COMPUTER ORGANIZATION AND DESIGN

RISC-V
Edition
The Hardware/Software Interface

Chapter 2
Instructions: Language of
the Computer
 §2.1 Introduction
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
Chapter 2 — Instructions: Language of the Computer — 2
The RISC-V Instruction Set
 Used as the example throughout the book
 Developed at UC Berkeley as open ISA
 Now managed by the RISC-V Foundation
(riscv.org)
 Typical of many modern ISAs
 See RISC-V Reference Data tear-out card
 Similar ISAs have a large share of embedded
core market
 Applications in consumer electronics, network/storage
equipment, cameras, printers, …

Chapter 2 — Instructions: Language of the Computer — 3

 §2.2 Operations of the Computer Hardware
Arithmetic Operations
 Add and subtract, three operands
 Two sources and one destination
add a, b, c // a gets b + c
 All arithmetic operations have this form
 Design Principle 1: Simplicity favours
regularity
 Regularity makes implementation simpler
 Simplicity enables higher performance at
lower cost

Chapter 2 — Instructions: Language of the Computer — 4

Arithmetic Example
 C code:
f = (g + h) - (i + j);
 Compiled RISC-V code:
add t0, g, h // temp t0 = g + h
add t1, i, j // temp t1 = i + j
sub f, t0, t1 // f = t0 - t1

Chapter 2 — Instructions: Language of the Computer — 5

 §2.3 Operands of the Computer Hardware
Register Operands
 Arithmetic instructions use register
operands

 RISC-V has a 32 × 64-bit register file

 Use for frequently accessed data
 64-bit data is called a “doubleword”

32 x 64-bit general purpose registers x0 to x30
 32-bit data is called a “word”

 Design Principle 2: Smaller is faster

 c.f. main memory: millions of locations

Chapter 2 — Instructions: Language of the Computer — 6

RISC-V Registers
 x0: the constant value 0
 x1: return address
 x2: stack pointer
 x3: global pointer
 x4: thread pointer
 x5 – x7, x28 – x31: temporaries
 x8: frame pointer
 x9, x18 – x27: saved registers
 x10 – x11: function arguments/results
 x12 – x17: function arguments

Chapter 2 — Instructions: Language of the Computer — 7

Register Operand Example
 C code:
f = (g + h) - (i + j);
 f, …, j in x19, x20, …, x23

 Compiled RISC-V code:

add x5, x20, x21
add x6, x22, x23
sub x19, x5, x6

Chapter 2 — Instructions: Language of the Computer — 8

Memory Operands
 Main memory used for composite data

Arrays, structures, dynamic data
 To apply arithmetic operations

Load values from memory into registers

Store result from register to memory
 Memory is byte addressed

Each address identifies an 8-bit byte
 RISC-V is Little Endian

Least-significant byte at least address of a word

c.f. Big Endian: most-significant byte at least address
 RISC-V does not require words to be aligned in
memory

Unlike some other ISAs
Chapter 2 — Instructions: Language of the Computer — 9
Memory Operand Example
 C code:
A[12] = h + A[8];
 h in x21, base address of A in x22

 Compiled RISC-V code:

 Index 8 requires offset of 64
8 bytes per doubleword


ld x9, 64(x22)
add x9, x21, x9
sd x9, 96(x22)

Chapter 2 — Instructions: Language of the Computer — 10

Registers vs. Memory
 Registers are faster to access than
memory
 Operating on memory data requires loads
and stores
 More instructions to be executed
 Compiler must use registers for variables
as much as possible
 Only spill to memory for less frequently used
variables
 Register optimization is important!

Chapter 2 — Instructions: Language of the Computer — 11

Immediate Operands
 Constant data specified in an instruction
addi x22, x22, 4

 Make the common case fast

 Small constants are common
 Immediate operand avoids a load instruction

Chapter 2 — Instructions: Language of the Computer — 12

Sign Extension
 Representing a number using more bits
 Preserve the numeric value
 Replicate the sign bit to the left
 c.f. unsigned values: extend with 0s
 Examples: 8-bit to 16-bit
 +2: 0000 0010 => 0000 0000 0000 0010
 –2: 1111 1110 => 1111 1111 1111 1110

 In RISC-V instruction set

 lb: sign-extend loaded byte
 lbu: zero-extend loaded byte

Chapter 2 — Instructions: Language of the Computer — 13

 §2.5 Representing Instructions in the Computer
Representing Instructions
 Instructions are encoded in binary
 Called machine code

 RISC-V instructions
 Encoded as 32-bit instruction words
 Small number of formats encoding operation code
(opcode), register numbers, …
 Regularity!

