Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===- MipsInstrFPU.td - Mips FPU Instruction Information -------*- C++ -*-===//
|
|
|
|
//
|
|
|
|
// The LLVM Compiler Infrastructure
|
|
|
|
//
|
|
|
|
// This file is distributed under the University of Illinois Open Source
|
|
|
|
// License. See LICENSE.TXT for details.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
//
|
|
|
|
// This file contains the Mips implementation of the TargetInstrInfo class.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Instructions
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
// ------------------------
|
|
|
|
// * 64bit fp:
|
|
|
|
// - 32 64-bit registers (default mode)
|
|
|
|
// - 16 even 32-bit registers (32-bit compatible mode) for
|
|
|
|
// single and double access.
|
|
|
|
// * 32bit fp:
|
|
|
|
// - 16 even 32-bit registers - single and double (aliased)
|
|
|
|
// - 32 32-bit registers (within single-only mode)
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Compare and Branch
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def SDT_MipsFPBrcond : SDTypeProfile<0, 3, [SDTCisSameAs<0, 2>, SDTCisInt<0>,
|
|
|
|
SDTCisVT<1, OtherVT>]>;
|
|
|
|
def SDT_MipsFPCmp : SDTypeProfile<0, 3, [SDTCisSameAs<0, 1>, SDTCisFP<0>,
|
|
|
|
SDTCisInt<2>]>;
|
2008-07-29 19:05:28 +00:00
|
|
|
def SDT_MipsFPSelectCC : SDTypeProfile<1, 4, [SDTCisInt<1>, SDTCisInt<4>,
|
|
|
|
SDTCisSameAs<0, 2>, SDTCisSameAs<2, 3>]>;
|
2009-05-27 17:23:44 +00:00
|
|
|
|
|
|
|
def MipsFPRound : SDNode<"MipsISD::FPRound", SDTFPRoundOp, [SDNPOptInFlag]>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def MipsFPBrcond : SDNode<"MipsISD::FPBrcond", SDT_MipsFPBrcond,
|
|
|
|
[SDNPHasChain]>;
|
2008-07-29 19:05:28 +00:00
|
|
|
def MipsFPCmp : SDNode<"MipsISD::FPCmp", SDT_MipsFPCmp>;
|
|
|
|
def MipsFPSelectCC : SDNode<"MipsISD::FPSelectCC", SDT_MipsFPSelectCC>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
|
|
|
// Operand for printing out a condition code.
|
|
|
|
let PrintMethod = "printFCCOperand" in
|
|
|
|
def condcode : Operand<i32>;
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Feature predicates.
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
def In32BitMode : Predicate<"!Subtarget.isFP64bit()">;
|
|
|
|
def IsSingleFloat : Predicate<"Subtarget.isSingleFloat()">;
|
|
|
|
def IsNotSingleFloat : Predicate<"!Subtarget.isSingleFloat()">;
|
2009-11-16 04:35:29 +00:00
|
|
|
def IsNotMipsI : Predicate<"!Subtarget.isMips1()">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Instruction Class Templates
|
|
|
|
//
|
2009-03-21 00:05:07 +00:00
|
|
|
// A set of multiclasses is used to address the register usage.
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//
|
2009-03-21 00:05:07 +00:00
|
|
|
// S32 - single precision in 16 32bit even fp registers
|
|
|
|
// single precision in 32 32bit fp registers in SingleOnly mode
|
|
|
|
// S64 - single precision in 32 64bit fp registers (In64BitMode)
|
|
|
|
// D32 - double precision in 16 32bit even fp registers
|
|
|
|
// D64 - double precision in 32 64bit fp registers (In64BitMode)
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//
|
2009-03-21 00:05:07 +00:00
|
|
|
// Only S32 and D32 are supported right now.
