Manuals / Freescale Semiconductor / Computer Equipment / Computer Hardware

Freescale Semiconductor SC140 specifications Fractional Double-Precision Multiplication

Models: SC140

1 760

Download 760 pages 48.94 Kb

Page 59

DALU

Figure 2-8 illustrates the use of these instructions in the case of a double-precision multiplication of

32-bit x 32-bit operands. The “Unsigned x Unsigned” operation is used to multiply or multiply-accumulate the unsigned low portion of one double-precision number with the unsigned low portion of the other double-precision number. The “Signed x Unsigned” and “Unsigned x Signed” operations are used to multiply or multiply-accumulate the signed high portion of one double-precision number with the unsigned low portion of the other double-precision number. The “Signed x Signed” operation is used to multiply or multiply-accumulate the two signed high portions of two signed double-precision numbers. The TFRx instructions in parentheses are optional instructions that are used only in case all 64 bits of the result are needed. Otherwise, the result is truncated to a 32-bit fraction.

32 bits

D0.h

D0.l

⋅

D1.h

D1.l

								=
mpyuu	D0,D1,D2							Unsigned ⋅ Unsigned
mpyuu	D0,D1,D2							D1.l ⋅ D0.l
(tfr	D2,D3)				Signed ⋅			D1.l ⋅ D0.l
(tfr	D2,D3)
							Unsigned			+
dmacsu	D0,D1,D2				D0.h ⋅ D1.l					+
										+


					Unsigned ⋅ Signed
					Unsigned ⋅ Signed
macus	D0,D1,D2
macus	D0,D1,D2				D1.h			⋅ D0.l
(tfr	D2,D4)				D1.h			⋅ D0.l		+
(tfr	D2,D4)									+

Signed ⋅ Signed

dmacss D0,D1,D2

S Ext

D1.h ⋅ D0.h

D2.e

D2.h

D2.l

D4.l

D3.l

64 bits

Figure 2-8. Fractional Double-Precision Multiplication

SC140 DSP Core Reference Manual

2-27

Image 59

Freescale Semiconductor SC140 specifications Fractional Double-Precision Multiplication