Chapter 2 — Instructions: Language of the Computer — 14

RISC-V R-format Instructions
funct7 rs2 rs1 funct3 rd opcode
7 bits 5 bits 5 bits 3 bits 5 bits 7 bits

 Instruction fields
 opcode: operation code
 rd: destination register number
 funct3: 3-bit function code (additional opcode)
 rs1: the first source register number
 rs2: the second source register number
 funct7: 7-bit function code (additional opcode)

Chapter 2 — Instructions: Language of the Computer — 15

R-format Example
funct7 rs2 rs1 funct3 rd opcode
7 bits 5 bits 5 bits 3 bits 5 bits 7 bits

add x9,x20,x21
0 21 20 0 9 51

0000000 10101 10100 000 01001 0110011

0000 0001 0101 1010 0000 0100 1011 0011two =

015A04B316

Chapter 2 — Instructions: Language of the Computer — 16

RISC-V I-format Instructions
immediate rs1 funct3 rd opcode
12 bits 5 bits 3 bits 5 bits 7 bits

 Immediate arithmetic and load instructions

 rs1: source or base address register number
 immediate: constant operand, or offset added to base address

2s-complement, sign extended
 Design Principle 3: Good design demands good
compromises
 Different formats complicate decoding, but allow 32-bit
instructions uniformly
 Keep formats as similar as possible

Chapter 2 — Instructions: Language of the Computer — 17

RISC-V S-format Instructions
imm[11:5] rs2 rs1 funct3 imm[4:0] opcode
7 bits 5 bits 5 bits 3 bits 5 bits 7 bits

 Different immediate format for store instructions

 rs1: base address register number
 rs2: source operand register number
 immediate: offset added to base address

Split so that rs1 and rs2 fields always in the same place

Chapter 2 — Instructions: Language of the Computer — 18

Stored Program Computers
The BIG Picture
 Instructions represented in
binary, just like data
 Instructions and data stored
in memory
 Programs can operate on
programs
 e.g., compilers, linkers, …
 Binary compatibility allows
compiled programs to work
on different computers
 Standardized ISAs

Chapter 2 — Instructions: Language of the Computer — 19

 §2.6 Logical Operations
Logical Operations
 Instructions for bitwise manipulation
Operation C Java RISC-V
Shift left << << slli
Shift right >> >>> srli
Bit-by-bit AND & & and, andi
Bit-by-bit OR | | or, ori
Bit-by-bit XOR ^ ^ xor, xori
Bit-by-bit NOT ~ ~
 Useful for extracting and inserting groups of
bits in a word
Chapter 2 — Instructions: Language of the Computer — 20
Shift Operations
funct6 immed rs1 funct3 rd opcode
6 bits 6 bits 5 bits 3 bits 5 bits 7 bits

 immed: how many positions to shift

 Shift left logical
 Shift left and fill with 0 bits
 slli by i bits multiplies by 2i
 Shift right logical
 Shift right and fill with 0 bits
 srli by i bits divides by 2i (unsigned only)

Chapter 2 — Instructions: Language of the Computer — 21

 AND Operations
 Useful to mask bits in a word
 Select some bits, clear others to 0
and x9,x10,x11

x10 00000000 00000000 00000000 00000000 00000000 00000000 00001101

11000000
x11 00000000 00000000 00000000 00000000 00000000 00000000 00111100
00000000

x9 00000000 00000000 00000000 00000000 00000000 00000000 00001100

00000000

Chapter 2 — Instructions: Language of the Computer — 22

 OR Operations
 Useful to include bits in a word
 Set some bits to 1, leave others unchanged
or x9,x10,x11