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
multiclass FFR1_1<bits<6> funct, string asmstr>
|
|
|
|
{
|
2009-03-21 00:05:07 +00:00
|
|
|
def _S32 : FFR<0x11, funct, 0x0, (outs FGR32:$fd), (ins FGR32:$fs),
|
|
|
|
!strconcat(asmstr, ".s $fd, $fs"), []>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
2009-05-27 17:23:44 +00:00
|
|
|
def _D32 : FFR<0x11, funct, 0x1, (outs FGR32:$fd), (ins AFGR64:$fs),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
!strconcat(asmstr, ".d $fd, $fs"), []>, Requires<[In32BitMode]>;
|
|
|
|
}
|
|
|
|
|
|
|
|
multiclass FFR1_2<bits<6> funct, string asmstr, SDNode FOp>
|
|
|
|
{
|
2009-03-21 00:05:07 +00:00
|
|
|
def _S32 : FFR<0x11, funct, 0x0, (outs FGR32:$fd), (ins FGR32:$fs),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
!strconcat(asmstr, ".s $fd, $fs"),
|
2009-03-21 00:05:07 +00:00
|
|
|
[(set FGR32:$fd, (FOp FGR32:$fs))]>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
|
|
|
def _D32 : FFR<0x11, funct, 0x1, (outs AFGR64:$fd), (ins AFGR64:$fs),
|
|
|
|
!strconcat(asmstr, ".d $fd, $fs"),
|
|
|
|
[(set AFGR64:$fd, (FOp AFGR64:$fs))]>, Requires<[In32BitMode]>;
|
|
|
|
}
|
|
|
|
|
|
|
|
class FFR1_3<bits<6> funct, bits<5> fmt, RegisterClass RcSrc,
|
|
|
|
RegisterClass RcDst, string asmstr>:
|
|
|
|
FFR<0x11, funct, fmt, (outs RcSrc:$fd), (ins RcDst:$fs),
|
|
|
|
!strconcat(asmstr, " $fd, $fs"), []>;
|
|
|
|
|
|
|
|
|
|
|
|
multiclass FFR1_4<bits<6> funct, string asmstr, SDNode FOp> {
|
2009-03-21 00:05:07 +00:00
|
|
|
def _S32 : FFR<0x11, funct, 0x0, (outs FGR32:$fd),
|
2008-07-09 04:45:36 +00:00
|
|
|
(ins FGR32:$fs, FGR32:$ft),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
!strconcat(asmstr, ".s $fd, $fs, $ft"),
|
2009-03-21 00:05:07 +00:00
|
|
|
[(set FGR32:$fd, (FOp FGR32:$fs, FGR32:$ft))]>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
2009-03-21 00:05:07 +00:00
|
|
|
def _D32 : FFR<0x11, funct, 0x1, (outs AFGR64:$fd),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
(ins AFGR64:$fs, AFGR64:$ft),
|
|
|
|
!strconcat(asmstr, ".d $fd, $fs, $ft"),
|
|
|
|
[(set AFGR64:$fd, (FOp AFGR64:$fs, AFGR64:$ft))]>,
|
|
|
|
Requires<[In32BitMode]>;
|
|
|
|
}
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Instructions
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
let ft = 0 in {
|
|
|
|
defm FLOOR_W : FFR1_1<0b001111, "floor.w">;
|
|
|
|
defm CEIL_W : FFR1_1<0b001110, "ceil.w">;
|
|
|
|
defm ROUND_W : FFR1_1<0b001100, "round.w">;
|
|
|
|
defm TRUNC_W : FFR1_1<0b001101, "trunc.w">;
|
|
|
|
defm CVTW : FFR1_1<0b100100, "cvt.w">;
|
|
|
|
|
|
|
|
defm FABS : FFR1_2<0b000101, "abs", fabs>;
|
|
|
|
defm FNEG : FFR1_2<0b000111, "neg", fneg>;
|
|
|
|
defm FSQRT : FFR1_2<0b000100, "sqrt", fsqrt>;
|
|
|
|
|
2009-03-21 00:05:07 +00:00
|
|
|
/// Convert to Single Precison
|
|
|
|
def CVTS_W32 : FFR1_3<0b100000, 0x2, FGR32, FGR32, "cvt.s.w">;
|
|
|
|
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
let Predicates = [IsNotSingleFloat] in {
|
|
|
|
/// Ceil to long signed integer
|
2009-03-21 00:05:07 +00:00
|
|
|
def CEIL_LS : FFR1_3<0b001010, 0x0, FGR32, FGR32, "ceil.l">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def CEIL_LD : FFR1_3<0b001010, 0x1, AFGR64, AFGR64, "ceil.l">;
|
|
|
|
|
|
|
|
/// Round to long signed integer
|
2009-03-21 00:05:07 +00:00
|
|
|
