Renesas M32R-FPU FMSUBFMSUB

3

3-47 M32R-FPU Software Manual (Rev.1.01)

INSTRUCTIONS

3.2 Instruction description

FMSUBFMSUB

[Mnemonic]

FMSUB Rdest,Rsrc1,Rsrc2

[Function]

Floating-point multiply and subtract

Rdest = Rdest - Rsrc1 * Rsrc2 ;

[Description]

This instruction is executed in the following 2 steps.

● Step 1

Multiply the floating-point single precision value stored in Rsrc1 by the floating-point single

precision value stored in Rsrc2.

The multiplication result is rounded toward 0 regardless of the value in the RM field of FPSR.

● Step 2

Subtract the result (rounded value) of Step 1 from the floating-point single precision value

stored in Rdest.

The subtraction result is rounded according to the RM field of FPSR.

The result of this operation is stored in Rdest. Exceptions are determined in both Step 1 and

Step 2. The DN bit of FPSR handles the conversion of denormalized numbers. The condition bit

(C) remains unchanged.

[EIT occurrence]

Floating-Point Exceptions (FPE)

• Unimplemented Operation Exception (UIPL)

• Invalid Operation Exception (IVLD)

• Overflow (OVF)

• Underflow (UDF)

• Inexact Exception (IXCT)

[Encoding]

floating-point Instructions

Floating-point multiply and subtract

[M32R-FPU Extended Instruction]

