Top Banner
RHK.S96 1 Lecture 11: Case Study— Tomasulo Algorithm Professor Randy H. Katz Computer Science 252 Spring 1996
52

Lecture 11: Case Study— Tomasulo Algorithm

Dec 02, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 1

Lecture 11: Case Study—Tomasulo Algorithm

Professor Randy H. KatzComputer Science 252

Spring 1996

Page 2: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 2

Review: Scoreboard Summary

• Speedup 1.7 from compiler; 2.5 by hand BUT slow memory (no cache)

• Limitations of 6600 scoreboard– No forwarding– Limited to instructions in basic block (small window)– Number of functional units(structural hazards)– Wait for WAR hazards– Prevent WAW hazards

Page 3: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 3

Another Dynamic Algorithm: Tomasulo Algorithm

• For IBM 360/91 about 3 years after CDC 6600• Goal: High Performance without special compilers• Differences between IBM 360 & CDC 6600 ISA

– IBM has only 2 register specifiers/instr vs. 3 in CDC 6600– IBM has 4 FP registers vs. 8 in CDC 6600

• Differences between Tomasulo Algorithm & Scoreboard– Control & buffers distributed with Function Units vs. centralized in

scoreboard; called “reservation stations”– Registers in instructions replaced by pointers to reservation station

buffer– HW renaming of registers to avoid WAR, WAW hazards– Common Data Bus broadcasts results to all FUs– Load and Stores treated as FUs as well

Page 4: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 4

LoadBuffer

FPRegisters

FP Op Queue

StoreBuffer

FP AddRes.Station

FP MulRes.Station

CommonDataBus

Tomasulo Organization

Page 5: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 5

Reservation Station Components

Op—Operation to perform in the unit (e.g., + or –)Qj, Qk—Reservation stations producing source registers Vj, Vk—Value of Source operandsRj, Rk—Flags indicating when Vj, Vk are ready

Busy—Indicates reservation station and FU is busy

Register result status—Indicates which functional unit will write each register, if one exists. Blank when no pending instructions that will write that register.

Page 6: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 6

Three Stages of Tomasulo Algorithm

1. Issue—get instruction from FP Op Queue If reservation station free, the scoreboard issues instr &

sends operands (renames registers).

2.Execution—operate on operands (EX) When both operands ready then execute;

if not ready, watch CDB for result

3.Write result—finish execution (WB) Write on Common Data Bus to all awaiting units;

mark reservation station available.

Page 7: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 7

Tomasulo Example Cycle 0

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 Load1 NoLD F2 45+ R3 Load2 NoMULTDF0 F2 F4 Load3 NoSUBD F8 F6 F2DIVD F10 F0 F6ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

0 FU

Page 8: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 8

Tomasulo Example Cycle 1

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 Load1 No 34+R2LD F2 45+ R3 Load2 NoMULTDF0 F2 F4 Load3 NoSUBD F8 F6 F2DIVD F10 F0 F6ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

1 FU Load1

Page 9: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 9

Tomasulo Example Cycle 2

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 Load1 Yes 34+R2LD F2 45+ R3 2 Load2 Yes 45+R3MULTDF0 F2 F4 Load3 NoSUBD F8 F6 F2DIVD F10 F0 F6ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

2 FU Load2 Load1

Page 10: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 10

Tomasulo Example Cycle 3

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 Load1 Yes 34+R2LD F2 45+ R3 2 Load2 Yes 45+R3MULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2DIVD F10 F0 F6ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 Yes MULTD R(F4) Load20 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

3 FU Mult1 Load2 Load1

Page 11: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 11

Tomasulo Example Cycle 4

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 Load2 Yes 45+R3MULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4DIVD F10 F0 F6ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 Yes SUBD M(34+R2) Load20 Add2 No

Add3 No0 Mult1 Yes MULTD R(F4) Load20 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

4 FU Mult1 Load2 M(34+R2) Add1

Page 12: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 12

Tomasulo Example Cycle 5

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 Load2 Yes 45+R3MULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4DIVD F10 F0 F6 5ADDD F6 F8 F2Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 Yes SUBD M(34+R2) Load20 Add2 No

Add3 No0 Mult1 Yes MULTD R(F4) Load20 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

5 FU Mult1 Load2 M(34+R2) Add1 Mult2

Page 13: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 13

Tomasulo Example Cycle 6

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk2 Add1 Yes SUBD M(34+R2) M(45+R3)0 Add2 Yes ADDD M(45+R3) Add1

Add3 No10 Mult1 Yes MULTD M(45+R3) R(F4)

0 Mult2 Yes DIVD M(34+R2) Mult1Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

6 FU Mult1 M(45+R3) Add2 Add1 Mult2

Page 14: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 14

Tomasulo Example Cycle 7

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk1 Add1 Yes SUBD M(34+R2) M(45+R3)0 Add2 Yes ADDD M(45+R3) Add1

Add3 No9 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

7 FU Mult1 M(45+R3) Add2 Add1 Mult2

Page 15: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 15

Tomasulo Example Cycle 8

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 Yes SUBD M(34+R2) M(45+R3)0 Add2 Yes ADDD M(45+R3) Add1