def ROUND_LS : FFR1_3<0b001000, 0x0, FGR32, FGR32, "round.l">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def ROUND_LD : FFR1_3<0b001000, 0x1, AFGR64, AFGR64, "round.l">;
|
|
|
|
|
|
|
|
/// Floor to long signed integer
|
2009-03-21 00:05:07 +00:00
|
|
|
def FLOOR_LS : FFR1_3<0b001011, 0x0, FGR32, FGR32, "floor.l">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def FLOOR_LD : FFR1_3<0b001011, 0x1, AFGR64, AFGR64, "floor.l">;
|
|
|
|
|
|
|
|
/// Trunc to long signed integer
|
2009-03-21 00:05:07 +00:00
|
|
|
def TRUNC_LS : FFR1_3<0b001001, 0x0, FGR32, FGR32, "trunc.l">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def TRUNC_LD : FFR1_3<0b001001, 0x1, AFGR64, AFGR64, "trunc.l">;
|
|
|
|
|
|
|
|
/// Convert to long signed integer
|
2009-03-21 00:05:07 +00:00
|
|
|
def CVTL_S : FFR1_3<0b100101, 0x0, FGR32, FGR32, "cvt.l">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def CVTL_D : FFR1_3<0b100101, 0x1, AFGR64, AFGR64, "cvt.l">;
|
|
|
|
|
|
|
|
/// Convert to Double Precison
|
|
|
|
def CVTD_S32 : FFR1_3<0b100001, 0x0, AFGR64, FGR32, "cvt.d.s">;
|
|
|
|
def CVTD_W32 : FFR1_3<0b100001, 0x2, AFGR64, FGR32, "cvt.d.w">;
|
|
|
|
def CVTD_L32 : FFR1_3<0b100001, 0x3, AFGR64, AFGR64, "cvt.d.l">;
|
|
|
|
|
|
|
|
/// Convert to Single Precison
|
|
|
|
def CVTS_D32 : FFR1_3<0b100000, 0x1, FGR32, AFGR64, "cvt.s.d">;
|
|
|
|
def CVTS_L32 : FFR1_3<0b100000, 0x3, FGR32, AFGR64, "cvt.s.l">;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// The odd-numbered registers are only referenced when doing loads,
|
|
|
|
// stores, and moves between floating-point and integer registers.
|
|
|
|
// When defining instructions, we reference all 32-bit registers,
|
|
|
|
// regardless of register aliasing.
|
|
|
|
let fd = 0 in {
|
|
|
|
/// Move Control Registers From/To CPU Registers
|
2009-05-27 17:23:44 +00:00
|
|
|
def CFC1 : FFR<0x11, 0x0, 0x2, (outs CPURegs:$rt), (ins CCR:$fs),
|
|
|
|
"cfc1 $rt, $fs", []>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
2009-05-27 17:23:44 +00:00
|
|
|
def CTC1 : FFR<0x11, 0x0, 0x6, (outs CCR:$rt), (ins CPURegs:$fs),
|
|
|
|
"ctc1 $fs, $rt", []>;
|
|
|
|
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def MFC1 : FFR<0x11, 0x00, 0x00, (outs CPURegs:$rt), (ins FGR32:$fs),
|
|
|
|
"mfc1 $rt, $fs", []>;
|
|
|
|
|
|
|
|
def MTC1 : FFR<0x11, 0x00, 0x04, (outs FGR32:$fs), (ins CPURegs:$rt),
|
2008-07-30 19:00:31 +00:00
|
|
|
"mtc1 $rt, $fs", []>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
}
|
|
|
|
|
2010-01-30 18:29:19 +00:00
|
|
|
def FMOV_S32 : FFR<0x11, 0b000110, 0x0, (outs FGR32:$fd), (ins FGR32:$fs),
|
|
|
|
"mov.s $fd, $fs", []>;
|
|
|
|
def FMOV_D32 : FFR<0x11, 0b000110, 0x1, (outs AFGR64:$fd), (ins AFGR64:$fs),
|
|
|
|
"mov.d $fd, $fs", []>;
|
|
|
|
|
2008-07-09 04:45:36 +00:00
|
|
|
/// Floating Point Memory Instructions
|
2009-11-16 04:35:29 +00:00
|
|
|
let Predicates = [IsNotSingleFloat, IsNotMipsI] in {
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def LDC1 : FFI<0b110101, (outs AFGR64:$ft), (ins mem:$addr),
|
|
|
|
"ldc1 $ft, $addr", [(set AFGR64:$ft, (load addr:$addr))]>;
|
|
|
|
|
|
|
|
def SDC1 : FFI<0b111101, (outs), (ins AFGR64:$ft, mem:$addr),
|
|
|
|
"sdc1 $ft, $addr", [(store AFGR64:$ft, addr:$addr)]>;
|
|
|
|
}
|
|
|
|
|
|
|
|
// LWC1 and SWC1 can always be emited with odd registers.