Contents

Reference Manual SC140 DSP CoreSC140 DSP Core Reference Manual Table of Contents SC140 DSP Core Reference Manual Control Registers SC140 DSP Core Reference Manual Program Control Instruction Set Accelerator Plug-In Programming Rules Appendix a SC140 DSP Core Instruction SetAppendix B StarCore RegistryXii List of Figures Xiv List of Tables Xvi SC140 DSP Core Reference Manual Xvii Xviii List of Examples SC140 DSP Core Reference Manual SC140 DSP Core Reference Manual Xxi Xxii About This Book Abbreviation Description Abbreviations used in this manual are listed belowAbbreviations ISR Revision Date Description Revision HistoryTarget Markets Chapter IntroductionArchitectural Differentiation Core Architecture Features Typical System-On-Chip Configuration SC140 platform Variable Length Execution Set Vles Software ModelSoC DSP expansion area System expansion areaCore Architecture Features Architecture Overview Chapter Core ArchitectureData Arithmetic Logic Unit Dalu Block Diagram of the SC140 CoreBit-Field Unit BFU Address Generation Unit AGUData Register File Multiply-Accumulate MAC UnitBit Mask Unit BMU Stack Pointer RegistersMemory Interface Program Sequencer Unit PseqEnhanced On-Chip Emulator EOnCE Instruction Set Accelerator Plug-in Isap InterfaceDalu Architecture DaluLimit EXT Dalu Programming ModelData Registers D0-D15 Operand Type Dn.e Dn.h Dn.l Write to Data RegistersRead from Data Registers Instruction Description Data Registers Access WidthDalu Arithmetic Instructions MAC Operand Type Data Width BitsDIV NEG Dalu Logical Instructions BFU Data Shifter/LimiterScaling Example ScalingLimiting Calculating the Ln BitLn Bit Calculation Scaling Mode Bits Defining the Ln bit CalculationLimiting with the Moves Instructions 10. Scaling and Limiting Interactions Scaling and Arithmetic Saturation Mode InteractionsSelected Special Six Other Dalu Instructions Mode Limiting Example11. Saturation and Rounding Interactions Dalu Arithmetic and RoundingData Representation Signed Fractional Data FormatsSigned Fractional Signed Integer Unsigned Integer Signed Integer12. Two’s Complement Word Representations Unsigned Multiplication MultiplicationDivision Unsigned ArithmeticUnsigned Comparison 13. Rounding Position in Relation to Scaling ModeScaling Mode High Portion Low Portion Rounding ModesConvergent Rounding No Scaling 2.6.2 Two’s Complement Rounding Two’s Complement Rounding No Scaling 14. Arithmetic Saturation Example Arithmetic Saturation ModeFractional Multi-Precision Arithmetic Multi-Precision Arithmetic SupportFractional Double-Precision Multiplication Fractional Mixed-Precision Multiplication Integer Multi-Precision Arithmetic10. Signed Integer Double-Precision Multiplication 11. Unsigned Integer Double-Precision Multiplication Viterbi Decoding SupportAGU Architecture Address Generation UnitUnit AAU AddressArithmetic Address Generation Unit 13. AGU Programming Model AGU Programming ModelStack Pointer Registers NSP, ESP Address Registers R0-R15Shadow Stack Pointer Registers Offset Registers N0-N3Base Address Registers B0-B7 Modifier Registers M0-M3Address Modifier Modes Modifier Control Register Mctl17. Address Modifier AM Bits Address Register Indirect Modes Addressing ModesRegister Direct Modes Address Generation Unit PC Relative Mode Special Addressing Modes Memory Access Misalignment Memory Access Width20. Addressing Modes Summary Access Type Aligned AddressAddressing Modes Summary 19. Memory Address AlignmentSpecial Address Register IndirectPC Relative Address Modifier Modes Linear Addressing ModeReverse-carry Addressing Mode Modulo Addressing Mode15. Modulo Addressing Example Modifier Mj Address Calculation Arithmetic Multiple Wrap-Around Modulo Addressing Mode21. Modulo Register Values for Modulo Addressing Mode 23. AGU Arithmetic Instructions Arithmetic Instructions on Address RegistersBit Mask Instructions 24. AGU Bit Mask Instructions BMU Bit Mask Test and Set Semaphore Support InstructionLabel BMTSET.W #mask,R0 JT label Move InstructionsSemaphore Hardware Implementation Example of Normal Usage of the Semaphoring MechanismMOVE.W 25. AGU Move Instructions16. Integer Move Instructions 17. Fractional Move Instructions 18. Bit Allocation in MOVE.L D0.eD1.e Memory Interface1.1 SC140 Bus Structure 1 SC140 Endian Support20. Basic Connection between SC140 Core and Memory Memory OrganizationRepresentation Type Value Data Moves26. Data Representation in Memory 22. Data Transfer in Big and Little Endian Modes Address Data Multi-Register MovesMulti-Register Transfer in Big and Little Endian Modes Instruction Word Transfers 25. Instruction Moves in Big and Little Endian Modes Memory Access Behavior in Big/Little Endian Modes Example MOVE.2F D0D1, R0 Example MOVE.L D0.ED1.E, A0Example MOVE.2L D0D1, R0 Example MOVE.F D0, R0Example VSL.2W D1D3, R0 + N0 Example VSL.4W D2D6D1D3, R0 + N0D6 = Example VSL.4F D2D6D1D3, R0 + N0Data = Example Push D0Example Push D0 Push D1 Example BMSET.W #$1234, A0Instruction Register Operands Big Little Endian 31. Control Instructions in Big and Little Endian ModesStatus Register SR Core Control RegistersStatus Register Description Name Description SettingsDescribes the various SR bits Exceptions I2-I0 Interrupt Mask Bits ReflectReserved Overflow Exception Enable BitDisable Interrupts Bit When this bit Exception Mode Bit SelectsRounding Mode Bit Selects the type Equation ModeS1-S0 Scaling Mode Bits Specify Scaling Mode BitName Description Settings Arithmetic Saturation Mode Selects Exception and Mode Register EMR Exception and Mode Register