Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

ECE/CS 472/572

Computer Architecture:

Background

Opening the Box
Capacitive multitouch LCD screen
3.8 V, 25 Watt-hour battery
Computer board
Inside the Processor
◼ Apple A5
What is Computer Architecture ?
◼ Computer Architecture is the science and art of
selecting and interconnecting hardware components
to create computers that meet functional,
performance and cost goals.
Application
Algorithm
Programming Language
Operating System
Circuits
Devices
Technology
Computer Architecture
How Computer Is Made
https://www.youtube.com/watch?v=UvluuAIiA50
Relative Performance
◼ Define: Performance = 1/Execution Time
◼ “X is n time faster than Y”
n== XY
YX
time Executiontime Execution
ePerformancePerformanc
◼ Example: time taken to run a program
◼ 10s on A, 15s on B
◼ Execution TimeB / Execution TimeA
= 15s / 10s = 1.5
◼ So A has a speedup of 1.5 over B
Instruction Count and CPI
◼ Instruction Count for a program
◼ Determined by program, ISA and compiler
◼ Average cycles per instruction (CPI)
◼ Determined by CPU hardware
◼ Different instructions may have different CPI
◼ Average CPI affected by instruction mix
Time Cycle ClockCPICount nInstructioTime CPU =
Uniprocessor Performance
Constrained by power, instruction-level parallelism, memory latency
Intel Tick-Tock Model
 Embedded processors
 Broadcom XLP-II: 20-core
 Cavium Octeon: 48-core
 Tilera Tile-Gx8072: 72-core
 Mobile devices – MPSoCs
 CPU, GPU, DSP, etc.
 (GP)GPUs
 Nvidia Kepler: 192x15
cores
 AMD Liverpool: 1152 cores
[1] http://www.theregister.co.uk/2010/02/03/intel_westmere_ep_preview/
[2] http://www.theregister.co.uk/2012/10/03/ibm_power7_plus_server_launch/
[3] http://www.theregister.co.uk/2012/09/04/oracle_sparc_t5_processor/
[4] http://www.intel.com/pressroom/archive/releases/2009/20091202comp_sm.htm
[5] http://www.scientificcomputing.com/news/2013/02/intel-xeon-phi-coprocessor/
Oracle SPARC T5: 16-core[3]IBM Power7+: 8-core[2]
Intel Westmere-EP: 6-core[1]
Intel SCC: 48-core[4] Intel Xeon Phi: 60-core[5]
Challenge: On-chip communication for parallel computing
Instruction Set
◼ The repertoire of instructions of a computer
◼ Different computers have different instruction sets
◼ But with many aspects in common
◼ Early computers had very simple instruction sets
◼ Simplified implementation
◼ Many modern computers also have simple
instruction sets
◼ CISC vs. RISC
The MIPS Instruction Set
◼ Large share of embedded core market
◼ Applications in consumer electronics, network/storage
equipment, cameras, printers, …
◼ Typical of many modern ISAs (see Appendixes E)
◼ We will examine two implementations of MIPS ISA
◼ A simplified version
◼ A more realistic pipelined version
◼ Simple subset, shows most aspects
◼ Memory reference instructions: lw, sw
◼ Arithmetic-logical instructions : add, sub, and, or, slt
◼ Control transfer instructions: beq, j
MIPS Instruction Examples
◼ C code:
g = h + A[8];
◼ g in $s1, h in $s2, base address of A in $s3
◼ Compiled MIPS code:
◼ Index 8 requires offset of 32
◼ 4 bytes per word
lw $t0, 32($s3) # load word
add $s1, $s2, $t0
offset base register
MIPS Instruction Examples
◼ C code:
A[12] = h + A[8];
◼ h in $s2, base address of A in $s3
◼ Compiled MIPS code:
◼ Index 8 requires offset of 32
lw $t0, 32($s3) # load word
add $t1, $s2, $t0
sw $t1, 48($s3) # store word
MIPS Instruction Examples
◼ Conditional
◼ Branch to a labeled instruction if a condition is
true; otherwise, continue sequentially
◼ beq rs, rt, L1
◼ If (rs == rt) branch to instruction labeled L1;
◼ Unconditional
◼ j L1
◼ Unconditional jump to instruction labeled L1
MIPS R-format Instructions
◼ Instruction fields
◼ op: operation code (opcode)
◼ rs: first source register number
◼ rt: second source register number
◼ rd: destination register number
◼ shamt: shift amount (00000 for now)
◼ funct: function code (extends opcode)
op rs rt rd shamt funct
6 bits 6 bits5 bits 5 bits 5 bits 5 bits
R-format Example
add $t0, $s1, $s2
0 17 18 8 0 32
000000 10001 10010 01000 00000 100000
000000100011001001000000001000002 = 0232402016
op rs rt rd shamt funct
6 bits 6 bits5 bits 5 bits 5 bits 5 bits
MIPS I-format Instructions
◼ Immediate arithmetic
◼ addi $s1, $s2, 20
◼ rs/rt: source/destination register number
◼ Constant: –2^15 to +2^15 – 1
◼ Load/store instructions
◼ lw $t0, 32($s3)
◼ rs/rt: source/destination register number
◼ Address: offset added to base address in rs
op rs rt constant or address
6 bits 5 bits 5 bits 16 bits
MIPS J-format Instructions
◼ Jump (j) targets could be anywhere in text segment
◼ j L1
◼ Encode full address in instruction
op address
6 bits 26 bits
◼ (Pseudo)Direct jump addressing
◼ Target address = PC31…28 : (address × 4)
Branch Addressing
◼ Branch instructions specify
◼ beq rs, rt, L1
◼ Opcode, two registers, target address
◼ Most branch targets are near branch
◼ Forward or backward
op rs rt constant or address
6 bits 5 bits 5 bits 16 bits
◼ PC-relative addressing
◼ Target address = PC + offset × 4
◼ PC already incremented by 4 by this time
Addressing Mode Summary