Adding more cores to a processor will not improve performance if (49)

No part of the program can be done in parallel with another

You can calculate execution time by the formula (34)

Instructions x CPI / Clock speed

Immediate operands are (72)

Data located in the instruction itself

When a program is loaded into memory, the text segment (104)

Contains the MIPS machine code, Contains the MIPS machine code

In general, Intel's arithmetic instructions (152)

A source and destination register be the same.

Which is an example of improving bandwidth? (30)

More open cashier lanes in the food court

The reason the "ugly" Intel architecture has survived over the years is that (157)

It has huge market forces behind it.

What is the fastest way of saving a variable in the $a0 register (138)

Add it to $0 and store it in another register.

A simple and efficient way to implement division is to (189)

use hardware to perform "grammar school" division

What is the primary difference between active matrix displays and liquid crystal displays? (17)

The active matrix display controls intensity of each pixel.

How is multiply overflow handled in MIPS? (188)

Software must check the hi half of the product

Which describes a symbol table? (125)

A list of needed and supplied program locations.

What is the job of an assembler (15)

Translate assembly code into machine language

Which is not a characteristic of Dynamic Random Access Memory (DRAM) (20)

It is one of the fastest types of random access memory

The difference between ASCII and Unicode is (109)

Unicode can represent more characters than ASCII.

An embedded computer is distinguished from other types of computers by the fact that it (6)

