Compaq EV68A - page 206

6–22 Privileged Architecture Library Code

21264/EV68A Hardware Reference Manual

PerformanceCounter Support

Forinstructions that cause a trap, the last cycle in the window is the 2nd cycle after

the trap.Mispredicted branches are included in this category.

For nontrappinginstructions that retire, the last cycle in the window is the2nd

cycle afterthe instruction retires.

For instructionsthat abort, the last cycle in the window is the 2nd cycle after the

trapthat caused the abort.

Forinstructions that are squashed (such as TRAPB), the last cycle in the window is

approximatelythe 2nd cycle after the squashed instruction would have aborted or

retired.

Every non-squashedvalid fetched instruction either aborts or retires, but not both.

In eithercase, the instruction may also trap.

PCTR0 is disabledfrom counting until PCTR_CTL is next written.

5. InterruptPALcode

When ISUM field PC[1:0]is set, execution of PCTR0's or PCTR1's interrupt PAL-

code is performed.

6. Operatingsystem interrupt handler

The handlershould first read the IPRs in Table 6–13 and then write PCTR_CTL to set

up the next interrupt.

Table 6–13 ProfileMe Mode Returned IPR Contents

IPRName RelevantFields Meaning

PMPC[63:0] All ProfiledPC.

I_STAT ICM Instructionwas in a new Icache fill stream.

TRP Instructioncaused a trap and was not in the shadowof

ayounger trapping instruction.

MIS Conditional branch mispredict.

TRAP TYPE Exception type code.

LSO Load-storeorder replay trap.

OVR Counter0 overcount.

PCTR_CTL VAL Instructionretiredvalid.

TAK Branchdirectionif instruction is a conditional branch.

PM_STALLED Instructi on stalled for at least one cycl e between fetch

and map stages of pipeline.

PM_KILLED_BM Instructionkilled during or before cycle in which it

wasmapped.

PCTR0[19:0] Counter0 value.

PCTR1[19:0] Counter1 value.