src11101 src20000 dest0011 00000100

FMSUB Rdest,Rsrc1,Rsrc2

Contents

Main RENESAS 32-BIT RISC SINGLE-CHIP MICROCOMPUTER REJ09B0112-0101Z M32R-FPU www.renesas.com Software Manual Keep safety first in your circuit designs! Notes regarding these materials REVISION HISTORY M32R-FPU Software Manual R1 (1) Table of contents CHAPTER 1 CPU PROGRAMMING MODEL CHAPTER 2 INSTRUCTION SET CHAPTER 3 INSTRUCTIONS APPENDIX (2) INDEX Page Page 1-2 1.1 CPU Register 1.1 CPU Register 1.2 General-purpose Registers 1-3 1.3 Control Registers 1-4 1.3.1 Processor Status Word Register: PSW (CR0) 1-5 1.3.2 Condition Bit Register: CBR (CR1) 1.3.3 Interrupt Stack Pointer: SPI (CR2) User Stack Pointer: SPU (CR3) 1.3.4 Backup PC: BPC (CR6) 1-6 1.3.5 Floating-point Status Register: FPSR (CR7) 1-7 1-8 1.3.6 Floating-point Exceptions (FPE) 1-9 1-10 1-11 1.4 Accumulator 1.4 Accumulator 1.5 Program Counter 1-12 1.6 Data Format Figure 1.6.1 Data Type 1.6.1 Data Type 1-13 1.6.2 Data Format 1-14 1-15 1.7 Addressing Mode 1.7 Addressing Mode Page Page 2-2 2.1 Instruction set overview 2.1.1 Load/store instructions 2-3 Page Page Page 2-7 Page 2-9 Fig. 2.1.2 DSP function instruction operation 1 (multiply, multiply and accumulate) 2-10 Note: The actual operation is processed in two steps. Refer to Chapter 3 for details. Fig. 2.1.3 DSP function instruction operation 2 (round off) Fig. 2.1.4 DSP function instruction operation 3 (transfer between accumulator and register) Page 2-12 2.2 Instruction format 2.2 Instruction format Fig. 2.2.2 Instruction format of M32R CPU Fig. 2.2.1 16-bit instruction and 32-bit instruction 2-13 2.2 Instruction format Page Page 3-2 3.1 Conventions for instruction description 3.1 Conventions for instruction description + - / % ++ -- Page 3-4 3.1 Conventions for instruction description 3-5 3.2 Instruction description ADD ADD Add ADD3 Add 3-operand 3-8 ADDI ADDI Rdest,#imm8 ADDI Add immediate imm8dest0100 ADDI Rdest,#imm8 3-9 ADDV ADDV Rdest,Rsrc src ADDV dest0000 ADDV Rdest,Rsrc1000 Add with overflow checking 3-10 ADDV3 ADDV3 Rdest,Rsrc,#imm16 src1000 dest1000 imm16 ADDV3 Add 3-operand with overflow checking ADDV3 Rdest,Rsrc,#imm16 3-11 ADDX src ADDX Rdest,Rsrc Add with carry 1001dest0000 ADDX Rdest,Rsrc ANDAND 11000000 AND Rdest,Rsrc AND srcdest AND Rdest,Rsrc AND3 AND 3-operand BCBC Bit clear M32R-FPU Extended Instruction 3-15 bit operation Bit clear BCLRBCLR 0 src0111 disp16 BCLR #bitpos,@(disp16,Rsrc) 1010 BCLR #bitpos,@(disp16,Rsrc) BEQBEQ 3-16 Branch on equal to 1011 src1 0000 src2 pcdisp16 BEQ Rsrc1,Rsrc2,pcdisp16 BEQZBEQZ 3-17 Branch on equal to zero 1011 0000 1000 src pcdisp16 BEQZ Rsrc,pcdisp16 BGEZBGEZ 3-18 Branch on greater than or equal to zero 1011 0000 1011 src pcdisp16 BGEZ Rsrc,pcdisp16 BGTZBGTZ 3-19 Branch on greater than zero 1011 0000 1101 src pcdisp16 BGTZ Rsrc,pcdisp16 BLBL Branch and link BLEZBLEZ 3-21 Branch on less than or equal to zero 1011 0000 1100 src pcdisp16 BLEZ Rsrc,pcdisp16 BLTZBLTZ 3-22 Branch on less than zero 1011 0000 1010 src pcdisp16 BLTZ Rsrc,pcdisp16 BNCBNC Branch on not C-bit BNEBNE 3-24 Branch on not equal