg() == MI->getOperand(1).getReg()) {
|
2008-01-07 01:35:02 +00:00
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable2Addr;
|
|
|
|
isTwoAddrFold = true;
|
|
|
|
} else if (i == 0) { // If operand 0
|
2013-05-30 13:19:42 +00:00
|
|
|
if (MI->getOpcode() == X86::MOV32r0) {
|
|
|
|
NewMI = MakeM0Inst(*this, X86::MOV32mi, MOs, MI);
|
|
|
|
if (NewMI)
|
|
|
|
return NewMI;
|
2012-08-23 04:57:36 +00:00
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable0;
|
|
|
|
} else if (i == 1) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable1;
|
|
|
|
} else if (i == 2) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable2;
|
2012-08-01 12:06:00 +00:00
|
|
|
} else if (i == 3) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable3;
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
// If table selected...
|
|
|
|
if (OpcodeTablePtr) {
|
|
|
|
// Find the Opcode to fuse
|
2010-10-07 23:08:41 +00:00
|
|
|
DenseMap<unsigned, std::pair<unsigned,unsigned> >::const_iterator I =
|
|
|
|
OpcodeTablePtr->find(MI->getOpcode());
|
2008-01-07 01:35:02 +00:00
|
|
|
if (I != OpcodeTablePtr->end()) {
|
2009-09-11 00:39:26 +00:00
|
|
|
unsigned Opcode = I->second.first;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
unsigned MinAlign = (I->second.second & TB_ALIGN_MASK) >> TB_ALIGN_SHIFT;
|
2009-07-15 06:10:07 +00:00
|
|
|
if (Align < MinAlign)
|
|
|
|
return NULL;
|
2009-09-11 01:01:31 +00:00
|
|
|
bool NarrowToMOV32rm = false;
|
2009-09-11 00:39:26 +00:00
|
|
|
if (Size) {
|
2012-05-07 22:10:26 +00:00
|
|
|
unsigned RCSize = getRegClass(MI->getDesc(), i, &RI, MF)->getSize();
|
2009-09-11 00:39:26 +00:00
|
|
|
if (Size < RCSize) {
|
|
|
|
// Check if it's safe to fold the load. If the size of the object is
|
|
|
|
// narrower than the load width, then it's not.
|
|
|
|
if (Opcode != X86::MOV64rm || RCSize != 8 || Size != 4)
|
|
|
|
return NULL;
|
|
|
|
// If this is a 64-bit load, but the spill slot is 32, then we can do
|
|
|
|
// a 32-bit load which is implicitly zero-extended. This likely is due
|
|
|
|
// to liveintervalanalysis remat'ing a load from stack slot.
|
2009-09-11 01:01:31 +00:00
|
|
|
if (MI->getOperand(0).getSubReg() || MI->getOperand(1).getSubReg())
|
|
|
|
return NULL;
|
2009-09-11 00:39:26 +00:00
|
|
|
Opcode = X86::MOV32rm;
|
2009-09-11 01:01:31 +00:00
|
|
|
NarrowToMOV32rm = true;
|
2009-09-11 00:39:26 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
if (isTwoAddrFold)
|
2009-09-11 00:39:26 +00:00
|
|
|
NewMI = FuseTwoAddrInst(MF, Opcode, MOs, MI, *this);
|
2008-01-07 01:35:02 +00:00
|
|
|
else
|
2009-09-11 00:39:26 +00:00
|
|
|
NewMI = FuseInst(MF, Opcode, i, MOs, MI, *this);
|
2009-09-11 01:01:31 +00:00
|
|
|
|
|
|
|
if (NarrowToMOV32rm) {
|
|
|
|
// If this is the special case where we use a MOV32rm to load a 32-bit
|
|
|
|
// value and zero-extend the top bits. Change the destination register
|
|
|
|
// to a 32-bit one.
|
|
|
|
unsigned DstReg = NewMI->getOperand(0).getReg();
|
|
|
|
if (TargetRegisterInfo::isPhysicalRegister(DstReg))
|
|
|
|
NewMI->getOperand(0).setReg(RI.getSubReg(DstReg,
|
2010-05-24 14:48:17 +00:00
|
|
|
X86::sub_32bit));
|
2009-09-11 01:01:31 +00:00
|
|
|
else
|
2010-05-24 14:48:17 +00:00
|
|
|
NewMI->getOperand(0).setSubReg(X86::sub_32bit);
|
2009-09-11 01:01:31 +00:00
|
|
|
}
|
2008-01-07 01:35:02 +00:00
|
|
|
return NewMI;
|
|
|
|
}
|
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
|
|
|
// No fusion
|
2010-07-09 20:43:09 +00:00
|
|
|
if (PrintFailedFusing && !MI->isCopy())
|
2010-01-05 01:29:29 +00:00
|
|
|
dbgs() << "We failed to fuse operand " << i << " in " << *MI;
|
2008-01-07 01:35:02 +00:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2011-09-15 21:42:23 +00:00
|
|
|
/// hasPartialRegUpdate - Return true for all instructions that only update
|
|
|
|
/// the first 32 or 64-bits of the destination register and leave the rest
|
|
|
|
/// unmodified. This can be used to avoid folding loads if the instructions
|
|
|
|
/// only update part of the destination register, and the non-updated part is
|
|
|
|
/// not needed. e.g. cvtss2sd, sqrtss. Unfolding the load from these
|
|
|
|
/// instructions breaks the partial register dependency and it can improve
|
|
|
|
/// performance. e.g.:
|
|
|
|
///
|
|
|
|
/// movss (%rdi), %xmm0
|
|
|
|
/// cvtss2sd %xmm0, %xmm0
|
|
|
|
///
|
|
|
|
/// Instead of
|
|
|
|
/// cvtss2sd (%rdi), %xmm0
|
|
|
|
///
|
2011-09-15 23:04:24 +00:00
|
|
|
/// FIXME: This should be turned into a TSFlags.
|
|
|
|
///
|
2011-09-15 21:42:23 +00:00
|
|
|
static bool hasPartialRegUpdate(unsigned Opcode) {
|
|
|
|
switch (Opcode) {
|
2011-11-15 01:15:30 +00:00
|
|
|
case X86::CVTSI2SSrr:
|
|
|
|
case X86::CVTSI2SS64rr:
|
|
|
|
case X86::CVTSI2SDrr:
|
|
|
|
case X86::CVTSI2SD64rr:
|
2011-09-15 21:42:23 +00:00
|
|
|
case X86::CVTSD2SSrr:
|
|
|
|
case X86::Int_CVTSD2SSrr:
|
|
|
|
case X86::CVTSS2SDrr:
|
|
|
|
case X86::Int_CVTSS2SDrr:
|
|
|
|
case X86::RCPSSr:
|
|
|
|
case X86::RCPSSr_Int:
|
|
|
|
case X86::ROUNDSDr:
|
2011-12-09 15:43:55 +00:00
|
|
|
case X86::ROUNDSDr_Int:
|
2011-09-15 21:42:23 +00:00
|
|
|
case X86::ROUNDSSr:
|
2011-12-09 15:43:55 +00:00
|
|
|
case X86::ROUNDSSr_Int:
|
2011-09-15 21:42:23 +00:00
|
|
|
case X86::RSQRTSSr:
|
|
|
|
case X86::RSQRTSSr_Int:
|
|
|
|
case X86::SQRTSSr:
|
|
|
|
case X86::SQRTSSr_Int:
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
2008-01-07 01:35:02 +00:00
|
|
|
|
2011-11-15 01:15:30 +00:00
|
|
|
/// getPartialRegUpdateClearance - Inform the ExeDepsFix pass how many idle
|
|
|
|
/// instructions we would like before a partial register update.
|
|
|
|
unsigned X86InstrInfo::
|
|
|
|
getPartialRegUpdateClearance(const MachineInstr *MI, unsigned OpNum,
|
|
|
|
const TargetRegisterInfo *TRI) const {
|
|
|
|
if (OpNum != 0 || !hasPartialRegUpdate(MI->getOpcode()))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
// If MI is marked as reading Reg, the partial register update is wanted.
|
|
|
|
const MachineOperand &MO = MI->getOperand(0);
|
|
|
|
unsigned Reg = MO.getReg();
|
|
|
|
if (TargetRegisterInfo::isVirtualRegister(Reg)) {
|
|
|
|
if (MO.readsReg() || MI->readsVirtualRegister(Reg))
|
|
|
|
return 0;
|
|
|
|
} else {
|
|
|
|
if (MI->readsRegister(Reg, TRI))
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
// If any of the preceding 16 instructions are reading Reg, insert a
|
|
|
|
// dependency breaking instruction. The magic number is based on a few
|
|
|
|
// Nehalem experiments.
|
|
|
|
return 16;
|
|
|
|
}
|
|
|
|
|
2013-10-14 22:19:03 +00:00
|
|
|
// Return true for any instruction the copies the high bits of the first source
|
|
|
|
// operand into the unused high bits of the destination operand.
|
|
|
|
static bool hasUndefRegUpdate(unsigned Opcode) {
|
|
|
|
switch (Opcode) {
|
|
|
|
case X86::VCVTSI2SSrr:
|
|
|
|
case X86::Int_VCVTSI2SSrr:
|
|
|
|
case X86::VCVTSI2SS64rr:
|
|
|
|
case X86::Int_VCVTSI2SS64rr:
|
|
|
|
case X86::VCVTSI2SDrr:
|
|
|
|
case X86::Int_VCVTSI2SDrr:
|
|
|
|
case X86::VCVTSI2SD64rr:
|
|
|
|
case X86::Int_VCVTSI2SD64rr:
|
|
|
|
case X86::VCVTSD2SSrr:
|
|
|
|
case X86::Int_VCVTSD2SSrr:
|
|
|
|
case X86::VCVTSS2SDrr:
|
|
|
|
case X86::Int_VCVTSS2SDrr:
|
|
|
|
case X86::VRCPSSr:
|
|
|
|
case X86::VROUNDSDr:
|
|
|
|
case X86::VROUNDSDr_Int:
|
|
|
|
case X86::VROUNDSSr:
|
|
|
|
case X86::VROUNDSSr_Int:
|
|
|
|
case X86::VRSQRTSSr:
|
|
|
|
case X86::VSQRTSSr:
|
|
|
|
|
|
|
|
// AVX-512
|
|
|
|
case X86::VCVTSD2SSZrr:
|
|
|
|
case X86::VCVTSS2SDZrr:
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
/// Inform the ExeDepsFix pass how many idle instructions we would like before
|
|
|
|
/// certain undef register reads.
|
|
|
|
///
|
|
|
|
/// This catches the VCVTSI2SD family of instructions:
|
|
|
|
///
|
|
|
|
/// vcvtsi2sdq %rax, %xmm0<undef>, %xmm14
|
|
|
|
///
|
|
|
|
/// We should to be careful *not* to catch VXOR idioms which are presumably
|
|
|
|
/// handled specially in the pipeline:
|
|
|
|
///
|
|
|
|
/// vxorps %xmm1<undef>, %xmm1<undef>, %xmm1
|
|
|
|
///
|
|
|
|
/// Like getPartialRegUpdateClearance, this makes a strong assumption that the
|
|
|
|
/// high bits that are passed-through are not live.
|
|
|
|
unsigned X86InstrInfo::
|
|
|
|
getUndefRegClearance(const MachineInstr *MI, unsigned &OpNum,
|
|
|
|
const TargetRegisterInfo *TRI) const {
|
|
|
|
if (!hasUndefRegUpdate(MI->getOpcode()))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
// Set the OpNum parameter to the first source operand.
|
|
|
|
OpNum = 1;
|
|
|
|
|
|
|
|
const MachineOperand &MO = MI->getOperand(OpNum);
|
|
|
|
if (MO.isUndef() && TargetRegisterInfo::isPhysicalRegister(MO.getReg())) {
|
|
|
|
// Use the same magic number as getPartialRegUpdateClearance.