Add3 No8 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

8 FU Mult1 M(45+R3) Add2 Add1 Mult2

Page 16: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 16

Tomasulo Example Cycle 9

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 Yes ADDD M()–M() M(45+R3)

Add3 No7 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

9 FU Mult1 M(45+R3) Add2 M()–M() Mult2

Page 17: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 17

Tomasulo Example Cycle 10

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No2 Add2 Yes ADDD M()–M() M(45+R3)

Add3 No7 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

10 FU Mult1 M(45+R3) Add2 M()–M() Mult2

Page 18: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 18

Tomasulo Example Cycle 11

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No1 Add2 Yes ADDD M()–M() M(45+R3)

Add3 No5 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

11 FU Mult1 M(45+R3) Add2 M()–M() Mult2

Page 19: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 19

Tomasulo Example Cycle 12

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 Yes ADDD M()–M() M(45+R3)

Add3 No4 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

12 FU Mult1 M(45+R3) Add2 M()–M() Mult2

Page 20: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 20

Tomasulo Example Cycle 13

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No3 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

13 FU Mult1 M(45+R3) (M–M)+M() M()–M() Mult2

Page 21: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 21

Tomasulo Example Cycle 14

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No2 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

14 FU Mult1 M(45+R3) (M–M)+M() M()–M() Mult2

Page 22: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 22

Tomasulo Example Cycle 15

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No1 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

15 FU Mult1 M(45+R3) (M–M)+M() M()–M() Mult2

Page 23: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 23

Tomasulo Example Cycle 16

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 Yes MULTD M(45+R3) R(F4)0 Mult2 Yes DIVD M(34+R2) Mult1

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

16 FU Mult1 M(45+R3) (M–M)+M() M()–M() Mult2

Page 24: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 24

Tomasulo Example Cycle 17

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 17 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 Yes DIVD M*F4 M(34+R2)

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

17 FU M*F4 M(45+R3) (M–M)+M() M()–M() Mult2

Page 25: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 25

Tomasulo Example Cycle 18

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 17 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No

40 Mult2 Yes DIVD M*F4 M(34+R2)Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

18 FU M*F4 M(45+R3) (M–M)+M() M()–M() Mult2

Page 26: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 26

Tomasulo Example Cycle 57

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 17 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No1 Mult2 Yes DIVD M*F4 M(34+R2)

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

57 FU M*F4 M(45+R3) (M–M)+M() M()–M() Mult2

Page 27: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 27

Tomasulo Example Cycle 58

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 17 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5 58ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 Yes DIVD M*F4 M(34+R2)

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

58 FU M*F4 M(45+R3) (M–M)+M() M()–M() Mult2

Page 28: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 28

Tomasulo Example Cycle 59

Instruction status Execution WriteInstruction j k Issue complete Result Busy AddressLD F6 34+ R2 1 3 4 Load1 NoLD F2 45+ R3 2 5 6 Load2 NoMULTDF0 F2 F4 3 16 17 Load3 NoSUBD F8 F6 F2 4 8 9DIVD F10 F0 F6 5 58 59ADDD F6 F8 F2 6 12 13Reservation Stations S1 S2 RS for j RS for k

Time Name Busy Op Vj Vk Qj Qk0 Add1 No0 Add2 No

Add3 No0 Mult1 No0 Mult2 No

Register result statusClock F0 F2 F4 F6 F8 F10 F12 ... F30

59 FU M*F4 M(45+R3) (M–M)+M() M()–M() M*F4/M

Page 29: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 29

Tomasulo Loop Example

Loop: LD F0 0 R1 MULTD F4 F0 F2 SD F4 0 R1 SUBI R1 R1 #8 BNEZ R1 Loop

• Multiply takes 4 clocks• Load have cache misses

Page 30: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 30

Loop Example Cycle 0Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 Load1 NoMULTDF4 F0 F2 1 Load2 NoSD F4 0 R1 1 Load3 No QiLD F0 0 R1 2 Store1 NoMULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 No SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

0 80 Qi

Page 31: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 31

Loop Example Cycle 1Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 Load2 NoSD F4 0 R1 1 Load3 No QiLD F0 0 R1 2 Store1 NoMULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 No SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

1 80 Qi Load1

Page 32: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 32

Loop Example Cycle 2Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 Load3 No QiLD F0 0 R1 2 Store1 NoMULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

2 80 Qi Load1 Mult1

Page 33: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 33

Loop Example Cycle 3Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

3 80 Qi Load1 Mult1

Page 34: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 34

Loop Example Cycle 4Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

4 72 Qi Load1 Mult1

Page 35: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 35

Loop Example Cycle 5Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

5 72 Qi Load1 Mult1

Page 36: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 36

Loop Example Cycle 6Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 Yes 72SD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 6 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

6 72 Qi Load1 Mult1

Page 37: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 37

Loop Example Cycle 7Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 Yes 72SD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 6 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 NoSD F4 0 R1 2 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 Yes MULTD R(F2) Load2 BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