x10 00000000 00000000 00000000 00000000 00000000 00000000 00001101

11000000
x11 00000000 00000000 00000000 00000000 00000000 00000000 00111100
00000000

x9 00000000 00000000 00000000 00000000 00000000 00000000 00111101

11000000

Chapter 2 — Instructions: Language of the Computer — 23

 XOR Operations
 Differencing operation
 Set some bits to 1, leave others unchanged
xor x9,x10,x12 // NOT operation

x10 00000000 00000000 00000000 00000000 00000000 00000000 00001101 11000000

x12 11111111 11111111 11111111 11111111 11111111 11111111 11111111

11111111

x9 11111111 11111111 11111111 11111111 11111111 11111111 11110010

00111111

Chapter 2 — Instructions: Language of the Computer — 24

 §2.7 Instructions for Making Decisions
Conditional Operations
 Branch to a labeled instruction if a condition is
true
 Otherwise, continue sequentially

 beq rs1, rs2, L1

 if (rs1 == rs2) branch to instruction labeled L1

 bne rs1, rs2, L1

 if (rs1 != rs2) branch to instruction labeled L1

Chapter 2 — Instructions: Language of the Computer — 25

Compiling If Statements
 C code:
if (i==j) f = g+h;
else f = g-h;
 f, g, … in x19, x20, …
 Compiled RISC-V code:
bne x22, x23, Else
add x19, x20, x21
beq x0,x0,Exit // unconditional
Else: sub x19, x20, x21
Exit: …
Assembler calculates addresses
Chapter 2 — Instructions: Language of the Computer — 26
Compiling Loop Statements
 C code:
while (save[i] == k) i += 1;

i in x22, k in x24, address of save in x25
 Compiled RISC-V code:
Loop: slli x10, x22, 2
add x10, x10, x25
lw x9, 0(x10)
bne x9, x24, Exit
addi x22, x22, 1
beq x0, x0, Loop
Exit: …

Chapter 2 — Instructions: Language of the Computer — 27

Basic Blocks
 A basic block is a sequence of instructions
with
 No embedded branches (except at end)
 No branch targets (except at beginning)

 A compiler identifies basic

blocks for optimization
 An advanced processor
can accelerate execution
of basic blocks

Chapter 2 — Instructions: Language of the Computer — 28

More Conditional Operations
 blt rs1, rs2, L1
 if (rs1 < rs2) branch to instruction labeled L1
 bge rs1, rs2, L1
 if (rs1 >= rs2) branch to instruction labeled L1
 Example
 if (a > b) a += 1;
 a in x22, b in x23
bge x23, x22, Exit // branch if b >= a
addi x22, x22, 1
Exit:

Chapter 2 — Instructions: Language of the Computer — 29

Signed vs. Unsigned
 Signed comparison: blt, bge
 Unsigned comparison: bltu, bgeu
 Example
 x22 = 1111 1111 1111 1111 1111 1111 1111 1111
 x23 = 0000 0000 0000 0000 0000 0000 0000 0001
 x22 < x23 // signed

–1 < +1
 x22 > x23 // unsigned

+4,294,967,295 > +1

Chapter 2 — Instructions: Language of the Computer — 30

Chapter 2 — Instructions: Language of the Computer — 31
Chapter 2 — Instructions: Language of the Computer — 32
Chapter 2 — Instructions: Language of the Computer — 33
 Branch Instruction Summary
Comparison Registers Signed Unsigned Zero
equal (eq) rs1 = rs2 beq beq beqz
not equal
rs1 ≠ rs2 bne bne bnez
(ne)
less than (lt) rs1 < rs2 blt bltu bltz
greater than
rs1 > rs2 bgt bgtu bgtz
(gt)
less or equal
rs1 ≤ rs2 ble bleu blez
(le)
greater or
rs1 ≥ rs2 bge bgeu bgez
equal (ge)

Chapter 2 — Instructions: Language of the Computer — 34

Chapter 2 — Instructions: Language of the Computer — 35
 §2.8 Supporting Procedures in Computer Hardware
Procedure Calling
 Steps required
1. Place parameters in registers x10 to x17
2. Transfer control to procedure
3. Acquire storage for procedure
4. Perform procedure’s operations
5. Place result in register for caller
6. Return to place of call (address in x1)