|
|
|
|
return 16;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-11-15 01:15:30 +00:00
|
|
|
void X86InstrInfo::
|
|
|
|
breakPartialRegDependency(MachineBasicBlock::iterator MI, unsigned OpNum,
|
|
|
|
const TargetRegisterInfo *TRI) const {
|
|
|
|
unsigned Reg = MI->getOperand(OpNum).getReg();
|
2013-10-14 22:19:03 +00:00
|
|
|
// If MI kills this register, the false dependence is already broken.
|
|
|
|
if (MI->killsRegister(Reg, TRI))
|
|
|
|
return;
|
2011-11-15 01:15:30 +00:00
|
|
|
if (X86::VR128RegClass.contains(Reg)) {
|
|
|
|
// These instructions are all floating point domain, so xorps is the best
|
|
|
|
// choice.
|
|
|
|
bool HasAVX = TM.getSubtarget<X86Subtarget>().hasAVX();
|
|
|
|
unsigned Opc = HasAVX ? X86::VXORPSrr : X86::XORPSrr;
|
|
|
|
BuildMI(*MI->getParent(), MI, MI->getDebugLoc(), get(Opc), Reg)
|
|
|
|
.addReg(Reg, RegState::Undef).addReg(Reg, RegState::Undef);
|
|
|
|
} else if (X86::VR256RegClass.contains(Reg)) {
|
|
|
|
// Use vxorps to clear the full ymm register.
|
|
|
|
// It wants to read and write the xmm sub-register.
|
|
|
|
unsigned XReg = TRI->getSubReg(Reg, X86::sub_xmm);
|
|
|
|
BuildMI(*MI->getParent(), MI, MI->getDebugLoc(), get(X86::VXORPSrr), XReg)
|
|
|
|
.addReg(XReg, RegState::Undef).addReg(XReg, RegState::Undef)
|
|
|
|
.addReg(Reg, RegState::ImplicitDefine);
|
|
|
|
} else
|
|
|
|
return;
|
|
|
|
MI->addRegisterKilled(Reg, TRI, true);
|
|
|
|
}
|
|
|
|
|
2008-12-03 18:43:12 +00:00
|
|
|
MachineInstr* X86InstrInfo::foldMemoryOperandImpl(MachineFunction &MF,
|
|
|
|
MachineInstr *MI,
|
2009-07-15 06:10:07 +00:00
|
|
|
const SmallVectorImpl<unsigned> &Ops,
|
2008-12-03 18:43:12 +00:00
|
|
|
int FrameIndex) const {
|
2011-01-26 02:03:37 +00:00
|
|
|
// Check switch flag
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NoFusing) return NULL;
|
|
|
|
|
2011-09-15 21:42:23 +00:00
|
|
|
// Unless optimizing for size, don't fold to avoid partial
|
|
|
|
// register update stalls
|
2012-12-30 10:32:01 +00:00
|
|
|
if (!MF.getFunction()->getAttributes().
|
|
|
|
hasAttribute(AttributeSet::FunctionIndex, Attribute::OptimizeForSize) &&
|
2011-09-15 21:42:23 +00:00
|
|
|
hasPartialRegUpdate(MI->getOpcode()))
|
|
|
|
return 0;
|
2009-12-18 07:40:29 +00:00
|
|
|
|
2008-02-08 21:20:40 +00:00
|
|
|
const MachineFrameInfo *MFI = MF.getFrameInfo();
|
2009-09-11 00:39:26 +00:00
|
|
|
unsigned Size = MFI->getObjectSize(FrameIndex);
|
2008-02-08 21:20:40 +00:00
|
|
|
unsigned Alignment = MFI->getObjectAlignment(FrameIndex);
|
2013-10-06 13:48:22 +00:00
|
|
|
// If the function stack isn't realigned we don't want to fold instructions
|
|
|
|
// that need increased alignment.
|
|
|
|
if (!RI.needsStackRealignment(MF))
|
|
|
|
Alignment = std::min(Alignment, TM.getFrameLowering()->getStackAlignment());
|
2008-01-07 01:35:02 +00:00
|
|
|
if (Ops.size() == 2 && Ops[0] == 0 && Ops[1] == 1) {
|
|
|
|
unsigned NewOpc = 0;
|
2009-09-11 00:39:26 +00:00
|
|
|
unsigned RCSize = 0;
|
2008-01-07 01:35:02 +00:00
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: return NULL;
|
2009-09-11 00:39:26 +00:00
|
|
|
case X86::TEST8rr: NewOpc = X86::CMP8ri; RCSize = 1; break;
|
2010-05-18 21:42:03 +00:00
|
|
|
case X86::TEST16rr: NewOpc = X86::CMP16ri8; RCSize = 2; break;
|
|
|
|
case X86::TEST32rr: NewOpc = X86::CMP32ri8; RCSize = 4; break;
|
|
|
|
case X86::TEST64rr: NewOpc = X86::CMP64ri8; RCSize = 8; break;
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
2009-09-11 00:39:26 +00:00
|
|
|
// Check if it's safe to fold the load. If the size of the object is
|
|
|
|
// narrower than the load width, then it's not.
|
|
|
|
if (Size < RCSize)
|
|
|
|
return NULL;
|
2008-01-07 01:35:02 +00:00
|
|
|
// Change to CMPXXri r, 0 first.
|
2008-01-11 18:10:50 +00:00
|
|
|
MI->setDesc(get(NewOpc));
|
2008-01-07 01:35:02 +00:00
|
|
|
MI->getOperand(1).ChangeToImmediate(0);
|
|
|
|
} else if (Ops.size() != 1)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
SmallVector<MachineOperand,4> MOs;
|
|
|
|
MOs.push_back(MachineOperand::CreateFI(FrameIndex));
|
2009-09-11 00:39:26 +00:00
|
|
|
return foldMemoryOperandImpl(MF, MI, Ops[0], MOs, Size, Alignment);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
2008-12-03 18:43:12 +00:00
|
|
|
MachineInstr* X86InstrInfo::foldMemoryOperandImpl(MachineFunction &MF,
|
|
|
|
MachineInstr *MI,
|
2009-07-15 06:10:07 +00:00
|
|
|
const SmallVectorImpl<unsigned> &Ops,
|
2008-12-03 18:43:12 +00:00
|
|
|
MachineInstr *LoadMI) const {
|
2011-01-26 02:03:37 +00:00
|
|
|
// Check switch flag
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NoFusing) return NULL;
|
|
|
|
|
2011-09-15 21:42:23 +00:00
|
|
|
// Unless optimizing for size, don't fold to avoid partial
|
|
|
|
// register update stalls
|
2012-12-30 10:32:01 +00:00
|
|
|
if (!MF.getFunction()->getAttributes().
|
|
|
|
hasAttribute(AttributeSet::FunctionIndex, Attribute::OptimizeForSize) &&
|
2011-09-15 21:42:23 +00:00
|
|
|
hasPartialRegUpdate(MI->getOpcode()))
|
|
|
|
return 0;
|
2009-12-18 07:40:29 +00:00
|
|
|
|
2008-07-12 00:10:52 +00:00
|
|
|
// Determine the alignment of the load.
|
2008-02-08 21:20:40 +00:00
|
|
|
unsigned Alignment = 0;
|
2008-07-12 00:10:52 +00:00
|
|
|
if (LoadMI->hasOneMemOperand())
|
2009-09-25 20:36:54 +00:00
|
|
|
Alignment = (*LoadMI->memoperands_begin())->getAlignment();
|
2009-09-21 18:30:38 +00:00
|
|
|
else
|
|
|
|
switch (LoadMI->getOpcode()) {
|
2011-11-19 22:34:59 +00:00
|
|
|
case X86::AVX2_SETALLONES:
|
2012-08-28 07:05:28 +00:00
|
|
|
case X86::AVX_SET0:
|
2010-08-12 20:20:53 +00:00
|
|
|
Alignment = 32;
|
|
|
|
break;
|
2011-09-29 05:10:54 +00:00
|
|
|
case X86::V_SET0:
|
2009-09-21 18:30:38 +00:00
|
|
|
case X86::V_SETALLONES:
|
|
|
|
Alignment = 16;
|
|
|
|
break;
|
|
|
|
case X86::FsFLD0SD:
|
|
|
|
Alignment = 8;
|
|
|
|
break;
|
|
|
|
case X86::FsFLD0SS:
|
|
|
|
Alignment = 4;
|
|
|
|
break;
|
|
|
|
default:
|
2011-06-10 01:13:01 +00:00
|
|
|
return 0;
|
2009-09-21 18:30:38 +00:00
|
|
|
}
|
2008-01-07 01:35:02 +00:00
|
|
|
if (Ops.size() == 2 && Ops[0] == 0 && Ops[1] == 1) {
|
|
|
|
unsigned NewOpc = 0;
|
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: return NULL;
|
|
|
|
case X86::TEST8rr: NewOpc = X86::CMP8ri; break;
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::TEST16rr: NewOpc = X86::CMP16ri8; break;
|
|
|
|
case X86::TEST32rr: NewOpc = X86::CMP32ri8; break;
|
|
|
|
case X86::TEST64rr: NewOpc = X86::CMP64ri8; break;
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
// Change to CMPXXri r, 0 first.
|
2008-01-11 18:10:50 +00:00
|
|
|
MI->setDesc(get(NewOpc));
|
2008-01-07 01:35:02 +00:00
|
|
|
MI->getOperand(1).ChangeToImmediate(0);
|
|
|
|
} else if (Ops.size() != 1)
|
|
|
|
return NULL;
|
|
|
|
|
2010-08-11 23:08:22 +00:00
|
|
|
// Make sure the subregisters match.
|
|
|
|
// Otherwise we risk changing the size of the load.
|
|
|
|
if (LoadMI->getOperand(0).getSubReg() != MI->getOperand(Ops[0]).getSubReg())
|
|
|
|
return NULL;
|
|
|
|
|
2010-07-08 22:41:28 +00:00
|
|
|
SmallVector<MachineOperand,X86::AddrNumOperands> MOs;
|
2009-09-21 18:30:38 +00:00
|
|
|
switch (LoadMI->getOpcode()) {
|
2011-09-29 05:10:54 +00:00
|
|
|
case X86::V_SET0:
|
2009-09-21 18:30:38 +00:00
|
|
|
case X86::V_SETALLONES:
|
2011-11-19 22:34:59 +00:00
|
|
|
case X86::AVX2_SETALLONES:
|
2012-08-28 07:05:28 +00:00
|
|
|
case X86::AVX_SET0:
|
2009-09-21 18:30:38 +00:00
|
|
|
case X86::FsFLD0SD:
|
2011-11-29 22:27:25 +00:00
|
|
|
case X86::FsFLD0SS: {
|
2011-09-29 05:10:54 +00:00
|
|
|
// Folding a V_SET0 or V_SETALLONES as a load, to ease register pressure.
|
2008-12-03 05:21:24 +00:00
|
|
|
// Create a constant-pool entry and operands to load from it.
|
|
|
|
|
2010-03-09 03:01:40 +00:00
|
|
|
// Medium and large mode can't fold loads this way.
|
|
|
|
if (TM.getCodeModel() != CodeModel::Small &&
|
|
|
|
TM.getCodeModel() != CodeModel::Kernel)
|
|
|
|
return NULL;
|
|
|
|
|
2008-12-03 05:21:24 +00:00
|
|
|
// x86-32 PIC requires a PIC base register for constant pools.
|
|
|
|
unsigned PICBase = 0;
|
2009-07-16 21:24:13 +00:00
|
|
|
if (TM.getRelocationModel() == Reloc::PIC_) {
|
2009-07-16 18:44:05 +00:00
|
|
|
if (TM.getSubtarget<X86Subtarget>().is64Bit())
|
|
|
|
PICBase = X86::RIP;
|
2009-07-16 21:24:13 +00:00
|
|
|
else
|
2010-07-10 09:00:22 +00:00
|
|
|
// FIXME: PICBase = getGlobalBaseReg(&MF);
|
2009-07-16 18:44:05 +00:00
|
|
|
// This doesn't work for several reasons.
|
|
|
|
// 1. GlobalBaseReg may have been spilled.
|
|
|
|
// 2. It may not be live at MI.