|
|
|
|
def LWC1 : FFI<0b110001, (outs FGR32:$ft), (ins mem:$addr), "lwc1 $ft, $addr",
|
|
|
|
[(set FGR32:$ft, (load addr:$addr))]>;
|
|
|
|
def SWC1 : FFI<0b111001, (outs), (ins FGR32:$ft, mem:$addr), "swc1 $ft, $addr",
|
|
|
|
[(store FGR32:$ft, addr:$addr)]>;
|
|
|
|
|
|
|
|
/// Floating-point Aritmetic
|
|
|
|
defm FADD : FFR1_4<0x10, "add", fadd>;
|
|
|
|
defm FDIV : FFR1_4<0x03, "div", fdiv>;
|
|
|
|
defm FMUL : FFR1_4<0x02, "mul", fmul>;
|
|
|
|
defm FSUB : FFR1_4<0x01, "sub", fsub>;
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Branch Codes
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Mips branch codes. These correspond to condcode in MipsInstrInfo.h.
|
|
|
|
// They must be kept in synch.
|
|
|
|
def MIPS_BRANCH_F : PatLeaf<(i32 0)>;
|
|
|
|
def MIPS_BRANCH_T : PatLeaf<(i32 1)>;
|
|
|
|
def MIPS_BRANCH_FL : PatLeaf<(i32 2)>;
|
|
|
|
def MIPS_BRANCH_TL : PatLeaf<(i32 3)>;
|
|
|
|
|
2008-07-09 04:45:36 +00:00
|
|
|
/// Floating Point Branch of False/True (Likely)
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
let isBranch=1, isTerminator=1, hasDelaySlot=1, base=0x8, Uses=[FCR31] in {
|
2008-07-28 19:11:24 +00:00
|
|
|
class FBRANCH<PatLeaf op, string asmstr> : FFI<0x11, (outs),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
(ins brtarget:$dst), !strconcat(asmstr, " $dst"),
|
|
|
|
[(MipsFPBrcond op, bb:$dst, FCR31)]>;
|
|
|
|
}
|
2008-07-28 19:11:24 +00:00
|
|
|
def BC1F : FBRANCH<MIPS_BRANCH_F, "bc1f">;
|
|
|
|
def BC1T : FBRANCH<MIPS_BRANCH_T, "bc1t">;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def BC1FL : FBRANCH<MIPS_BRANCH_FL, "bc1fl">;
|
|
|
|
def BC1TL : FBRANCH<MIPS_BRANCH_TL, "bc1tl">;
|
|
|
|
|
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Flag Conditions
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Mips condition codes. They must correspond to condcode in MipsInstrInfo.h.
|
|
|
|
// They must be kept in synch.
