Chapter 4
Basic Instructions
4.1 Copying Data
mov Instructions
• mov (“move”) instructions are really copy
instructions, like simple assignment
statements in a high-level language
• Format: mov destination, source
register
or
memory
register,
memory
or
immediate
Operand Restrictions
• Operands must be same size
• Can’t move from memory to memory
– mov nbr1, nbr2
illegal if nbr1 and nbr2 reference doublewords
in memory
– Instead use a register
mov eax, nbr2
mov nbr1, eax
• Can only move one byte, word or
doubleword at a time
Effect on Flags
• In general, an instruction may have one of
three effects:
– no flags are altered
– specific flags are given values depending on
the results of the instruction
– some flags may be altered, but their settings
cannot be predicted
• No mov instruction changes any flag
Machine Code
• Depends on operand type(s), with several
different opcodes used for mov instructions
• Word-size and doubleword-size
instructions use same opcodes, but wordsize instructions have 66 prefix byte
• Object and source code from listing file
B0 9B
mov al, 155
66| B8 009B
mov ax, 155
B8 0000009B
mov eax, 155
ModR/M Byte
• Part of the object code for many
instructions
• Used to encode specific registers
• Used to distinguish between instructions
that share the same opcode
• Used to specify memory modes
ModR/M Fields
•
•
•
•
mod (mode), 2 bits
reg (register), 3 bits
r/m (register/memory), 3 bits
Examples of encodings
– mod = 00 and r/m = 101 combined always
means direct memory addressing
– reg = 011 means the EBX register in a 32-bit
instruction
xchg Instruction
• Swaps the values referenced by its two
operands
– Can’t have both operands in memory
• Does not alter any flag
4.2 Integer Addition and
Subtraction Instructions
add Instruction
• Format: add destination, source
• The integer at source is added to the
integer at destination and the sum
replaces the old value at destination
• SF, ZF, OF, CF, PF and AF flags are set
according to the value of the result of the
operation
– Example: CF = 1 if there is a carry out of the
sum
Addition Example
• Before
EAX: 00000075
ECX: 000001A2
• Instruction
add eax, ecx
• After
EAX: 00000217
ECX: 000001A2
SF=0 ZF=0 CF=0 OF=0
sub Instruction
• Format: sub destination, source
• The integer at source is subtracted from
the integer at destination and the
difference replaces the old value at
destination
• SF, ZF, OF, CF, PF and AF flags are set
according to the value of the result of the
operation
– Example: ZF = 1 if the difference is zero
Subtraction Example
• Before
doubleword at Dbl: 00000100
• Instruction
sub Dbl, 2
• After
Dbl: 000000FE
SF=0 ZF=0 CF=0 OF=0
Instruction Encoding
• Opcode depends on operand types
• The ModR/M byte distinguishes
– Between operand types
– Between add, sub and other operations for
certain operand types
• An small immediate operand is sometimes
encoded as a byte even in a 32-bit
instruction
Increment and Decrement Instructions
• inc
destination
– Adds 1 to destination
• dec
destination
– Subtracts 1 from destination
• Each sets same flags as add or sub
except for CF which isn’t changed
neg Instruction
• neg
destination
• Negates (takes the 2's complement of) its
operand
– A positive value gives a negative result
– A negative value will become positive
– Zero remains 0
• Affects same flags as add and sub
Programming in Assembly Language
• Start with a design
• Plan register usage
– Decide what registers will be used for what
variables in the design
– There are only a few available registers
• Plan memory usage
4.3 Multiplication Instructions
Multiplication Instruction Mnemonics
• mul for unsigned multiplication
– Operands treated as unsigned numbers
• imul for signed multiplication
– Operands treated as signed numbers and
result is positive or negative depending on the
signs of the operands
mul Instruction Format
• mul
source
• Single operand may be byte, word,
doubleword or quadword in register or
memory (not immediate) and specifies one
factor
• Location of other factor is implied
– AL for byte-size source
– AX for word source
– EAX for doubleword source
– RAX for quadword source
mul Instruction Operation
• When a byte source is multiplied by the
value in AL, the product is put in AX
• When a word source is multiplied by the
value in AX, the product is put in DX:AX
– The high-order 16 bits in DX and the loworder 16 bits in AX
• When a doubleword source is multiplied
by the value in EAX, the product is put in
EDX:EAX
• Product of two quadwords in RAX:DAX
Double-Length Product
• The “double-length” product ensures that
the result will always fit in the destination
location
• If significant bits of the product actually
“spill over” into the high-order half (AH, DX
or EDX), then CF and OF are both set to 1
• If the high-order half is not significant, then
CF and OF are both cleared to 0
– For unsigned multiplication, this is when the
high-order half is all 0’s
mul Instruction Example
• Before
EAX: 00000005
EBX: 00000002
EDX: ????????
