Iss, Execute stages, Ret Instruction retire

Introduction

Iss

read,

address

generation, and

instruction

issue

Ex1

DC1

EX1

Ex2

DC2

EX1

Ret

Fwr

Load/store

pipeline

Data

processing

pipeline

Floating point

pipeline

Mispredicted direct branches

Exception flush and mispredicted

indirect branches

Figure 1-4 Cortex-R4F Issue and Execution pipeline stages

The names of the common pipeline stages and their functions are:

Iss	Register read and instruction issue to execute stages.
Ex	Execute stages.
Wr	Write-back of data from the execution pipelines.
Ret	Instruction retire.

The names of the load/store pipeline stages and their functions are:

DC1	First stage of data memory access.
DC2	Second stage of data memory access.

The names of the floating point pipeline stages and their functions are:

F0	Floating point register read.
F1	First stage of floating point execution.
F2	Second stage of floating point execution.
Fwr	Floating point writeback.

The pipeline structure provides a pipelined 2-cycle memory access and single-cycle load-use penalty. This enables integration with slow RAM blocks and maintains good CPI at reasonable frequencies.

ARM DDI 0363E	Copyright © 2009 ARM Limited. All rights reserved.	1-18
ID013010	Non-Confidential, Unrestricted Access

Page 39

Image 39

ARM R4F, r1p3 manual Iss, Execute stages, Write-back of data from the execution pipelines, Ret Instruction retire