to 1011 src1 0001 src2 pcdisp16 BNE Rsrc1,Rsrc2,pcdisp16 BNEZBNEZ 3-25 Branch on not equal to zero 1011 0000 1001 src pcdisp16 BNEZ Rsrc,pcdisp16 BRABRA Branch BSETBSET 3-27 bit operation Instructions 0 src0110 disp16 BTSTBTST Bit test CLRPSWCLRPSW 3-29 bit operation Instructions imm8 Clear PSW CMP Compare 3-31 CMPI CMPI Rsrc,#imm16 compare instruction Compare immediate 1000 0000 0100 src imm16 CMPI Rsrc,#imm16 CMPU Compare unsigned 3-33 CMPUI CMPUI Rsrc,#imm16 compare instruction Compare unsigned immediate 1000 0000 0101 src imm16 CMPUI Rsrc,#imm16 DIV Divide DIV Rdest,Rsrc 3-35 DIVU DIVU Rdest,Rsrc Divide unsigned dest1001 src0001 00000000 00000000 DIVU Rdest,Rsrc 3-36 FADDFADD FADD Rdest,Rsrc1,Rsrc2 Floating-point add src11101 src20000 dest0000 00000000 FADD Rdest,Rsrc1,Rsrc2 FADDFADD 3-37 Floating-point addd 3-38 FCMPFCMP src11101 src20000 dest0000 00001100 FCMP Rdest,Rsrc1,Rsrc2 FCMP Rdest,Rsrc1,Rsrc2 floating point Instructions Floating-point compare FCMPFCMP 3-39 Floating-point compare 3-40 FCMPEFCMPE src11101 src20000 dest0000 00001101 FCMPE Rdest,Rsrc1,Rsrc2 FCMPE Rdest,Rsrc1,Rsrc2 Floating-point compare with exception if unordered FCMPEFCMPE 3-41 Floating-point compare with exception if unordered 3-42 FDIVFDIV FDIV Rdest,Rsrc1,Rsrc2 Floating-point divide src11101 src20000 dest0010 00000000 FDIV Rdest,Rsrc1,Rsrc2 FDIVFDIV 3-43 DIV0: Zero Divide Exception Floating-point divide 3-44 FMADDFMADD FMADD Rdest,Rsrc1,Rsrc2 Floating-point multiply and add src11101 src20000 dest0011 00000000 FMADD Rdest,Rsrc1,Rsrc2 FMADDFMADD 3-45 Value after Addition Operation Floating-point multiply and add FMADDFMADD 3-46 Value after Addition Operation Floating-point multiply and add 3-47 FMSUBFMSUB FMSUB Rdest,Rsrc1,Rsrc2 Floating-point multiply and subtract src11101 src20000 dest0011 00000100 FMSUB Rdest,Rsrc1,Rsrc2 FMSUBFMSUB 3-48 Value after Subtraction Operation Floating-point multiply and subtract FMSUBFMSUB 3-49 Value after Subtraction Operation Floating-point multiply and subtract 3-50 FMULFMUL FMUL Rdest,Rsrc1,Rsrc2 Floating-point multiply src11101 src20000 dest0001 00000000 FMUL Rdest,Rsrc1,Rsrc2 FMULFMUL 3-51 Floating-point multiply 3-52 FSUBFSUB FSUB Rdest,Rsrc1,Rsrc2 Floating-point subtract src11101 src20000 dest0000 00000100 FSUB Rdest,Rsrc1,Rsrc2 FSUBFSUB 3-53 Floating-point subtract 3-54 FTOIFTOI FTOI Rdest,Rsrc Float to Integer src1101 00000000 dest0100 00001000 FTOI Rdest,Rsrc FTOIFTOI 3-55 Float to Integer 3-56 FTOSFTOS FTOS Rdest,Rsrc Float to short src1101 00000000 dest0100 00001100 FTOS Rdest,Rsrc FTOSFTOS 3-57 Float to short 3-58 ITOFITOF ITOF Rdest,Rsrc Integer to float src1101 00000000 dest0100 00000000 ITOF Rdest,Rsrc JLJL 3-59 Jump and link JL Rsrc 11100001 JL Rsrcsrc1100 JMPJMP 3-60 Jump JMP Rsrc JMP Rsrc11110001 1100 src LDLD Load LD24LD24 3-62 Load 24-bit immediate LD24 Rdest,#imm24 dest1110 imm24 LD24 Rdest,#imm24 LDBLDB Load byte LDHLDH Load halfword LDILDI Load immediate LDUBLDUB Load unsigned byte LDUHLDUH Load unsigned halfword LOCKLOCK 3-68 Load locked LOCK Rdest,@Rsrc dest0010 LOCK Rdest,@Rsrcsrc1101 MACHIMACHI 