DLX Instruction Fetch Execute Addr Calc Instr Decode

  • Slides: 46
Download presentation

Επεξεργαστής DLX Instruction Fetch Execute Addr. Calc Instr. Decode Reg. Fetch Memory Access Write

Επεξεργαστής DLX Instruction Fetch Execute Addr. Calc Instr. Decode Reg. Fetch Memory Access Write Back MUX RD RD WB Data RD MUX Sign Imm Extend MEM/WB Μνήμη Δεδομένων EX/MEM ALU MUX RS 2 ID/EX Imm Opcode RS 1 Return PC (Addr + 8) Branch? Reg File IF/ID PC Μνήμη Adder Εντολών 4 Adder Next PC

Τεχνικές ελάττωσης stalls. CPI = Ideal CPI + Structural stalls + RAW stalls +

Τεχνικές ελάττωσης stalls. CPI = Ideal CPI + Structural stalls + RAW stalls + WAR stalls + WAW stalls + Control stalls Στο 1ο παράδειγμα (in order execution) είχαμε 2 RAW stalls και κανένα άλλο stall. Στο 2ο παράδειγμα (out order execution) είχαμε 1 Structural stall και κανένα άλλο stall. Θα μελετήσουμε δύο ειδών τεχνικές Δυναμικές δρομολόγηση εντολών (hardware) • Scoreboard (ελάττωση RAW stalls) • Register Renaming α)Tomasulo β)Reorder Buffer (ελάττωση WAR και WAW stalls) • Branch prediction (ελάττωση Control stalls) Στατικές (shoftware/compiler) • Loop Unrolling • Software Pipelining • Trace Scheduling

Αρχιτεκτονική Scoreboard (CDC 6600) Registers FP Mult FP Divide FP Add Integer SCOREBOARD Functional

Αρχιτεκτονική Scoreboard (CDC 6600) Registers FP Mult FP Divide FP Add Integer SCOREBOARD Functional Units FP Mult Memory

Four Stages of Scoreboard Control • Issue—decode εντολή & έλεγξε για structural hazards (ID

Four Stages of Scoreboard Control • Issue—decode εντολή & έλεγξε για structural hazards (ID 1) – Instructions issued in program order (for hazard checking) – Don’t issue if structural hazard – Don’t issue if instruction is output dependent on any previously issued but uncompleted instruction (no WAW hazards) • Read operands—περίμενε μέχρι να μην υπάρχουν data hazards, μετά διάβασε μεταβλητές (ID 2) – All real dependencies (RAW hazards) resolved in this stage, since we wait for instructions to write back data. – No forwarding of data in this model!

Four Stages of Scoreboard Control • Execution— εκτέλεση στην αριθμητική μονάδα (EX) – The

Four Stages of Scoreboard Control • Execution— εκτέλεση στην αριθμητική μονάδα (EX) – The functional unit begins execution upon receiving operands. When the result is ready, it notifies the scoreboard that it has completed execution. • Write result—Τέλος Εκτέλεσης (WB) – Stall μέχρι να μην υπάρχει WAR hazards με προηγούμενες εντολές: Παράδειγμα: DIV. D F 0, F 2, F 4 ADD. D F 10, F 8 SUB. D F 8, F 14 CDC 6600 scoreboard would stall SUBD until ADDD reads operands

3 μέρη του Scoreboard • Instruction status: Σε ποιο από τα 4 στάδια είναι

3 μέρη του Scoreboard • Instruction status: Σε ποιο από τα 4 στάδια είναι η εντολή • Functional unit status: —Δείχνει την κατάσταση του functional unit (FU). 9 πεδία για κάθε functional unit Busy: Indicates whether the unit is busy or not Op: Operation to perform in the unit (e. g. , + or –) Fi: Destination register Fj, Fk: Source-register numbers Qj, Qk: Functional units producing source registers Fj, Fk Rj, Rk: Flags indicating when Fj, Fk are ready • Register result status—Δείχνει ποιο functional unit θα γράψει κάθε register, αν υπάρχει κάποια. Blank when no pending instructions will write that register

Παράδειγμα Scoreboard

Παράδειγμα Scoreboard

Detailed Scoreboard Pipeline Control Instruction status Wait until Bookkeeping Issue Busy (FU)=No and result(D)=0

Detailed Scoreboard Pipeline Control Instruction status Wait until Bookkeeping Issue Busy (FU)=No and result(D)=0 (struct hazard, WAW) Busy(FU) yes; Op(FU) op; Fi(FU) `D’; Fj(FU) `S 1’; Fk(FU) `S 2’; Qj Result(‘S 1’); Qk Result(`S 2’); Rj not Qj; Rk not Qk; Result(‘D’) FU; Read operands Rj and Rk (RAW) Rj No; Rk No Execution complete Functional unit done Write result f((Fj(f) Fi(FU) or Rj(f)=No) & f(if Qj(f)=FU then Rj(f) Yes); (Fk(f) Fi(FU) or f(if Qk(f)=FU then Rj(f) Yes); Rk( f )=No)) Result(Fi(FU)) 0; Busy(FU) No (WAR)

Παράδειγμα Scoreboard : Κύκλος 2 • Issue Δεύτερο LD; Structural Hazard.