Contents

Cortex-R4 and Cortex-R4F Cortex-R4 and Cortex-R4F Technical Reference ManualCopyright 2009 ARM Limited. All rights reserved Chapter Introduction Cortex-R4 and Cortex-R4F Technical Reference Manual Chapter Processor Initialization, Resets, and Clocking Chapter AC Characteristics Appendix B ECC Schemes List of Tables Adfsr and Aifsr bit functions Dirty register format, with ECC 11-12 Example interlocks 14-11 Table A-18 FPU signals Table C-1 List of Figures Cache operations C7 format for Set and Way Vector Catch Register format 11-20 Feedback on Preface Identifies the major revision of the product Using this bookAbout this book Product revision status Typographical Conventions Signals Timing diagrams Other publications Further readingThis section lists publications by ARM and by third parties ARM publications Feedback on this book FeedbackARM welcomes feedback on this product and its documentation Feedback on this product Introduction About the processor About the architecture System control coprocessor on Interrupt handling on Components of the processorThis section describes the main components of the processor Debug on Prefetch unit Data Processing UnitFloating Point Unit Load/store unit Instruction and data caches Memory Protection UnitTCM interfaces AXI slave interface Error correction and detection5 L2 AXI interfaces AXI master interface Real-time debug facilities DebugSystem performance monitoring ETM interface Low interrupt latency System control coprocessorInterrupt handling VIC port Changes Return from exception using data from the stack APB Debug interface Processor has the following interfaces for external accessExternal interfaces of the processor APB Debug interface ETM interface Test interface Shutdown mode Power managementRun mode Standby mode Configurable options Configurable optionsAtcm MPU BtcmVFP FPU includeda B1TCMPCEN Configurable options at reset Feature Options RegisterAtcmpcen B0TCMPCEN Atcmrmw AtcmecenB0TCMECEN B1TCMECEN Execution pipeline stages Names of the pipeline stages and their functions areInstruction decode Write-back of data from the execution pipelines First stage of data memory accessIss Execute stages Redundant core comparison Test features Product documentation, design flow, and architecture DocumentationDesign flow Architectural information Build configurationConfiguration inputs Software configuration Advanced Microcontroller Bus Architecture protocol Processor identification Product revision information Revision field, Debug ID Register Variant field, Main ID RegisterRevision field, Main ID Register Variant field, Debug ID Register Exceptions on Programmer’s ModelRegisters on Program status registers on About the programmer’s model Thumb state Switching stateInstruction set states ARM state Operating modes Data types Little-endian format Memory formatsByte-invariantbig-endian format Little-endian format Byte-invariant big-endian format Register set Registers Register mode identifiers Mode Mode identifier General registers and program counter N, Z, C, and V bits Program status registers IT bits Q bit J bit DNM bitsGE bits Non-maskable fast interrupts on E bitA bit I and F bits Modification of PSR bits by MSR instructions PSR mode bit valuesM40 Mode Visible state registers Thumb M bits ARM DDI 0363E Exception entry and exit summary Reset on Interrupts on Aborts onExceptions Exception entry and exit summary Leaving an exception Taking an exception Reset InterruptsInterrupt request Fast interrupt request Program status registers on Non-maskable fast interruptsNon-maskable fast interrupts Interrupt controller Interrupt entry flowchart Interrupt entry sequence Precise aborts AbortsPrefetch aborts Data aborts Aborts in Strongly Ordered and Device memory Imprecise aborts Supervisor call instruction Abort handler Breakpoint instruction Undefined instruction Exception vectors Acceleration of execution environments Architecture Reference Manual Unaligned and mixed-endian data access support Big-endian instruction support Initialization on Processor Initialization, Resets, and ClockingResets on Reset modes on 2 CRS InitializationCaches on TCM on 1 MPU Caches 5 TCMPreloading TCMs DMA into TCM Write to TCM directly from debuggerPreloading TCMs with parity or ECC Using TCMs from reset NSYSPORESET ResetsNRESET PRESETDBGn Power-on reset Reset modes Processor reset Normal operationHalt operation Clock gating ClockingAXI interface clocking Clock gating AXI interface clocking System Control Coprocessor System control coprocessor functional groups About the system control coprocessor System control coprocessor register functions Function Register/operation Reference to descriptionFcse PID System performance Performance monitoring System control and configurationConfiguration Region TCM control TCM Status Cache control and configuration MPU control and configuration System performance monitor TCM control and configuration System performance monitor