Primarily runs a single, predetermined application

studbud

Assembly Midterm Flashcards | StudBud

Assembly Midterm

264 termsStudBud Administration28 plays

quiz 1, 2, 3 merged

Flashcards264 cardsFPractice quizPHost liveL

Questions 264

Answers

1
Adding more cores to a processor will not improve performance if (49)
More cores means more power consumption
Program execution speed depends on the clock frequency
The cores are all the same size
No part of the program can be done in parallel with another
2
In a 32-bit register, overflow can occur when adding a positive and negative number (179)
F
T
3
Which MIPS registers should not be used by the programmer? (105)
$ra
$v0, $v1
$sp
$at
4
Adjusting exponents and adding fractional parts describes (202)
Floating point multiplication
Integer multiplication
Integer addition
Floating point addition
5
In single-precision binary floating point notation, the fraction is ______ bits long (198)
11
8
23
52
6
You can calculate execution time by the formula (34)
Instructions / CPI / Clock speed
Instructions x CPI * Clock speed
Instructions / (CPI Clock speed)
Instructions x CPI / Clock speed
7
Memory which retains its contents while powered off is called (22)
DRAM
Volatile
SRAM
Non-volatile
8
To refer to each of the 32 MIPS processor registers, an instruction needs (83)
32 Bits
5 Bits
6 Bits
4 Bits
9
The percentage of good dies on a wafer is called (27)
Profit
The Beneficial Margin
The Yield
The Success Rate
10
Immediate operands are (72)
Data located in registers
Data located in the instruction itself
Data located in cache
Data located in main memory
11
Which is the fastest way to convert an array index to a memory address location (144)
shift left 2 in hardware
shift left 2 in software
successive adds of four
Shift right in hardware.
12
When a program is loaded into memory, the text segment (104)
Contains the word processor data for the text editor.
Contains the MIPS machine code
Contains the MIPS machine code
Holds the program's source code
13
A disadvantage of recursion is that it can consume a lot of (104)
Heap memory
Memory on the stack.
Registers
Execution cycles
14
In general, Intel's arithmetic instructions (152)
Execute faster than others'.
A source and destination register be the same.
Are not backward compatible.
Are a feat of modern engineering.
15
One of Hennessey and Patterson's design rules is (46)
Make the common case fast
Get the product to market as soon as possible
Make the instruction set as compact as possible
Memory is cheap
16
To move a unicode character from memory to a register, which MIPS instruction should be used?
lw
lb
lhu
lh
17
If one subroutine calls another, we must at least preserve (138)
The frame pointer.
The $sp register
The $s registers
The return address register
18
Which device accepts 16-bit immediate data and produces a 32-bit output? (255)
The hazard detector
Sign Extender
Register file
Program Counter
19
Machine Code optimized for speed has the fewest instructions (140)
true
false
20
The easiest conversion is from (82)
Decimal to Binary
Hexadecimal to binary
Binary to decimal
Hexadecimal to decimal
21
Unlike in mathematics, floating point addition is not associative (229)
F
T
22
Which is an example of improving bandwidth? (30)
More open cashier lanes in the food court
More streamlined procedures for cashiers in the food court
Moving dorms closer class so students can get there faster
Faster cash registers in the cashier lanes
23
The most popular instruction set for embedded devices is the _____ ISA (146)
Verilog
Motorola
MIPS
Acorn RISC Machine
24
The reason the "ugly" Intel architecture has survived over the years is that (157)
It followed IBM's market strategy.
Intel successfully lobbied for congressional support.
It is not backward compatible.
It has huge market forces behind it.
25
What is the fastest way of saving a variable in the $a0 register (138)
Push it onto the stack
Use the $at register.
Add it to $0 and store it in another register.
Write it to disk
26
A simple and efficient way to implement division is to (189)
Look up the answers in a division table
use hardware to perform "grammar school" division
Group bits into fours and perform hexadecimal division
Run multiply hardware with reversed voltages
27
While the C language is compiled and linked, the Java language is __________ (145)
Linked
Parsed
Executed
Compiled and Interpreted
28
Cycles Per Instruction (CPI) must be an integer value (35)
True
False
29
What is the primary difference between active matrix displays and liquid crystal displays? (17)
LCD pixels are more expensive than active matrix pixels.
The LCD makes its own light.
The active matrix display controls intensity of each pixel.
The active matrix display consumes less power.
30
Which describe two steps in the chip fabrication process? (26)
Testing and resting
Slicing and dicing
Waxing and faxing
Slipping and shipping
31
A conditional branch is appropriate for (92)
Jumping to a new location
Termination of a program
A subroutine call
A loop
32
Inside the CPU, what activity does the datapath perform? (19)
Translates assembly instructions into binary code
Moves data to and from the central processor
Tells the memory what to do
Arithmetic operations
33
How is multiply overflow handled in MIPS? (188)
The sign-bit logic prevents overflow
It triggers an exception
The EPC register contains the result of the overflow
Software must check the hi half of the product
34
Which describes a symbol table? (125)
A table that contains no information.
A list of all globally declared data in the program.