|
2009-09-21 18:30:38 +00:00
|
|
|
return NULL;
|
2009-07-16 21:24:13 +00:00
|
|
|
}
|
2008-12-03 05:21:24 +00:00
|
|
|
|
2009-09-21 18:30:38 +00:00
|
|
|
// Create a constant-pool entry.
|
2008-12-03 05:21:24 +00:00
|
|
|
MachineConstantPool &MCP = *MF.getConstantPool();
|
2011-07-18 04:54:35 +00:00
|
|
|
Type *Ty;
|
2010-08-12 20:20:53 +00:00
|
|
|
unsigned Opc = LoadMI->getOpcode();
|
2011-11-29 22:27:25 +00:00
|
|
|
if (Opc == X86::FsFLD0SS)
|
2009-09-21 18:30:38 +00:00
|
|
|
Ty = Type::getFloatTy(MF.getFunction()->getContext());
|
2011-11-29 22:27:25 +00:00
|
|
|
else if (Opc == X86::FsFLD0SD)
|
2009-09-21 18:30:38 +00:00
|
|
|
Ty = Type::getDoubleTy(MF.getFunction()->getContext());
|
2012-08-28 07:05:28 +00:00
|
|
|
else if (Opc == X86::AVX2_SETALLONES || Opc == X86::AVX_SET0)
|
2012-01-13 06:12:41 +00:00
|
|
|
Ty = VectorType::get(Type::getInt32Ty(MF.getFunction()->getContext()), 8);
|
2009-09-21 18:30:38 +00:00
|
|
|
else
|
|
|
|
Ty = VectorType::get(Type::getInt32Ty(MF.getFunction()->getContext()), 4);
|
2011-07-25 23:05:32 +00:00
|
|
|
|
2012-08-28 07:30:47 +00:00
|
|
|
bool IsAllOnes = (Opc == X86::V_SETALLONES || Opc == X86::AVX2_SETALLONES);
|
2011-07-25 23:05:32 +00:00
|
|
|
const Constant *C = IsAllOnes ? Constant::getAllOnesValue(Ty) :
|
|
|
|
Constant::getNullValue(Ty);
|
2009-09-21 18:30:38 +00:00
|
|
|
unsigned CPI = MCP.getConstantPoolIndex(C, Alignment);
|
2008-12-03 05:21:24 +00:00
|
|
|
|
|
|
|
// Create operands to load from the constant pool entry.
|
|
|
|
MOs.push_back(MachineOperand::CreateReg(PICBase, false));
|
|
|
|
MOs.push_back(MachineOperand::CreateImm(1));
|
|
|
|
MOs.push_back(MachineOperand::CreateReg(0, false));
|
|
|
|
MOs.push_back(MachineOperand::CreateCPI(CPI, 0));
|
2009-04-08 21:14:34 +00:00
|
|
|
MOs.push_back(MachineOperand::CreateReg(0, false));
|
2009-09-21 18:30:38 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
default: {
|
2012-11-27 18:09:26 +00:00
|
|
|
if ((LoadMI->getOpcode() == X86::MOVSSrm ||
|
|
|
|
LoadMI->getOpcode() == X86::VMOVSSrm) &&
|
|
|
|
MF.getRegInfo().getRegClass(LoadMI->getOperand(0).getReg())->getSize()
|
|
|
|
> 4)
|
|
|
|
// These instructions only load 32 bits, we can't fold them if the
|
|
|
|
// destination register is wider than 32 bits (4 bytes).
|
|
|
|
return NULL;
|
|
|
|
if ((LoadMI->getOpcode() == X86::MOVSDrm ||
|
|
|
|
LoadMI->getOpcode() == X86::VMOVSDrm) &&
|
|
|
|
MF.getRegInfo().getRegClass(LoadMI->getOperand(0).getReg())->getSize()
|
|
|
|
> 8)
|
|
|
|
// These instructions only load 64 bits, we can't fold them if the
|
|
|
|
// destination register is wider than 64 bits (8 bytes).
|
|
|
|
return NULL;
|
|
|
|
|
2008-12-03 05:21:24 +00:00
|
|
|
// Folding a normal load. Just copy the load's address operands.
|
|
|
|
unsigned NumOps = LoadMI->getDesc().getNumOperands();
|
2010-07-08 22:41:28 +00:00
|
|
|
for (unsigned i = NumOps - X86::AddrNumOperands; i != NumOps; ++i)
|
2008-12-03 05:21:24 +00:00
|
|
|
MOs.push_back(LoadMI->getOperand(i));
|
2009-09-21 18:30:38 +00:00
|
|
|
break;
|
|
|
|
}
|
2008-12-03 05:21:24 +00:00
|
|
|
}
|
2009-09-11 00:39:26 +00:00
|
|
|
return foldMemoryOperandImpl(MF, MI, Ops[0], MOs, 0, Alignment);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2008-10-16 01:49:15 +00:00
|
|
|
bool X86InstrInfo::canFoldMemoryOperand(const MachineInstr *MI,
|
|
|
|
const SmallVectorImpl<unsigned> &Ops) const {
|
2011-01-26 02:03:37 +00:00
|
|
|
// Check switch flag
|
2008-01-07 01:35:02 +00:00
|
|
|
if (NoFusing) return 0;
|
|
|
|
|
|
|
|
if (Ops.size() == 2 && Ops[0] == 0 && Ops[1] == 1) {
|
|
|
|
switch (MI->getOpcode()) {
|
|
|
|
default: return false;
|
2011-01-26 02:03:37 +00:00
|
|
|
case X86::TEST8rr:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::TEST16rr:
|
|
|
|
case X86::TEST32rr:
|
|
|
|
case X86::TEST64rr:
|
|
|
|
return true;
|
2011-04-30 23:00:05 +00:00
|
|
|
case X86::ADD32ri:
|
|
|
|
// FIXME: AsmPrinter doesn't know how to handle
|
|
|
|
// X86II::MO_GOT_ABSOLUTE_ADDRESS after folding.
|
|
|
|
if (MI->getOperand(2).getTargetFlags() == X86II::MO_GOT_ABSOLUTE_ADDRESS)
|
|
|
|
return false;
|
|
|
|
break;
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (Ops.size() != 1)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
unsigned OpNum = Ops[0];
|
|
|
|
unsigned Opc = MI->getOpcode();
|
2008-01-07 07:27:27 +00:00
|
|
|
unsigned NumOps = MI->getDesc().getNumOperands();
|
2008-01-07 01:35:02 +00:00
|
|
|
bool isTwoAddr = NumOps > 1 &&
|
2011-06-28 19:10:37 +00:00
|
|
|
MI->getDesc().getOperandConstraint(1, MCOI::TIED_TO) != -1;
|
2008-01-07 01:35:02 +00:00
|
|
|
|
|
|
|
// Folding a memory location into the two-address part of a two-address
|
|
|
|
// instruction is different than folding it other places. It requires
|
|
|
|
// replacing the *two* registers with the memory location.
|
2010-10-07 23:08:41 +00:00
|
|
|
const DenseMap<unsigned, std::pair<unsigned,unsigned> > *OpcodeTablePtr = 0;
|
2011-01-26 02:03:37 +00:00
|
|
|
if (isTwoAddr && NumOps >= 2 && OpNum < 2) {
|
2008-01-07 01:35:02 +00:00
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable2Addr;
|
|
|
|
} else if (OpNum == 0) { // If operand 0
|
2013-05-30 13:19:42 +00:00
|
|
|
if (Opc == X86::MOV32r0)
|
|
|
|
return true;
|
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable0;
|
|
|
|
} else if (OpNum == 1) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable1;
|
|
|
|
} else if (OpNum == 2) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable2;
|
2012-08-31 22:12:16 +00:00
|
|
|
} else if (OpNum == 3) {
|
|
|
|
OpcodeTablePtr = &RegOp2MemOpTable3;
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2010-10-08 03:54:52 +00:00
|
|
|
if (OpcodeTablePtr && OpcodeTablePtr->count(Opc))
|
|
|
|
return true;
|
2012-11-28 02:35:17 +00:00
|
|
|
return TargetInstrInfo::canFoldMemoryOperand(MI, Ops);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
bool X86InstrInfo::unfoldMemoryOperand(MachineFunction &MF, MachineInstr *MI,
|
|
|
|
unsigned Reg, bool UnfoldLoad, bool UnfoldStore,
|
2009-02-11 21:51:19 +00:00
|
|
|
SmallVectorImpl<MachineInstr*> &NewMIs) const {
|
2010-10-07 23:08:41 +00:00
|
|
|
DenseMap<unsigned, std::pair<unsigned,unsigned> >::const_iterator I =
|
|
|
|
MemOp2RegOpTable.find(MI->getOpcode());
|
2008-01-07 01:35:02 +00:00
|
|
|
if (I == MemOp2RegOpTable.end())
|
|
|
|
return false;
|
|
|
|
unsigned Opc = I->second.first;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
unsigned Index = I->second.second & TB_INDEX_MASK;
|
|
|
|
bool FoldedLoad = I->second.second & TB_FOLDED_LOAD;
|
|
|
|
bool FoldedStore = I->second.second & TB_FOLDED_STORE;
|
2008-01-07 01:35:02 +00:00
|
|
|
if (UnfoldLoad && !FoldedLoad)
|
|
|
|
return false;
|
|
|
|
UnfoldLoad &= FoldedLoad;
|
|
|
|
if (UnfoldStore && !FoldedStore)
|
|
|
|
return false;
|
|
|
|
UnfoldStore &= FoldedStore;
|
|
|
|
|
2011-06-28 19:10:37 +00:00
|
|
|
const MCInstrDesc &MCID = get(Opc);
|
2012-05-07 22:10:26 +00:00
|
|
|
const TargetRegisterClass *RC = getRegClass(MCID, Index, &RI, MF);
|
2010-07-02 20:36:18 +00:00
|
|
|
if (!MI->hasOneMemOperand() &&
|
|
|
|
RC == &X86::VR128RegClass &&
|
|
|
|
!TM.getSubtarget<X86Subtarget>().isUnalignedMemAccessFast())
|
|
|
|
// Without memoperands, loadRegFromAddr and storeRegToStackSlot will
|
|
|
|
// conservatively assume the address is unaligned. That's bad for
|
|
|
|
// performance.
|
|
|
|
return false;
|
2010-07-08 22:41:28 +00:00
|
|
|
SmallVector<MachineOperand, X86::AddrNumOperands> AddrOps;
|
2008-01-07 01:35:02 +00:00
|
|
|
SmallVector<MachineOperand,2> BeforeOps;
|
|
|
|
SmallVector<MachineOperand,2> AfterOps;
|
|
|
|
SmallVector<MachineOperand,4> ImpOps;
|
|
|
|
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
|
|
|
|
MachineOperand &Op = MI->getOperand(i);
|
2010-07-08 22:41:28 +00:00
|
|
|
if (i >= Index && i < Index + X86::AddrNumOperands)
|
2008-01-07 01:35:02 +00:00
|
|
|
AddrOps.push_back(Op);
|
2008-10-03 15:45:36 +00:00
|
|
|
else if (Op.isReg() && Op.isImplicit())
|
2008-01-07 01:35:02 +00:00
|
|
|
ImpOps.push_back(Op);
|
|
|
|
else if (i < Index)
|
|
|
|
BeforeOps.push_back(Op);
|
|
|
|
else if (i > Index)
|
|
|
|
AfterOps.push_back(Op);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Emit the load instruction.
|
|
|
|
if (UnfoldLoad) {
|
2009-10-09 18:10:05 +00:00
|
|
|
std::pair<MachineInstr::mmo_iterator,
|
|
|
|
MachineInstr::mmo_iterator> MMOs =
|
|
|
|
MF.extractLoadMemRefs(MI->memoperands_begin(),
|
|
|
|
MI->memoperands_end());
|
|
|
|
loadRegFromAddr(MF, Reg, AddrOps, RC, MMOs.first, MMOs.second, NewMIs);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (UnfoldStore) {
|
|
|
|
// Address operands cannot be marked isKill.