registers System validation ARM DDI 0363E System control coprocessor registers Register allocationRAZ Number Enable Undefined MPU Region Access Read/writeUndefined MPU Region Size Read/write Control Undefined MPU Memory Region Read/write ARM DDI 0363E Undefined C3-c11 C12 Performance Monitor Read/write Event Select Read/write Unpredictable Performance MonitorCount Undefined Writes Context ID Read/write C14 User Enable Read/writeUndefined C1-c15 C12 C0-c15 C13 RAZ,ignore 7shows the arrangement of bits in the register 2 c0, Main ID Register 3 c0, Cache Type Register Main ID Register bit functionsBits Field Function TCM Type Register bit functions Bits Field Function To access the Cache Type Register, read CP15 with4 c0, TCM Type Register Cache Type Register bit functions Bits Field Function TCM Type Register bit functions To access the TCM Type Register, read CP15 withTo access the MPU Type Register, read CP15 with 5 c0, MPU Type Register Processor Feature Registers C0, Processor Feature Register 0, PFR06 c0, Multiprocessor ID Register Processor Feature Register 1 bit functions To access the Processor Feature Register 0 read CP15 withC0, Processor Feature Register 1, PFR1 Processor Feature Register 0 bit functions 3124 Reserved To access the Processor Feature Register 1 read CP15 with8 c0, Debug Feature Register Debug Feature Register 0 bit functions C0, Memory Model Feature Register 0, MMFR0 To access the Debug Feature Register 0 read CP15 with9 c0, Auxiliary Feature Register Memory Model Feature Registers 10 Memory Model Feature Register 0 bit functions C0, Memory Model Feature Register 1, MMFR1 11 Memory Model Feature Register 1 bit functions 16 Memory Model Feature Register 1 format C0, Memory Model Feature Register 2, MMFR2 WFIDMB C0, Memory Model Feature Register 3, MMFR3 13 Memory Model Feature Register 3 bit functions3112 Reserved Instruction Set Attributes Registers C0, Instruction Set Attributes Register 0, ISAR014 Instruction Set Attributes Register 0 bit functions 20 Instruction Set Attributes Register 1 format C0, Instruction Set Attributes Register 1, ISAR1 C0, Instruction Set Attributes Register 2, ISAR2 ITEIndicates support for if then instructions 16 Instruction Set Attributes Register 2 bit functions PSRIndicates support for PSR instructions C0, Instruction Set Attributes Register 3, ISAR3 17 Instruction Set Attributes Register 3 bit functionsThumb instruction sets 23 Instruction Set Attributes Register 4 format C0, Instruction Set Attributes Register 4, ISAR4 C0, Instruction Set Attributes Registers 12 c0, Current Cache Size Identification Register18 Instruction Set Attributes Register 4 bit functions 8KB 4KB 3130 Reserved 13 c0, Current Cache Level ID Register 15 c1, System Control Register 14 c0, Cache Size Selection Register Nmfi 23 System Control Register bit functionsAFE TRE = strict alignment fault checking enabled Enables L1 data cache= data caching disabled. This is the reset value = data caching enabled Auxiliary Control Registers C1, Auxiliary Control Register24 Auxiliary Control Register bit functions Deolp AxiscenAxiscuen Dilsm Dnch RsdisDbwr Dlfo C15, Secondary Auxiliary Control Register 25 Secondary Auxiliary Control Register bit functions UFC DoofmacsIXC OFC Atcmecc 17 c1, Coprocessor Access RegisterPrimary input RMWENRAM1 defines the reset value Primary input RMWENRAM0 defines the reset value Fault Status and Address Registers C5, Data Fault Status RegisterAll other encodings for these FSR bits are Reserved C5, Instruction Fault Status Register 28 Data Fault Status Register bit functionsTo use the Dfsr read or write CP15 with 29 Instruction Fault Status Register bit functions To access the Ifsr read or write CP15 withC5, Auxiliary Fault Status Registers There are two auxiliary fault status registers = Btcm C6, Data Fault Address Register30 Adfsr and Aifsr bit functions = Atcm 19 c6, MPU memory region programming registers C6, Instruction Fault Address Register C6, MPU Region Base Address Registers C6, MPU Region Size and Enable Registers C6, MPU Region Access Control Registers 158 Sub-region disable32 Region Size Register bit functions TEX 33 MPU Region Access Control Register bit functions 35 MPU Memory Region Number Register bit functions 34 Access data permission bit encodingC6, MPU Memory Region Number Register UNP Cache operations Point of Coherency PoCPoint of Unification PoU Set and Way format Invalidate and clean operations 37 Widths of the set field for L1 cache sizes Size Set 37shows the cache sizes and the resultant bit range for SetAddress format 36 Functional bits of c7 for Set and Way 38 Functional bits of c7 for address format Bits Field Data Synchronization Barrier operationData Memory Barrier operation 21 c9, Btcm Region Register To access the Btcm Region Register, read or write CP15 with 22 c9, Atcm Region Register39 Btcm Region Register bit functions 40 Atcm Region Register bit functions To access the Atcm Region Register, read or