A list of needed and supplied program locations.
The list of ASCII characters used in the program code.
35
Which operation is unsupported in the ARM-7 instruction set? (146)
Divide
atomic swap
Shift right and left.
load halfword signed
36
The rectangular area on a wafer containing the circuitry for a single chip is a(n) (27)
Module
Die
Imprint
Mask
37
Which is the most complex operation for floating point calculations?
Subtraction
Rounding
Division
Multiplication
38
Which instruction field is 6-bits long? (82)
Op Code
Rs
Imm
Rt
39
What is the job of an assembler (15)
Translate assembly code into machine language
Link together code modules into a single executable module
Translate source code in to assembly code
Run assembly code
40
Two important laws in computer performance are (19)
Ampere's and Ohm's
Moore's and Ampere's
Moore's and Amdahl's
Amdahl's and Planck's
41
Which is not in the hierarchical layers of a system? (13)
I/O Software
Applications Software
Hardware
Systems Software
42
Which is not a characteristic of Dynamic Random Access Memory (DRAM) (20)
It operates on a slower clock than the CPU
It takes very little power
It is the most dense in terms of bits per area on a chip
It is one of the fastest types of random access memory
43
Integer addition and subtraction use the same add-and-carry logic in the CPU. (178)
true
false
44
A terabyte is 2^___ bytes (5)
30
10
20
40
45
The typical MIPS immediate instruction (e.g., ADDI, SUBI) references how many registers? (72)
Five
Two
Three
Four
46
The difference between ASCII and Unicode is (109)
Apple only uses ASCII
ASCII is popular in foreign languages and Unicode is not
Unicode is used by Java and ASCII is used by Microsoft.
Unicode can represent more characters than ASCII.
47
The total time required for a computer to complete a task, including overhead is called (29)
CPU Time
Application Time
Execution Time
Absolute Time
48
An embedded computer is distinguished from other types of computers by the fact that it (6)
Is built into an incredibly small space
Has strict power requirements
Primarily runs a single, predetermined application
Must survive harsh operating conditions
49
Negative exponents in floating point binary are represented as twos-complement numbers (200)
F
T
50
The ALU zero output is used in handling which instruction? (256)
ADD
SLT
JAL
BNE
51
The BEQ and BNE instructions are (90)
Not Jumps
Offsets
Unconditional Branches
Labels
52
The CPU time spent in the O.S. performing tasks on behalf of a program is called (32)
Execution Time
System CPU Time
Wasted Time
User CPU Time
53
If subroutine A calls subroutine B (100)
Subroutine A must preserve $ra.
Subroutine B must preserve $ra.
Subroutine A is a leaf routine.
Subroutine A is a non-leaf subroutine.
54
What do we call the binary representation of machine instructions? (14)
Bytes
Assembly Language
Source Code
Machine Language
55
The J-type instruction has an immediate field which is ____ bits long (113)
16
6
5
26
56
Which spreadsheet tools can read this sheet and export to xlsx format?
LibreOffice
Excel
Numbers
Google Docs
57
Which is not a MIPS design principle? (161)
Smaller is faster.
Speed = Efficiency.
Good design demands good compromises.
Simplicity favors regularity.
58
How many five-bit inputs does the Register File have (253)
Four
Five
Two
Three
59
The chip manufacturing process begins with (26)
Checking the wafer for defects
Slices of silicon about 0.1" thick
Photolithography
A silicon ingot
60
Most liquid crystal displays (LCDs) now rely on ______ technology (18)
Advanced Module
Amplitude Modulation
Faster Raster
Active Matrix
61
Which instruction has a 26-bit immediate field? (270)
J
ADD
BEQ
ADDI
62
Dynamically Linked Libraries (DLLs) (129)
. Are rarely used today because main memory is cheap.
Were invented by IBM in the 1960s.
Are not linked until runtime.
Tend to make programs take more space in memory
63
Which is not one of the eight great ideas in Computer Architecture? (11)
Reliability via Redundancy
Make the Common Case Fast
Performance via Pipelining
Fail Early
64
Which is not used in the execution of a BEQ instruction? (254)
Register file
Sign extender
Data memory
ALU
65
The 64-bit ARM v8 is an extension of the design of the 32-bit ARM v7. (159)
false
true
66
Processing data for multimedia, especially graphics, has led to (181)
wider cache buses
three-color processors
quad-precision floating-point instructions
Long registers with carry partitions
67
To adjust the stack pointer to make room for three 32-bit values,the correct instruction is(99)
addi $sp, $sp, 0xc
addi $sp, $sp, 0xfff4
addi $sp, $sp, 12
addi $sp, $sp, -3
68
When binary floating point numbers are multiplied, what does not happen? (206)
Fractional parts are multiplied
The result is rounded
The result is normalized
Exponents are made equal
69
The big problem with the switch to multi-processing is that (43)
More heat-sinking is required
Programmers have to think differently
The chips are more difficult to manufacture
Performance measures are very different
70
In a recursive procedure (101)
Local storage must be allocated on the stack.
$ra does not have to be preserved.
$t register values must be preserved
JAL is never used.
71
A sequence of instructions without branches in the middle is call a(n) (93)
Code Segment
Subroutine
Text
Basic Code Block
72
MUX is the abbreviation for which device? (247)
Memory access
Multiplexer
Memory storage
Multiplier
73
Which signed byte has the smallest value? (95)
0x01
0x0f
0x7f
0x80
74
ARMS processors can load blocks of memory into multiple registers with one instruction (148)
false
true
75