3-69 Multiply-accumulate high-order halfwords MACHI Rsrc1,Rsrc2 src10011 MACHI Rsrc1,Rsrc2src20100 MACLOMACLO 3-70 Multiply-accumulate low-order halfwords MACLO Rsrc1,Rsrc2 src10011 MACLO Rsrc1,Rsrc2src20101 MACWHIMACWHI 3-71 Multiply-accumulate word and high-order halfword MACWHI Rsrc1,Rsrc2 src10011 MACWHI Rsrc1,Rsrc2src20110 MACWLOMACWLO 3-72 Multiply-accumulate word and low-order halfword MACWLO Rsrc1,Rsrc2 src10011 MACWLO Rsrc1,Rsrc2src20111 MULMUL 3-73 MUL Rdest,Rsrc dest0001 MUL Rdest,Rsrcsrc0110 MULHIMULHI 3-74 Multiply high-order halfwords MULHI Rsrc1,Rsrc2 MULHI Rsrc1,Rsrc2src10011 src20000 MULLOMULLO 3-75 Multiply low-order halfwords MULLO Rsrc1,Rsrc2 src10011 MULLO Rsrc1,Rsrc2src20001 MULWHIMULWHI 3-76 DSP function instruction word and high-order halfword MULWHI Rsrc1,Rsrc2 MULWLOMULWLO 3-77 DSP fucntion instruction word and low-order halfword MULWLO Rsrc1,Rsrc2 MVMV 3-78 Move register MV Rdest,Rsrc dest0001 MV Rdest,Rsrcsrc1000 MVFACHIMVFACHI 3-79 Move high-order word MVFACHI Rdest dest0101 MVFACHI Rdest00001111 MVFACLOMVFACLO 3-80 Move low-order word MVFACLO Rdest dest0101 MVFACLO Rdest00011111 MVFACMIMVFACMI 3-81 DSP function instruction Move middle-order word MVFACMI Rdest MVFCMVFC 3-82 Move from control register MVFC Rdest,CRsrc dest0001 MVFC Rdest,CRsrcsrc1001 MVTACHIMVTACHI 3-83 Move high-order word to accumulator MVTACHI Rsrc src0101 MVTACHI Rsrc00000111 MVTACLOMVTACLO 3-84 Move low-order word to accumulator MVTACLO Rsrc src0101 MVTACLO Rsrc00010111 MVTCMVTC 3-85 Move to control register MVTC Rsrc,CRdest dest0001 MVTC Rsrc,CRdestsrc1010 NEGNEG 3-86 Negate NEG Rdest,Rsrc dest0000 NEG Rdest,Rsrcsrc0011 NOPNOP 3-87 No operation NOP NOP00000111 00000000 NOTNOT 3-88 Logical NOT NOT Rdest,Rsrc dest0000 NOT Rdest,Rsrcsrc1011 OROR 3-89 OR OR Rdest,Rsrc dest0000 OR Rdest,Rsrcsrc1110 OR3OR3 3-90 OR 3-operand OR3 Rdest,Rsrc,#imm16 dest1000 src1110 imm16 OR3 Rdest,Rsrc,#imm16 RACRAC 3-91 Round accumulator RAC RAC00000101 00001001 RACRAC 3-92 [Supplement] This instruction is executed in two steps as shown below: <step 1> <step 2> DSP function instruction RACHRACH 3-93 Round accumulator halfword RACH RACH00000101 00001000 RACHRACH 3-94 [Supplement] This instruction is executed in two steps, as shown below. <proccess 1> <proccess 2> DSP function instruction REMREM 3-95 Remainder REM Rdest,Rsrc dest1001 src0010 00000000 00000000 REM Rdest,Rsrc REMUREMU 3-96 Remainder unsigned REMU Rdest,Rsrc dest1001 src0011 00000000 00000000 REMU Rdest,Rsrc RTERTE 3-97 EIT-related instruction Return from EIT RTE SETHSETH Set high-order 16-bit SETPSWSETPSW 3-99 Bit Operation Instructions 00010111 imm8 SETPSW #imm8 Set PSW SLLSLL 3-100 Shift left logical SLL Rdest,Rsrc dest0001 src0100 SLL Rdest,Rsrc SLL3SLL3 Shift left logical 3-operand SLLISLLI 3-102 Shift left logical immediate SLLI Rdest,#imm5 dest0101 imm5010 SLLI Rdest,#imm5 SRASRA 3-103 Shift right arithmetic SRA Rdest,Rsrc dest0001 src0010 SRA Rdest,Rsrc SRA3SRA3 Shift right arithmetic 3-operand SRAISRAI 3-105 Shift right arithmetic immediate SRAI Rdest,#imm5 dest0101 imm5001 SRAI