7 72 Qi Load2 Mult2

Page 38: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 38

Loop Example Cycle 8Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 Yes 72SD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 6 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 Yes MULTD R(F2) Load2 BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

8 72 Qi Load2 Mult2

Page 39: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 39

Loop Example Cycle 9Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 Load1 Yes 80MULTDF4 F0 F2 1 2 Load2 Yes 72SD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 6 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load1 SUBI R1 R1 #80 Mult2 Yes MULTD R(F2) Load2 BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

9 64 Qi Load2 Mult2

Page 40: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 40

Loop Example Cycle 10Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 Load2 Yes 72SD F4 0 R1 1 3 Load3 No QiLD F0 0 R1 2 6 10 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R14 Mult1 Yes MULTD M(80) R(F2) SUBI R1 R1 #80 Mult2 Yes MULTD R(F2) Load2 BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

10 64 Qi Load2 Mult2

Page 41: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 41

Loop Example Cycle 11Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R13 Mult1 Yes MULTD M(80) R(F2) SUBI R1 R1 #84 Mult2 Yes MULTD M(72) R(F2) BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

11 64 Qi Mult2

Page 42: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 42

Loop Example Cycle 12Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R12 Mult1 Yes MULTD M(80) R(F2) SUBI R1 R1 #83 Mult2 Yes MULTD M(72) R(F2) BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

12 64 Qi Mult2

Page 43: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 43

Loop Example Cycle 13Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R11 Mult1 Yes MULTD M(80) R(F2) SUBI R1 R1 #82 Mult2 Yes MULTD M(72) R(F2) BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

13 64 Qi Mult2

Page 44: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 44

Loop Example Cycle 14Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 Mult1MULTDF4 F0 F2 2 7 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD M(80) R(F2) SUBI R1 R1 #81 Mult2 Yes MULTD M(72) R(F2) BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

14 64 Qi Mult2

Page 45: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 45

Loop Example Cycle 15Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 M(80)*R(F2)MULTDF4 F0 F2 2 7 15 Store2 Yes 72 Mult2SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 No SUBI R1 R1 #80 Mult2 Yes MULTD M(72) R(F2) BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

15 64 Qi Mult2

Page 46: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 46

Loop Example Cycle 16Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 M(80)*R(F2)MULTDF4 F0 F2 2 7 15 16 Store2 Yes 72 M(72)*R(72)SD F4 0 R1 2 8 Store3 NoReservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

16 64 Qi Mult1

Page 47: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 47

Loop Example Cycle 17Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 M(80)*R(F2)MULTDF4 F0 F2 2 7 15 16 Store2 Yes 72 M(72)*R(72)SD F4 0 R1 2 8 Store3 Yes 64 Mult1Reservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

17 64 Qi Mult1

Page 48: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 48

Loop Example Cycle 18Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 18 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 Yes 80 M(80)*R(F2)MULTDF4 F0 F2 2 7 15 16 Store2 Yes 72 M(72)*R(72)SD F4 0 R1 2 8 Store3 Yes 64 Mult1Reservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

18 56 Qi Mult1

Page 49: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 49

Loop Example Cycle 19Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 18 19 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 NoMULTDF4 F0 F2 2 7 15 16 Store2 Yes 72 M(72)*R(72)SD F4 0 R1 2 8 Store3 Yes 64 Mult1Reservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

19 56 Qi Mult1

Page 50: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 50

Loop Example Cycle 20Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 18 19 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 NoMULTDF4 F0 F2 2 7 15 16 Store2 Yes 72 M(72)*R(72)SD F4 0 R1 2 8 20 Store3 Yes 64 Mult1Reservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

20 56 Qi Mult1

Page 51: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 51

Loop Example Cycle 21Instruction status ExecutionWriteInstruction j k iteration Issue completeResult Busy AddressLD F0 0 R1 1 1 9 10 Load1 NoMULTDF4 F0 F2 1 2 14 15 Load2 NoSD F4 0 R1 1 3 18 19 Load3 Yes 64 QiLD F0 0 R1 2 6 10 11 Store1 NoMULTDF4 F0 F2 2 7 15 16 Store2 NoSD F4 0 R1 2 8 20 21 Store3 Yes 64 Mult1Reservation Stations S1 S2 RS for jRS for k

Time Name Busy Op Vj Vk Qj Qk Code:0 Add1 No LD F0 0 R10 Add2 No MULTDF4 F0 F20 Add3 No SD F4 0 R10 Mult1 Yes MULTD R(F2) Load3 SUBI R1 R1 #80 Mult2 No BNEZ R1 Loop

Register result statusClock R1 F0 F2 F4 F6 F8 F10 F12... F30

21 56 Qi Mult1

Page 52: Lecture 11: Case Study— Tomasulo Algorithm

RHK.S96 52

Tomasulo Summary

• Prevents Register as bottleneck• Avoids WAR, WAW hazards of Scoreboard• Allows loop unrolling in HW• Not limited to basic blocks (provided branch

prediction)• Lasting Contributions

– Dynamic scheduling– Register renaming– Load/store disambiguation

• Next: More branch prediction