How many bits in a MIPS instruction are allocated for immediate data (84)
16
15
12
8
76
Negative numbers can be represented in binary (75)
False
True
77
In binary, 1011 + 1 = (179)
11
0x1010
0xC
0x1100
78
In the instruction "add $s0, $s1, $s2", $s1 and $s2 are (67)
Immediate Values
Outputs
Operands
Memory Locations
79
Performance and execution time are (30)
Proportionally Related
The Same Thing
Exponentially Related
Inversely Related
80
Which number below is the hex representation of a negative floating point number? (202)
0x08010000
0xC0100000
0x20400000
0x10800000
81
The market in which Intel processors do not do well is (158)
Laptops
Web servers
Servers
Cell phones
82
Which area of performance improvement is out of the scope of this course (9)
Algorithms
Memory Systems
The I/O System
Programming Languages
83
The info. tech. industry accounts for about __ of the U.S.A.'s gross national product (3)
60
30
10
50
84
What is the input to the Java Virtual Machine (131)
Java source code
Java bytecode
Dynamically linked libraries
MSIL
85
Which is not a difference between the MIPS and ARM processor (146)
The number of registers
The number of data addressing modes
The year of first product announcement
The use of the zero register
86
The two operands of the division process are (189)
The dividend and the remainder
The quotient and the dividend
Can be stored in conventional registers
The quotient and the remainder
87
Moore's Law (11)
Underscores the importance of parallelism
Says to optimize tasks that are frequently performed
Expects chip capacity to change about every two years
Says performance improvements are limited by the problem
88
Two methods of indexing lists of numbers are (141)
Arrays and pointers
Tables and arrays
Increments and decrements.
. in-order and out-of-order
89
The main reason for the shift from uniprocessors to multi-core processors was (40)
The Internet
Personal Mobile Devices
iPhones and iPads
The Power Wall
90
The execution time of a program is related to (35)
Number of instructions times the average CPI for the program
The number of instructions divided by the clock speed
The average CPI times the clock speed
The number of instructions times the clock speed
91
Execution time is the only valid measure of performance (52)
False
True
92
Machine A is 4x faster than machine B. B runs an app in 30 sec. How fast can A run it? (33)
7.5 sec
120 sec
10 sec
90 sec
93
The interface between low-level hardware and software is called (78)
Cache
The interpreter
The instruction set architecture
The operating system
94
What best describes what a linker does? (124)
Compresses executable code size.
Resolves references among separate code modules.
Calculates load parameters for the program.
Ensures instructions are on 4-byte boundaries.
95
Twos-complement is a method of (182)
Expressing negative numbers
Multiplying numbers
Adding logarithms
Performing addition
96
A floating point number is composed of (197)
Exponent, Fraction, Logarithm
Sign, Characteristic, Mantissa
Sign, Extent
1 bit for the sign 8 for the exponent & 23 for the fraction
97
How many registers does a branch instruction read from the register file? (258)
One
Two
Three
Five
98
The C code "d = a - e" could be represented in MIPS assembly as (65)
SUB d, e, a
sub $t0, $t2, $t1 #where d=$t0, a=$t1 and e=$t2
SUB a, e, d
sub $t0, $t1, $t2 #where d=$t0, a=$t1 and e=$t2
99
Which memory is non-volatile? (23)
ROM
SRAM
Main Memory
Cache
100
The time it takes a computer to complete a task is an example of (29)
Response Time
Clock Frequency
Consistency
Bandwidth
101
In binary, 1011 - 100 = (179)
1010
11
0x7
1100
102
Addition can be speeded up by (182)
widening the address bus
adding smaller numbers
increasing the processor voltage
Performing add operations in parallel with more hardware
103
The two parts of the integer division result are (189)
The dividend and the divisor
The dividend and the remainder
The quotient and the dividend
Stored in the HI and LO registers
104
Yield is primarily affected by (28)
N and P Doping
Etching Chemistry
Transistor Size
Wafer Defects
105
The jal instruction is unique because it also (97)
Writes to the $ra register
Resets the $pc register
Always executes (i.e., it is not conditional)
Does not update the program counter
106
Storage and use of decimal numbers for computation in computers is (107)
Never used anymore.
Becoming more likely with the advent of multicore processors
Sometimes useful in financial applications.
Unpopular because the standards are complicated.
107
ARM's NEON extensions (223)
Support quad-precision calculations
Support 64-bit floating point numbers
Provide simultaneous multithreading support
Provide multimedia support
108
Each time the cost of computing goes down (4)
The cost of big IT project failure goes up.
The risks associated with processor failure increase.
New opportunities put processors into new areas of life.
It's more difficult for hackers to create effective exploits
109
To get a better picture of CPU performance (47)
Customers are allowed to assess CPU performance on their own
A suite of benchmark programs is used
Industry has standards for honesty and integrity in testing
Tests are run multiple times
110
Multiplying two 32 bit numbers (184)
Must be 64 bits long
Cannot overflow a 32-bit register
Can overflow a 64-bit register
Can result in a 64-bit product
111
What MIPS registers bound the local variable storage area for a subroutine? (103)
The stack pointer and the frame pointer.
The stack pointer and the return address.
$ra and $sp
The frame pointer and the global pointer.
112