|
|
|
|
def MIPS_FCOND_F : PatLeaf<(i32 0)>;
|
|
|
|
def MIPS_FCOND_UN : PatLeaf<(i32 1)>;
|
|
|
|
def MIPS_FCOND_EQ : PatLeaf<(i32 2)>;
|
|
|
|
def MIPS_FCOND_UEQ : PatLeaf<(i32 3)>;
|
|
|
|
def MIPS_FCOND_OLT : PatLeaf<(i32 4)>;
|
|
|
|
def MIPS_FCOND_ULT : PatLeaf<(i32 5)>;
|
|
|
|
def MIPS_FCOND_OLE : PatLeaf<(i32 6)>;
|
|
|
|
def MIPS_FCOND_ULE : PatLeaf<(i32 7)>;
|
|
|
|
def MIPS_FCOND_SF : PatLeaf<(i32 8)>;
|
|
|
|
def MIPS_FCOND_NGLE : PatLeaf<(i32 9)>;
|
|
|
|
def MIPS_FCOND_SEQ : PatLeaf<(i32 10)>;
|
|
|
|
def MIPS_FCOND_NGL : PatLeaf<(i32 11)>;
|
|
|
|
def MIPS_FCOND_LT : PatLeaf<(i32 12)>;
|
|
|
|
def MIPS_FCOND_NGE : PatLeaf<(i32 13)>;
|
|
|
|
def MIPS_FCOND_LE : PatLeaf<(i32 14)>;
|
|
|
|
def MIPS_FCOND_NGT : PatLeaf<(i32 15)>;
|
|
|
|
|
|
|
|
/// Floating Point Compare
|
|
|
|
let hasDelaySlot = 1, Defs=[FCR31] in {
|
2009-03-21 00:05:07 +00:00
|
|
|
def FCMP_S32 : FCC<0x0, (outs), (ins FGR32:$fs, FGR32:$ft, condcode:$cc),
|
2008-07-28 19:11:24 +00:00
|
|
|
"c.$cc.s $fs, $ft", [(MipsFPCmp FGR32:$fs, FGR32:$ft, imm:$cc),
|
2009-03-21 00:05:07 +00:00
|
|
|
(implicit FCR31)]>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
|
|
|
def FCMP_D32 : FCC<0x1, (outs), (ins AFGR64:$fs, AFGR64:$ft, condcode:$cc),
|
2008-07-28 19:11:24 +00:00
|
|
|
"c.$cc.d $fs, $ft", [(MipsFPCmp AFGR64:$fs, AFGR64:$ft, imm:$cc),
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
(implicit FCR31)]>, Requires<[In32BitMode]>;
|
|
|
|
}
|
|
|
|
|
2008-07-29 19:05:28 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// Floating Point Pseudo-Instructions
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
|
|
|
// For some explanation, see Select_CC at MipsInstrInfo.td. We also embedd a
|
|
|
|
// condiciton code to enable easy handling by the Custom Inserter.
|
2009-10-29 18:10:34 +00:00
|
|
|
let usesCustomInserter = 1, Uses=[FCR31] in {
|
2008-07-29 19:05:28 +00:00
|
|
|
class PseudoFPSelCC<RegisterClass RC, string asmstr> :
|
|
|
|
MipsPseudo<(outs RC:$dst),
|
|
|
|
(ins CPURegs:$CmpRes, RC:$T, RC:$F, condcode:$cc), asmstr,
|
|
|
|
[(set RC:$dst, (MipsFPSelectCC CPURegs:$CmpRes, RC:$T, RC:$F,
|
|
|
|
imm:$cc))]>;
|
|
|
|
}
|
|
|
|
|
|
|
|
// The values to be selected are fp but the condition test is with integers.
|
2009-03-21 00:05:07 +00:00
|
|
|
def Select_CC_S32 : PseudoSelCC<FGR32, "# MipsSelect_CC_S32_f32">;
|
|
|
|
def Select_CC_D32 : PseudoSelCC<AFGR64, "# MipsSelect_CC_D32_f32">,
|
|
|
|
Requires<[In32BitMode]>;
|
2008-07-29 19:05:28 +00:00
|
|
|
|
|
|
|
// The values to be selected are int but the condition test is done with fp.
|
2009-03-21 00:05:07 +00:00
|
|
|
def Select_FCC : PseudoFPSelCC<CPURegs, "# MipsSelect_FCC">;
|
2008-07-29 19:05:28 +00:00
|
|
|
|
|
|
|
// The values to be selected and the condition test is done with fp.