write CP15 with23 c9, TCM Selection Register 24 c11, Slave Port Control Register RAZ/UNP 25 c13, Fcse PID Register26 c13, Context ID Register 312 Reserved 27 c13, Thread and Process ID Registers C15, nVAL IRQ Enable Set Register Validation RegistersCcnt overflow IRQ request Ccnt overflow FIQ request C15, nVAL FIQ Enable Set Register C15, nVAL Reset Enable Set Register C15, nVAL Debug Request Enable Set RegisterCcnt overflow reset request Ccnt overflow debug request C15, nVAL IRQ Enable Clear Register 48 nVAL IRQ Enable Clear Register format C15, nVAL FIQ Enable Clear Register 303 Reserved UNP or Sbzp C15, nVAL Reset Enable Clear Register 51 nVAL Debug Request Enable Clear Register format C15, nVAL Debug Request Enable Clear Register B0000 4kB B0001 8kB B0011 16kB B0111 32kB B1111 64kB C15, nVAL Cache Size Override Register50 nVAL Cache Size Override Register 318 Reserved 53 Correctable Fault Location Register cache Correctable Fault Location Register 52 Correctable Fault Location Register cache C15, Build Options 2 Register To access the Build Options 1 Register, write CP15 withBuild Options Registers C15, Build Options 1 Register Atcmes 55 Build Options 2 RegisterNoicache Nodcache Nompu BtcmesNoie Nofpu Axibusparity To access the Build Options 2 Register, write CP15 withNoharderrorcach Dcachees Prefetch Unit About the prefetch unit Disabling program flow prediction Branch prediction Configuring the branch predictor Branch predictorIncorrect predictions and correction Return stack Events and Performance Monitor Event Bit position Update ValueAbout the events Event bus interface bit functions ETMEXTOUT1 ETMEXTOUT0 Dual issue case B1, B2, F2 load/store, F2D Non-cacheable access on AXI master busInstruction cache access But with different attributes Dual issue case a branch TCM correctable ECC error reported by prefetch unit Yes TCM correctable ECC error reported by load/store unit Yes About the PMU 2shows how the bit values correspond with the Pmnc Register Performance monitoring registersPerformance monitoring registers are described 1 c9, Performance Monitor Control Register Pmnc Register bit functions 2 c9, Count Enable Set Register To access the Cntens Register, read or write CP15 with 3 c9, Count Enable Clear RegisterCntens Register bit functions Bits Field Function Cntenc Register bit functions Bits Field Function To access the Cntenc Register, read or write CP15 withCycle counter enable clear 4 c9, Overflow Flag Status Register To access the Flag Register, read or write CP15 with 5shows how the bit values correspond with the Flag Register5 c9, Software Increment Register 313 Reserved RAZ on reads, Sbzp on writes Increment Counter To access the Swincr Register, read or write CP15 with6 c9, Performance Counter Selection Register Swincr Register bit functions Bits Field Function 7 c9, Cycle Count Register 8 c9, Event Selection RegisterSEL EVTSELx Register bit functions To access the EVTSELx Register, read or write CP15 with 10 c9, User Enable Register 9 c9, Performance Monitor Count RegistersUseren Register bit functions 11 c9, Interrupt Enable Set Register Ccnt overflow interrupt enable10 Intens Register bit functions Bits Field Function 12 c9, Interrupt Enable Clear Register Ccnt overflow interrupt enable bit11 Intenc Register bit functions Bits Field Function To access the Intenc Register, read or write CP15 with Use of the event bus and counters Event bus interface MPU faults on MPU software-accessible registers onMemory Protection Unit Default memory map About the MPUTrue Subregions Memory regionsRegion base address Region size Region Region access permissionsRegion attributes Overlapping regions Example of using subregions Example of using regions that overlap TCM regions Background regions Memory types Using memory types ARM DDI 0363E TEX20, C, and B encodings Description Memory Type Shareable? Region attributes 1BB Cacheable memory policies On page 7-2shows the default memory map MPU interaction with memory systemFollowing code is an example of disabling the MPU Permission fault MPU faultsBackground fault Permission fault Alignment fault Background fault On page 4-5shows the CP15 registers that control the MPU MPU software-accessible registers Level One Memory System About the L1 memory system L1 memory system block diagram About the error detection and correction schemes Parity Bit ECC onParity Read-Modify-Write Error checking and correctionHard errors Bit ECC Error correction Correct inlineCorrect-and-retry MPU faults Fault handlingFaults Classes of fault that can occur are External faults Cache and TCM parity and ECC errorsTCM external faults Abort exceptions Fault status informationDebug events Precise and imprecise aborts Precise abort exceptions Imprecise abort exceptionsUsage models Correctable errors Types of aborts Conditions Source Precise FatalAXI ARM DDI 0363E TCM attributes and permissions About the TCMs Atcm and Btcm configuration TCM internal error detection