Chapter 2 — Instructions: Language of the Computer — 36

Registers

Chapter 2 — Instructions: Language of the Computer — 37

Procedure Call Instructions
 Procedure call: jump and link
jal x1, ProcedureLabel
 Address of following instruction put in x1

 Jumps to target address

 Procedure return: jump and link register

jalr x0, 0(x1)
 Like jal, but jumps to 0 + address in x1

 Use x0 as rd (x0 cannot be changed)

 Can also be used for computed jumps


e.g., for case/switch statements
Chapter 2 — Instructions: Language of the Computer — 38
Leaf Procedure Example
 C code:
int leaf_example (
int g, int h,
int i, int j)
{
int f;
f = (g + h) - (i + j);
return f;
}

Arguments g, …, j in x10, …, x13

f in x20

temporaries x5, x6

Need to save x5, x6, x20 on stack
Chapter 2 — Instructions: Language of the Computer — 39
Leaf Procedure Example
 RISC-V code:
leaf_example:
addi sp,sp,-12 Save x5, x6, x20 on stack
sw x5,8(sp)
sw x6,4(sp)
sw x20,0(sp
add x5,x10,x11 x5 = g + h
add x6,x12,x1 x6 = i + j
sub x20,x5,x6 f = x5 – x6
addi x10,x20,0 copy f to return register
lw x20,0(sp) Resore x5, x6, x20 from stack
lw x6,4(sp)
lw x5,12(sp)
addi sp,sp,12
jalr x0,0(x1) Return to caller

Chapter 2 — Instructions: Language of the Computer — 40

Local Data on the Stack

Chapter 2 — Instructions: Language of the Computer — 41

Register Usage
 x5 – x7, x28 – x31: temporary registers
 Not preserved by the callee

 x8 – x9, x18 – x27: saved registers

 If used, the callee saves and restores them

Chapter 2 — Instructions: Language of the Computer — 42

Non-Leaf Procedures
 Procedures that call other procedures
 For nested call, caller needs to save on the
stack:
 Its return address
 Any arguments and temporaries needed after
the call
 Restore from the stack after the call

Chapter 2 — Instructions: Language of the Computer — 43

Non-Leaf Procedure Example
 C code:
int fact (int n)
{
if (n < 1) return f;
else return n * fact(n - 1);
}

 Argument n in x10
 Result in x10

Chapter 2 — Instructions: Language of the Computer — 44

Leaf Procedure Example
 RISC-V code:
fact:
addi sp,sp,-8 Save return address and n on stack
sd x1,4(sp)
sd x10,0(sp)
addi x5,x10,-1 x5 = n - 1
bge x5,x0,L1 if n >= 1, go to L1
addi x10,x0,1 Else, set return value to 1
addi sp,sp,8 Pop stack, don’t bother restoring values
jalr x0,0(x1) Return
L1: addi x10,x10,-1 n=n-1
jal x1,fact call fact(n-1)
addi x6,x10,0 move result of fact(n - 1) to x6
ld x10,0(sp) Restore caller’s n
ld x1,4(sp) Restore caller’s return address
addi sp,sp,8 Pop stack
mul x10,x10,x6 return n * fact(n-1)
jalr x0,0(x1) return

Chapter 2 — Instructions: Language of the Computer — 45

Memory Layout
 Text: program code
 Static data: global
variables
 e.g., static variables in C,
constant arrays and strings
 x3 (global pointer)
initialized to address
allowing ±offsets into this
segment
 Dynamic data: heap
 E.g., malloc in C, new in
Java
 Stack: automatic storage
Chapter 2 — Instructions: Language of the Computer — 46
Local Data on the Stack

 Local data allocated by callee


e.g., C automatic variables
 Procedure frame (activation record)

Used by some compilers to manage stack storage
Chapter 2 — Instructions: Language of the Computer — 47
 §2.9 Communicating with People
Character Data
 Byte-encoded character sets
 ASCII: 128 characters