EMREMR Description Describes the EMR fieldsIllegal Execution Set Indicates whether an IlstExample 3-1. Clearing an EMR Bit PLL and Clock RegistersClearing EMR Bits Bmclr #$fffb,EMR.LDebugging System Emulation and Debug EOnCECascading Multiple SC140 EOnCE Modules in a SoC Jtag Interface Signal DescriptionsSignal Name Signal Description Overview of the Combined Jtag and EOnCE InterfaceJtag Instructions Jtag Scan PathsLoadgpr Select-DR Scan Path Select-IR Scan Path TAP Controller State Machine Jtag Scan PathsEnabling the EOnCE Module Activating the EOnCE Through the Jtag PortReading/Writing EOnCE Registers Through Jtag Debugrequest and Enableeonce CommandsReading and Writing EOnCE Registers Via Jtag EOnCE register read capture operation through Jtag EOnCE register write operation through JtagCore Interface EOnCE SignalsEOnCE Signals Jtag Signals Main Capabilities of the EOnCE ModuleDebug State EOnCE Dedicated InstructionsSoftware Downloading Debug ExceptionExecuting an Instruction while in Debug State Software Downloading Event type Occurs when EOnCE EventsEOnCE Event Types Event type EOnCE ActionsEvent and Action Summary EOnCE Event and Action SummaryEOnCE Controller EOnCE Enabling and Power ConsiderationsEOnCE Module Internal Architecture Address Control Decoder Logic Receive Register Command RegisterTransmit Register Update Signal from the TAP Controller Address Monitor and Control RegisterEvent Counter Event Counter Register Set Shows a block diagram of the event counterEvent Detection Unit EDU EventD Event0 Event1 Event2 Event5 Count event EE5..0Address Buses XDBxx Data BusesEEi Address Event Detection Channel EdcaEdca Register Set Edcd register set is shown below Data Event Detection Channel EdcdEvent External Event 6,7 Count Event EventDOptional External Event Detection Address Channels Event Selector ES10. Event Selector Register Set EE40 ES block diagram is shown in FigureTrace Unit Event0..Event5 External Event6, Event7 EventD Count eventEOnCE Module Internal Architecture Trace Buffer TB Off-Core Change of Flow and Interrupt TracingChange of Flow Trace Unit Programming Model Writing to the Trace BufferReading the Trace Buffer Tbbuff 11. Trace Buffer Register Set EOnCE Register AddressingOffset 12 displays the EOnCE register addressing offsets12. EOnCE Register Addressing Offsets EDCA4REFA Real-Time Jtag Access Reading or Writing EOnCE Registers Using Core SoftwareReal-Time Data Transfer General EOnCE Register IssuesEOnCE Register Addressing 13 describes the ECR fields EOnCE Command Register ECRRead/Write Command Specifies EOnCE Controller RegistersEOnCE Status Register ESR 16 displays the bit configuration of the ESRName Description 14 describes the ESR fields14. ESR Description Section DREE3 Name Description Reserved EOnCE Monitor and Control Register Emcr15 describes the Emcr fields 15. Emcr DescriptionDebugerst EOnCE Transmit Register Etrsmt EOnCE Receive Register ErcvDetecting Entry into Debug State EE SignalsEE Signals as Outputs Detection by the Event Detection ChannelsEE Signals Control Register Eectrl EE Signals as InputsEeddef 16. Eectrl DescriptionEE2DEF Length Control Bits Description Core Command Register CorecmdLength control bits are described in -17, below 17. Length Control BitsPC Breakpoint Detection Register Pcdetect PC of the Exception Execution Set PcexcpPC of the Next Execution Set Pcnext PC of Last Execution Set PclastEvent Counter Control Register Ecntctrl Event Counter RegistersReserved for Test Extended Mode of Operation Bit18 describes the Ecntctrl fields 18. Ecntctrl DescriptionEvents to be Counted Determines Event Counter Enable Used toEvent Counter Value Register Ecntval Extension Counter Value Register Ecntext EC Signals19 describes the EDCAiCTRL fields Event Detection Unit EDU Channels and RegistersAddress Event Detection Channel Edca Edca Control Registers EDCAiCTRLComparators Selection Used to Event Detection Channel EDCAiComparator a Condition Selection Access Type Selection These bitsComparator B Condition Selection Edca Mask Register EDCAiMASK Edca Reference Value Registers a and B EDCAiREFA, EDCAiREFB20. Edcdctrl Description Data Event Detection Channel EdcdEdcd Control Register Edcdctrl 20 describes the Edcdctrl fieldsAWS Access Width SelectionComparator Condition Selection Access Type Selection The ATS bitEdcd Mask Register Edcdmask Event Selector ES RegistersEvent Selector Control Register Eselctrl Edcd Reference Value Register Edcdref21. Eselctrl Description Eselctrl fields are described in TableEvent Selector Mask Debug State Register Eseldm 24 displays the bit configuration of EseldmEvent Selector Mask Debug Exception Register Eseldi Event Selector Mask Enable Trace Register EseletbTrace Buffer Control Register Tbctrl Event Selector Mask Disable Trace Register EseldtbTrace Unit Registers Upon a trace event, trace the counter value Ecntval This mode is usefull only with the Tcount mode22. Allowed tracing mode combinations Trace modeTrace Buffer Extension Counter Trace Buffer Counter ModeTbctrl fields are described in the following table 23. Tbctrl DescriptionTrace Issue of Execution Sets Enable Trace Loops Mode Enables tracingTrace Buffer Enable Mode Enables Trace Mark Instruction ModeTrace Buffer Register Tbbuff Trace Buffer Read Pointer Register TbrdTrace Buffer Write Pointer Register Tbwr Trace Unit Registers Pipeline Chapter Program ControlIllustrates the five instruction pipeline stages Instruction Pipeline StagesPipeline Stage Description Pipeline ExamplePipeline Stages Overview Instruction Cycle