and correctionHandling TCM parity errors on Handling TCM ECC errors on Handling TCM parity errors Handling TCM ECC errorsTCM arbitration External TCM errors TCM initializationTCM port protocol AXI slave interfaces for TCMs Store buffer About the caches Store buffer draining Cache maintenance operationsStore buffer merging Store buffer behavior Error build options Cache error detection and correction Address decoder faults Handling cache parity errorsCache parity error behavior Value Behavior Cache ECC error behavior Value Behavior Handling cache ECC errors Errors on evictions Errors on instruction cache readErrors on data cache read Errors on data cache write Invalidate all instruction cache Clean data cache by set/way Tag RAM Cache RAM organizationTag RAM Dirty RAM on Data RAM on Organization of a dirty RAM line Dirty RAMData RAM Cache sizes and tag RAM organization Tag RAM organization Nonsequential read operation performed with one RAM access Data RAM sizes without parity or ECC implemented 13 Data cache RAM bits, with parity Description 15 Data cache data RAM sizes with ECC Cache size Data RAMs Cache interaction with memory systemDisabling or enabling all of the caches Following code is an example of enabling caches Disabling or enabling error checking Disabling or enabling instruction cacheDisabling or enabling data cache MCR p15 R0, c1 Write System Control Register Internal exclusive monitor Memory types and L1 memory system behavior Data-cache error events Error detection eventsTCM error events Instruction-cache error events ARM DDI 0363E Level Two Interface About the L2 interface AXI master interface 1shows the AXI master interface attributesAttribute Value Comments Write response Identifiers for AXI bus accessesOutstanding write/read access on different IDs Outstanding write accesses with the same ID Aruserm and Awuserm encodings Eviction bufferMemory attributes Arcachem and Awcachem encodings Encoding a Meaning Memory system implications for AXI accesses AXI master interface transfers Strongly Ordered and Device transactions onLinefills on Address20 Restrictions on AXI transfersStrongly Ordered and Device transactions Non-cacheable Ldrb Ldrh from Strongly Ordered or Device memory Address30 LDR or LDM that transfers one register LDM5, Strongly Ordered or Device memory Address40 LDM that transfers five registers Strh to Strongly Ordered or Device memory Address20 Strb to Strongly Ordered or Device memory Address40 STR or STM of one register STM of seven registersFirst Wstrbm Linefills Cache line write-back evictionNon-cacheable reads 15 LDR or LDM1 from Non-cacheable Normal memory Address20 14 Ldrh from Non-cacheable Normal memory Address20 Non-cacheable or write-through writes AXI transaction splitting 20 AXI transaction splitting, data in two cache lines Normal write merging Example 9-1 Write merging ARM DDI 0363E AXI slave interface for cache RAMs AXI slave interface TCM parity and ECC support Cache parity and ECC supportAXI slave control 25 AXI slave interface attributes AXI slave characteristics Enabling or disabling AXI slave accesses Accessing RAMs using the AXI slave interface TCM RAM access on Cache RAM access on26 RAM region decode AxUSERS bit One-hot RAM select RAM selected TCM RAM accessARADDRSMSB, see Table 27 TCM chip-select decode Btcm ports Cache RAM access Memory map when accessing the cache RAMsThis section contains the following 0001 Bank 0010 0100 1000 Data RAM access0010 Bank 0100 1000 31 Cache data RAM bank/address decode Inputs ARADDRS1815 35 Data format, data cache, with ECC Data bit Description 34 Data format, instruction cache, with ECC Tag RAM access ARM DDI 0363E 43 Dirty register format, with ECC Data bit Description Dirty RAM access ARADDRS1815 = 4b1111 Other examples of accessing cache RAMs About power control on Power management on Power Control About power control Shutdown mode Run modeStandby mode Dormant mode Communication to the Power Management Controller Debug state on Cache debug on Debug Protocol converter Debug systemsDebug host Protocol converter Debug target Debug host Programming the debug unit About the debug unitHalting debug-mode debugging Monitor debug-mode debugging All other state information associated with the debug unit Coprocessor registers summary 11.3.2 CP14 access permissionsDebug register interface Coprocessor registers Debug memory-mapped registers Offset Register Access Mnemonic Description Hex NumberMemory-mapped registers Instruction Mnemonic Description Memory addresses for breakpoints and watchpoints Privilege of memory access permission Power domainsEffects of resets on debug registers APB port access permissions Other registers External debug interface access permissions RegistersLock Other Debug registers 11.4.2 CP14 c0, Debug ID Register Accessing debug registersDebug register descriptions 6shows the CP14 debug register map Debug ID Register functions WRPBRP 11.4.4 CP14 c0, Debug Self Address Offset