95 graphic, 33 control
 Latin-1: 256 characters

ASCII, +96 more graphic characters
 Unicode: 32-bit character set
 Used in Java, C++ wide characters, …
 Most of the world’s alphabets, plus symbols
 UTF-8, UTF-16: variable-length encodings

Chapter 2 — Instructions: Language of the Computer — 48

Byte/Halfword/Word Operations
 RISC-V byte/halfword/word load/store
 Load byte/halfword/word: Sign extend to 64 bits in rd

lb rd, offset(rs1)

lh rd, offset(rs1)

lw rd, offset(rs1)
 Load byte/halfword/word unsigned: Zero extend to 64 bits in rd

lbu rd, offset(rs1)

lhu rd, offset(rs1)

lwu rd, offset(rs1)
 Store byte/halfword/word: Store rightmost 8/16/32 bits

sb rs2, offset(rs1)

sh rs2, offset(rs1)

sw rs2, offset(rs1)

Chapter 2 — Instructions: Language of the Computer — 49

String Copy Example
 C code:
 Null-terminated string
void strcpy (char x[], char y[])
{ size_t i;
i = 0;
while ((x[i]=y[i])!='\0')
i += 1;
}

Chapter 2 — Instructions: Language of the Computer — 50

String Copy Example
 RISC-V code:
strcpy:
addi sp,sp,-4 // adjust stack for 1
doubleword
sd x19,0(sp) // push x19
add x19,x0,x0 // i=0
L1: add x5,x19,x11 // x5 = addr of y[i]
lbu x6,0(x5) // x6 = y[i]
add x7,x19,x10 // x7 = addr of x[i]
sb x6,0(x7) // x[i] = y[i]
beq x6,x0,L2 // if y[i] == 0 then exit
addi x19,x19, 1 // i = i + 1
jal x0,L1 // next iteration of loop
L2: ld x19,0(sp) // restore saved x19
addi sp,sp,4 // pop 1 doubleword from
stack
jalr x0,0(x1) // and return

Chapter 2 — Instructions: Language of the Computer — 51

 §2.10 RISC-V Addressing for Wide Immediates and Addresses
32-bit Constants
 Most constants are small
 12-bit immediate is sufficient
 For the occasional 32-bit constant
lui rd, constant
 Copies 20-bit constant to bits [31:12] of rd
 Extends bit 31 to bits [63:32]
 Clears bits [11:0] of rd to 0

lui x19, 976 // 0x003D0

0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0011 1101 0000 0000 0000 0000

addi x19,x19,128 // 0x500

0000 0000 0000 0000 0000 0000 0000 0000 0000 0000 0011 1101 0000 0101 0000 0000

Chapter 2 — Instructions: Language of the Computer — 52

Branch Addressing
 Branch instructions specify
 Opcode, two registers, target address
 Most branch targets are near branch
 Forward or backward
 SB format:
imm imm
[10:5] rs2 rs1 funct3 [4:1] opcode

imm[12] imm[11]

 PC-relative addressing
 Target address = PC + immediate × 2

Chapter 2 — Instructions: Language of the Computer — 53

Jump Addressing
 Jump and link (jal) target uses 20-bit
immediate for larger range
 UJ format:
imm[10:1] imm[19:12] rd opcode
5 bits 7 bits
imm[20] imm[11]

 For long jumps, eg, to 32-bit absolute address

 lui: load address[31:12] to temp register
 jalr: add address[11:0] and jump to target

Chapter 2 — Instructions: Language of the Computer — 54

RISC-V Addressing Summary

Chapter 2 — Instructions: Language of the Computer — 55

RISC-V Encoding Summary

Chapter 2 — Instructions: Language of the Computer — 56

 §2.11 Parallelism and Instructions: Synchronization
Synchronization
 Two processors sharing an area of memory
 P1 writes, then P2 reads
 Data race if P1 and P2 don’t synchronize

Result depends of order of accesses
 Hardware support required
 Atomic read/write memory operation
 No other access to the location allowed between the
read and write
 Could be a single instruction
 E.g., atomic swap of register ↔ memory
 Or an atomic pair of instructions