OperationAddress Generation Instruction Pre-Fetch and FetchInstruction Dispatch Example 5-1. Four SC140 Instructions in an Execution Set Instruction GroupingExecution Instruction Grouping Methods Grouping TypesPrefix Grouping Serial GroupingTwo-Word Prefix Prefix TypesPrefix Instructions Conditional ExecutionOne-Word Low Register Prefix For examplePrefix Selection Algorithm Assembly Syntax MeaningLow Register Prefix Selection Algorithm Instruction Reordering Within an Execution Set Example 5-5. Set of 2 Two-word Instructions Requiring a NOP Example 5-4. Conditional Vles Having Two SubgroupsInstruction Timing Instruction Categories Timing Summary Sequential Instruction TimingCompare Shift Test Dalu Instruction TimingMove Instruction Timing Bit Mask Instruction TimingNon-Loop Change-of-Flow Instructions Example 5-6. Delayed Change-of-Flow and Its Delay SlotChange-Of-Flow Instruction Timing Loop Change-Of-Flow Instructions Direct, PC-Relative, and Conditional COFCOF Execution Cycles Delayed COFExample 5-7. Subroutine Call Timing Example 5-7 shows a case when a stall cycle is addedHighest cycle count of instructions grouped with Call Number of Cycles Needed by Change-of-Flow InstructionsMemory Access Timing Memory Access Examples Example 5-8. Parallel Execution of Two Move Instructions MOVE.LMemory Stall Conditions Implicit Push/Pop Memory TimingLoop Start Address Registers SAn Hardware LoopsLoop Programming Model Status Register SR Loop Flag Bits Loop Notation and EncodingLoop Counter Registers LCn Location Functionality Loop Initiation and ExecutionLpmarka and Lpmarkb Bits in Short and Long Loops Loop TypeLoop Iteration and Termination Loop NestingInstruction Operation Loop Control Instructions10 lists the loop instructions 10. Loop Control InstructionsExample 5-14. Short Loop, Two Execution Sets Example 5-13. Long Loop DisassemblyExample 5-12. Long Loop Example 5-16. Nested Loop Following is an example of a nested loopExample 5-15. Short Loop, One Execution Set 1 SC140 Single Stack Memory Use Stack SupportLoop Timing Shows the stack structure 2 SC140 Dual Stack Memory UseEven Register De File Odd Register Do File Stack Support Instructions11. Stack Push/Pop Instructions 12. Even and Odd Registers14. Stack Move Instructions Addressing Mode DescriptionShadow Stack Pointer Registers 13. Stack Memory MapFast Return from Subroutines Working Modes Normal Working ModeException Working Mode Working Mode EXP bit Active SPDual-stack Rtos Typical Working Mode Usage ScenariosSingle-stack Rtos Working Mode TransitionsFrom Exception to Normal mode From Normal to Exception modeWorking Modes 16. Processing State Change Instructions Processing StatesProcessing State Change Instructions Processing State Transitions 10. Core State DiagramProcessing State Transitions Description Reset Processing StateExecution State 17. Processing State TransitionsWait Processing State 18. Exit Wait Processing State due to an Interrupt or NMI Stop Processing StateException Processing 11. Core-PIC Interface SC140Programming Exception Routine Addresses Interrupt Vector AddressVector Base Address Register Exception Address Priority Type Description Offset Highest Return From Exception Instructions19. Exception Vector Address Table Interrupt Priority Level Maskable InterruptsNon-Maskable Interrupts NMI Internal ExceptionsIllegal Execution Set Illegal ExceptionIllegal Instruction Debug Exception Exception Interface to the PipelineDalu Overflow Trap ExceptionExample 5-17. Basic Exception Timing Exception Mode ExecutionException Timing 20. Exception PipelineException Processing 12 provides a flow chart for Example 21. Pipeline Example Introduction Instruction Set Accelerator Plug-InSC140Core Isap SC140 Schematic ConnectionSingle Isap Data MemoryMultiple Isap Core to Multiple Isap Connection SchematicBinary Encoding Words Bits Isap Memory AccessIsap instructions and instruction encoding Isap Encoding FieldsTo understand this, look at the following lines of code Example 6-1. Isap memory accessISAP-core register transfers Example 6-3. ISAP-Core register transfers Immediate Data Transfer to Isap registersFollowing line of code Example 6-2. ISAP-Core register transfersVles that uses an implicit Isap ID string Core Assembly Syntax with an IsapIdentification of Isap instructions Working with One IsapOne Isap in a Multi-Line Vles Working with Multiple ISAPsOne Isap in a Single-Line Vles Example 6-5. Multiple Isap coding An Example of the Definition Flexibility of an IsapMultiple ISAPs in a Multi-Line Vles Example 6-7. Conditional Execution Example Example 6-6. Conditional Execution ExampleIsap Functions that Interact With the Core Programming RulesRules for implicit AGU instructions Grouping rules for explicit Isap instructionsD.2, D.3 Sequencing rules for T bit updateProgramming Rules Vles Grouping Semantics Vles Sequencing SemanticsVles Grouping Semantics Grouping Rules SC140 Pipeline ExposureProgramming Rule Notation Register Read/Write Sequencing RulesRegister Aliasing Status Bit UpdatesInstruction Words MOVE-like InstructionsChange-Of-Flow Destinations Delayed COF InstructionsDelay Slot AGU Arithmetic InstructionsHardware Loop Detection Enabled LoopStatic Programming Rules Hardware LoopsRule G.G.3 General Grouping RulesRule G.G.1 Rule G.G.2Example 7-7 Duplicate Stack Pointer Destinations Rule G.G.4Example 7-5 Duplicate PC Destinations Example 7-6 Duplicate Address Pointer Register DestinationsExample 7-10 Duplicate Status Bit Destinations Rule G.G.4 ExceptionsExample 7-8 Duplicate Register Destinations Example 