Register To use the Debug ID Register, read CP14 c0 with11.4.3 CP14 c0, Debug ROM Address Register To use the Debug ROM Address Register, read CP14 c0 with Debug Self Address Offset Register functions Debug Self Address Offset Register format 10 Debug Status and Control Register functions 11.4.5 CP14 c1, Debug Status and Control Register Flag is set to 1 on entry to debug state Execute ARM instruction enable bit = disabled, this is the reset valueARM MOE DTR access mode Data Transfer Register Reserved RAZ Watchpoint Fault Address Register12 Watchpoint Fault Address Register functions Vector Catch Register SVC 14 Debug State Cache Control Register functions Bits Field Reset Description ValueDebug State Cache Control Register Instruction Transfer Register Debug Run Control Register 15 Debug Run Control Register functions315 Reserved Breakpoint Control Registers Breakpoint Value Registers 17 Breakpoint Control Registers functions BVR2220 Meaning +0 is accessedCorresponding instruction address breakpoint 18 Meaning of BVR bits Watchpoint Value Registers Watchpoint Control Registers19 Watchpoint Value Registers functions Bits Description 20 Watchpoint Control Registers functions 11 Watchpoint Control Registers format Operating System Lock Status Register Accessed 21 OS Lock Status Register functions Authentication Status Register 23 Prcr functions Device Power-down and Reset Control RegisterDevice Power-down and Reset Status Register = Dbgnopwrdwn is LOW. This is the reset value 24 Prsr functions 15 Prsr format Processor ID Registers Management registers Claim Registers Claim Tag Set Register27 Claim Tag Set Register functions Bits Field Function Lock Status Register Lock Access Register318 Reserved RAZ or Sbzp Claim tag clear Reset value is Claim Tag Clear Register Device Type Register Debug Identification Registers30 Device Type Register functions Bits Value Description 32 Fields in the Peripheral Identification RegistersField Size Description 33 Peripheral ID Register 0 functions 37 Peripheral ID Register 4 functions 34 Peripheral ID Register 1 functions35 Peripheral ID Register 2 functions 36 Peripheral ID Register 3 functions 1022 1020Component Identification Register 1021 Software debug event Debug events Watchpoint debug events Halting debug eventBehavior of the processor on debug events Debug event priority Debug exception Four CP15 registers that record abort information are 40shows the values in the link register after exceptionsFollowing sections describe Effect of debug exceptions on CP15 registers and Wfar Avoiding unrecoverable states 41 Read PC value after debug state entry Debug statePrivilege on Entering debug state 41 Read PC value after debug state entry Debug event Behavior of the PC and Cpsr in debug state Privilege Accessing registers and memoryExecuting instructions in debug state Writing to the Cpsr in debug state Exceptions in debug state Coprocessor instructionsEffect of debug state on non-invasive debug Effects of debug events on processor registers Leaving debug state Precise Data abortImprecise Data Abort Imprecise Data Aborts on entry and exit from debug state Sets the DSCR1 core restarted flag to Cache coherency in debug state Cache debugThis section describes cache debug. It consists Cache pollution in debug state External debug interface APB signalsMiscellaneous debug signals This section describes the miscellaneous debug signals Non-invasive debug permitted Authentication signalsChanging the authentication signals 42 Authentication signal restrictions Dbgen a Issue an Instruction Synchronization Barrier ISB instruction Example 11-1 Executing an ARM instruction through the ITR Using the debug functionality Debug communications channel Rules for accessing the DCC Example 11-4 Target to host data transfer host end Software access to the DCCDebugger access to the DCC Example 11-5shows the code for host-to-target data transfer Example 11-6 Polling the DCC host end Programming breakpoints and watchpointsThis section describes the following operations Programming simple breakpoints and the byte address select Example 11-7 Setting a simple breakpoint Setting a simple aligned watchpoint Example 11-8 Setting a simple aligned watchpoint Setting a simple unaligned watchpoint Not required Example 11-9 Setting a simple unaligned watchpoint45shows some examples Single-stepping Debug state entry Example 11-10 Single-stepping off an instructionExample 11-11 Entering debug state Example 11-12 Leaving debug state Debug state exit Example 11-13 Reading an ARM register Accessing registers and memory in debug stateThis section describes the following Reading and writing registers through the DCC Example 11-16shows the code for reading the Cpsr Reading the PC in debug stateExample 11-15shows the code to read the PC Reading the Cpsr in debug state Example 11-17 Writing the Cpsr Example 11-19 Checking for an abort after memory accessReading memory Example 11-18shows the code for