Chapter 2 — Instructions: Language of the Computer — 57

Synchronization in RISC-V
 Load reserved: lr.d rd,(rs1)
 Load from address in rs1 to rd
 Place reservation on memory address
 Store conditional: sc.d rd,(rs1),rs2
 Store from rs2 to address in rs1
 Succeeds if location not changed since the lr.d

Returns 0 in rd
 Fails if location is changed

Returns non-zero value in rd

Chapter 2 — Instructions: Language of the Computer — 58

Synchronization in RISC-V
 Example 1: atomic swap (to test/set lock variable)
again: lr.d x10,(x20)
sc.d x11,(x20),x23 // X11 = status
bne x11,x0,again // branch if store failed
addi x23,x10,0 // X23 = loaded value

 Example 2: lock
addi x12,x0,1 // copy locked value
again: lr.d x10,(x20) // read lock
bne x10,x0,again // check if it is 0
yet
sc.d x11,(x20),x12 // attempt to store
bne x11,x0,again // branch if fails
 Unlock:
sd x0,0(x20) // free lock
Chapter 2 — Instructions: Language of the Computer — 59
 §2.12 Translating and Starting a Program
Translation and Startup

Many compilers produce

object modules directly

Static linking

Chapter 2 — Instructions: Language of the Computer — 60

Producing an Object Module
 Assembler (or compiler) translates program into
machine instructions
 Provides information for building a complete
program from the pieces
 Header: described contents of object module
 Text segment: translated instructions
 Static data segment: data allocated for the life of the
program
 Relocation info: for contents that depend on absolute
location of loaded program
 Symbol table: global definitions and external refs
 Debug info: for associating with source code

Chapter 2 — Instructions: Language of the Computer — 61

Linking Object Modules
 Produces an executable image
1. Merges segments
2. Resolve labels (determine their
addresses)
3. Patch location-dependent and external
refs
 Could leave location dependencies for fixing
by a relocating loader

But with virtual memory, no need to do this

Program can be loaded into absolute location in
virtual memory space
Chapter 2 — Instructions: Language of the Computer — 62
Loading a Program
 Load from image file on disk into memory
1. Read header to determine segment sizes
2. Create virtual address space
3. Copy text and initialized data into memory

Or set page table entries so they can be faulted in
4. Set up arguments on stack
5. Initialize registers (including sp, fp, gp)
6. Jump to startup routine

Copies arguments to x10, … and calls main

When main returns, do exit syscall

Chapter 2 — Instructions: Language of the Computer — 63

Dynamic Linking
 Only link/load library procedure when it is
called
 Requires procedure code to be relocatable
 Avoids image bloat caused by static linking of
all (transitively) referenced libraries
 Automatically picks up new library versions

Chapter 2 — Instructions: Language of the Computer — 64

Lazy Linkage

Indirection table

Stub: Loads routine ID,

Jump to linker/loader

Linker/loader code

Dynamically
mapped code

Chapter 2 — Instructions: Language of the Computer — 65

Static library

Chapter 2 — Instructions: Language of the Computer — 66

Dynamic library

Chapter 2 — Instructions: Language of the Computer — 67

 Starting Java Applications

Simple portable
instruction set for
the JVM

Compiles
Interprets
bytecodes of
bytecodes
“hot” methods
into native
code for host
machine

Chapter 2 — Instructions: Language of the Computer — 68

 §2.13 A C Sort Example to Put It All Together
C Sort Example
 Illustrates use of assembly instructions
for a C bubble sort function
 Swap procedure (leaf)
void swap(int v[],
int k)
{
int temp;
temp = v[k];
v[k] = v[k+1];
v[k+1] = temp;
}
 v in x10, k in x11, temp in x5
Chapter 2 — Instructions: Language of the Computer — 69
The Procedure Swap
swap:
slli x6,x11,2 // reg x6 = k * 4
add x6,x10,x6 // reg x6 = v + (k * 4)
ld x5,0(x6) // reg x5 (temp) = v[k]
ld x7,4(x6) // reg x7 = v[k + 1]
sd x7,0(x6) // v[k] = reg x7
sd x5,4(x6) // v[k+1] = reg x5 (temp)
jalr x0,0(x1) // return to calling routine

