lecture 2: computer architecture: a science oftradeoffs
TRANSCRIPT
![Page 1: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/1.jpg)
Lecture 2:
Computer Architecture: A Science ofTradeoffs
![Page 2: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/2.jpg)
Outline
The ISA
The Microarchitecture
The System level
![Page 3: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/3.jpg)
The ISA(visible to the software)
Dynamic/Static Interface
What at compile time, what at run time
Rich instruction set vs. simpler instruction set
Fixed-length, Uniform Decode vs. …
Condition codes vs. …
Load/Store vs. …
Help for the Programmer vs. help for the uarchitectAddressing modes
Data types
Unaligned accesses
![Page 4: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/4.jpg)
The ISA (continued)
Hardware interlocks vs. …
VLIW vs. …
0,1,2,3 address machine
Compatibility vs. a new ISA
Precise exceptions vs. …
Vectored interrupts vs Polling
Address space (16-bit 32-bit 64-bit)
![Page 5: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/5.jpg)
Two example ISAs
LC-3b (from EE 360n)Load/store, 3-address machine
Few opcodes, addressing modes, data types
Fixed length, uniform decode
x86Not load/store, 2-address machine
Variable length, decoding nightmare
Prefixes
Many opcodes, addressing modes, data types
Dense encoding (comprehensive instructions)
Mod/RM, SIB, addressing mode offset, literal operand
Virtual memory PLUS segmentation
Memory-mapped PLUS I/O instructions
![Page 6: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/6.jpg)
The Microarchitecture(underneath the hood)
CPI vs. cycle time (or, IPC vs. frequency)
in-order vs. out-of-order execution
Speculate vs. stand around and wait
Issue-width
ASIC vs. programmed control
Use of chip real estateBetter branch predictor
Accelerators
Microcode
Pipeline depth
Cache structures
![Page 7: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/7.jpg)
Microarchitecture (continued)
Fault toleranceYesterday: Tandem (Two cores in lock step)
Tomorrow: repeat to remove errors (or 2 cores lockstep)
On-chip latencynear neighbor communication
Branch target buffersaves a bubble on a taken branch
Partitioning – until when?
![Page 8: Lecture 2: Computer Architecture: A Science ofTradeoffs](https://reader036.vdocuments.mx/reader036/viewer/2022082407/5697bf771a28abf838c813a0/html5/thumbnails/8.jpg)
The System
• MP granularity (loosely coupled vs. tightly)
• Distributed shared memory versus centralized
• Interconnection structure
• Use of commodity vs tailored parts
• What do we integrate on chip• Memory controller
• Analog devices