Contents

Main 21264/EV68A Hardware Reference Manual Table of Contents Preface 1 Introduction 2 Internal Architecture 21264/EV68A Hardware Reference Manual 3 Hardware Interface 4 Cache and ExternalInterfaces 5 Internal Processor Registers 21264/EV68A Hardware Reference Manual 6 PrivilegedArchitecture Library Code 7 Initialization and Configuration 21264/EV68A Hardware Reference Manual 8 Error Detectionand Error Handling 9 Electrical Data 10 ThermalManagement 11 Testabilityand Diagnostics A Alpha Instruction Set B 21264/EV68A Boundary-ScanRegister C Serial Icache Load PredecodeValues D PALcode Restrictionsand Guidelines E 21264/EV68A-to-Bcache Pin Interface Glossary Index Figures 21264/EV68A Hardware Reference Manual Tables Page Page Page Preface Audience Content Page Terminologyand Conventions Page A capital X represents any valid value. Page Page Page Introduction 1.1 The Arc hitecture Page 1.2 21264/EV68A Microprocessor Fe atures Page Internal Architecture 2.1 21264/EV68A Microarchitecture Page InternalArchitecture Figure 21 2126 4/EV68A Block Diagram 2.1.1.2 Branch P redictor Page Page Page Page 2.1.2 Integer Execution Unit Page Page Page Page 2.1.7 SROM Interface 2.2 Pipeline Orga nization Page Page 2.2.1 Pipeline Aborts 2.3 Instructio n Issue Rules 2.3.1 Instruction Group Definitions 2.3.2 Ebox Slotting Page 2.3.3 Instruction Latencies Instruction Retire Rules 2.4 Instruction RetireRules 2.5 Retire of Op erate Instructions into R31/F31 Load Instructionsto R31 and F31 2.6 Load Instructions to R31 and F3 1 2.6.1 NormalPre fetch:L DBU, LDF,LDG, LDL, LDT,LDWU, HW_LDL In structions 2.6.2 Prefetch with Modify Intent: LDS Instruction 2.7 Speci al Cases of Alpha Instruction Execution 2.7.1 Load Hit Speculation Special Cases of AlphaInstruction Execution Special Cases of Alpha Instruct ion Execution 2.7.2 Floating-Point StoreInstructions 2.7.3 CMOV Instruction Memory andI/O Address Space Instructions 2.8 Memory and I/O Address Space Instruction s 2.8.1 MemoryAddress Space Load Instructions 2.8.2 I/OAddress Space Load Instructions Memory and I/O Address Space Instructions 2.8.3 MemoryAddress Space Store Instructions Memory andI/O Address Space Instructions 2.8.4 I/O Address Space Store Instructions MAF Memory Address SpaceMerging Rules 2.9 MAF Memory Address Space Merging Rules 2.10 InstructionOrdering Replay Traps 2.11 R eplay Traps 2.11.1 Mbox Order Traps 2.12 I/O Writ e Buffer and the WMB Instruction I/O Write Buffer and the WMB Instruction I/O Write Buffer andthe WMB Instruction Performance MeasurementSupportPerformance Counters 2.13 Performance Measurement SupportPerformanceCounters 2.14 Floating-Point Control Register 21264/EV68A Hardware Reference Manual Floating-PointControl Register The floating-pointcontrol register fields are described in Table 214. AMASK and IMPLVER Instruction Values 2.15 AMASK and IMPLVERInstruction Values Design Examples 2.15.1 AMASK 2.15.2 IMPLVER 2.16 Design Examples InternalArchitecture Design Examples Figure 213 Typical Multiprocessor Configuration Figure 212 Typical Uniprocessor Configuration Page Hardware Interface 3.1 21264/EV68A Microproces sor Logic Symbol 32 Hardware Interface 21264/EV68A MicroprocessorLogic Symbol Figure 31 21264/EV68A Microprocessor Logic Symbol 3.2 21264/EV68ASignal Names and Functions Table31 defines the 21264/EV68A signal types referred to in this section. Page Page Table33 lists signals by function and provides an abbreviated description. Page 3.3 Pin Assignments Page Page Hardware Interface Table 35 Pin Li stSo rted by PGA Location Table 34 PinList Sorted by Signal Name (Continued) Page Page Hardware Interface Table36 lists the 21264/EV68A ground and power (VSS and VDD, respectively) pin list. Table 36 Ground