Παράδειγμα Scoreboard : Κύκλος 2 • Issue Δεύτερο LD; Structural Hazard.

Παράδειγμα Scoreboard : Κύκλος 3 • Issue MULT; In order issue.

Παράδειγμα Scoreboard : Κύκλος 3 • Issue MULT; In order issue.

Παράδειγμα Scoreboard : Κύκλος 62 • In-order issue; out-of-order execute & commit

Παράδειγμα Scoreboard : Κύκλος 62 • In-order issue; out-of-order execute & commit

CDC 6600 Scoreboard • Speedup 1. 7 from compiler; 2. 5 by hand BUT

CDC 6600 Scoreboard • Speedup 1. 7 from compiler; 2. 5 by hand BUT slow memory (no cache) limits benefit • Limitations of 6600 scoreboard: – No forwarding hardware – Limited to instructions in basic block (small window) – Small number of functional units (structural hazards), especially integer/load store units – Do not issue on structural hazards – Wait for WAR hazards – Prevent WAW hazards

DLX Instruction Fetch Execute Addr Calc Instr DecodeDLX Instruction Fetch Execute Addr Calc Instr Decode
Introduction aux microprocesseurs Fetch decode execute fetch decodeIntroduction aux microprocesseurs Fetch decode execute fetch decode
Introduction aux microprocesseurs Fetch decode execute fetch decodeIntroduction aux microprocesseurs Fetch decode execute fetch decode
Instruction Cycle Two steps Fetch Execute Fetch CycleInstruction Cycle Two steps Fetch Execute Fetch Cycle
Instruction Execution fetch cycle hmtfas fetch execute cycleInstruction Execution fetch cycle hmtfas fetch execute cycle
Instruction Cycle Main Topics Instruction fetch Instruction executeInstruction Cycle Main Topics Instruction fetch Instruction execute
The Fetch Decode Execute Cycle The History ofThe Fetch Decode Execute Cycle The History of
Instruction Fetch 4 Address PC Instr 31 0Instruction Fetch 4 Address PC Instr 31 0
Advanced Topics 1 Processor Instruction fetch n DecodeAdvanced Topics 1 Processor Instruction fetch n Decode
Studentsid name addr age GPA sid name addrStudentsid name addr age GPA sid name addr
Studentsid name addr age GPA sid name addrStudentsid name addr age GPA sid name addr
The DLX Architecture CS 448 Chapter 2 DLXThe DLX Architecture CS 448 Chapter 2 DLX
Upregulation of homeodomain genes DLX 1 and DLXUpregulation of homeodomain genes DLX 1 and DLX
Floating PointMulticycle Pipelining in DLX Completion of DLXFloating PointMulticycle Pipelining in DLX Completion of DLX
Floating PointMulticycle Pipelining in DLX Completion of DLXFloating PointMulticycle Pipelining in DLX Completion of DLX
Processor Microarchitecture Fetch Decode ExecuteWriteback Register Files ALUProcessor Microarchitecture Fetch Decode ExecuteWriteback Register Files ALU
Chapter 3 e Fetch and Decode Stages McChapter 3 e Fetch and Decode Stages Mc
Slicer 3 Webservices Infrastructure Fetch Medical Informatics FetchSlicer 3 Webservices Infrastructure Fetch Medical Informatics Fetch
Central Processing Unit Fetch Cycle FETCH Programming CounterCentral Processing Unit Fetch Cycle FETCH Programming Counter
DLX Instruction Format I type instruction 6 bitsDLX Instruction Format I type instruction 6 bits
DLX Instruction Format I type instruction 6 bitsDLX Instruction Format I type instruction 6 bits
RISC 5 Decode EXecute Writeback Store Memory ALURISC 5 Decode EXecute Writeback Store Memory ALU
Addressing Instruction Fetch Bottlenecks by Using an InstructionAddressing Instruction Fetch Bottlenecks by Using an Instruction
Instruction Fetch VHDL ASIC NO 4 Instruction DecodingInstruction Fetch VHDL ASIC NO 4 Instruction Decoding