Chapter 2 — Instructions: Language of the Computer — 70

The Sort Procedure in C
 Non-leaf (calls swap)
void sort (int v[], size_t n)
{
size_t i, j;
for (i = 0; i < n; i += 1) {
for (j = i – 1;
j >= 0 && v[j] > v[j + 1];
j -= 1) {
swap(v,j);
}
}
}

v in x10, n in x11, i in x19, j in x20

Chapter 2 — Instructions: Language of the Computer — 71

The Outer Loop
 Skeleton of outer loop:
 for (i = 0; i <n; i += 1) {

li x19,0 // i = 0
for1tst:
bge x19,x11,exit1 // go to exit1 if x19 ≥ x11 (i≥n)

(body of outer for-loop)

addi x19,x19,1 // i += 1
j for1tst // branch to test of outer loop
exit1:

Chapter 2 — Instructions: Language of the Computer — 72

The Inner Loop
 Skeleton of inner loop:
 for (j = i − 1; j >= 0 && v[j] > v[j + 1]; j − = 1) {
addi x20,x19,-1 // j = i −1
for2tst:
blt x20,x0,exit2 // go to exit2 if X20 < 0 (j < 0)
slli x5,x20,2 // reg x5 = j * 4
add x5,x10,x5 // reg x5 = v + (j * 4)
ld x6,0(x5) // reg x6 = v[j]
ld x7,4(x5) // reg x7 = v[j + 1]
ble x6,x7,exit2 // go to exit2 if x6 ≤ x7
mv x21, x10 // copy parameter x10 into x21
mv x22, x11 // copy parameter x11 into x22
mv x10, x21 // first swap parameter is v
mv x11, x20 // second swap parameter is j
jal x1,swap // call swap
addi x20,x20,-1 // j –= 1
j for2tst // branch to test of inner loop
exit2:

Chapter 2 — Instructions: Language of the Computer — 73

Preserving Registers
 Preserve saved registers:
addi sp,sp,-20 // make room on stack for 5 regs
sd x1,16(sp) // save x1 on stack
sd x22,12(sp) // save x22 on stack
sd x21,8(sp) // save x21 on stack
sd x20,4(sp) // save x20 on stack
sd x19,0(sp) // save x19 on stack

 Restore saved registers:

exit1:
sd x19,0(sp) // restore x19 from stack
sd x20,4(sp) // restore x20 from stack
sd x21,8(sp) // restore x21 from stack
sd x22,12(sp) // restore x22 from stack
sd x1,16(sp) // restore x1 from stack
addi sp,sp, 20 // restore stack pointer
jalr x0,0(x1)

Chapter 2 — Instructions: Language of the Computer — 74

 Effect of Compiler Optimization
Compiled with gcc for Pentium 4 under Linux

3 Relative Performance 140000 Instruction count

2.5 120000
100000
2
80000
1.5
60000
1
40000
0.5 20000
0 0
none O1 O2 O3 none O1 O2 O3

180000 Clock Cycles 2 CPI

160000
140000 1.5
120000
100000
1
80000
60000
40000 0.5
20000
0 0
none O1 O2 O3 none O1 O2 O3

Chapter 2 — Instructions: Language of the Computer — 75

Effect of Language and Algorithm
3 Bubblesort Relative Performance
2.5

1.5

0.5

0
C/ none C/ O1 C/ O2 C/ O3 J ava/ int J ava/ J IT

2.5 Quicksort Relative Performance

2

1.5

0.5

0
C/ none C/ O1 C/ O2 C/ O3 J ava/ int J ava/ J IT

3000 Quicksort vs. Bubblesort Speedup

2500

2000

1500

1000

500

0
C/ none C/ O1 C/ O2 C/ O3 J ava/ int J ava/ J IT

Chapter 2 — Instructions: Language of the Computer — 76

Lessons Learnt
 Instruction count and CPI are not good
performance indicators in isolation
 Compiler optimizations are sensitive to the
algorithm
 Java/JIT compiled code is significantly
faster than JVM interpreted
 Comparable to optimized C in some cases
 Nothing can fix a dumb algorithm!