|
2010-07-08 22:41:28 +00:00
|
|
|
for (unsigned i = 1; i != 1 + X86::AddrNumOperands; ++i) {
|
2008-01-07 01:35:02 +00:00
|
|
|
MachineOperand &MO = NewMIs[0]->getOperand(i);
|
2008-10-03 15:45:36 +00:00
|
|
|
if (MO.isReg())
|
2008-01-07 01:35:02 +00:00
|
|
|
MO.setIsKill(false);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Emit the data processing instruction.
|
2011-06-28 19:10:37 +00:00
|
|
|
MachineInstr *DataMI = MF.CreateMachineInstr(MCID, MI->getDebugLoc(), true);
|
2012-12-19 21:31:56 +00:00
|
|
|
MachineInstrBuilder MIB(MF, DataMI);
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2008-01-07 01:35:02 +00:00
|
|
|
if (FoldedStore)
|
2009-05-13 21:33:08 +00:00
|
|
|
MIB.addReg(Reg, RegState::Define);
|
2008-01-07 01:35:02 +00:00
|
|
|
for (unsigned i = 0, e = BeforeOps.size(); i != e; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(BeforeOps[i]);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (FoldedLoad)
|
|
|
|
MIB.addReg(Reg);
|
|
|
|
for (unsigned i = 0, e = AfterOps.size(); i != e; ++i)
|
2009-02-18 05:45:50 +00:00
|
|
|
MIB.addOperand(AfterOps[i]);
|
2008-01-07 01:35:02 +00:00
|
|
|
for (unsigned i = 0, e = ImpOps.size(); i != e; ++i) {
|
|
|
|
MachineOperand &MO = ImpOps[i];
|
2009-05-13 21:33:08 +00:00
|
|
|
MIB.addReg(MO.getReg(),
|
|
|
|
getDefRegState(MO.isDef()) |
|
|
|
|
RegState::Implicit |
|
|
|
|
getKillRegState(MO.isKill()) |
|
2009-06-30 08:49:04 +00:00
|
|
|
getDeadRegState(MO.isDead()) |
|
|
|
|
getUndefRegState(MO.isUndef()));
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
// Change CMP32ri r, 0 back to TEST32rr r, r, etc.
|
|
|
|
switch (DataMI->getOpcode()) {
|
|
|
|
default: break;
|
|
|
|
case X86::CMP64ri32:
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP64ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP32ri:
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP32ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP16ri:
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP16ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP8ri: {
|
|
|
|
MachineOperand &MO0 = DataMI->getOperand(0);
|
|
|
|
MachineOperand &MO1 = DataMI->getOperand(1);
|
|
|
|
if (MO1.getImm() == 0) {
|
2012-08-21 08:16:16 +00:00
|
|
|
unsigned NewOpc;
|
2008-01-07 01:35:02 +00:00
|
|
|
switch (DataMI->getOpcode()) {
|
2012-08-21 08:16:16 +00:00
|
|
|
default: llvm_unreachable("Unreachable!");
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP64ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP64ri32: NewOpc = X86::TEST64rr; break;
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP32ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP32ri: NewOpc = X86::TEST32rr; break;
|
2010-05-18 21:54:15 +00:00
|
|
|
case X86::CMP16ri8:
|
2008-01-07 01:35:02 +00:00
|
|
|
case X86::CMP16ri: NewOpc = X86::TEST16rr; break;
|
|
|
|
case X86::CMP8ri: NewOpc = X86::TEST8rr; break;
|
|
|
|
}
|
2008-01-11 18:10:50 +00:00
|
|
|
DataMI->setDesc(get(NewOpc));
|
2008-01-07 01:35:02 +00:00
|
|
|
MO1.ChangeToRegister(MO0.getReg(), false);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
NewMIs.push_back(DataMI);
|
|
|
|
|
|
|
|
// Emit the store instruction.
|
|
|
|
if (UnfoldStore) {
|
2012-05-07 22:10:26 +00:00
|
|
|
const TargetRegisterClass *DstRC = getRegClass(MCID, 0, &RI, MF);
|
2009-10-09 18:10:05 +00:00
|
|
|
std::pair<MachineInstr::mmo_iterator,
|
|
|
|
MachineInstr::mmo_iterator> MMOs =
|
|
|
|
MF.extractStoreMemRefs(MI->memoperands_begin(),
|
|
|
|
MI->memoperands_end());
|
|
|
|
storeRegToAddr(MF, Reg, true, AddrOps, DstRC, MMOs.first, MMOs.second, NewMIs);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
X86InstrInfo::unfoldMemoryOperand(SelectionDAG &DAG, SDNode *N,
|
2009-02-11 21:51:19 +00:00
|
|
|
SmallVectorImpl<SDNode*> &NewNodes) const {
|
2008-07-17 19:10:17 +00:00
|
|
|
if (!N->isMachineOpcode())
|
2008-01-07 01:35:02 +00:00
|
|
|
return false;
|
|
|
|
|
2010-10-07 23:08:41 +00:00
|
|
|
DenseMap<unsigned, std::pair<unsigned,unsigned> >::const_iterator I =
|
|
|
|
MemOp2RegOpTable.find(N->getMachineOpcode());
|
2008-01-07 01:35:02 +00:00
|
|
|
if (I == MemOp2RegOpTable.end())
|
|
|
|
return false;
|
|
|
|
unsigned Opc = I->second.first;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
unsigned Index = I->second.second & TB_INDEX_MASK;
|
|
|
|
bool FoldedLoad = I->second.second & TB_FOLDED_LOAD;
|
|
|
|
bool FoldedStore = I->second.second & TB_FOLDED_STORE;
|
2011-06-28 19:10:37 +00:00
|
|
|
const MCInstrDesc &MCID = get(Opc);
|
2012-05-07 22:10:26 +00:00
|
|
|
MachineFunction &MF = DAG.getMachineFunction();
|
|
|
|
const TargetRegisterClass *RC = getRegClass(MCID, Index, &RI, MF);
|
2011-06-28 19:10:37 +00:00
|
|
|
unsigned NumDefs = MCID.NumDefs;
|
2008-07-27 21:46:04 +00:00
|
|
|
std::vector<SDValue> AddrOps;
|
|
|
|
std::vector<SDValue> BeforeOps;
|
|
|
|
std::vector<SDValue> AfterOps;
|
2013-05-25 02:42:55 +00:00
|
|
|
SDLoc dl(N);
|
2008-01-07 01:35:02 +00:00
|
|
|
unsigned NumOps = N->getNumOperands();
|
2009-09-25 20:36:54 +00:00
|
|
|
for (unsigned i = 0; i != NumOps-1; ++i) {
|
2008-07-27 21:46:04 +00:00
|
|
|
SDValue Op = N->getOperand(i);
|
2010-07-08 22:41:28 +00:00
|
|
|
if (i >= Index-NumDefs && i < Index-NumDefs + X86::AddrNumOperands)
|
2008-01-07 01:35:02 +00:00
|
|
|
AddrOps.push_back(Op);
|
2009-03-04 19:23:38 +00:00
|
|
|
else if (i < Index-NumDefs)
|
2008-01-07 01:35:02 +00:00
|
|
|
BeforeOps.push_back(Op);
|
2009-03-04 19:23:38 +00:00
|
|
|
else if (i > Index-NumDefs)
|
2008-01-07 01:35:02 +00:00
|
|
|
AfterOps.push_back(Op);
|
|
|
|
}
|
2008-07-27 21:46:04 +00:00
|
|
|
SDValue Chain = N->getOperand(NumOps-1);
|
2008-01-07 01:35:02 +00:00
|
|
|
AddrOps.push_back(Chain);
|
|
|
|
|
|
|
|
// Emit the load instruction.
|
|
|
|
SDNode *Load = 0;
|
|
|
|
if (FoldedLoad) {
|
2009-08-10 22:56:29 +00:00
|
|
|
EVT VT = *RC->vt_begin();
|
2009-11-16 21:56:03 +00:00
|
|
|
std::pair<MachineInstr::mmo_iterator,
|
|
|
|
MachineInstr::mmo_iterator> MMOs =
|
|
|
|
MF.extractLoadMemRefs(cast<MachineSDNode>(N)->memoperands_begin(),
|
|
|
|
cast<MachineSDNode>(N)->memoperands_end());
|
2010-07-02 20:36:18 +00:00
|
|
|
if (!(*MMOs.first) &&
|
|
|
|
RC == &X86::VR128RegClass &&
|
|
|
|
!TM.getSubtarget<X86Subtarget>().isUnalignedMemAccessFast())
|
|
|
|
// Do not introduce a slow unaligned load.
|
|
|
|
return false;
|
2011-09-14 02:36:58 +00:00
|
|
|
unsigned Alignment = RC->getSize() == 32 ? 32 : 16;
|
|
|
|
bool isAligned = (*MMOs.first) &&
|
|
|
|
(*MMOs.first)->getAlignment() >= Alignment;
|
2009-09-25 18:54:59 +00:00
|
|
|
Load = DAG.getMachineNode(getLoadRegOpcode(0, RC, isAligned, TM), dl,
|
2013-04-19 22:22:57 +00:00
|
|
|
VT, MVT::Other, AddrOps);
|
2008-01-07 01:35:02 +00:00
|
|
|
NewNodes.push_back(Load);
|
2009-10-09 18:10:05 +00:00
|
|
|
|
|
|
|
// Preserve memory reference information.
|
|
|
|
cast<MachineSDNode>(Load)->setMemRefs(MMOs.first, MMOs.second);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
// Emit the data processing instruction.
|
2009-08-10 22:56:29 +00:00
|
|
|
std::vector<EVT> VTs;
|
2008-01-07 01:35:02 +00:00
|
|
|
const TargetRegisterClass *DstRC = 0;
|
2011-06-28 19:10:37 +00:00
|
|
|
if (MCID.getNumDefs() > 0) {
|
2012-05-07 22:10:26 +00:00
|
|
|
DstRC = getRegClass(MCID, 0, &RI, MF);
|
2008-01-07 01:35:02 +00:00
|
|
|
VTs.push_back(*DstRC->vt_begin());
|
|
|
|
}
|
|
|
|
for (unsigned i = 0, e = N->getNumValues(); i != e; ++i) {
|
2009-08-10 22:56:29 +00:00
|
|
|
EVT VT = N->getValueType(i);
|
2011-06-28 19:10:37 +00:00
|
|
|
if (VT != MVT::Other && i >= (unsigned)MCID.getNumDefs())
|
2008-01-07 01:35:02 +00:00
|
|
|
VTs.push_back(VT);
|
|
|
|
}
|
|
|
|
if (Load)
|
2008-07-27 21:46:04 +00:00
|
|
|
BeforeOps.push_back(SDValue(Load, 0));
|
2008-01-07 01:35:02 +00:00
|
|
|
std::copy(AfterOps.begin(), AfterOps.end(), std::back_inserter(BeforeOps));
|
2013-04-19 22:22:57 +00:00
|
|
|
SDNode *NewNode= DAG.getMachineNode(Opc, dl, VTs, BeforeOps);
|
2008-01-07 01:35:02 +00:00
|
|
|
NewNodes.push_back(NewNode);
|
|
|
|
|
|
|
|
// Emit the store instruction.
|
|
|
|
if (FoldedStore) {
|
|
|
|
AddrOps.pop_back();
|
2008-07-27 21:46:04 +00:00
|
|
|
AddrOps.push_back(SDValue(NewNode, 0));
|
2008-01-07 01:35:02 +00:00
|
|
|
AddrOps.push_back(Chain);
|
2009-11-16 21:56:03 +00:00
|
|
|
std::pair<MachineInstr::mmo_iterator,
|
|
|
|
MachineInstr::mmo_iterator> MMOs =
|
|
|
|
MF.extractStoreMemRefs(cast<MachineSDNode>(N)->memoperands_begin(),
|
|
|
|
cast<MachineSDNode>(N)->memoperands_end());
|
2010-07-02 20:36:18 +00:00
|
|
|
if (!(*MMOs.first) &&
|
|
|
|
RC == &X86::VR128RegClass &&
|
|
|
|
!TM.getSubtarget<X86Subtarget>().isUnalignedMemAccessFast())
|
|
|
|
// Do not introduce a slow unaligned store.