and Power (VSS and VDD) Pin Lis t Hardware Interface Mechanical Specifications 3.4 Mechanical Specifications Figure 32 Package Dimensions Page Page Page Cache and External Interfaces 4.1 Introduct ion to the External Interfaces Introductionto the External Interfaces Cache and External Interfaces Introduction to the Externa l Interfaces x Figure 41 2126 4/EV68A System and Bcache Interfaces 4.1.1 System Interface x 4.2 Physical Address Considerat ions Physical AddressConsiderations Physical Address Considerations Bcache Structure 4.3 Bcache Structure 4.3.1 Bcache Interface Signals 4.3.2 SystemDuplicate Tag Stores 4.4 Victim Data Buffer 4.5 Cac he Coherency Page 4.5.3 Cache Block State Transitions Page Page 4.5.5 DcacheStates and Duplicate Tags Lock Mechanism 4.6 Loc k Mechanism Lock Mechanism 4.6.1 In-Order Processingof LDx_L/STx_C Instructions 4.6.2 Internal Eviction of LDx_L Blocks 4.6.3 Liveness and Fairness 4.6.4 Managing Speculative Store Issues with MultiprocessorSystems 4.7 Syste m Port 4.7.1 System Port Pins 4.7.2 Programming the SystemInterface Clocks 4.7.3 21264/EV68A-to-SystemCommands Table411 shows the command format for page hit mode (21264/EV68A-to-system). Table412 describes the field definitions for Tables 410 and 411. 4.7.4 21264/EV68A-to-SystemCommands Descriptions Page Page 4.7.5 ProbeResponse Commands(Command[4:0] = 00001) 4.7.6 SysAck and 21264/EV68A-to-SystemCommands Flow Control 4.7.7 System-to-21264/EV68A Commands Table420 describes the system-to-21264/EV68A probe commands fields descriptions. Table422 lists the next cache block state selected by Probe[2:0]. Page Page 4.7.8 Data Movement In and Out of the 21264/EV68A Page Page Page Page Page Page 4.7.9 Nonexistent Memory Processing Table432 shows each 21264/EV68A command, with NXM addresses, and the appro- priatesystem response. Page 21264/EV68Afails the STx_C instruction as defined in the AlphaArchitecture Handbook,Version 4 becausethe 21264/EV68A does not yet own the block. 4.8 Bcache Port 4.8.1 Bcache Port Pins 4.8.2 Bcache Clocking Page Page 4.8.3 Bcache Transactions Page 4.8.4 Pin Descriptions Page Page 4.8.5 Bcache Banking 4.8.6 Disabling the Bcache for Debugging Interrupts 4.9 Interrupts Internal Processor Registers MboxIPRs Table 51 Internal Processor Registers (Con tinued) 5.1 Ebox IPRs 5.1.1 Cycle Counter Register CC 5.1.2 Cycle Counter Control Register CC_CTL 5.1.3 Virtual AddressRegister VA 5.1.4 Virtual AddressControl Register VA_CTL 5.1.5 Virtual AddressFormat Register VA_FORM 5.2 Ibox IPRs 5.2.1 ITB Tag Array Write Register ITB_TAG 5.2.2 ITB PTE Array Write Register ITB_PTE 5.2.3 ITB Invalidate All Process (ASM=0) Register ITB_IAP 5.2.4 ITB Invalidate All Register ITB_IA 5.2.5 ITB Invalidate Single Register ITB_IS 5.2.6 ProfileMe PC Register PMPC 5.2.7 Exception Address Register EXC_ADDR 5.2.8 Instruction VirtualAddress Format Register IVA_FORM 5.2.9 Interrupt Enable and CurrentProcessor Mode Register IER_CM Table55 describes the interrupt enable and current processormode register fields. 5.2.10 Software Interrupt Request Register SIRR 5.2.11 Interrupt Summary Register ISUM Table57 describes the interrupt summary register fields. 5.2.12 Hardware InterruptClear Register HW_INT_CLR 5.2.13 Exception Summary Register EXC_SUM Table59 describes the exception summary register fields. 5.2.14 PAL Base Register PAL_BASE 5.2.15 Ibox Control Register I_CTL Table511 describes the Ibox control register fields. Page 5.2.16 Ibox StatusRegister I_STAT Table512 describes the Ibox status register fields. Page 5.2.17 Icache Flush Register IC_FLUSH 5.2.18 Icache Flush ASM Register IC_FLUSH_ASM 5.2.19 Clear Virtual-to-PhysicalMap Register CLR_MAP 5.2.20 Sleep Mode Register SLEEP 5.2.21 Process Context Register PCTX Table514 describes the process context register fields. 