Intel's instruction set is characterized by (149)
Simplicity.
Reverse engineering.
Backward compatibility.
Regularity.
113
When MIPS detects an overflow condition a(n) _____ occurs (180)
branch
dump
exception
halt
114
Which device determines which instruction to read next? (253)
The PC
The instruction Multiplexer
The Write-back logic
The control unit
115
What operation wound combine 0x80000000 with 0xC0000000 to produce 0x40000000? (89)
XOR
NOR
XAND
NAND
116
Normalization is (205)
Converting the number from binary to hexadecimal
Setting the exponents so they are equal
The same as rounding
Adjusting the exponent so that the number is >=1 and
117
The principle of prediction is based on the idea that (12)
It is better to ask for forgiveness than ask for permission.
When the going gets tough, the tough get going.
If you don't succeed at first, try, try again.
You should eat dessert first.
118
The MIPS and ARM instruction sets differ widely on how they perform (147)
Multiplication.
Addition.
Memory addressing.
Register addressing.
119
Performance is inversely proportional to execution time (30)
False
True
120
If machines run the same program with the same CPI, what does execution time depend on? (38)
The Paging Algorithm
Clock Cycle Time
Cache
Number of Memory Accesses
121
Which is not one of the five classic components of a computer (17)
Memory
Display
Control
Datapath
122
The compiler can affect CPI (39)
True
False
123
Which instruction is commonly used for a subroutine call? (96)
j
jr $ra
call
jump and link (jal)
124
Which is not a length of floating point data? (221)
32 bits
16 bits
8 bits
64 bits
125
The primary technology used in fabricating computer chips is (41)
3D printing
Tunneling
NAND-flash
Complementary Metal Oxide Semiconductor (CMOS)
126
The advantage of capacitive sensing is that (19)
It allows you to point with toes as well as your fingers
It requires less power than other types of touch screens
It allows multiple touches to be detected at the same time
It can more precisely locate the point of touch
127
In binary, subtracting a large positive number from a small positive number produces (75)
An overflow condition
A zero in the leftmost bit of the result
An underflow condition
A one in the left bit of the result
128
Which of the following types of memory is nonvolatile? (23)
Flash
DIMM
SRAM
DRAM
129
Parallel execution strategies which work for integer types also work for floating point types
T
F
130
Which is not a feature of the MIPS processor? (244)
There are only two conditional branch instructions
The only memory accesses are the load and store instructions
The set arithmetic instructions are simple building-blocks
Complex instructions are longer than simpler instructions
131
Which is not a step in multiplying two floating point numbers? (209)
Aligning the exponents
Multiplying the fractional parts
Rounding
Normalizing
132
How can a CPI of less than 1.0 be achieved? (39)
Issuing multiple instructions per clock cycle
Shorter instructions
Executing fewer instructions
Faster clock cycle time
133
In the last 30 years, tech. revolutions begun by a new computer design were cut short by (3)
Foreign Competition
Laws
Even Faster Computers
Presidential Decree
134
the NaN notion can result from (199)
Subtracting two numbers which are nearly the same
Performing an operation with a NaN value
A sign miscompare
Incorrect use of associativity
135
What is the job of an assembler? (15)
It executes machine instructions
Translates assembly instructions into machine instructions
It translates source code into byte code
It optimizes loop structures
136
The MIPS J instruction uses ______ addressing (116)
Shift right two bits
Register
Replace bottom 28 bits of PC
PC-relative
137
Which is not a factor in CPU power consumption? (41)
Instruction Length
Supply Voltage
Switching Frequency
Capacitive Load
138
In determining the sign of the quotient, what rule must always hold? (193)
Dividend = Quotient * Divisor + Remainder
The remainder's sign bit = XOR of the operands' sign bits
Sign of Remainder = sign of quotient
The remainder is always positive
139
What happens in edge-triggered clocking? (249)
The clock pulse starts on the falling edge
State is recorded on the rising edge
Instruction execution determines when the clock pulse starts
Combinational logic continues after the pulse
140
SPEC benchmarks are (46)
Unreliable because testers cheat
A long-standing set of standard test programs
Obsolete because of recent innovations in chip design
A brand new way to measure performance
141
If by mistake you divide a positive number by a negative number using unsigned divide
The remainder is the dividend
The remainder will be zero
The quotient is zero and the remainder is the divisor
An exception occurs
142
On which of the following does execution time not depend? (34)
Clock Cycle Time
Average Cycles per Instruction
Number of Instructions Executed
Amount of System Memory
143
If four-byte alignment is required, at which address could an integer not be stored? (70)
0x0041000E
0x0041791C
0x00ABFEDC
0x00123454
144
How many registers does an R-type instruction read from the register file? (258)
Two
three-color processors
Five
One
145
Which of the following is a pseudoinstruction? (137)
addi
lgb
move
add
146
Which kind of parallelism is not discussed in computer organization and architecture? (10)
Data-Level Parallelism
Thread-Level Parallelism
Circuit-Level Parallelism
Instruction-Level Parallelism
147
Which set of binary numbers are the two's complement of each other? (76)
00000001 11111111
00000001 00000000
00000011 11111100
00000000 11111111