|
|
|
|
return false;
|
2011-09-14 02:36:58 +00:00
|
|
|
unsigned Alignment = RC->getSize() == 32 ? 32 : 16;
|
|
|
|
bool isAligned = (*MMOs.first) &&
|
|
|
|
(*MMOs.first)->getAlignment() >= Alignment;
|
2009-09-25 18:54:59 +00:00
|
|
|
SDNode *Store = DAG.getMachineNode(getStoreRegOpcode(0, DstRC,
|
|
|
|
isAligned, TM),
|
2013-04-19 22:22:57 +00:00
|
|
|
dl, MVT::Other, AddrOps);
|
2008-01-07 01:35:02 +00:00
|
|
|
NewNodes.push_back(Store);
|
2009-10-09 18:10:05 +00:00
|
|
|
|
|
|
|
// Preserve memory reference information.
|
|
|
|
cast<MachineSDNode>(Load)->setMemRefs(MMOs.first, MMOs.second);
|
2008-01-07 01:35:02 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
unsigned X86InstrInfo::getOpcodeAfterMemoryUnfold(unsigned Opc,
|
2009-10-30 22:18:41 +00:00
|
|
|
bool UnfoldLoad, bool UnfoldStore,
|
|
|
|
unsigned *LoadRegIndex) const {
|
2010-10-07 23:08:41 +00:00
|
|
|
DenseMap<unsigned, std::pair<unsigned,unsigned> >::const_iterator I =
|
|
|
|
MemOp2RegOpTable.find(Opc);
|
2008-01-07 01:35:02 +00:00
|
|
|
if (I == MemOp2RegOpTable.end())
|
|
|
|
return 0;
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
bool FoldedLoad = I->second.second & TB_FOLDED_LOAD;
|
|
|
|
bool FoldedStore = I->second.second & TB_FOLDED_STORE;
|
2008-01-07 01:35:02 +00:00
|
|
|
if (UnfoldLoad && !FoldedLoad)
|
|
|
|
return 0;
|
|
|
|
if (UnfoldStore && !FoldedStore)
|
|
|
|
return 0;
|
2009-10-30 22:18:41 +00:00
|
|
|
if (LoadRegIndex)
|
* Combines Alignment, AuxInfo, and TB_NOT_REVERSABLE flag into a
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@139311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-08 18:35:57 +00:00
|
|
|
*LoadRegIndex = I->second.second & TB_INDEX_MASK;
|
2008-01-07 01:35:02 +00:00
|
|
|
return I->second.first;
|
|
|
|
}
|
|
|
|
|
2010-01-22 03:34:51 +00:00
|
|
|
bool
|
|
|
|
X86InstrInfo::areLoadsFromSameBasePtr(SDNode *Load1, SDNode *Load2,
|
|
|
|
int64_t &Offset1, int64_t &Offset2) const {
|
|
|
|
if (!Load1->isMachineOpcode() || !Load2->isMachineOpcode())
|
|
|
|
return false;
|
|
|
|
unsigned Opc1 = Load1->getMachineOpcode();
|
|
|
|
unsigned Opc2 = Load2->getMachineOpcode();
|
|
|
|
switch (Opc1) {
|
|
|
|
default: return false;
|
|
|
|
case X86::MOV8rm:
|
|
|
|
case X86::MOV16rm:
|
|
|
|
case X86::MOV32rm:
|
|
|
|
case X86::MOV64rm:
|
|
|
|
case X86::LD_Fp32m:
|
|
|
|
case X86::LD_Fp64m:
|
|
|
|
case X86::LD_Fp80m:
|
|
|
|
case X86::MOVSSrm:
|
|
|
|
case X86::MOVSDrm:
|
|
|
|
case X86::MMX_MOVD64rm:
|
|
|
|
case X86::MMX_MOVQ64rm:
|
|
|
|
case X86::FsMOVAPSrm:
|
|
|
|
case X86::FsMOVAPDrm:
|
|
|
|
case X86::MOVAPSrm:
|
|
|
|
case X86::MOVUPSrm:
|
|
|
|
case X86::MOVAPDrm:
|
|
|
|
case X86::MOVDQArm:
|
|
|
|
case X86::MOVDQUrm:
|
2011-09-15 22:15:52 +00:00
|
|
|
// AVX load instructions
|
|
|
|
case X86::VMOVSSrm:
|
|
|
|
case X86::VMOVSDrm:
|
|
|
|
case X86::FsVMOVAPSrm:
|
|
|
|
case X86::FsVMOVAPDrm:
|
2011-09-14 02:36:58 +00:00
|
|
|
case X86::VMOVAPSrm:
|
|
|
|
case X86::VMOVUPSrm:
|
|
|
|
case X86::VMOVAPDrm:
|
|
|
|
case X86::VMOVDQArm:
|
|
|
|
case X86::VMOVDQUrm:
|
2011-07-14 18:50:58 +00:00
|
|
|
case X86::VMOVAPSYrm:
|
|
|
|
case X86::VMOVUPSYrm:
|
|
|
|
case X86::VMOVAPDYrm:
|
|
|
|
case X86::VMOVDQAYrm:
|
|
|
|
case X86::VMOVDQUYrm:
|
2010-01-22 03:34:51 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
switch (Opc2) {
|
|
|
|
default: return false;
|
|
|
|
case X86::MOV8rm:
|
|
|
|
case X86::MOV16rm:
|
|
|
|
case X86::MOV32rm:
|
|
|
|
case X86::MOV64rm:
|
|
|
|
case X86::LD_Fp32m:
|
|
|
|
case X86::LD_Fp64m:
|
|
|
|
case X86::LD_Fp80m:
|
|
|
|
case X86::MOVSSrm:
|
|
|
|
case X86::MOVSDrm:
|
|
|
|
case X86::MMX_MOVD64rm:
|
|
|
|
case X86::MMX_MOVQ64rm:
|
|
|
|
case X86::FsMOVAPSrm:
|
|
|
|
case X86::FsMOVAPDrm:
|
|
|
|
case X86::MOVAPSrm:
|
|
|
|
case X86::MOVUPSrm:
|
|
|
|
case X86::MOVAPDrm:
|
|
|
|
case X86::MOVDQArm:
|
|
|
|
case X86::MOVDQUrm:
|
2011-09-15 22:15:52 +00:00
|
|
|
// AVX load instructions
|
|
|
|
case X86::VMOVSSrm:
|
|
|
|
case X86::VMOVSDrm:
|
|
|
|
case X86::FsVMOVAPSrm:
|
|
|
|
case X86::FsVMOVAPDrm:
|
2011-09-14 02:36:58 +00:00
|
|
|
case X86::VMOVAPSrm:
|
|
|
|
case X86::VMOVUPSrm:
|
|
|
|
case X86::VMOVAPDrm:
|
|
|
|
case X86::VMOVDQArm:
|
|
|
|
case X86::VMOVDQUrm:
|
2011-07-14 18:50:58 +00:00
|
|
|
case X86::VMOVAPSYrm:
|
|
|
|
case X86::VMOVUPSYrm:
|
|
|
|
case X86::VMOVAPDYrm:
|
|
|
|
case X86::VMOVDQAYrm:
|
|
|
|
case X86::VMOVDQUYrm:
|
2010-01-22 03:34:51 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Check if chain operands and base addresses match.
|
|
|
|
if (Load1->getOperand(0) != Load2->getOperand(0) ||
|
|
|
|
Load1->getOperand(5) != Load2->getOperand(5))
|
|
|
|
return false;
|
|
|
|
// Segment operands should match as well.
|
|
|
|
if (Load1->getOperand(4) != Load2->getOperand(4))
|
|
|
|
return false;
|
|
|
|
// Scale should be 1, Index should be Reg0.
|
|
|
|
if (Load1->getOperand(1) == Load2->getOperand(1) &&
|
|
|
|
Load1->getOperand(2) == Load2->getOperand(2)) {
|
|
|
|
if (cast<ConstantSDNode>(Load1->getOperand(1))->getZExtValue() != 1)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
// Now let's examine the displacements.
|
|
|
|
if (isa<ConstantSDNode>(Load1->getOperand(3)) &&
|
|
|
|
isa<ConstantSDNode>(Load2->getOperand(3))) {
|
|
|
|
Offset1 = cast<ConstantSDNode>(Load1->getOperand(3))->getSExtValue();
|
|
|
|
Offset2 = cast<ConstantSDNode>(Load2->getOperand(3))->getSExtValue();
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool X86InstrInfo::shouldScheduleLoadsNear(SDNode *Load1, SDNode *Load2,
|
|
|
|
int64_t Offset1, int64_t Offset2,
|
|
|
|
unsigned NumLoads) const {
|
|
|
|
assert(Offset2 > Offset1);
|
|
|
|
if ((Offset2 - Offset1) / 8 > 64)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
unsigned Opc1 = Load1->getMachineOpcode();
|
|
|
|
unsigned Opc2 = Load2->getMachineOpcode();
|
|
|
|
if (Opc1 != Opc2)
|
|
|
|
return false; // FIXME: overly conservative?
|
|
|
|
|
|
|
|
switch (Opc1) {
|
|
|
|
default: break;
|
|
|
|
case X86::LD_Fp32m:
|
|
|
|
case X86::LD_Fp64m:
|
|
|
|
case X86::LD_Fp80m:
|
|
|
|
case X86::MMX_MOVD64rm:
|
|
|
|
case X86::MMX_MOVQ64rm:
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
EVT VT = Load1->getValueType(0);
|
|
|
|
switch (VT.getSimpleVT().SimpleTy) {
|
2010-06-22 22:16:17 +00:00
|
|
|
default:
|
2010-01-22 03:34:51 +00:00
|
|
|
// XMM registers. In 64-bit mode we can be a bit more aggressive since we
|
|
|
|
// have 16 of them to play with.
|
|
|
|
if (TM.getSubtargetImpl()->is64Bit()) {
|
|
|
|
if (NumLoads >= 3)
|
|
|
|
return false;
|
2010-06-22 22:16:17 +00:00
|
|
|
} else if (NumLoads) {
|
2010-01-22 03:34:51 +00:00
|
|
|
return false;
|
2010-06-22 22:16:17 +00:00
|
|
|
}
|
2010-01-22 03:34:51 +00:00
|
|
|
break;
|
|
|
|
case MVT::i8:
|
|
|
|
case MVT::i16:
|
|
|
|
case MVT::i32:
|
|
|
|
case MVT::i64:
|
2010-01-22 23:49:11 +00:00
|
|
|
case MVT::f32:
|
|
|
|
case MVT::f64:
|
2010-01-22 03:34:51 +00:00
|
|
|
if (NumLoads)
|
|
|
|
return false;
|
2010-06-22 22:16:17 +00:00
|
|
|
break;
|
2010-01-22 03:34:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2013-06-23 09:00:28 +00:00
|
|
|
bool X86InstrInfo::shouldScheduleAdjacent(MachineInstr* First,
|
|
|
|
MachineInstr *Second) const {
|
|
|
|
// Check if this processor supports macro-fusion. Since this is a minor
|
|
|
|
// heuristic, we haven't specifically reserved a feature. hasAVX is a decent
|
|
|
|
// proxy for SandyBridge+.