reading a byte of memory Example 11-21 Reading a word of memory Example 11-20 Reading a block of bytes of memory Fast register read/write Example 11-22 Changing the DTR access modeExample 11-23 Reading registers in stall mode Example 11-24 Writing registers in stall mode Example 11-25 Reading a block of words of memory Fast memory read/write Example 11-27 Reading a coprocessor register Accessing coprocessor registers 11-70 Debugging systems with energy management capabilities Emulating power down 11-72 FPU Programmer’s Model About the FPU programmer’s model FPU functionalityAbout the VFPv3-D16 architecture FPU views of the register bank General-purpose registers System registers VFPv3 architecture describes the following system registers1shows the VFP system registers in the Cortex-R4F FPU This is a change in VFPv3 compared to VFPv2 All hardware ID information is privileged access onlyFpsid is privileged access only Mvfr registers are privileged access only Rmode Floating-Point Status and Control Register, FpscrFpscr Register bit functions DNM DNM IXE RAZ UFE OFE DZE IOE IDC Floating-Point Exception Register, FpexcLEN IDE RAZ MVFR0 Register bit functions Media and VFP Feature Registers, MVFR0 and MVFR1Floating-Point Exception Register bit functions DEX Full denormal arithmetic supported for VFP MVFR1 Register bit functions Default NaN mode Full-compliance mode Flush-to-zero mode Default NaN modeFull-compliance mode Flush-to-zero mode NaN handling Compliance with the Ieee 754 standardComplete implementation of the Ieee 754 standard Ieee 754 standard implementation choices CDP ComparisonsUnderflow QNaN and SNaN handling Exceptions Integration Test Registers About Integration Test Registers Programming and reading Integration Test Registers Software access using APB Itmiscin Register ItctrlItetmif Itmiscout Processor integration testing 13121110 Performing integration testing Using the Integration Test Registers Itetmif Register bit assignments Bits Name Function Itetmif Register ETM interface Itmiscout Register Miscellaneous Outputs Itmiscin Register Miscellaneous InputsItmiscout Register bit assignments Bits Name Function Etmextout Integration Mode Control Register ItctrlBits Name Function Dbgrestart Intmode Bits Access Reset value Name Function7shows the fields of the Itctrl Register RAZ/SBZP Multiplies on Divide on Branches on Cycle Timings and Interlock Behavior Dual issue on Instruction execution overview About cycle timings and interlock behavior Definition of terms Following sequence where R1 is a Late Reg takes two cyclesFlag-setting instructions Conditional instructions Assembler language syntax Takes three cycles because of the result latency of R1 Register interlock examplesInstruction Behavior Sequence Takes two cycles because there are no register dependencies Data processing instructions Cycle counts if destination is not PCCycle counts if destination is the PC Example interlocks ShifterRegister controlled shifts QADD, QDADD, QSUB, and Qdsub instructions SEL Media data-processing USAD8 instruction Sum of Absolute Differences SADInstruction sequence Behavior Result of the USAD8 instruction Umlals Multiplies 14-13 Divide Branches 10 Branch instruction cycle timing behaviorExample instruction Cycles Memory Comments All MSR instructions to the Spsr Mode changingProcessor state updating instructions All MRS instructions 13shows the cycle timing behavior for loads to the PC Single load and store instructions Base register update 13 Cycle timing behavior for loads to the PCExample instruction Cycles Memory Result Comments Latency 14-19 Load and Store Double instructions Register offset, then 3-issue cycles Write-back Load and Store Multiple instructions Correct condition prediction and incorrect Load Multiples, where the PC is in the register listCorrect condition prediction and correct Return stack prediction 14-23 RFE and SRS instructions Clrex Synchronization instructions Some instructions such as cache operations take more cycles Coprocessor instructions Prefetch Abort Undefined Instruction SVC formerly SWI Miscellaneous instructions Floating-point register transfer instructions Blocking and serializingSerializing 2,2 Floating-point load/store instructionsBit aligned address Not aligned 4,5 2,3 Floating-point single-precision data processing instructions Floating-point double-precision data processing instructions Dual issue Dual issue rules Permitted combinations onDual issue rules Dual issue First instruction Second instruction Case Permitted combinations28 Permitted instruction combinations Case F2Db Case F2stbAny single-precision CDPi, excluding Multiply-accumulate instructionso Processor timing on Processor timing parameters on AC Characteristics Processor timing Clock uncertainty 50% Processor timing parametersInput port timing parameters Clock uncertainty 10% 3shows the timing parameters for the interrupt input ports 4shows the input timing parameters for the AXI