148
Each ARM instruction begins with _____ (147)
A valid bit.
An indication of how long it is.
Bits that tell the length of the instruction.
Condition codes
149
The sign of an integer divide is (192)
The XOR of the dividend and divisor signs
he OR of the dividend and divisor signs
The AND of the dividend and divisor signs
The NOR of the dividend and divisor signs
150
A major difficulty in determining performance is that it (26)
Takes a long time to do accurately
Can be measured in different ways
Can be affected by other components in the computer
Is largely independent of the processor itself
151
Inlining saves (138)
The overhead of a procedure call
Space on the heap.
Programmers from themselves
variables onto the stack.
152
Intel's widest register is ____ bits wide (152)
512
32
128
256
153
Which is not a type of MIPS instruction? (113)
I-type
J-type
R-type
B-Type
154
The role of the personal computer is now being replaced by the (7)
Personal Mobile Device
Cloud
Super Computer
Wearable Computer
155
With the arrival of multimedia applications in the late 1990s, Intel introduced (150)
Micropipelining.
Special registers for audio and video data.
Single Instruction Multiple Data (SIMD) instructions.
A brand new microarchitecture.
156
Which results in better performance? (34)
Decreased Clock Rate
Increased Instruction Count
Decreased CPI
Increased CPI
157
Which is not a MIPS type of instruction? (82)
B
I
R
J
158
Adding sign-extended, shifted-left-2 immediate data to the program counter is done in the
J instruction
BEQ instrustion
ADDI instruction
JAL instruction
159
The length of a single-precision floating point binary number is ____ bits (200)
128
16
64
32
160
The problem with one's-complement representation of negative numbers was (79)
There were more positive numbers than negative
It was more difficult to implement than twos-complement
It confused the programmers
There is no such thing as one's complement
161
Which has the least influence on the performance of a computer running a program? (10)
The Processor
The Algorithm Chosen
The Power Supply Wattage
The I/O System and Devices
162
New instructions can be added to the instruction set without changing the compiler. (227)
T
F
163
To add two numbers, you would use what device? (252)
Cache
Branch control
ALU
Control Unit
164
The EPC register is used for (182)
performing a crash dump
Storing the PC when an exception occurs
handling environmental protection faults
resuming a halted program
165
Which signed number is the same as 0x80? (78)
0x7FFFFF80
0xff80
0xff08
0x80FFFFFF
166
Which register is used in allocation of memory on the heap? (104)
$fp
$gp
The global pointer
$sp
167
The abstract interface between the hardware and the lowest level of software is called the (22)
Instruction set architecture
Operating system
Machine design specification
Application programming interface
168
MIPS supports single-precision floating point, but not double-precision floating point (211)
T
F
169
What is not an advantage of high-level code (16)
It is closer to natural language
It improves programmer productivity
It produces the most compact run-time code
It allows machine independence
170
What is the generic name of instructions which move data between the processor and memory? (68)
Data transfer instructions
Arithmetic instructions
Memory Control
Branch instructions
171
How many Branch instructions does the MIPS processor recognize? (90)
Four (BNE, BEQ, BEZ, BNZ)
One (BRN)
Three (BNE, BEQ, BEZ)
Two (BNE, BEQ)
172
Alongside the computer revolution is the ______ revolution (4)
Electrical
Travel
Information
Leisure
173
A key to the future of computation is (43)
Making computer science courses harder
Creating lower-power processors
Doubling the size of computer science programs
Understanding parallelism
174
What is the vocabulary of commands understood by the processor? (62)
The assembly language
The intermediate code
The byte code
The instruction set
175
After an exception, the EPC register contains (182)
The reason why the exception occurred
The location of the crash dump
The system exit instruction
the PC when the exception occurred
176
Idling computers consume a tiny fraction of the power they use at full load (50)
True
False
177
The notion that integrated circuit resources double every 18-24 months is known as (11)
Moore's Law
Boyle's Law
Amdahl's Law
Ohm's Law
178
MIPS conditional branch instructions require the comparison of (90)
rs and rd
rs and rt
rt and rd
One register to zero
179
Registers in the current Intel processors are the same length (153)
false
true
180
The contents of a memory in register $s1 may be expressed as: (64)
Regs[$s1]
Mem[$s1]
4($s1)
Mem[Regs[$s1]]
181
The measure of the frequency of instructions of certain types across program(s) is the (39)
Instruction Mix
Instruction Set Architecture
Average CPI
Program Count
182
In MIPS, data is moved to and from memory using (68)
R-type add and subtract instructions
Branch instructions
I-type add and subtract instructions
Load and Store instructions
183
In single-precision binary floating point notation, the exponent is ______ bits long (198)
23
11
Eight
52
184
All MIPS instructions have (81)
A function code
Immediate data
Length of 32 bits
Embedded memory addresses
185
To create all conditions for branches, MIPS uses which set of instructions? (93)
The $0 register, slt/slti and bne/beq
sll, srl, beq, bne, and the $zero register
blt, bgt, lgbt
j, jne, jeq, jal and the $zero register
186
Another name for the fractional part of a floating point number is the (204)
Denominator
Significand
Numerator