|
|
|
|
if (!TM.getSubtarget<X86Subtarget>().hasAVX())
|
|
|
|
return false;
|
|
|
|
|
|
|
|
enum {
|
|
|
|
FuseTest,
|
|
|
|
FuseCmp,
|
|
|
|
FuseInc
|
|
|
|
} FuseKind;
|
|
|
|
|
|
|
|
switch(Second->getOpcode()) {
|
|
|
|
default:
|
|
|
|
return false;
|
|
|
|
case X86::JE_4:
|
|
|
|
case X86::JNE_4:
|
|
|
|
case X86::JL_4:
|
|
|
|
case X86::JLE_4:
|
|
|
|
case X86::JG_4:
|
|
|
|
case X86::JGE_4:
|
|
|
|
FuseKind = FuseInc;
|
|
|
|
break;
|
|
|
|
case X86::JB_4:
|
|
|
|
case X86::JBE_4:
|
|
|
|
case X86::JA_4:
|
|
|
|
case X86::JAE_4:
|
|
|
|
FuseKind = FuseCmp;
|
|
|
|
break;
|
|
|
|
case X86::JS_4:
|
|
|
|
case X86::JNS_4:
|
|
|
|
case X86::JP_4:
|
|
|
|
case X86::JNP_4:
|
|
|
|
case X86::JO_4:
|
|
|
|
case X86::JNO_4:
|
|
|
|
FuseKind = FuseTest;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
switch (First->getOpcode()) {
|
|
|
|
default:
|
|
|
|
return false;
|
|
|
|
case X86::TEST8rr:
|
|
|
|
case X86::TEST16rr:
|
|
|
|
case X86::TEST32rr:
|
|
|
|
case X86::TEST64rr:
|
|
|
|
case X86::TEST8ri:
|
|
|
|
case X86::TEST16ri:
|
|
|
|
case X86::TEST32ri:
|
|
|
|
case X86::TEST32i32:
|
|
|
|
case X86::TEST64i32:
|
|
|
|
case X86::TEST64ri32:
|
|
|
|
case X86::TEST8rm:
|
|
|
|
case X86::TEST16rm:
|
|
|
|
case X86::TEST32rm:
|
|
|
|
case X86::TEST64rm:
|
|
|
|
case X86::AND16i16:
|
|
|
|
case X86::AND16ri:
|
|
|
|
case X86::AND16ri8:
|
|
|
|
case X86::AND16rm:
|
|
|
|
case X86::AND16rr:
|
|
|
|
case X86::AND32i32:
|
|
|
|
case X86::AND32ri:
|
|
|
|
case X86::AND32ri8:
|
|
|
|
case X86::AND32rm:
|
|
|
|
case X86::AND32rr:
|
|
|
|
case X86::AND64i32:
|
|
|
|
case X86::AND64ri32:
|
|
|
|
case X86::AND64ri8:
|
|
|
|
case X86::AND64rm:
|
|
|
|
case X86::AND64rr:
|
|
|
|
case X86::AND8i8:
|
|
|
|
case X86::AND8ri:
|
|
|
|
case X86::AND8rm:
|
|
|
|
case X86::AND8rr:
|
|
|
|
return true;
|
|
|
|
case X86::CMP16i16:
|
|
|
|
case X86::CMP16ri:
|
|
|
|
case X86::CMP16ri8:
|
|
|
|
case X86::CMP16rm:
|
|
|
|
case X86::CMP16rr:
|
|
|
|
case X86::CMP32i32:
|
|
|
|
case X86::CMP32ri:
|
|
|
|
case X86::CMP32ri8:
|
|
|
|
case X86::CMP32rm:
|
|
|
|
case X86::CMP32rr:
|
|
|
|
case X86::CMP64i32:
|
|
|
|
case X86::CMP64ri32:
|
|
|
|
case X86::CMP64ri8:
|
|
|
|
case X86::CMP64rm:
|
|
|
|
case X86::CMP64rr:
|
|
|
|
case X86::CMP8i8:
|
|
|
|
case X86::CMP8ri:
|
|
|
|
case X86::CMP8rm:
|
|
|
|
case X86::CMP8rr:
|
|
|
|
case X86::ADD16i16:
|
|
|
|
case X86::ADD16ri:
|
|
|
|
case X86::ADD16ri8:
|
|
|
|
case X86::ADD16ri8_DB:
|
|
|
|
case X86::ADD16ri_DB:
|
|
|
|
case X86::ADD16rm:
|
|
|
|
case X86::ADD16rr:
|
|
|
|
case X86::ADD16rr_DB:
|
|
|
|
case X86::ADD32i32:
|
|
|
|
case X86::ADD32ri:
|
|
|
|
case X86::ADD32ri8:
|
|
|
|
case X86::ADD32ri8_DB:
|
|
|
|
case X86::ADD32ri_DB:
|
|
|
|
case X86::ADD32rm:
|
|
|
|
case X86::ADD32rr:
|
|
|
|
case X86::ADD32rr_DB:
|
|
|
|
case X86::ADD64i32:
|
|
|
|
case X86::ADD64ri32:
|
|
|
|
case X86::ADD64ri32_DB:
|
|
|
|
case X86::ADD64ri8:
|
|
|
|
case X86::ADD64ri8_DB:
|
|
|
|
case X86::ADD64rm:
|
|
|
|
case X86::ADD64rr:
|
|
|
|
case X86::ADD64rr_DB:
|
|
|
|
case X86::ADD8i8:
|
|
|
|
case X86::ADD8mi:
|
|
|
|
case X86::ADD8mr:
|
|
|
|
case X86::ADD8ri:
|
|
|
|
case X86::ADD8rm:
|
|
|
|
case X86::ADD8rr:
|
|
|
|
case X86::SUB16i16:
|
|
|
|
case X86::SUB16ri:
|
|
|
|
case X86::SUB16ri8:
|
|
|
|
case X86::SUB16rm:
|
|
|
|
case X86::SUB16rr:
|
|
|
|
case X86::SUB32i32:
|
|
|
|
case X86::SUB32ri:
|
|
|
|
case X86::SUB32ri8:
|
|
|
|
case X86::SUB32rm:
|
|
|
|
case X86::SUB32rr:
|
|
|
|
case X86::SUB64i32:
|
|
|
|
case X86::SUB64ri32:
|
|
|
|
case X86::SUB64ri8:
|
|
|
|
case X86::SUB64rm:
|
|
|
|
case X86::SUB64rr:
|
|
|
|
case X86::SUB8i8:
|
|
|
|
case X86::SUB8ri:
|
|
|
|
case X86::SUB8rm:
|
|
|
|
case X86::SUB8rr:
|
|
|
|
return FuseKind == FuseCmp || FuseKind == FuseInc;
|
|
|
|
case X86::INC16r:
|
|
|
|
case X86::INC32r:
|
|
|
|
case X86::INC64_16r:
|
|
|
|
case X86::INC64_32r:
|
|
|
|
case X86::INC64r:
|
|
|
|
case X86::INC8r:
|
|
|
|
case X86::DEC16r:
|
|
|
|
case X86::DEC32r:
|
|
|
|
case X86::DEC64_16r:
|
|
|
|
case X86::DEC64_32r:
|
|
|
|
case X86::DEC64r:
|
|
|
|
case X86::DEC8r:
|
|
|
|
return FuseKind == FuseInc;
|
|
|
|
}
|
|
|
|
}
|
2010-01-22 03:34:51 +00:00
|
|
|
|
2006-10-20 17:42:20 +00:00
|
|
|
bool X86InstrInfo::
|
2008-08-14 22:49:33 +00:00
|
|
|
ReverseBranchCondition(SmallVectorImpl<MachineOperand> &Cond) const {
|
2006-10-21 05:52:40 +00:00
|
|
|
assert(Cond.size() == 1 && "Invalid X86 branch condition!");
|
2008-08-29 23:21:31 +00:00
|
|
|
X86::CondCode CC = static_cast<X86::CondCode>(Cond[0].getImm());
|
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873 91177308-0d34-0410-b5e6-96231b3b80d8
2008-10-21 03:29:32 +00:00
|
|
|
if (CC == X86::COND_NE_OR_P || CC == X86::COND_NP_OR_E)
|
|
|
|
return true;
|
2008-08-29 23:21:31 +00:00
|
|
|
Cond[0].setImm(GetOppositeBranchCondition(CC));
|
2006-10-21 05:52:40 +00:00
|
|
|
return false;
|
2006-10-20 17:42:20 +00:00
|
|
|
}
|
|
|
|
|
2008-10-27 07:14:50 +00:00
|
|
|
bool X86InstrInfo::
|
2009-02-06 17:17:30 +00:00
|
|
|
isSafeToMoveRegClassDefs(const TargetRegisterClass *RC) const {
|
|
|
|
// FIXME: Return false for x87 stack register classes for now. We can't
|
2008-10-27 07:14:50 +00:00
|
|
|
// allow any loads of these registers before FpGet_ST0_80.
|
2009-02-06 17:17:30 +00:00
|
|
|
return !(RC == &X86::CCRRegClass || RC == &X86::RFP32RegClass ||
|
|
|
|
RC == &X86::RFP64RegClass || RC == &X86::RFP80RegClass);
|
2008-10-27 07:14:50 +00:00
|
|
|
}
|
|
|
|
|
2008-09-30 00:58:23 +00:00
|
|
|
/// getGlobalBaseReg - Return a virtual register initialized with the
|
|
|
|
/// the global base register value. Output instructions required to
|
|
|
|
/// initialize the register in the function entry block, if necessary.
|
2008-09-23 18:22:58 +00:00
|
|
|
///
|
2010-07-10 09:00:22 +00:00
|
|
|
/// TODO: Eliminate this and move the code to X86MachineFunctionInfo.
|
|
|
|
///
|
2008-09-30 00:58:23 +00:00
|
|
|
unsigned X86InstrInfo::getGlobalBaseReg(MachineFunction *MF) const {
|
|
|
|
assert(!TM.getSubtarget<X86Subtarget>().is64Bit() &&
|
|
|
|
"X86-64 PIC uses RIP relative addressing");
|
|
|
|
|
|
|
|
X86MachineFunctionInfo *X86FI = MF->getInfo<X86MachineFunctionInfo>();
|
|
|
|
unsigned GlobalBaseReg = X86FI->getGlobalBaseReg();
|
|
|
|
if (GlobalBaseReg != 0)
|
|
|
|
return GlobalBaseReg;
|
|
|
|
|
2010-07-10 09:00:22 +00:00
|
|
|
// Create the register. The code to initialize it is inserted
|
|
|
|
// later, by the CGBR pass (below).
|
2008-09-23 18:22:58 +00:00
|
|
|
MachineRegisterInfo &RegInfo = MF->getRegInfo();
|
2012-05-20 18:43:00 +00:00
|
|
|
GlobalBaseReg = RegInfo.createVirtualRegister(&X86::GR32_NOSPRegClass);
|
2008-09-30 00:58:23 +00:00
|
|
|
X86FI->setGlobalBaseReg(GlobalBaseReg);
|
|
|
|
return GlobalBaseReg;
|
2008-09-23 18:22:58 +00:00
|
|
|
}
|
2010-03-25 17:25:00 +00:00
|
|
|
|
2010-03-29 23:24:21 +00:00
|
|
|
// These are the replaceable SSE instructions. Some of these have Int variants
|
|
|
|
// that we don't include here. We don't want to replace instructions selected
|
|
|
|
// by intrinsics.