7-9 Duplicate SR/EMR Register DestinationsExample 7-15. Dalu Register Use Exceeds Four Times Prefix Grouping RulesRule G.G.5 Following rules only apply to prefix-grouped VlesExample 7-17 Two-Word Instructions Exceed Two Rule G.P.1Example 7-16 Vles Extension Words Exceed Two Example 7-18. Vles Has Mutually Exclusive Instructions Rule G.P.3Rule G.P.4 Rule G.P.5Example 7-21. IFc Having Two Subgroups Rule G.P.6Rule G.P.7 Example 7-20. Data Source Use of Nn and Mn RegistersExample 7-24. Isap instructions in same IFc group Rule G.P.8Rule G.P.9 Example 7-25. Mctl Write to R0-R7 Use AGU RulesRule A.1 Example 7-26. Rn, Nn, Mn Write to AGU Use Rule A.2Rule A.3 Example 7-28. LCn Write to MOVE-like Use Rule A.4Example 7-27. Rn or Nn Write to MOVE-like Use Rule A.7 Delayed COF RulesExample 7-29. Nmid Update to EMR Read Example 7-30. Instructions in a Delay SlotRule D.3 Example 7-31. Instructions in a Rted Delay SlotExample RTE/D with SR Updates Rule D.2Rule D.6 Rule D.4Rule D.5 Rule D.5aRule T.1 Status Bit RulesRule D.8 Rule D.9Rule SR.2 Rule T.2.aRule T.2.b Rule T.2.cStatic Programming Rules Example 7-43. SR Write to SR Status Bit Use Rule SR.4 Example 7-44. SR Write to SR Status Bit UpdateRule SR.3 Rule SR.4a Example 7-45. Dovf Update to SR Read or WriteExample 7-46. Dovf Update grouped with Move-like SR updates DI and EI DOENn and DOENSHn Loop Nesting RulesRule SR.7 Rule L.N.1Example 7-50. Nested DOENn/DOENSHn Instructions Rule L.N.2Rule L.N.3 Example 7-49. Nested Loops with Ordered IndexExample 7-53. Changing a loop type Example 7-52. Loopend between Doen and LoopendExample 7-54. Instructions at the End of Long Loops Loop LA RulesRule L.L.1 Rule L.L.2Example 7-56. Instructions in Short Loops Rule L.L.3Rule L.L.4 LA of a short loop cannot be at LA-1 of a long loopRule L.D.1 Loop Sequencing RulesRule L.L.5 Rule L.L.6Example 7-60. LCn Write at the Start of Short Loop n Rule L.D.3Rule L.D.5 Rule L.D.6Rule L.D.9 Rule L.D.7Rule L.D.8 Rule L.C.3 Loop COF RulesRule L.C.1 Rule L.C.2Example 7-68. Bc/Jc at LA-3 of a Long Loop Rule L.C.5Bc or Jc instruction is not allowed at LA-3 of a long loop Example 7-69. Loop COF Destination in the Same Loop Rule L.C.7Example 7-71. Subroutine Call to End of Loops Rule L.C.9Rule L.C.10 Example 7-70. Loop COF at End of Nested Long LoopsRule L.G.3 General Looping RulesRule L.C.11 Rule L.C.12Rule A.2a Dynamic Programming RulesAGU Dynamic Rules Rule L.G.5Rule D.7 Memory Access RulesRule A.5 Rule A.6Rule L.N.6 RAS RulesLoop Rules Rule J.4Example 7-82. SR.2 Across a COF Boundary Example 7-83. A.2 from a Delay Slot to a COF DestinationRule Detection Across COF Boundaries Cycle-Based COF RulesVLES-Based COF Rules Rule SR.4b Example 7-85. EMR access at the start of an exceptionRule Detection Across Exception Boundaries Rule SR.2aExample 7-86. Mctl Write to R0-R7 Use Rule A.1aRule J.2 COF destination cannot be a delay slotProgramming Guidelines Rule J.1Rule J.5 Rules Not Detected Across COF BoundariesGood Programming Practices Source Code PracticesBinary Code Practices Software Development Practices Lpmark RulesLpmark Instruction Type Prefix Grouping Rules Static Programming RulesDynamic Programming Rules General Grouping RulesLoop LA Rules Loop Nesting RulesExample 7-93. Active LCn Write at the End of Long Loops Lpmark Rule L.L.2Lpmark Rule L.L.3 Lpmark Rule L.L.5Lpmark Rule L.D.6 Loop Sequencing RulesLpmark Rule L.L.6 Lpmark Rule L.D.2 + L.D.3Example 7-97. Active LCn Read at the Start of a Loop Loop COF RulesLpmark Rule L.C.2 COF instructions are not allowed at LPB of a long loopExample 7-99. Bc/Jc at the Start of a Loop Lpmark Rule L.C.3 + L.C.5Example 7-98. COF Instructions at LPB of a Long Loop Example 7-101. Subroutine Call to End of Loops Lpmark Rule L.C.9Lpmark Rule L.C.10 Example 7-100. Loop COF at End of Nested Long LoopsRule Detection Across Exception Boundaries Lpmark Programming GuidelinesGeneral Looping Rules Lpmark Rule L.C.1 Example 7-104. COF Destination to Loop Delay SlotsNOP Definition Is encoded as Grouping ExamplesIs assembler mapped to the IFF prefix and encoded as Is assembler mapped to the IFT prefix and encoded asIs encoded ignoring the NOP subgroup as NOP Definition Appendix a SC140 DSP Core Instruction Set Convention Definition ConventionsTable A-1. Instruction Conventions Abbreviation Register Name Table A-2. Operations SyntaxTable A-3. Register Abbreviations Operator DescriptionTable A-4. Assembler Syntax Brackets as Isap indicatorsBrackets as address indicators Addressing Mode Definition Notation in Instruction Field Addressing Mode NotationTable A-5. Addressing Mode Notation for the EA Operand Table A-6. Addressing Mode Notation for the ea OperandDefinition for the field is Data Representation in Memory for the ExamplesEncoding Notation Prefix Word Encoding Ccc Instruction Formats and OpcodesInstruction Fields AaaPrefix Words Cycles Type Example, 2-w prefix + 2 grouped instruction words, aaa =If true D0, D2, A0, if false D1, D3, A1 If true, all the setHigh data register is used for the op1 field E2 is set Last, or to last-1 ExampleFirst execution set of the loop High data register is used for the op3 field E3 is setDSP Core Instruction Set Instruction Sub-types Instruction TypesTable A-7. Dalu Arithmetic Instructions MAC Table A-8. Dalu Logical Instructions BFU Table A-9. AGU Arithmetic Instructions Table A-11. AGU Stack Support Instructions Table A-10. AGU Move InstructionsTable A-13. AGU Non-Loop Change-of-Flow Instructions Table A-12. AGU Bit-Mask Instructions BMUTable A-16. Prefix Instructions Table A-15. AGU Program