Exponent
187
Which language will have the most lines of code to implement a sort algorithm? (66)
C#
C++
Python
Assembly code
188
Energy efficiency can be an important performance goal (50)
False
True
189
A major problem in measuring performance of a computer is (28)
It is impossible to match conditions from run to run
It can be measured several ways
Variations in the manufacturing process
Laughing at a English translation of the instruction manual
190
In which situation does an overflow occur? (180)
Adding a pos and neg # and getting a neg result.
Adding two positive numbers and getting a negative result.
Adding two negative numbers and getting a negative result.
Adding a pos and neg # and getting a pos result.
191
The Intel instruction set has special instructions for string move and string compare (154)
true
false
192
A branch instruction is a(n) ____ instruction (113)
K-Type
I-type
R-type
B-Type
193
What is the output of a Java compiler (131)
Java bytecode
A .jar file.0
Dynamically linked libraries
Assembly instructions
194
Computers write numbers to memory differently than characters (178)
false
true
195
What is the output of a linker? (126)
The binary executable.
Assembly language instructions
Microsoft Intermediate Language.
Java bytecode.
196
Which has the least lines of code? (226)
Unoptimized DGEMM written in C
Optimized DGEMM written in conventional X86 assembly
Optimized DGEMM written in C
Optimized DGEMM written in conventional AVX assembly
197
The general name for the unit used to operate on or hold data within a processor is called a
Datapath element
Memory bank
Register
Pipeline stage
198
An invalid mathematical operation which can happen when computers do calculations is (189)
Adding zero
Adding zero to a negative number
Multiplying by zero
Arithmetic overflow
199
The concept of cloud computing is most closely associated with (7)
Vague Computing
Transaction Processing
Warehouse Scale Computing
Personal Mobile Devices
200
Silicon is a (25)
Transductor
Insulator
Semiconductor
Conductor
201
MMX and SSE (224)
Are Intel multi-media instruction extensions
Add about 150 instructions to the x-86 instruction set
Are MIPS multi-media instruction extensions
Add about 100 instructions to the ARM instruction set
202
In MIPS Assembly Language, which is the instruction for a single-precision operation? (212)
smul
muls
mul.s
adds
203
A problem with flash memory is that (23)
The bits are easy to write but difficult to read
The number of writes are limited
It loses its memory when power is turned off
It is cheaper but slower than magnetic media
204
The standard for floating point notation is (198)
ISO-9000
NFPA-1002
ASA-400
IEEE-754
205
Faster multipliers (188)
Use much higher clock speeds
Are easier to implement than faster dividers
Work better with smaller numbers
Perform shifts and adds sequentially
206
Which 16-bit hex value would represent a negative number? (74)
0x8001
0x100f
0x0001
0x7777
207
Some character sets use 16-bit characters. The would be loaded by which MIPS command? (110)
LBU
LHU
LB
LW
208
Multiple SPEC ratios are combined using (47)
The Geometric Mean
Logarithms
Covariance
The Arithmetic Mean
209
The LW instruction requires that ___ register(s) be read. (245)
Zero
Two
One
Three
210
In a 32-bit MIPS instruction, the location of bit 0 (the least significant bit) (74)
Is on the left
Varies by memory manufacturer
Is on the right
Depends on the instruction
211
The length of the product of an m-bit multiplicand and an n-bit multiplier is without sign bit
64 bits
up to m + n bits
2n bits
2m bits
212
In multimedia arithmetic, an overflow often results in (181)
a blue screen
Achieving a maximum value, but no greater
a black screen
an exception
213
Which of the following does not result in performance gains? (45)
Memory Hierarchies
Pipelining
Higher Voltage Requirements
Prediction
214
In twos complement notation, the most negative value is (79)
0x8000
0x0000
0xfffe
0x0001
215
Which instruction does the MIPS processor not have? (87)
SUBI
ADD
ADDI
Shift Left Logical (SLL)
216
The length of a double-precision floating point binary number is ____ bits (200)
64
16
128
32
217
When does combinational logic happen? (249)
Between the rising edges of two adjacent clock pulses
After Instruction Fetch and before Instruction Decode
During Data Write
During instruction Fetch
218
The iPad processor is the (21)
ARM A5
Intel Xeon
AMD Barcelona
Core i7
219
Early microprocessors which disappeared from the market did so because (158)
They were too expensive.
Slow clock speeds
Insufficient address bits
Unreliable performance
220
The J instruction immediate field is _____ bits. (113)
15
16
26
28
221
When floating point operations were introduced in the Intel processor line, (150)
Integer processing slowed down as a result
The instruction length was increased to 32-bits.
They did not affect the instruction set.
They were not performed in the CPU itself.
222
The shamt field in the MIPS instruction is use for which instruction? (88)
SLL
XOR
SLLI
SLT
223
The average number of clock cycles per instruction (CPI) in a program is an integer value
True
False
224
Normalized scientific notation (196)
Means the number left of the point is not zero
Makes it difficult to multiply numbers
Applies to binary but not decimal numbers
Is widely used in integer calculations
225
What is the native, binary form of instructions used by the processor? (81)
Intermediate Language
Machine instructions
Assembly language
Byte Code
226
Which instruction calculates an address with the ALU and then uses that address in main memory?
ADDI
SLT
BEQ
LW
227