master portClock uncertainty 60% Rreadys 5shows the input timing parameters for the AXI slave port 7shows the input timing parameters for the ETM input ports 6shows the input timing parameters for the debug input ports 8shows the timing parameters for the test input ports Clock uncertainty 65%Clock uncertainty 40% 11shows the timing parameters for the interrupt output ports Output ports timing parameters Write response channel Clock uncertainty 60% 13shows the timing parameters for the AXI slave output ports BRESPS10 Clock uncertainty 45% 16shows the timing parameters for the test output ports Fpidc 18shows the timing parameters for the FPU output signals 15-13 FPU signals on page A-23 Processor Signal Descriptions About the processor signal descriptions AnyFrom any clock Global signals Table A-1 Global signalsSignal Direction Clocking Description Information Configuration signalsTable A-2shows the processor configuration signals Table A-2 Configuration signals Tie High for odd parity Tie LOW for even parity RMWENRAM10b Interrupt signals, including VIC interface signals L2 interface signals Table A-4 AXI master port signals for the L2 interfaceAXI master port Protection signals provide addition information about a bus Identification tag for the write data group of signalsIdentification tag for the write response signal Identification tag for the read address group of signals AXI slave port AXI master port error detection signalsTable A-5 AXI master port error detection signals Table A-6 AXI slave port signals for the L2 interface One to 16. a four bit binary value minus one determines Protection information, privileged/normal access. AWPROT0AXI specification Protection information, privileged/normal access. ARPROT0 Table A-7 AXI slave port error detection signals AXI slave port error detection signalsATCM, one hot. AWUSERS30 signal is not part Standard AXI specification TCM interface signals Table A-8shows the Atcm port signalsTable A-9shows the B0TCM port signals Table A-10shows the B1TCM port signals Table A-10 B1TCM port signals= DMA Write parity or ECC code for B1TCM B1TCM RAM access is sequentialAddress for B1TCM data RAM Write data for B1TCM data RAM Dual core interface signals Table A-11shows the dual redundant core interface signalsTable A-11 Dual core interface signals Debug interface signals Table A-13shows the debug miscellaneous signalsTable A-12 Debug interface signals Input Tie-off Debug self-address offset valid Table A-13 Debug miscellaneous signalsInput Tie-off Debug ROM physical address valid Input Tie-off Debug self-address offset ETM interface signals Table A-14shows the ETM interface signalsTable A-14 ETM interface signals Test signals Table A-15shows the test signalsTable A-15 Test signals Table A-16shows the Mbist signals Mbist signals Table A-17shows the validation signals Validation signals Table A-18 FPU signals FPU signals ECC scheme selection guidelines on page B-2 ECC Schemes ECC scheme selection guidelines Clarified little-endian format NCPUHALT removed from timing diagram Added sectionsRevisions Clarified byte-invariant big-endian format Change Location Table C-1 Differences between issue B and issue C No technical changes Table C-2 Differences between issue C and issue D Base register write-back GlossaryAbort, and an internal or External Abort See also Data Abort, External Abort and Prefetch Abort See Advanced Microcontroller Bus Architecture See also Advanced High-performance BusSee Advanced High-performance Bus That are divisible by four and two respectively Completed transfer Active read transactionActive transfer Active write transaction Write ID width Read ID widthRead issuing capability Write ID capability See also Beat See also Burst Bus Accesses are expected to be word-alignedAccessed in parallel during a cache look-up See also Word-invariant See also Dirty See also Clean See Embedded Trace Macrocell Precision and the fraction is all zeros An instruction that is architecturally UndefinedBy individual implementations Option chosen does not affect software compatibility Result of attempting to access invalid instruction memory Serviced while normal program execution is suspendedSee also Halt mode See Cold reset See Boundary scan chain Or equal to 1 and is not restricted to being a power of twoSee Should Be One See Should Be Zero User trap handler is executed See Debug test access portIncrement of +2 Destination precision A processor Expected behavior for an unaligned accessBit for the exception is set Processor-specific. a victim is also known as a cast out Cache terminology diagram

DC2

Figure 1-4 Cortex-R4F Issue and Execution pipeline stages

Iss

Execute stages.

Write-back of data from the execution pipelines.

Ret

Instruction retire.

First stage of data memory access.

DC2

Floating point register read.

First stage of floating point execution.

Second stage of floating point execution.

Fwr

Floating point writeback.

ARM DDI 0363E