|
2012-03-09 07:45:21 +00:00
|
|
|
static const uint16_t ReplaceableInstrs[][3] = {
|
2010-08-12 02:08:52 +00:00
|
|
|
//PackedSingle PackedDouble PackedInt
|
2010-03-30 22:46:53 +00:00
|
|
|
{ X86::MOVAPSmr, X86::MOVAPDmr, X86::MOVDQAmr },
|
|
|
|
{ X86::MOVAPSrm, X86::MOVAPDrm, X86::MOVDQArm },
|
|
|
|
{ X86::MOVAPSrr, X86::MOVAPDrr, X86::MOVDQArr },
|
|
|
|
{ X86::MOVUPSmr, X86::MOVUPDmr, X86::MOVDQUmr },
|
|
|
|
{ X86::MOVUPSrm, X86::MOVUPDrm, X86::MOVDQUrm },
|
|
|
|
{ X86::MOVNTPSmr, X86::MOVNTPDmr, X86::MOVNTDQmr },
|
|
|
|
{ X86::ANDNPSrm, X86::ANDNPDrm, X86::PANDNrm },
|
|
|
|
{ X86::ANDNPSrr, X86::ANDNPDrr, X86::PANDNrr },
|
|
|
|
{ X86::ANDPSrm, X86::ANDPDrm, X86::PANDrm },
|
|
|
|
{ X86::ANDPSrr, X86::ANDPDrr, X86::PANDrr },
|
|
|
|
{ X86::ORPSrm, X86::ORPDrm, X86::PORrm },
|
|
|
|
{ X86::ORPSrr, X86::ORPDrr, X86::PORrr },
|
|
|
|
{ X86::XORPSrm, X86::XORPDrm, X86::PXORrm },
|
|
|
|
{ X86::XORPSrr, X86::XORPDrr, X86::PXORrr },
|
2010-08-12 20:20:53 +00:00
|
|
|
// AVX 128-bit support
|
|
|
|
{ X86::VMOVAPSmr, X86::VMOVAPDmr, X86::VMOVDQAmr },
|
|
|
|
{ X86::VMOVAPSrm, X86::VMOVAPDrm, X86::VMOVDQArm },
|
|
|
|
{ X86::VMOVAPSrr, X86::VMOVAPDrr, X86::VMOVDQArr },
|
|
|
|
{ X86::VMOVUPSmr, X86::VMOVUPDmr, X86::VMOVDQUmr },
|
|
|
|
{ X86::VMOVUPSrm, X86::VMOVUPDrm, X86::VMOVDQUrm },
|
|
|
|
{ X86::VMOVNTPSmr, X86::VMOVNTPDmr, X86::VMOVNTDQmr },
|
|
|
|
{ X86::VANDNPSrm, X86::VANDNPDrm, X86::VPANDNrm },
|
|
|
|
{ X86::VANDNPSrr, X86::VANDNPDrr, X86::VPANDNrr },
|
|
|
|
{ X86::VANDPSrm, X86::VANDPDrm, X86::VPANDrm },
|
|
|
|
{ X86::VANDPSrr, X86::VANDPDrr, X86::VPANDrr },
|
|
|
|
{ X86::VORPSrm, X86::VORPDrm, X86::VPORrm },
|
|
|
|
{ X86::VORPSrr, X86::VORPDrr, X86::VPORrr },
|
|
|
|
{ X86::VXORPSrm, X86::VXORPDrm, X86::VPXORrm },
|
|
|
|
{ X86::VXORPSrr, X86::VXORPDrr, X86::VPXORrr },
|
2011-07-14 18:50:58 +00:00
|
|
|
// AVX 256-bit support
|
|
|
|
{ X86::VMOVAPSYmr, X86::VMOVAPDYmr, X86::VMOVDQAYmr },
|
|
|
|
{ X86::VMOVAPSYrm, X86::VMOVAPDYrm, X86::VMOVDQAYrm },
|
|
|
|
{ X86::VMOVAPSYrr, X86::VMOVAPDYrr, X86::VMOVDQAYrr },
|
|
|
|
{ X86::VMOVUPSYmr, X86::VMOVUPDYmr, X86::VMOVDQUYmr },
|
|
|
|
{ X86::VMOVUPSYrm, X86::VMOVUPDYrm, X86::VMOVDQUYrm },
|
2011-11-15 05:55:35 +00:00
|
|
|
{ X86::VMOVNTPSYmr, X86::VMOVNTPDYmr, X86::VMOVNTDQYmr }
|
|
|
|
};
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const uint16_t ReplaceableInstrsAVX2[][3] = {
|
2011-11-15 05:55:35 +00:00
|
|
|
//PackedSingle PackedDouble PackedInt
|
2011-11-09 09:37:21 +00:00
|
|
|
{ X86::VANDNPSYrm, X86::VANDNPDYrm, X86::VPANDNYrm },
|
|
|
|
{ X86::VANDNPSYrr, X86::VANDNPDYrr, X86::VPANDNYrr },
|
|
|
|
{ X86::VANDPSYrm, X86::VANDPDYrm, X86::VPANDYrm },
|
|
|
|
{ X86::VANDPSYrr, X86::VANDPDYrr, X86::VPANDYrr },
|
|
|
|
{ X86::VORPSYrm, X86::VORPDYrm, X86::VPORYrm },
|
|
|
|
{ X86::VORPSYrr, X86::VORPDYrr, X86::VPORYrr },
|
|
|
|
{ X86::VXORPSYrm, X86::VXORPDYrm, X86::VPXORYrm },
|
2011-11-29 05:37:58 +00:00
|
|
|
{ X86::VXORPSYrr, X86::VXORPDYrr, X86::VPXORYrr },
|
|
|
|
{ X86::VEXTRACTF128mr, X86::VEXTRACTF128mr, X86::VEXTRACTI128mr },
|
|
|
|
{ X86::VEXTRACTF128rr, X86::VEXTRACTF128rr, X86::VEXTRACTI128rr },
|
|
|
|
{ X86::VINSERTF128rm, X86::VINSERTF128rm, X86::VINSERTI128rm },
|
|
|
|
{ X86::VINSERTF128rr, X86::VINSERTF128rr, X86::VINSERTI128rr },
|
|
|
|
{ X86::VPERM2F128rm, X86::VPERM2F128rm, X86::VPERM2I128rm },
|
|
|
|
{ X86::VPERM2F128rr, X86::VPERM2F128rr, X86::VPERM2I128rr }
|
2010-03-29 23:24:21 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
// FIXME: Some shuffle and unpack instructions have equivalents in different
|
|
|
|
// domains, but they require a bit more work than just switching opcodes.
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const uint16_t *lookup(unsigned opcode, unsigned domain) {
|
2010-03-29 23:24:21 +00:00
|
|
|
for (unsigned i = 0, e = array_lengthof(ReplaceableInstrs); i != e; ++i)
|
|
|
|
if (ReplaceableInstrs[i][domain-1] == opcode)
|
|
|
|
return ReplaceableInstrs[i];
|
2011-11-15 06:39:01 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-03-09 07:45:21 +00:00
|
|
|
static const uint16_t *lookupAVX2(unsigned opcode, unsigned domain) {
|
2011-11-15 06:39:01 +00:00
|
|
|
for (unsigned i = 0, e = array_lengthof(ReplaceableInstrsAVX2); i != e; ++i)
|
|
|
|
if (ReplaceableInstrsAVX2[i][domain-1] == opcode)
|
|
|
|
return ReplaceableInstrsAVX2[i];
|
2010-03-29 23:24:21 +00:00
|
|
|
return 0;
|
|
|
|
}
|
2010-03-25 17:25:00 +00:00
|
|
|
|
2010-03-29 23:24:21 +00:00
|
|
|
std::pair<uint16_t, uint16_t>
|
2011-09-27 22:57:18 +00:00
|
|
|
X86InstrInfo::getExecutionDomain(const MachineInstr *MI) const {
|
2010-03-29 23:24:21 +00:00
|
|
|
uint16_t domain = (MI->getDesc().TSFlags >> X86II::SSEDomainShift) & 3;
|
2011-11-15 05:55:35 +00:00
|
|
|
bool hasAVX2 = TM.getSubtarget<X86Subtarget>().hasAVX2();
|
2011-11-15 06:39:01 +00:00
|
|
|
uint16_t validDomains = 0;
|
|
|
|
if (domain && lookup(MI->getOpcode(), domain))
|
|
|
|
validDomains = 0xe;
|
|
|
|
else if (domain && lookupAVX2(MI->getOpcode(), domain))
|
|
|
|
validDomains = hasAVX2 ? 0xe : 0x6;
|
|
|
|
return std::make_pair(domain, validDomains);
|
2010-03-29 23:24:21 +00:00
|
|
|
}
|
2010-03-25 17:25:00 +00:00
|
|
|
|
2011-09-27 22:57:18 +00:00
|
|
|
void X86InstrInfo::setExecutionDomain(MachineInstr *MI, unsigned Domain) const {
|
2010-03-29 23:24:21 +00:00
|
|
|
assert(Domain>0 && Domain<4 && "Invalid execution domain");
|
|
|
|
uint16_t dom = (MI->getDesc().TSFlags >> X86II::SSEDomainShift) & 3;
|
|
|
|
assert(dom && "Not an SSE instruction");
|
2012-03-09 07:45:21 +00:00
|
|
|
const uint16_t *table = lookup(MI->getOpcode(), dom);
|
2011-11-23 04:03:08 +00:00
|
|
|
if (!table) { // try the other table
|
|
|
|
assert((TM.getSubtarget<X86Subtarget>().hasAVX2() || Domain < 3) &&
|
|
|
|
"256-bit vector operations only available in AVX2");
|
2011-11-15 06:39:01 +00:00
|
|
|
table = lookupAVX2(MI->getOpcode(), dom);
|
2011-11-23 04:03:08 +00:00
|
|
|
}
|
2010-03-29 23:24:21 +00:00
|
|
|
assert(table && "Cannot change domain");
|
|
|
|
MI->setDesc(get(table[Domain-1]));
|
2010-03-25 17:25:00 +00:00
|
|
|
}
|
2010-04-26 23:37:21 +00:00
|
|
|
|
|
|
|
/// getNoopForMachoTarget - Return the noop instruction to use for a noop.
|
|
|
|
void X86InstrInfo::getNoopForMachoTarget(MCInst &NopInst) const {
|
|
|
|
NopInst.setOpcode(X86::NOOP);
|
|
|
|
}
|
2010-07-10 09:00:22 +00:00
|
|
|
|
2011-03-05 08:00:22 +00:00
|
|
|
bool X86InstrInfo::isHighLatencyDef(int opc) const {
|
|
|
|
switch (opc) {
|
2010-10-19 18:58:51 +00:00
|
|
|
default: return false;
|
|
|
|
case X86::DIVSDrm:
|
|
|
|
case X86::DIVSDrm_Int:
|
|
|
|
case X86::DIVSDrr:
|
|
|
|
case X86::DIVSDrr_Int:
|
|
|
|
case X86::DIVSSrm:
|
|
|
|
case X86::DIVSSrm_Int:
|
|
|
|
case X86::DIVSSrr:
|
|
|
|
case X86::DIVSSrr_Int:
|
|
|
|
case X86::SQRTPDm:
|
|
|
|
case X86::SQRTPDr:
|
|
|
|
case X86::SQRTPSm:
|
|
|
|
case X86::SQRTPSr:
|
|
|
|
case X86::SQRTSDm:
|
|
|
|
case X86::SQRTSDm_Int:
|
|
|
|
case X86::SQRTSDr:
|
|
|
|
case X86::SQRTSDr_Int:
|
|
|
|
case X86::SQRTSSm:
|
|
|
|
case X86::SQRTSSm_Int:
|
|
|
|
case X86::SQRTSSr:
|
|
|
|
case X86::SQRTSSr_Int:
|
2011-09-15 22:15:52 +00:00
|
|
|
// AVX instructions with high latency
|
|
|
|
case X86::VDIVSDrm:
|
|
|
|
case X86::VDIVSDrm_Int:
|
|
|
|
case X86::VDIVSDrr:
|
|
|
|
case X86::VDIVSDrr_Int:
|
|
|
|
case X86::VDIVSSrm:
|
|
|
|
case X86::VDIVSSrm_Int:
|
|
|
|
case X86::VDIVSSrr:
|
|
|
|
case X86::VDIVSSrr_Int:
|
|
|
|
case X86::VSQRTPDm:
|
|
|
|
case X86::VSQRTPDr:
|
|
|
|
case X86::VSQRTPSm:
|
|
|
|
case X86::VSQRTPSr:
|
|
|
|
case X86::VSQRTSDm:
|
|
|
|
case X86::VSQRTSDm_Int:
|
|
|
|
case X86::VSQRTSDr:
|
|
|
|
case X86::VSQRTSSm:
|
|
|
|
case X86::VSQRTSSm_Int:
|
|
|
|
case X86::VSQRTSSr:
|
2013-09-02 07:41:01 +00:00
|
|
|
case X86::VSQRTPDZrm:
|
|
|
|
case X86::VSQRTPDZrr:
|
|
|
|
case X86::VSQRTPSZrm:
|
|
|
|
case X86::VSQRTPSZrr:
|
|
|
|
case X86::VSQRTSDZm:
|
|
|
|
case X86::VSQRTSDZm_Int:
|
|
|
|
case X86::VSQRTSDZr:
|
|
|
|
case X86::VSQRTSSZm_Int:
|
|
|
|
case X86::VSQRTSSZr:
|
|
|
|
case X86::VSQRTSSZm:
|
|
|
|
case X86::VDIVSDZrm:
|
|
|
|
case X86::VDIVSDZrr:
|
|
|
|
case X86::VDIVSSZrm:
|
|
|
|
case X86::VDIVSSZrr:
|
2013-09-02 07:12:29 +00:00
|
|
|
|
|
|
|
case X86::VGATHERQPSZrm:
|
|
|
|
case X86::VGATHERQPDZrm:
|
|
|
|
case X86::VGATHERDPDZrm:
|
|
|
|
case X86::VGATHERDPSZrm:
|
|
|
|
case X86::VPGATHERQDZrm:
|
|
|
|
case X86::VPGATHERQQZrm:
|
|
|
|
case X86::VPGATHERDDZrm:
|
2013-09-02 07:41:01 +00:00
|
|
|
case X86::VPGATHERDQZrm:
|
|
|
|
case X86::VSCATTERQPDZmr:
|
|
|
|
case X86::VSCATTERQPSZmr:
|
|
|
|
case X86::VSCATTERDPDZmr:
|
|
|
|
case X86::VSCATTERDPSZmr:
|
|
|
|
case X86::VPSCATTERQDZmr:
|
|
|
|
case X86::VPSCATTERQQZmr:
|
|
|
|
case X86::VPSCATTERDDZmr:
|
|
|
|
case X86::VPSCATTERDQZmr:
|
2010-10-19 18:58:51 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-03-05 08:00:22 +00:00
|
|
|
bool X86InstrInfo::
|
|
|
|
hasHighOperandLatency(const InstrItineraryData *ItinData,
|
|
|
|
const MachineRegisterInfo *MRI,
|
|
|
|
const MachineInstr *DefMI, unsigned DefIdx,
|
|
|
|
const MachineInstr *UseMI, unsigned UseIdx) const {
|
|
|
|
return isHighLatencyDef(DefMI->getOpcode());
|
|
|
|
}
|
|
|
|
|
2010-07-10 09:00:22 +00:00
|
|
|
namespace {
|
|
|
|
/// CGBR - Create Global Base Reg pass. This initializes the PIC
|
|
|
|
/// global base register for x86-32.