Rdest,#imm5 SRLSRL 3-106 Shift right logical SRL Rdest,Rsrc dest0001 src0000 SRL Rdest,Rsrc SRL3SRL3 Shift right logical 3-operand SRLISRLI 3-108 Shift right logical immediate SRLI Rdest,#imm5 dest0101 imm5000 SRLI Rdest,#imm5 STST 3-109 Store (1) ST Rsrc1,@Rsrc2 (2) ST Rsrc1,@+Rsrc2 (3) ST Rsrc1,@-Rsrc2 (4) ST Rsrc1,@(disp16,Rsrc2) STST STBSTB Store byte STHSTH Store halfword [M32R-FPU Extended Mnemonic] SUBSUB 3-113 Subtract SUB Rdest,Rsrc dest0000 0010 src SUB Rdest,Rsrc SUBVSUBV 3-114 Subtract with overflow checking SUBV Rdest,Rsrc dest0000 0000 src SUBV Rdest,Rsrc SUBXSUBX 3-115 Subtract with borrow SUBX Rdest,Rsrc dest0000 0001 src SUBX Rdest,Rsrc TRAPTRAP 3-116 EIT-related instruction Trap TRAP #imm4 UNLOCKUNLOCK 3-117 Store unlocked UNLOCK Rsrc1,@Rsrc2 src10010 UNLOCK Rsrc1,@Rsrc2src20101 UTOFUTOF 3-118 Floating Point Instructions Unsigned integer to float UTOF Rdest,Rsrc XORXOR 3-119 Exclusive OR XOR Rdest,Rsrc dest0000 XOR Rdest,Rsrcsrc1101 XOR3XOR3 3-120 Exclusive OR 3-operand XOR3 Rdest,Rsrc,#imm16 dest1000 1101 src imm16 XOR3 Rdest,Rsrc,#imm16 Page APPENDICES-2 Appendix1 Hexadecimal Instraction Code FPU extended instruction (b0-b3 = 1101, b8-b11 = 0000) 0 1101 0000 b16-b19 b24-b27 b0-b3b8-b11 APPENDIX 1 Appendix 1 Hexadecimal Instraction Code 1000 1001 1010 1011 1100 1101 1110 1111 01 234 567 1 2 3 4 5 6 7 0 APPENDICES-4 Appendix 2 Instruction List The M32R-FPU instruction list is shown below (in alphabetical order). APPENDICES-5 APPENDICES-6 Page APPENDICES-8 Appendix 3 Pipeline Processing Appendix Figure 3.1.1 Instructions and Pipeline Process Appendix Figure 3.1.1 shows each instruction type and the pipeline process. Appendix 3.1 Instructions and Pipeline Processing APPENDICES-9 APPENDICES-10 Appendix Figure 3.2.1 Pipeline Flow with no Stall (1) Appendix 3.2 Pipeline Basic Operation APPENDICES-11 Appendix Figure 3.2.2 Pipeline Flow with no Stall (2) APPENDICES-12 LD R3,@R4 <Case 1> An instruction which requires several cycles is executed in E DIV R1,R2 ADD R3,R4 APPENDICES-13 Appendix Figure 3.2.4 Pipeline Flow with Stalls (2) APPENDICES-14 Appendix Figure 3.2.5 Pipeline Flow with Stalls (3) APPENDICES-15 Appendix Figure 3.2.6 Pipeline Flow with Stalls (4) APPENDICES-16 Appendix Figure 3.2.7 Pipeline Flow with Stalls (5) APPENDICES-17 Appendix 4 Instruction Execution Time APPENDIX 4 Appendix 4 Instruction Execution Time APPENDICES-18 Appendix 5 IEEE754 Specification Overview Appendix 5.1 Floating Point Formats APPENDICES-19 APPENDICES-20 Appendix 5.2 Rounding Appendix 5.3 Exceptions APPENDICES-21 APPENDICES-22 APPENDICES-23 Appendix 6 M32R-FPU Specification Supplemental Explanation APPENDICES-24 APPENDICES-25 (2) When underflow occurs in Step 1 <When EU = 0, DN = 1: UDF occurs> <When EU = 0, DN = 0: UDF and UIPL occur> <When EU = 1: UDF occurs> APPENDICES-26 APPENDICES-27 APPENDICES-28 Appendix 6.2 Rules concerning Generation of QNaN in M32R-FPU APPENDICES-29 Appendix 7 Precautions Appendix 7.1 Precautions to be taken when aligning data APPENDIX 7 Appendix 7 Precautions Page Page INDEX-2 Symbol A B C D L M O P E R S T U