For a given method, which language will have the most lines of code?
C/C++
Java
Assembly
C#
228
The chip fabrication process begins with (26)
A Package
A Die
An Ingot
A Wafer
229
The iPad accelerometer and gyroscope (20)
Significantly shorten battery life
Were originally in the iPad design but are now of little use
Provide three-axis information to the CPU
Sense six degree-of-freedom motion
230
Current IEEE 802.11 wireless technology allows for data transfer rates of up to (24)
10 megabits per second
100 gigabits per second
1 gigabit per second
1 megabit per second
231
Understanding the hardware layer (8)
Provides for a Steady Income
Is the Key to an Early Retirement
Builds Character
Makes you a Better Programmer
232
The fastest access to data is via (64)
Main Memory
Cache
Registers
Disk
233
Which action does a MIPS branch instruction not do? (254)
Sign extend the immediate data
Write to the register file
Subtract the contents of one register from another
Shift the immediate data left two bits
234
An increase in which of the following has the greatest effect on CPU power consumption (41)
Switching Frequency
Voltage
Capacitive Load
CPI
235
Using many adders to speed up division (194)
Is difficult because the result is unknown
Works because fast multiplies can check the answe
Works because of Moore's law
Won't work because decisions are made after the subtractions
236
Why do the addresses of integers stored in main memory need the last two bits set to 0? (69)
There is an alignment restriction
The cache can operate more efficiently
Data can be located anywhere in memory
The data bus is only 30 bits wide
237
Which of the following indicators are inversely proportional to performance? (31)
Bandwidth
Execution time
Instructions per second
Memory bits transferred per second
238
Instruction Set Architecture has to do with (22)
The layout of circuits on the CPU chip
The size and shape of the computer
The interface between the lowest level software and hardware
Die Yield
239
Intel's 1978 microprocessor, the 8086 had a 16-bit architecture with (150)
An accumulator.
A hardware stack.
A single 16-bit register
Fewer registers than are common now.
240
Compilers need no assistance optimizing vector code for routines like matrix multiply (227)
T
F
241
CPU A has a CPI of 3 and CPU B has a CPI of 2. What is the performance difference? (38)
1.5
0.67
0.33
It cannot be determined
242
Applications running on servers are particularly sensitive to (32)
Cache Memory
I/O Time
The Amount of Main Memory
Disk Performance
243
The vocabulary of the CPU is the (62)
Instruction Set
Datapath
Programs
Architecture
244
The most common use of processor arithmetic is for (184)
calculating loop variables
graphics calculations
Address calculations
floating point multiplies
245
Twos complement is (179)
A way of representing negative numbers
A way to speed up multiplication.
A way of preventing overflow
A way of adding positive numbers
246
Which is not a field in the MIPS instruction format? (88)
rt
shamt
precision
op
247
Subword parallelism is another name for what happens in (222)
SIMD
Floating point multiply
Unsigned addition
Simultaneous Multi-Threading
248
Which processor uses instructions which are of different lengths (155)
Samsung
ARM
AMD
MIPS
249
Besides cost, what is important in selecting a computer for a particular application? (29)
The size of the screen
Limitations in performance
The speed of the network card
Range of power supply voltages the machine will tolerate
250
Which is not an element of the MIPS CPU (246)
Arithmetic Logic Unit
Program Counter
Register File
Flash memory
251
Memory alignment means (69)
Similar data items are placed next to each other in memory
Dynamic RAM is faster than Static RAM
More than one word can be read from memory at once
The bottom two bits of a MIPS instruction address must be 0
252
In general, a more complex instruction set means (148)
Faster execution time
More compact code
Shorter clock cycles
Lower cycles per instruction
253
Which is not a "great idea" in computer architecture? (12)
Memory Hierarchy
Branch Prediction
Flash Cache
Pipelining
254
In binary, 1111...10001 - 1111...11000 = (180)
0000...01000
0xfffffff9
1111...11000
1000...01001
255
Which instruction is commonly used for a subroutine call? (96)
bne
jr $ra
jal
beq
256
AVX is (225)
AMD's improvement on Intel's SIMD instructions
Intel's Advanced Vector Extensions
Addition and Vector multiplication
Capable of performing 2 double-precision operations at once
257
Double precision floating point numbers may not accurately represent decimal numbers (218)
T
F
258
Because binary representation involves only 0's and 1's, carrying in addition is unnecessary.
F
T
259
Only theoretical mathematicians care about floating point accuracy. (231)
F
T
260
Instructions * CPI / GHz is (35)
Run time in nanoseconds
Average CPI
The number of instructions executed
Run time in seconds
261
The first step in adding two floating point numbers is (205)
Express the numbers in binary fractions
Make the exponents equal
Compare the signs
Make the fractional parts equal
262
The MIPS instruction set has many more data addressing modes than the ARM set. (147)
false
true
263
A large number (more than 200) instructions were added to the intel instruction set (150)
Because of the demand for audio and video processing
In 2006 when silicon prices dropped.
In 1978 when the 8080 first went to market
In 1985 with the first 16-bit processor
264
The Arithmetic Logic Unit is not used for which set of instructions (246)
Load (LW, LH, LB)
Store (SW, SH, SB)
Branch (BEQ, BNE)
J-types