• Instruction
mul ebx
• After
EAX: 0000000A
EBX: 00000002
EDX: 00000000
CF=OF=0
imul Instruction Formats
• imul source
• imul register, source
• imul register, source, immediate
imul source
• “Single-operand format”
• Similar to mul source except for signed
operands
• CF=OF=0 if each bit in the high-order half
is the same as the sign bit in the low-order
half
• CF=OF=1 otherwise (the bits in the highorder half are significant)
Single-Operand Example
• Before
AX: ??05
byte at Factor: FF
• Instruction
imul Factor
• After
AX: FFFB
CF=OF=0
imul register,source
• “Two-operand format”
• Source operand can be in a register, in
memory, or immediate
• Register contains other factor, and also
specifies the destination
• Both operands must be word-size or
doubleword-size, not byte-size
• Product must “fit” in destination register
– CF and OF are cleared to 0 if result fits
– CF and OF are set to 1 if it doesn’t fit
Two-operand Example
• Before
EBX: 0000000A
• Instruction
imul ebx, 10
• After
EBX: 00000064
CF=OF=0
imul register,source,immediate
• “Three-operand format”
• The two factors are given by source
(register or memory) and the immediate
value
• The first operand, a register, specifies the
destination for the product
• Operands register and source are the same
size, both 16-bit or both 32-bit (not 8-bit)
• If the product will fit in the destination
register, then CF and OF are cleared to 0; if
not, they are set to 1
Three-Operand Example
• Before
word at Value: 08F2
BX: ????
• Instruction
imul bx, Value, 1000
• After
BX: F150
CF=OF=1
4.4 Division Instructions
Division Instruction Formats
• idiv source
for signed operands
• div source
for unsigned operands
• source identifies the divisor
– Byte, word, doubleword or quadword
– In memory or register, but not immediate
Implicit Dividend for div and idiv
• Byte source divided into word in AX
• Word source divided into doubleword in
DX:AX
• Doubleword source divided into quadword
in EDX:EAX
• Quadword source divided into RDX:RAX
Results of div and idiv
• Byte-size divisor:
quotient in AL and remainder in AH
• Word-size divisor:
quotient in AX and remainder in DX
• Doubleword-size divisor:
quotient in EAX and remainder in EDX
• Quadword-size divisor:
quotient in RAX and remainder in RDX
Results of div and idiv
• All division operations satisfy the relation
dividend = quotient*divisor + remainder
– For signed division, the remainder will have
same sign as dividend
Flag Settings
• Division instructions do not set flags to any
meaningful values
• They may change previously set values of
AF, CF, OF, PF, SF or ZF
Unsigned Division Example
• Before
EDX: 00 00 00 00
EAX: 00 00 00 64
EBX: 00 00 00 0D
• Instruction
div ebx
; 100/13
• After
EDX: 00000009
EAX: 00000007
100 = 7 * 13 + 9
Signed Division Example
• Before
EDX: FF FF FF FF
EAX: FF FF FF 9C
ECX: 00 00 00 0D
• Instruction
idiv ecx ; -100/13
• After
EDX: FFFFFFF7
–100 = (–7) * 13 + (–9)
EAX: FFFFFFF9
Errors in Division
• Caused by
– Dividing by 0, or
– Quotient too large to fit in destination
• Triggers an exception
– The interrupt handler routine that services this
exception may vary from system to system
– When a division error occurs for a program
running under Visual Studio, an error window
pops up
Preparing for Division
• Dividend must be extended to double
length
• Example
– Copy a doubleword dividend to EAX
– Extend dividend to EDX:EAX
• For unsigned division, use mov edx, 0
• For signed division, use cdq instruction
– Finally use div or idiv instruction
Convert Instructions
• No operand
• cbw
sign extends the byte in AL to the word in AX
• cwd
sign extends the word in AX to the doubleword
in DX:AX
• cdq
sign extends the doubleword in EAX to the
quadword in EDX:EAX
• cqo
sign extends the quadword in RAX to RDX:RAX