5.2.22 Performance Counter Control Register PCTR_CTL Table515 describes the performance counter control registerfields. Page 5.3 Mbox IPRs 5.3.1 DTB Tag Array Write Registers 0 and1 DTB_TAG0, DTB_TAG1 5.3.2 DTB PTE Array Write Registers 0 and 1 DTB_PTE0, DTB_PTE1 5.3.3 DTB Alternate Processor Mode Register DTB_ALTMODE 5.3.4 Dstream TB Invalidate A ll Process (ASM=0) Register DTB_IAP 5.3.5 Dstream TB Invalidate All Register DTB_IA 5.3.6 Dstream TB Invalidate Single Registers 0 and 1 DTB_IS0,1 5.3.7 Dstream TB Address Space Number Registers 0 and 1 DTB_ASN0,1 5.3.8 Memory Management Status Register MM_STAT the Ibox EXC_SUM register. 5.3.9 Mbox Control Register M_CTL Table519 describes the Mbox control register fields. erencesto superpagesresult in access violations. 5.3.10 Dcache Control Register DC_CTL Table520 describes the Dcache control register fields. 5.3.11 Dcache Status Register DC_STAT 5.4 CboxCSRsandIPRs 5.4.1 Cbox Data Register C_DATA Figure 535shows the Cbox data register. Table522 describes the Cbox data register fields. 5.4.2 Cbox Shift Register C_SHFT Figure 536shows the Cbox shift register. Page Page Page 5.4.4 Cbox WRITE_MANY ChainDescription The WRITE_MANY chainorder is contained in Table 525. Note the following: CSRs is containedin C hapter3. are indicatedin italics and have two leading asterisks. chain. Table525 describes the Cbox WRITE_MANY chain order from LSB to MSB. Page Page Page Privileged Architecture Library Code 6.1 PALcode Description PALmode Environment 6.2 PALmodeEnvironment Required PALcodeFunction Codes 6.3 Required PALcodeFunction Codes 6.4 Opcodes Reserved for PALcode 6.4.1 HW_LD Instruction Opcodes Reserved for PALcode Table63 describes the HW_LD instruction fields. 6.4.2 HW_ST Instruction Opcodes Reserved forP ALcode 6.4.3 HW_RET Instruction Opcodes Reserved for PALcode Table65 describes the HW_RET instruction fields. 6.4.4 HW_MFPR and HW_MTPR Instructions Internal ProcessorRegister Access Mechanisms 6.5 Internal Processor Register Access Mechanisms Internal ProcessorR egisterAccess Mechanisms 6.5.1 IPR Scoreboard Bits 6.5.2 Hardware Structure of Explicitly Written IPRs Internal ProcessorRegister Access Mechanisms 6.5.3 Hardware Structure of Implicitly Written IPRs 6.5.4 IPR Access Ordering Internal ProcessorR egisterAccess Mechanisms 6.5.5 Correct Ordering of Explicit Writers Followed by Implicit Readers PALshadow Registers 6.5.6 Correct Ordering of Explicit Readers Followed by Implicit Writers 6.6 PALshadow Registers 6.7 PALcode Emulationof the FPCR PALcode Entry Points 6.7.1 StatusFlags 6.7.2 MF_FPCR 6.8 PALcode EntryPoints 6.8.1 CALL_PALEntry Points PALcode Entry Points 6.8.2 PALcode Exception Entry Points Translation Buffer(TB) Fill Flows 6.9 TranslationBuffer (TB) Fill Flows 6.9.1 DTBFill Figure 65 shows single-missDTB instructions flow. Translation Buffer (TB) Fill Flows Translation Buffer(TB) Fill Flows 6.9.2 ITB Fill 6.10 Performance CounterSupport 6.10.1 General Precautions 6.10.2 Aggregate Mode Programming Guid elines Page E 6.10.3 ProfileMe Mo de Programming Guidelines Page Page E Table614 shows the counter modes that are used with ProfileMe mode. Initialization and Configuration 7.1 Power-Up Reset Flowand the Reset_L and DCOK_H Pins Page 7.1.1 Power Sequencing and Reset State for SignalPins 7.1.2 Clock Forwarding and System Clock Ratio Configuration Page 7.1.3 PLL Ramp Up 7.1.4 BiST and SROM Load and the TestStat_HPin 7.1.5 Clock Forward Reset and System Interface Initialization Fault Reset Flow 7.2 Fault Reset Flow Energy Star Certificationand Sleep Mode Flow 7.3 Energy Star Certi fication and Sleep Mode Flow Energy Star Certificationand Sleep Mode Flow Warm Reset Flow Table77 describes each signal and constraint for the sleep modesequence. 