|
2009-03-21 00:05:07 +00:00
|
|
|
def Select_FCC_S32 : PseudoFPSelCC<FGR32, "# MipsSelect_FCC_S32_f32">;
|
|
|
|
def Select_FCC_D32 : PseudoFPSelCC<AFGR64, "# MipsSelect_FCC_D32_f32">,
|
|
|
|
Requires<[In32BitMode]>;
|
2008-07-29 19:05:28 +00:00
|
|
|
|
2009-05-27 17:23:44 +00:00
|
|
|
def MOVCCRToCCR : MipsPseudo<(outs CCR:$dst), (ins CCR:$src),
|
|
|
|
"# MOVCCRToCCR", []>;
|
|
|
|
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-09 04:45:36 +00:00
|
|
|
// Floating Point Patterns
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
//===----------------------------------------------------------------------===//
|
2008-07-30 19:00:31 +00:00
|
|
|
def fpimm0 : PatLeaf<(fpimm), [{
|
2009-11-11 23:09:33 +00:00
|
|
|
return N->isExactlyValue(+0.0);
|
|
|
|
}]>;
|
|
|
|
|
|
|
|
def fpimm0neg : PatLeaf<(fpimm), [{
|
|
|
|
return N->isExactlyValue(-0.0);
|
2008-07-30 19:00:31 +00:00
|
|
|
}]>;
|
|
|
|
|
2009-03-21 00:05:07 +00:00
|
|
|
def : Pat<(f32 fpimm0), (MTC1 ZERO)>;
|
2009-11-11 23:09:33 +00:00
|
|
|
def : Pat<(f32 fpimm0neg), (FNEG_S32 (MTC1 ZERO))>;
|
2008-07-30 19:00:31 +00:00
|
|
|
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
def : Pat<(f32 (sint_to_fp CPURegs:$src)), (CVTS_W32 (MTC1 CPURegs:$src))>;
|
|
|
|
def : Pat<(f64 (sint_to_fp CPURegs:$src)), (CVTD_W32 (MTC1 CPURegs:$src))>;
|
2008-07-30 19:00:31 +00:00
|
|
|
|
2009-03-21 00:05:07 +00:00
|
|
|
def : Pat<(i32 (fp_to_sint FGR32:$src)), (MFC1 (TRUNC_W_S32 FGR32:$src))>;
|
Several changes to Mips backend, experimental fp support being the most
important.
- Cleanup in the Subtarget info with addition of new features, not all support
yet, but they allow the future inclusion of features easier. Among new features,
we have : Arch family info (mips1, mips2, ...), ABI info (o32, eabi), 64-bit
integer
and float registers, allegrex vector FPU (VFPU), single float only support.
- TargetMachine now detects allegrex core.
- Added allegrex (Mips32r2) sext_inreg instructions.
- *Added Float Point Instructions*, handling single float only, and
aliased accesses for 32-bit FPUs.
- Some cleanup in FP instruction formats and FP register classes.
- Calling conventions improved to support mips 32-bit EABI.
- Added Asm Printer support for fp cond codes.
- Added support for sret copy to a return register.
- EABI support added into LowerCALL and FORMAL_ARGS.
- MipsFunctionInfo now keeps a virtual register per function to track the
sret on function entry until function ret.
- MipsInstrInfo FP support into methods (isMoveInstr, isLoadFromStackSlot, ...),
FP cond codes mapping and initial FP Branch Analysis.
- Two new Mips SDNode to handle fp branch and compare instructions : FPBrcond,
FPCmp
- MipsTargetLowering : handling different FP classes, Allegrex support, sret
return copy, no homing location within EABI, non 32-bit stack objects
arguments, and asm constraint for float.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53146 91177308-0d34-0410-b5e6-96231b3b80d8
2008-07-05 19:05:21 +00:00
|
|
|
|
2008-07-30 19:00:31 +00:00
|
|
|
def : Pat<(i32 (bitconvert FGR32:$src)), (MFC1 FGR32:$src)>;
|
2009-03-21 00:05:07 +00:00
|
|
|
def : Pat<(f32 (bitconvert CPURegs:$src)), (MTC1 CPURegs:$src)>;
|
2009-05-27 17:23:44 +00:00
|
|
|
|
|
|
|
let Predicates = [In32BitMode] in {
|
|
|
|
def : Pat<(f32 (fround AFGR64:$src)), (CVTS_D32 AFGR64:$src)>;
|
|
|
|
def : Pat<(f64 (fextend FGR32:$src)), (CVTD_S32 FGR32:$src)>;
|
|
|
|
}
|
|
|
|
|
|
|
|
// MipsFPRound is only emitted for MipsI targets.
|
|
|
|
def : Pat<(f32 (MipsFPRound AFGR64:$src)), (CVTW_D32 AFGR64:$src)>;
|
|
|
|
|