Lesson 3--Intro to ISA Flashcards

1
Q

3 types of instructions

A
  1. Data or memory access (load, store, etc.)
  2. Control flow (branch)
  3. Arithmetic or logic (add, subtract, etc.)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

ISA design

A

Operation: fundamental RISC-like minimal unit of work
Instruction: Fundamental unit of encoding. Refers to a parallel set of operations
Bundle: memory aligned encoding unit

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

True or False: VLIW exposes a scheduler in the compiler

A

True. Superscalar hides this.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Memory

A

Off chip, not specialized, slow

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Registers

A

On chip, connected to the logic of the processor, fast

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Dependencies

A

RAW: True dependence
WAR: False dependence
WAW: False dependence

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

VLIW Design Principles

A
  • -VLIW instructions consist of parallel operations. Compiler must guarantee safe parallel operations.
  • -Simplification of HW to improve energy efficiency.
  • -Latency and functional unit are exposed.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

True or False: VLIW’s advantages come largely from having an intelligent compiler that can schedule many instructions simultaneously.

A

True

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

True or False: VLIW-compatible code is easily portable to hardware otehr than the chip it is designed for

A

False

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

True or False: VLIW has some problems with the inflexibility of its compiler-first design

A

True

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Role of VLIW compilers

A
  1. Gather independent operations and bundle them into sections.
  2. Resolve all structural and resource hazards.
  3. If the VLIW is clustered, the compiler assigns operations to the clusters and generates cross cluster register references.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Horizontal Decisions

A

Decisions within an instruction word

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Vertical Decisions

A

Decisions across pipelined instructions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Fixed overhead encoding

A

Prepend mask bits to a VLIW instruction to identify “what goes where”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Distributed encoding advantages

A
  • -Explicitly insert stop bit for delimiter
  • -Distributed and incremental encoding cost.
  • -No need to compute next PC.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Distributed encoding disadvantages

A
  • -More complex.

- -Bits are embedded as part of the instruction, so have to get them back to NOPs.

17
Q

Fixed overhead encoding advantages

A

–Simple scheme.

18
Q

Fixed overhead encoding disadvantages

A

–Space overhead for the longest possible VLIW instruction.

19
Q

Template-based encoding advantages

A
  • -Low overhead

- -No penalty for short instructions.

20
Q

Template-based encoding disadvantages

A

Limited number of templates