Control InstructionsInstruction Definition Layout InstInstructions ABS Instruction Single Source/Destination Data RegisterADC Dc,Dd ADCAdd Long With Carry Dalu Dc + Dd + C → DdRegister/Memory Address Before Dc,Dd Data Register PairsAdd d0,d1,d2 ADDAdd Dalu Operation Assembler SyntaxAdd d1,d0,d2 #u5 Da,DbDa,Da Data Register Pairs Add2 d0,d1 ADD2Add Two 16-Bit Values Dalu Single Source Data Register JJJAdda #s16,rx,Rn AddaAdd AGU Adda #u5,RxAddress Register Adda r0,r1#s16 Rrrr AGU Source RegisterAGU Source/Destination Register Rx1 + Rx → Rx ADDL1A Add With One-Bit Arithmetic Shift Left ADDL1ASource Operand AGU Addl1a r0,r1ADDL1A rx,Rx ADDL2A rx,Rx ADDL2A Add With Two-Bit Arithmetic Shift Left ADDL2AAddl2a r0,r1 Rx2 + Rx → RxADDL2A rx,Rx Addnc.w #$ca3e,d1,d2 ADDNC.WAdd Without Changing ADDNC.W Carry Bit DaluInstruction Words Cycles Type RndDa + Dn → Dn ADRAdd and Round Dalu Adr d3,d4ADR Da,Dn Da,Dn Bitwise and Dalu#0u16,Da,Dn #u16$0000,Da,Dn#$ff2e0000,d2,d1 D2,d1#$0ff2e,d2,d1 #u16 #0u16#u16$0000 #u16,DR.L #$a70e,d1.h#u16 DR.L → DR.L #u16 DR.H → DR.HCycles Type Opcode Data/Address RegisterAND.W #u16,SP+s16 AND.W #u16,RnAND.W #u16,SP-u5 AND.W #u16,a16And.w #$54a1,r7 S16 A16ASL Da,Dn ASLAsl d0,d1 Da 1→ DnASL Da,Dn ASL2A Rx ASL2AAsl2a r0 Rx2 → RxAsla Rx AslaAsla r0 Rx1 → RxAsll Da,Dn AsllMultiple-Bit Arithmetic Shift Left Dalu Asll #u5,DnAsll d0,d1 Asll Da16 → Dn AslwWord Arithmetic Shift Left 16 Bits Dalu Aslw Aslw d0,d1Aslw Da,Dn ASR Da,Dn ASRAsr d5,d3 Da1 → DnRegister/Memory Address Before After Asra Rx AsraAsra r2 Multiple-Bit Arithmetic Shift Right Dalu AsrrAsrr d3,d5 Asrr #$3,d5#u5 Asrw Da,Dn Asrw d5,d0Asrw Da,Dn BF label If T==0, then PC + displacement → PCBranch If False AGU BF lblInstruction Words Cycles1 Type Opcode Displacement labelBFD label BFD Branch If False Using a Delay Slot AGU OperationBFD BFD lblLabel Displacement ~DR.Hi → DR.Hi Bmchg~C1.Hi → C1.Hi i denotes bits=1 in #u16 ~C1.Li → C1.LiControl Registers Bmchg #$f0f0,d1.hClears the Ln bit in the destination data register Iiiiiiiiiiiiiiii 16-bit unsigned immediate data Bit-Masked Change a BMCHG.WBMCHG.W #u16,SP-u5 Bmchg.w #$661f,$800cBit signed SP address offset Bmclr #u16,DR.H Bmclr Bit-Masked Clear a 16-Bit Operand BMU Bmclr OperationBmclr #u16,C1.H Bmclr #u16,C1.LBmclr #$b646,d7.l #u16 Bit Operand in Memory BMU Operation Assembler Syntax BMCLR.WBit-Masked Clear a BMCLR.W #u16,SP-u5 Bmset #u16,DR.L Bmset #u16,C1.HBmset #u16,C1.L Bmset #u16,DR.HBmset #$2436,d1.l Bit Operand in Memory BMU BMSET.W$800C Bmset.w #$f111,$800cRegister/Memory Address Before Immediate Bmtset #u16,DR.L BmtsetBmtset #$111f,d1.l Bmtset #u16,DR.HBmtset #$4238,d4.l Bit-Masked Test and Set a BMTSET.W BMTSET.WBmtset.w #$4328,$c BMTSET.W #u16,SP-u5 Bmtstc #$8a59,d7.h Bmtstc$0$0024A60000 $00E40000 BMTSTC.W #u16,Rn BMTSTC.WBMTSTC.W #u16,SP-u5 BMTSTC.W #u16,SP+s16Bmtstc.w #$8A59,r0 BMTSTC.W #u16,SP-u5 Bmtsts #u16,DR.L BmtstsBmtsts #u16,C1.L Bmtsts #u16,DR.HBmtsts #$24a6,d7.h BMTSTS.W #u16,Rn BMTSTS.WBMTSTS.W #u16,SP-u5 BMTSTS.W #u16,SP+s16Bmtsts.w #$0428,r0 BMTSTS.W #u16,SP-u5 BRA label PC + displacement → PCBRA Branch AGUAAAAAAAAAA0 Source Code Comments BradBrad label $0000 000A $0000 000E → LFn PC + displacement → PCBreak Break labelEncoding is the displacement with bit Bsr label BSRBranch to Subroutine AGU Status and Conditions Changed by Instruction None ExampleBSR Bsrd label PC + displacement → PC, next* PC→RASNext* PC → SP SR → SP + 4 SP + 8 → SP Bsrd label BT label If T==1, then PC + displacement → PCBranch If True AGU BT lblRegister/Memory Address Before BT After BTD label BTD Branch If True Using a Delay Slot AGU OperationBTD BTD lbl$0035 $0000 $0006 $002A $001A $0016 CLB Da,Dn CLBCount Leading Bits Dalu Clb d3,d7CLB Da,Dn → Dn CLRClear a Data Register Dalu Clr d1Source Data Register Destination Data RegisterIf Da == Dn, then 1→ T, else 0 → T CmpeqCompare for Equal Dalu Cmpeq d2,d3118 CMPEQ.W #u5,Dn CMPEQ.WCMPEQ.W Compare for Equal Dalu Cmpeq.w #$5,d3CMPEQ.W Cmpeqa rx,Rx Cmpeqa Compare for Equal AGU Cmpeqa OperationCmpeqa r1,r2 If rx == Rx, then 1 → T, else 0 → TCmpeqa rx,Rx Dn Da → T CmpgtCompare for Greater Than Dalu Cmpgt d2,d3124 CMPGT.W #s16,Dn CMPGT.WCmpgt.w #$8002,d2 CMPGT.W #u5,DnCMPGT.W #u5,Dn Rx rx → T CmpgtaCompare for Greater Than AGU Cmpgta Cmpgta r2,r3Cmpgta rx,Rx Cmphi Da,Dn CmphiUnsigned Compare for Higher Dalu Cmphi Cmphi d1,d0130 Cmphia rx,Rx CmphiaUnsigned Compare for Higher AGU Cmphia Cmphia r0,r1Cmphia rx,Rx Label ContContinue to the Next Loop Iteration AGU LabelCycles1 Type Opcode Contd label ContdCycles1 Type Debug Enter Debug Mode AGUDebug Debugev Signal a Debug Event AGU DebugevRx 1 → Rx DecaDecrement a Register AGU Deca r0Bit unsigned immediate data = 1, set by the assembler Deceq Dn Deceq d7Dn 1 → Dn if Dn==0, then 1→ T, else 0 → T 142 Deceqa Rx Deceqa Decrement and Set T If Equal Zero DeceqaDeceqa r0 Rx 1 → Rx if Rx==0, then 1 → T, else 0 → T Deceqa RxDecge Dn DecgeExample decge Dn 1 → Dn Dn≥0 → TSC140 DSP Core Reference Manual 145 Decgea Rx DecgeaDecgea r4 Rx 1 → Rx Rx ≥ 0 → TDecgea Rx SR18 Determines execution working modeSR19 Set disable interrupt bit → DIPage Then 2 * Dn + C + Da & $FF Ffff 0000 → Dn DIVDivide Iteration Dalu If Dn39 ⊕ Da39 =Div d2,d1 DIV Da,Dn Dc signed, Dd signed DmacssDmacss d2,d3,d5 Dn16 + Dc.H * Dd.H → DnDmacss Dc,Dd,Dn 1 1 Dmacsu d2,d3,d5 Dmacsu Multiply Signed By Unsigned and DmacsuAccumulate With Right Shifted Data Register Dalu 156 DOENn #u16 Do Enable Long Loop AGUDoen2 d0 DOENn #u6#u6 Loop IdentifierDOENSHn #u16 Do Enable Short Loop AGU DOENSHnDoensh2 d0 DOENSHn #u6$00E4 $A0E4 DOSETUPn label Setup Long Loop DOSETUPnDosetup1 label PC + displacement → SAnEncoding is the displacement with → DI SR19 Clears disable interrupt bit164 Da ⊕ Dn → Dn EORBitwise Exclusive or Dalu Eor d4,d5EOR Da,Dn EOR #u16,DR.H Eor #$5,d5.lEOR #u16,DR.L EOR #u16,DR.L EOR #u16,DR.H Bitwise Exclusive or on EOR.WEor.w #$aaaa,r0 Extract #U6,#u6,Db,Dn ExtractExtract Extract Signed Bit Field Dalu Extract #$c,#$e,d2,d4Jjj Single Source/Destination Data Register Extractu #U6,#u6,Db,Dn ExtractuExtractu Extract Unsigned Bit Field Extractu #$c,#$e,d2,d4Extractu #U6,#u6,Da,Dn IADDNC.W #s16,Dn Iaddnc.w #$a002,d2Else treat as NOP Execute group/subgroup unconditionally Conditionally Execute a Group or Subgroup Prefix IFcIf T == Then execute group/subgroup Else treat as NOP If T == Then execute group/subgroupCcc Conditional execution of the entire execution set Ift move.w #$ffff,d0Illegal Illegal Imac ±Da,Db,Dn ImacImac d4,d5,d6 Dn ± Da.L * Db.L → DnAccumulation Notation Imac -d4,d5,d6182 Imaclhuu Da,Db,Dn Dn + Da.L * Db.H → DnImaclhuu Da,Db,Dn Imacus d3,d4,d0 ImacusInteger Multiply Accumulate Unsigned By Signed Dalu Operation Assembler Syntax$0002 x -64 $FFC0 -128 $FF80 +0 $0000 -128 $FF80 Impy Da,Db,Dn ImpyInteger Multiply Dalu Da.L * Db.L → DnD1,D1 D3,D3 D5,D5 D7,D7 IMPY.W #s16,Dn Impy.w #$fffe,d3#s16 * Dn.L → Dn +16 Da.H * Db.L → Dn ImpyhluuInteger Multiply Upper Impyhluu Impyhluu d4,d3,d0Impyhluu Da,Db,Dn Register Bit Name Description Address ImpysuImpysu d3,d5,d1 Impysu Da,Db,DnImpysu Da,Db,Dn Impyuu Da,Db,Dn ImpyuuImpyuu d5,d3,d1 196 INC Dn INCINC Increment a Data Register By One Dalu Operation Inc d0Inc d15 INC.F Dn Inc.f d15Dn + $0000010000 → Dn INC.F Dn Rx + 1 → Rx IncaIncrement Register AGU Inca r0Inca Rx Insert #U6,#u6,Db,Dn InsertInsert Bit Field Dalu Insert #12,#22,d6,d7Insert #U6,#u6,Db,Dn JF Rn Jump If False AGUJF lbl JF labelBit absolute long address JFD Rn JFDJFD label $00E0 $00 0000 $0000 $00 0000 002A $00 0000 001A JMP label JMPJump AGU Jmp labelJMP label Jmpd label Jump Using a Delay Slot AGUJmpd Example jmpd lblAaaaaaaaaaaaaaaaAAAAAAAAAAAAAAAAA JSR label JSRJump to Subroutine AGU Jsr r6Absolute long address Jsrd Rn Example jsrd r6Next* PC → RAS Rn → PC Jsrd labelJsrd label JT Rn Jump If True AGUJt r0 JT labelJT label JTD label Jump If True Using Delay Slot AGUJTD Example jtd r0JTD label LCn 1 → LCn Else next PC → PC → LFn → SLF LPMARKx End-of-Loop Mark Prefix LPMARKx OperationIf LCn Then SAn → PC LCn 1 → LCn Else next PC → PC → LFn If LCn Then SAn → PCLCn Description Status and Conditions that Affect Lpmark ExecutionTable A-17. Combinations of LPMARKx Use LFnInstruction Disassembled Instruction Comments Status and Conditions Changed by Lpmark ExecutionPrefix Formats and Opcodes Insertion of lpmarkb by assemblerLsll Da,Dn LsllMultiple-Bit Bitwise Shift Left Dalu Lsll d4,d2$00E4 $0$FF 8765 Dn1 → Dn 0 → Dn39 LSRBitwise Shift Right One Bit Dalu Lsr d4Rx1 → Rx 0 → Rx31 LsraBitwise Shift Right By One Bit AGU Lsra r2Lsrr #u5,Dn LsrrMultiple-Bit Bitwise Shift Right Dalu Lsrr Da,DnBefore After Lsrr d4,d2Bit unsigned immediate data Lsrw Da,Dn LsrwWord Bitwise Shift Right Dalu Lsrw d4,d2Lsrw Da,Dn MAC #s16,Da,Dn MACSigned Fractional Multiply-Accumulate Dalu Mac d4,d5,d6Mac #$1000,d5,d6 SC140 DSP Core Reference Manual 235 Macr ±Da,Db,Dn MacrMacr d4,d5,d6 RndDn ± Da.H * Db.H → Dn000 0000 0000 1000 $0008 Instruction Formats and Opcodes 238 Macsu Dc,Dd,Dn MacsuMacsu d0,d1,d4 Dn + Dc.H * Dd.L → Dn111 1111 1111 1111 $FFFF Instruction Formats and Opcodes Macus Dc,Dd,Dn MacusDn + Dc.L * Dd.H → Dn Macus Dc,Dd,Dn Macuu d2,d3,d1 MacuuFractional Multiply-Accumulate Unsigned By Unsigned Dalu Operation Assembler SyntaxMacuu Dc,Dd,Dn PC → trace buffer MarkPush the PC into the Trace Buffer AGU Mark If Dg Dh, then Dg → Dh MAXTransfer Maximum Signed Value Dalu Max d0,d4If Dg.L Dh.L, then Dg.L → Dh.L MAX2Max2 d0,d4 If Dg.H Dh.H, then Dg.H → Dh.H00 D0,D4 01 D1,D5 10 D2,D6 11 D3,D7 248 MAX2VIT Da,Db MAX2VITIf Da.L Db.L, then 0 → VFn, Da.L → Db.L Else 1 → VFnMax2vit d4,d2 Maxm Dg,Dh Maxm Transfer Maximum Absolute Value Dalu Maxm OperationMaxm d2,d6 00 D0,D4 01 D1,D5 10 D2,D6 11 D3,D7 252 MIN Dg,Dh MINTransfer Minimum Signed Value Dalu Min d1,d5Description MOVE.2FMove Two Fractional Words from MOVE.2F Memory to a Register Pair AGUMOVE.2F EA,DaDb DaDb Data Register PairsMOVE.2L DaDb,EA MOVE.2L EA,DaDb MOVE.2LMove.2l d0d1,r0 Da,Db ↔ EARead/Write Notation MOVE.2W EA,DaDb MOVE.2W DaDb,EA MOVE.2WMove.2w d0d1,r0 EA ↔ DaDb$FF Ffff AF44 EA → DaDbDcDd MOVE.4FMove Four Fractional Words from MOVE.4F Memory to a Register Quad AGUDaDbDcDd Data Register Quad Move.4f r0,d0d1d2d3MOVE.4W EA,DaDbDcDd MOVE.4W DaDbDcDd,EA MOVE.4WEA ↔ DaDbDcDd Move.4w d0d1d2d3,r0 Byte Move AGU MOVE.BMOVE.B DR,SP+s15 Move.b d3,r7+$3MOVE.B DR,ea MOVE.B SP+s15,DRMOVE.B a16,DR S15 A32To/from Memory AGU Operation Assembler Syntax MOVE.FMove Fractional Word MOVE.F Db,ea Move.f $54,d10MOVE.F SP+s15,Db MOVE.F #s16,Db MOVE.F a16,Db SC140 DSP Core Reference Manual 271 Move Long Word AGU MOVE.LGeneral Registers Cccc#u32 #s32MOVE.L Da.EDb.E,SP+s15 MOVE.L SP+s15,De.EMOVE.L Da.EDb.E,a32 Move.l d0.ed1.e,$1224MOVE.L SP+s15,Do.E MOVE.L a32,De.EDa.EDb.E ff Data Register Extension Pair Data Register278 Move Long AGU MOVE.L Rn+s15,DR MOVE.L DR,Rn+s15 MOVE.L a32,DR MOVE.L DR,a32MOVE.L a16,C4 MOVE.L C4,a16 MOVE.L Rn+u3,DR MOVE.L DR,Rn+u3MOVE.L SP+s15,C4 MOVE.L C4,SP+s15 Move.l d0,r0MOVE.L EA,DR MOVE.L DR,EA Read/Write Notation Rrr Address RegisterUnsigned 3-bit offset MOVE.W #s16,SP-u5 MOVE.W #s7,DRMOVE.W #s16,C4 MOVE.W #s16,a16Move.w #$0050,r7 MOVE.W #s7,DR MOVE.W #s16,C4 Sa16 #s7MOVE.W a16,C4 MOVE.W C4,a16 MOVE.WMove Integer Word AGU MOVE.W a32,DR MOVE.W DR,a32MOVE.W Rn,C3 MOVE.W C3,Rn MOVE.W Rn+u3,DR MOVE.W DR,Rn+u3MOVE.W Rn+s15,DR MOVE.W DR,Rn+s15 MOVE.W Rn+Rr,DR MOVE.W DR,Rn+RrMove.w d1,r7+4 MOVE.W a32,DR MOVE.W DR,a32 Write Sss0 Movef Rq,Rn MOVEc Conditional Address Register Move AGU MOVEc OperationMovet r0,r1 Movet Rq,RnQqq Address Register MOVES.2F DaDb,EA MOVES.2F Move Two Fractional Words to MOVES.2FDaDb → EA $7FFF $7EAC MOVES.4F DaDbDcDd,EA Moves.4f d0d1d2d3,r0DaDbDcDd → EA $7FFF Move Fractional Word to MOVES.FMoves.f d0,r0 MOVES.F Db,a16 304 Moves.l d0,r0 MOVES.LMove Long to Memory With Scaling and Saturation AGU OperationMOVES.L Db,EA Memory AGU Operation Assembler Syntax MOVEU.BMove Unsigned Byte from Moveu.b $0053,d10 MOVEU.B a16,DR 310 MOVEU.L #u32,Db MOVEU.LMoveu.l #$fffffff8,d3 #u32 → Db31IIIIIIIIIIIIIIII16 MOVEU.W #u16,Db.H Moveu.w #$2345,d10.l#u16 → Db3116 #u16 → Db150Iiiiiiiiiiiiiiii Bit unsigned immediate data #u16Memory to a Register AGU Operation MOVEU.WMove Unsigned Word from Moveu.w r7+2,d10 MOVEU.W a16,C4 318 MPY Da,Db,Dn MPYMpy d4,d5,d6 Da.H * Db.H → DnMpy d6,d6,d7 SC140 DSP Core Reference Manual 321 Mpyr Da,Db,Dn MpyrMpyr d4,d5,d6 RndDa.H * Db.H → DnRegister/Memory Address Before After L6D6 $0000324 Mpysu Dc,Dd,Dn MpysuMpysu d4,d5,d6 Dc.H * Dd.L → Dn326 Mpyus Dc,Dd,Dn MpyusDc.L * Dd.H → Dn 328 Mpyuu Dc,Dd,Dn MpyuuMpyuu d4,d5,d6 Dc.L * Dd.L → Dn330 NEG Dn NEGNegate Dalu Neg d3NEG Dn Nop NOPNo Operation Prefix No operation~Da → Dn NotBitwise Complement Dalu Not d4,d5SC140 DSP Core Reference Manual 335 ~DR.L → DR.L Binary Inversion of a 16-Bit Operand BMUNot D0.L Not DR.L Memory BMU Operation Assembler Syntax NOT.WBinary Inversion of a 16-Bit Operand Not.w r1 Or Da,Dn Bitwise Inclusive or DaluExample or d3,d0 Da Dn → DnSC140 DSP Core Reference Manual 341 Or #u16,DR.L Or #$0f0a,d0.l#u16 DR.L → DR.L #u16 DR.H → DR.HOr #u16,DR.L Or #u16,DR.H OR.W #u16,a16 OR.W #u16,RnOR.W #u16,SP-u5 OR.W #u16,SP+s16OR.W #u16,Rn Or.w #$f01a,r1346 POP Do SP 8 → De SP 8 → SPSP 4 → Do SP 8 → SP POP DePop d3 Extension Pairs, Even Registers, and Loop Start Registers EeeeeNSP 4 → Do NSP 8 → ΝSP NSP 8 → De NSP 8 → ΝSPPopn Do Popn d6.ed7.ePopn De 352 Push Do De → SP SP + 8 → SPDo → SP + 4 SP + 8 → SP Push DePush d0.ed1.e SC140 DSP Core Reference Manual 355 Pushn Do De → NSP NSP + 8 → ΝSPDo → NSP + 4 NSP + 8 → ΝSP Pushn DePushn d0.ed1.e Pushn De Pushn Do RND Da,Dn RNDRound Dalu RndDa → DnRnd d2,d1 Rnd d1,d5RND Da,Dn ROL Dn Rol d5Dn3801 → Dn391 Dn39 → C → Dn0ROL Dn ROR Dn Ror d15Dn39-11 → Dn38-0 → Dn39 Dn0 → CROR Dn SP 4 → SR SP 8 → SP → Nmid RTESP 8 → PC Instruction Words Cycles1 Type RteRted Trap Example rtedIf RAS valid, then RAS → PC RTSReturn From Subroutine AGU RtsRTS Rtsd RtsdRtsd Register Address Bit Name Description EMR3 Restore PC from Stack AGURtstk ClearedExample rtstk SP 8 → SP RtstkdSP 8 → PC Example rtstkd SAT.F Da,Dn SAT.FSat.f d2,d3 If Da $007FFFFFFF then $007FFF0000 → DnSAT.F Da,Dn SAT.L Dn SAT.LSat.l d6 SC140 DSP Core Reference Manual 381 SBC Dc,Dd SBCSubtract With Borrow Dalu Db Dc C → DdSBC Dc,Dd RndDn Da → Dn SBRSubtract And Round Dalu Sbr d3,d00010 1010 1110 0111 0000 0000 1000$2AE7 Skipls label If LCn ≤ Then PC + displacement → PC Skipls label → LFnSkipls Skipls labelSkipls label Enter the stop processing state StopStop Stop Instruction Processing AGU Operation SUB #u5,Dn SUBSubtract Dalu Sub d1,d0,d2Sub d0,d1,d2 SC140 DSP Core Reference Manual 391 Sub2 d0,d1 SUB2Subtract Two 16-Bit Values Dalu SUB2 Da,Dn Suba #u5,Rx SubaSubtract AGU Suba r1,r0Suba Dn Da → Dn SublShift Left and Subtract Dalu Subl d0,d1$0$FF Ffff Fffe SUBNC.W #s16,Dn SUBNC.WSubnc.w #$15,d0 Dn #s16 → DnSUBNC.W #s16,Dn Sxt.w d3,d2 Sign-Extension DaluSxt.b d3,d0 Sxt.l d3 Sxta.w r3 Sign-Extension AGUSxta.b r3,r1 SXTA.B Tfr d7,d6 TFRTransfer Data Register to Data Register Dalu Tfr d15,d14TFR Da,Dn Tfra rx,Rx TfraTfra r0,r1 Rx → RxSC140 DSP Core Reference Manual 407 Else Rn → ESP To/from a Register AGUIf Srexp = Then NSP → Rn Else ESP → Rn If Srexp = Then Rn → NSPTfra r0,osp Tfrt Da, Dn Tfrt d14,d15If T=1, then Da → Dn If T=0, then Da → DnTfrt TRAPn Trap Execute a Software Exception AGU Trap OperationTrap If Dn == 0, then 1 → T, else 0 → T TsteqTest for Equal to Zero Dalu Tsteq d1TSTEQA.W Rx TSTEQA.x Test for Equal to Zero AGU TSTEQA.x OperationTsteqa.w r4 Tsteqa.l r1TSTEQA.W TSTEQA.L If Dn = 0, then 1 → T, else 0 → T TstgeTest for Greater Than Or Equal to Zero Dalu Tstge d4TESTGEA.L Rx Tstgea.l r7If Rx ≥ 0, then 1 → T, else 0 → T TSTGEA.L Rx Tstgt Dn Tstgt Test for Greater Than Zero Dalu Tstgt OperationTstgt d6 If Dn 0, then 1 → T, else 0 → ΤTstgta Rx Tstgta Test for Greater Than Zero AGU Tstgta OperationTstgta r2 If Rx 0, then 1 → T, else 0 → ΤViterbi Shift Left Move AGU Word Big Endian ModeLittle Endian Mode VSLVSL.2F D1D3,Rn+N0 VSL.4W D2D6D1D3,Rn+N0VSL.4F D2D6D1D3,Rn+N0 VSL.2W D1D3,Rn+N0After Big Endian Vsl.2w d1d3,r0+n0After Little Endian VSL.4W Response Enters the low-power standby Wait processingWait StateWait Zxt.w d3,d6 Zero Extension DaluZxt.b d2,d5 Zxt.l d0 Zxta.w r4 Zero Extension AGUZxta.b r3,n2 ZXTA.B ZXTA.x 432 Appendix B StarCore Registry Using the StarCore RegistrySet Version SoC / platform Table B-1. Scid AssignmentsHex Bits Instruction Cores Example Index Ecnten Eeddef Eselctrl 4-26ESELDI 4-26ESELDM 4-26ESELDTB 4-26 Eseletb MOVES.F A-299 MOVES.L A-301 MOVEU.B A-307 Macsu A-239 Macus A-241 Macuu A-243 Mpysu A-325 Mpyus A-327 Mpyuu A-329 Rtstk A-374 Index Index SC140 DSP Core Reference Manual