|
|
|
|
struct CGBR : public MachineFunctionPass {
|
|
|
|
static char ID;
|
2010-08-06 18:33:48 +00:00
|
|
|
CGBR() : MachineFunctionPass(ID) {}
|
2010-07-10 09:00:22 +00:00
|
|
|
|
|
|
|
virtual bool runOnMachineFunction(MachineFunction &MF) {
|
|
|
|
const X86TargetMachine *TM =
|
|
|
|
static_cast<const X86TargetMachine *>(&MF.getTarget());
|
|
|
|
|
|
|
|
assert(!TM->getSubtarget<X86Subtarget>().is64Bit() &&
|
|
|
|
"X86-64 PIC uses RIP relative addressing");
|
|
|
|
|
|
|
|
// Only emit a global base reg in PIC mode.
|
|
|
|
if (TM->getRelocationModel() != Reloc::PIC_)
|
|
|
|
return false;
|
|
|
|
|
2010-09-17 20:24:24 +00:00
|
|
|
X86MachineFunctionInfo *X86FI = MF.getInfo<X86MachineFunctionInfo>();
|
|
|
|
unsigned GlobalBaseReg = X86FI->getGlobalBaseReg();
|
|
|
|
|
|
|
|
// If we didn't need a GlobalBaseReg, don't insert code.
|
|
|
|
if (GlobalBaseReg == 0)
|
|
|
|
return false;
|
|
|
|
|
2010-07-10 09:00:22 +00:00
|
|
|
// Insert the set of GlobalBaseReg into the first MBB of the function
|
|
|
|
MachineBasicBlock &FirstMBB = MF.front();
|
|
|
|
MachineBasicBlock::iterator MBBI = FirstMBB.begin();
|
|
|
|
DebugLoc DL = FirstMBB.findDebugLoc(MBBI);
|
|
|
|
MachineRegisterInfo &RegInfo = MF.getRegInfo();
|
|
|
|
const X86InstrInfo *TII = TM->getInstrInfo();
|
|
|
|
|
|
|
|
unsigned PC;
|
|
|
|
if (TM->getSubtarget<X86Subtarget>().isPICStyleGOT())
|
2012-04-20 06:31:50 +00:00
|
|
|
PC = RegInfo.createVirtualRegister(&X86::GR32RegClass);
|
2010-07-10 09:00:22 +00:00
|
|
|
else
|
2010-09-17 20:24:24 +00:00
|
|
|
PC = GlobalBaseReg;
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2010-07-10 09:00:22 +00:00
|
|
|
// Operand of MovePCtoStack is completely ignored by asm printer. It's
|
|
|
|
// only used in JIT code emission as displacement to pc.
|
|
|
|
BuildMI(FirstMBB, MBBI, DL, TII->get(X86::MOVPC32r), PC).addImm(0);
|
2011-01-26 02:03:37 +00:00
|
|
|
|
2010-07-10 09:00:22 +00:00
|
|
|
// If we're using vanilla 'GOT' PIC style, we should use relative addressing
|
|
|
|
// not to pc, but to _GLOBAL_OFFSET_TABLE_ external.
|
|
|
|
if (TM->getSubtarget<X86Subtarget>().isPICStyleGOT()) {
|
|
|
|
// Generate addl $__GLOBAL_OFFSET_TABLE_ + [.-piclabel], %some_register
|
|
|
|
BuildMI(FirstMBB, MBBI, DL, TII->get(X86::ADD32ri), GlobalBaseReg)
|
|
|
|
.addReg(PC).addExternalSymbol("_GLOBAL_OFFSET_TABLE_",
|
|
|
|
X86II::MO_GOT_ABSOLUTE_ADDRESS);
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
virtual const char *getPassName() const {
|
|
|
|
return "X86 PIC Global Base Reg Initialization";
|
|
|
|
}
|
|
|
|
|
|
|
|
virtual void getAnalysisUsage(AnalysisUsage &AU) const {
|
|
|
|
AU.setPreservesCFG();
|
|
|
|
MachineFunctionPass::getAnalysisUsage(AU);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
|
|
|
char CGBR::ID = 0;
|
|
|
|
FunctionPass*
|
|
|
|
llvm::createGlobalBaseRegPass() { return new CGBR(); }
|
2012-06-01 16:27:21 +00:00
|
|
|
|
|
|
|
namespace {
|
|
|
|
struct LDTLSCleanup : public MachineFunctionPass {
|
|
|
|
static char ID;
|
|
|
|
LDTLSCleanup() : MachineFunctionPass(ID) {}
|
|
|
|
|
|
|
|
virtual bool runOnMachineFunction(MachineFunction &MF) {
|
|
|
|
X86MachineFunctionInfo* MFI = MF.getInfo<X86MachineFunctionInfo>();
|
|
|
|
if (MFI->getNumLocalDynamicTLSAccesses() < 2) {
|
|
|
|
// No point folding accesses if there isn't at least two.
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
MachineDominatorTree *DT = &getAnalysis<MachineDominatorTree>();
|
|
|
|
return VisitNode(DT->getRootNode(), 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Visit the dominator subtree rooted at Node in pre-order.
|
|
|
|
// If TLSBaseAddrReg is non-null, then use that to replace any
|
|
|
|
// TLS_base_addr instructions. Otherwise, create the register
|
|
|
|
// when the first such instruction is seen, and then use it
|
|
|
|
// as we encounter more instructions.
|
|
|
|
bool VisitNode(MachineDomTreeNode *Node, unsigned TLSBaseAddrReg) {
|
|
|
|
MachineBasicBlock *BB = Node->getBlock();
|
|
|
|
bool Changed = false;
|
|
|
|
|
|
|
|
// Traverse the current block.
|
|
|
|
for (MachineBasicBlock::iterator I = BB->begin(), E = BB->end(); I != E;
|
|
|
|
++I) {
|
|
|
|
switch (I->getOpcode()) {
|
|
|
|
case X86::TLS_base_addr32:
|
|
|
|
case X86::TLS_base_addr64:
|
|
|
|
if (TLSBaseAddrReg)
|
|
|
|
I = ReplaceTLSBaseAddrCall(I, TLSBaseAddrReg);
|
|
|
|
else
|
|
|
|
I = SetRegister(I, &TLSBaseAddrReg);
|
|
|
|
Changed = true;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Visit the children of this block in the dominator tree.
|
|
|
|
for (MachineDomTreeNode::iterator I = Node->begin(), E = Node->end();
|
|
|
|
I != E; ++I) {
|
|
|
|
Changed |= VisitNode(*I, TLSBaseAddrReg);
|
|
|
|
}
|
|
|
|
|
|
|
|
return Changed;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Replace the TLS_base_addr instruction I with a copy from
|
|
|
|
// TLSBaseAddrReg, returning the new instruction.
|
|
|
|
MachineInstr *ReplaceTLSBaseAddrCall(MachineInstr *I,
|
|
|
|
unsigned TLSBaseAddrReg) {
|
|
|
|
MachineFunction *MF = I->getParent()->getParent();
|
|
|
|
const X86TargetMachine *TM =
|
|
|
|
static_cast<const X86TargetMachine *>(&MF->getTarget());
|
|
|
|
const bool is64Bit = TM->getSubtarget<X86Subtarget>().is64Bit();
|
|
|
|
const X86InstrInfo *TII = TM->getInstrInfo();
|
|
|
|
|
|
|
|
// Insert a Copy from TLSBaseAddrReg to RAX/EAX.
|
|
|
|
MachineInstr *Copy = BuildMI(*I->getParent(), I, I->getDebugLoc(),
|
|
|
|
TII->get(TargetOpcode::COPY),
|
|
|
|
is64Bit ? X86::RAX : X86::EAX)
|
|
|
|
.addReg(TLSBaseAddrReg);
|
|
|
|
|
|
|
|
// Erase the TLS_base_addr instruction.
|
|
|
|
I->eraseFromParent();
|
|
|
|
|
|
|
|
return Copy;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Create a virtal register in *TLSBaseAddrReg, and populate it by
|
|
|
|
// inserting a copy instruction after I. Returns the new instruction.
|
|
|
|
MachineInstr *SetRegister(MachineInstr *I, unsigned *TLSBaseAddrReg) {
|
|
|
|
MachineFunction *MF = I->getParent()->getParent();
|
|
|
|
const X86TargetMachine *TM =
|
|
|
|
static_cast<const X86TargetMachine *>(&MF->getTarget());
|
|
|
|
const bool is64Bit = TM->getSubtarget<X86Subtarget>().is64Bit();
|
|
|
|
const X86InstrInfo *TII = TM->getInstrInfo();
|
|
|
|
|
|
|
|
// Create a virtual register for the TLS base address.
|
|
|
|
MachineRegisterInfo &RegInfo = MF->getRegInfo();
|
|
|
|
*TLSBaseAddrReg = RegInfo.createVirtualRegister(is64Bit
|
|
|
|
? &X86::GR64RegClass
|
|
|
|
: &X86::GR32RegClass);
|
|
|
|
|
|
|
|
// Insert a copy from RAX/EAX to TLSBaseAddrReg.
|
|
|
|
MachineInstr *Next = I->getNextNode();
|
|
|
|
MachineInstr *Copy = BuildMI(*I->getParent(), Next, I->getDebugLoc(),
|
|
|
|
TII->get(TargetOpcode::COPY),
|
|
|
|
*TLSBaseAddrReg)
|
|
|
|
.addReg(is64Bit ? X86::RAX : X86::EAX);
|
|
|
|
|
|
|
|
return Copy;
|
|
|
|
}
|
|
|
|
|
|
|
|
virtual const char *getPassName() const {
|
|
|
|
return "Local Dynamic TLS Access Clean-up";
|
|
|
|
}
|
|
|
|
|
|
|
|
virtual void getAnalysisUsage(AnalysisUsage &AU) const {
|
|
|
|
AU.setPreservesCFG();
|
|
|
|
AU.addRequired<MachineDominatorTree>();
|
|
|
|
MachineFunctionPass::getAnalysisUsage(AU);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
}
|
|
|
|
|
|
|
|
char LDTLSCleanup::ID = 0;
|
|
|
|
FunctionPass*
|
|
|
|
llvm::createCleanupLocalDynamicTLSPass() { return new LDTLSCleanup(); }
|