7.4 Warm Reset Flow Array Initialization 7.5 Array Initialization 7.6 Initialization Mode Pr ocessing Initialization Mode Processing External InterfaceInitialization 7.7 ExternalInte rface Initialization 7.8 Internal Processor Register Power-UpReset State Internal ProcessorRegister Power-Up Reset State IEEE 1149.1 Test PortReset 7.9 IEEE 1149.1 Test Port Reset 7.10 Reset State Machine Initialization and Configuration Reset State Machine Figure 75 2126 4/EV68A Reset State Machine State Diagram Table 711 21264/E V68AR eset State Machine State Descriptions StateName Description Page Phase-Lock Loop (PLL) FunctionalDescription 7.11 Phase-Lock Loop (PLL) Functional Description 7.11.1 DifferentialReference Clocks 7.11.2 PLLOutput Clocks Phase-Lock Loop(PLL) Functional Description Error Detection and Error Handling Data Error Correction Code 8.1 Data Error Correction Code 8.2 Icache Data or Tag Parity Error 8.3 Dcache Tag Parity Error 8.4 DcacheData Single- Bit Correctable ECC Error 8.5 Dcach e Store Second Error 8.6 Dcach e Duplicate TagPar ity Error 8.7 Bcache Tag Parity Error 8.8 Controlling Bcache Block Parity Calculat ion 8.9 Bcache Data Single-Bit Correctable ECC Error Page 8.10 Memory/System Port Single-Bit DataCorrectable ECC Error Page 8.11 Bcache Data Single-Bit Correctable ECC Error on a Probe 8.12 Double-BitFill Errors Error Case Summary 8.13 Error Case Summary Table83 summarizes the various error cases and their ramifications. Error CaseSummary D.36 Page Electrical Data 9.1 Electr ical Characteristics 9.2 DC Characteristics symbolindicates current flowing into a 21264/EV68A pin. Page Power Supply Sequencing andAvoiding Potential Failure Mechanisms 9.3 Power Supply Sequencingand Avoiding PotentialFailure Mech- anisms 9.4 AC Characteristics Page 98 ElectricalDat a Table 913 AC Specifications (Continued) Page Page Thermal Management 10.1 Operating Temperature Operating Temperature 10.2 Heat Sink Specifications 104 Thermal Management Figure 102shows the heat sink type 2, along with its approximate dimensions. Figure 102 T ype2 H eatSi nk Thermal Management Figure 103 Ty pe3 H eatSi nk 10.3 Thermal Design Considerations Testability and Diagnostics 11.1 Test P ins 11.2 SROM/Serial Diagnostic TerminalPort IEEE 1149.1 Port 11.3 IEEE 114 9.1 Port 114 Testability and Diagnostics TestStat_H Pin Scan SequenceScan Sequence Table 113 TAP Controller State Machine 11.4 TestStat_H Pin Note: A system designerm ay sample the TestStat_Hpin on the first rising edge Power-Up Self-Test and Initialization 11.5 Power-Up Self-Test and Initialization 11.5.1 Built-in Self-Test 11.5.2 SROM Initialization Power-Up Self-Test and Initialization Notes on IEEE 1149.1Operation and Compliance 11.6 Noteson IEEE 1149.1 Operationand Compliance Page A Alpha Instruction Set A.1 Alpha Instru ction Summary Page Page Page Page Page Page Reserved Opcodes A.2 Reserved Opcodes A.2.1 Opcodes Reserved for Compaq IEEE Floating-PointInstructions A.2.2 Opcodes Reserved for PALcode A.3 IEEE Floating-Point Inst ructions IEEE Floating-Point Instr uctions VAX Floating-Point Instructions A.4 VAX Floating-Point Instructions A.5 Independent Floating-Point Instructions Opcode Summary A.6 Opcode Summary Required PALcodeFunction Codes A.7 Required PALcodeFunction Codes A.8 IEEE Floating-Point Co nformance Page Page See Section2.14 for information about the floating-point control register (FPCR). 21264/EV68A Boundary-Scan Register B 21264/EV68A Boundary-Scan Register This appendixcontains the BSDL description of the 21264/EV68A boundary-scan reg- ister. B.1 Boundary-ScanRegister B.1.1 BSDL Description of the Alpha 21264/EV68A Boundary-Scan Register B2 21264/EV68A Boundary-Scan Register 21264/EV68A Boundary-Scan Register B4 21264/EV68A Boundary-Scan Register 21264/EV68A Boundary-Scan Register B6 21264/EV68A Boundary-Scan Register 21264/EV68A Boundary-Scan Register B8 21264/EV68A Boundary-Scan Register 21264/EV68A Boundary-Scan Register B10 21264/EV68A Boundary-Scan Register 21264/EV68A Boundary-Scan Register B12 21264/EV68A Boundary-Scan Register Page Page PALcode Restrictions and Guidelines D PALcode Restrictions and Guidelines D.1 Restriction 1 : Reset Sequence Required by Retire Logic and Mapper Page Page Page Page Page PALcode Restrictions and Guidelines Restriction 2 : No Multiple Writers to IPRs in Same Scoreboard Group D.2 Restriction 2 : No Multi ple Writers to IPRs in Same Scoreboard Group D.3 Restriction 4 :NoWritersandReaderstoIPRsinSameScore- board Group Guideline 6 : Avoid Consecutive Read-Modify-Write-Read-Modify-Write D.4 Guideline 6 :Avoid Consec utive Read-Modify-Write-Read- Modify-Write D.5 Restriction 7 : Replay Trap, Interrupt Code Sequence,and STF/ ITOF Restriction 9 : PALmodeIstream Address Ranges D.6 Restriction 9 : PALmode Istream AddressRanges D.7 Restriction 10: Duplicate IPR Mode Bits Restriction 11: IboxIPR Update Synchronization D.8 Restriction 11: Ibox IPR UpdateSynchronization D.9 Restriction 12: MFPR of Implicitly-Written IPRs EXC_ADDR, IVA_FORM, and EXC_SUM D.10 Restriction 13 : DTB Fill Flow Collision D.11 Restriction 14 : HW_RET Page Restriction 22: HW_RET/STALLAfter HW_MTPR IS0/IS1 D.18 Restriction22: HW_RET/STALL After HW_MTPR IS0/IS1 D.19 Restriction 23: HW_ST/P/CONDITIO NAL DoesN ot Clear the Lock Flag Restriction 24: HW_RET/STALL After HW_MTPR IC_FLUSH, IC_FLUSH_ASM, D.20 Restriction 24: HW_RET/STALL After HW_MTPR IC_FLUSH, IC_FLUSH_ASM, CLEAR_MAP D.21 Restriction25: HW_MTPR ITB_IA After Reset D.22 Guideline 26: C onditional Branches in PALcode Restriction 27: Reset of Force-FailLock Flag State in PALcode D.23 Restriction2 7: Reset of Force-Fail Lock Flag State in PALcode D.25 Guid eline 29 : JSR, JMP,RET, and JSR_COR in PALcode D.26 Restriction 30 : HW_MTPR and HW_MFPR to the Cbox CSR Restriction30 : HW_MTPR and HW_MFPRto the Cbox CSR Restriction 31 : I_CTL[VA_48]Update D.27 Restriction 31 : I_CTL[VA_48] Update D.28 Restriction 32 : PCTR_CTL Update Restriction 33 : HW_LD Physical/LockU se D.29 Restriction 33 : HW_LD Physical/Lock U se D.30 Restriction34 : Writing Multiple ITB Entries in theSame PAL- code Flow D.31 Guideline 35 : HW_INT_CLR Update D.32 Restriction 36 : Updating I_CTL[SDE] D.33 Restriction 37 : Updating VA_CTL[VA_48] D.35 Guideline 39: Writing Multiple DTB Entries in the Same PAL Flow D.36 Restriction 40: Scrubbing a Single-Bit Error Restriction 40: Scrubbinga Single-Bit Error Page Restriction 46: AvoidingLivelocks in Speculative Load CRD Handlers D.42 Restriction 46: Avoiding Livelocks in SpeculativeLoad CRD Handlers D.43 Restriction 47: Cache Eviction for Single-Bit Cache Er rors PALcode Restrictions and Guidelines Restriction 47: Cache Evictionfor Single-Bit Cache Errors Page E 21264/EV68A-to-Bcache Pin Interface This appendixprovides the pin interface b etween the21264/EV68A and Bcache SSRAMs. E.1 Forwarding ClockPin Groupings Late-Write Non-Bursting SSRAMs E.2 Late-Write Non-Bursting SSRAM s Unused Bcachetag pins should be pulled to ground through a200-ohm resistor. Data Pin Usage Tag Pin Usage Dual-Data Rate SSRAMs E.3 Dual-Data Rate SSRAMs Data Pin Usage Dual-Data RateSSRAMs Tag Pin Usage Unused Bcachetag pins should be pulled to ground through a200-ohm resister. Dual-Data Rate SSRAMs Page Glossary Page Page Page Page Page Page nn Page Page Page Page Page Page Page Page Page Page Index Numerics A B C D E F G H I Page J L M N O P R S T U V W X