Chapter 2 — Instructions: Language of the Computer — 77

 §2.14 Arrays versus Pointers
Arrays vs. Pointers
 Array indexing involves
 Multiplying index by element size
 Adding to array base address
 Pointers correspond directly to memory
addresses
 Can avoid indexing complexity

Chapter 2 — Instructions: Language of the Computer — 78

 Example: Clearing an Array
Array -> x10, size -> x11, I ->x5 Array -> x10, size -> x11, I ->x5
clear1(int array[], int size) { clear2(int *array, int size) {
int i; int *p;
for (i = 0; i < size; i += 1) for (p = &array[0]; p < &array[size];
array[i] = 0; p = p + 1)
} *p = 0;
}

li x5,0 // i = 0 mv x5,x10 // p = address

loop1: // of array[0]
slli x6,x5,2 // x6 = i * 4 slli x6,x11,2 // x6 = size * 4
add x7,x10,x6 // x7 = address add x7,x10,x6 // x7 = address
// of array[i] // of array[size]
sd x0,0(x7) // array[i] = 0 loop2:
addi x5,x5,1 // i = i + 1 sd x0,0(x5) // Memory[p] = 0
blt x5,x11,loop1 // if (i<size) addi x5,x5,4 // p = p + 4
// go to loop1 bltu x5,x7,loop2
// if (p<&array[size])
// go to loop2

Chapter 2 — Instructions: Language of the Computer — 79

Comparison of Array vs. Ptr
 Multiply “strength reduced” to shift
 Array version requires shift to be inside
loop
 Part of index calculation for incremented i
 c.f. incrementing pointer
 Compiler can achieve same effect as
manual use of pointers
 Induction variable elimination
 Better to make program clearer and safer

Chapter 2 — Instructions: Language of the Computer — 80

 §2.16 Real Stuff: MIPS Instructions
MIPS Instructions
 MIPS: commercial predecessor to RISC-V
 Similar basic set of instructions

32-bit instructions

32 general purpose registers, register 0 is always 0

32 floating-point registers

Memory accessed only by load/store instructions

Consistent use of addressing modes for all data sizes
 Different conditional branches

For <, <=, >, >=

RISC-V: blt, bge, bltu, bgeu

MIPS: slt, sltu (set less than, result is 0 or 1)

Then use beq, bne to complete the branch

Chapter 2 — Instructions: Language of the Computer — 81

Instruction Encoding

Chapter 2 — Instructions: Language of the Computer — 82

 §2.17 Real Stuff: x86 Instructions
The Intel x86 ISA
 Evolution with backward compatibility
 8080 (1974): 8-bit microprocessor

Accumulator, plus 3 index-register pairs
 8086 (1978): 16-bit extension to 8080

Complex instruction set (CISC)
 8087 (1980): floating-point coprocessor

Adds FP instructions and register stack
 80286 (1982): 24-bit addresses, MMU

Segmented memory mapping and protection
 80386 (1985): 32-bit extension (now IA-32)

Additional addressing modes and operations

Paged memory mapping as well as segments

Chapter 2 — Instructions: Language of the Computer — 83

The Intel x86 ISA
 Further evolution…

i486 (1989): pipelined, on-chip caches and FPU

Compatible competitors: AMD, Cyrix, …

Pentium (1993): superscalar, 64-bit datapath

Later versions added MMX (Multi-Media eXtension)
instructions

The infamous FDIV bug

Pentium Pro (1995), Pentium II (1997)

New microarchitecture (see Colwell, The Pentium Chronicles)

Pentium III (1999)

Added SSE (Streaming SIMD Extensions) and associated
registers

Pentium 4 (2001)

New microarchitecture

Added SSE2 instructions

Chapter 2 — Instructions: Language of the Computer — 84

The Intel x86 ISA
 And further…

AMD64 (2003): extended architecture to 64 bits

EM64T – Extended Memory 64 Technology (2004)

AMD64 adopted by Intel (with refinements)

Added SSE3 instructions

Intel Core (2006)

Added SSE4 instructions, virtual machine support

AMD64 (announced 2007): SSE5 instructions

Intel declined to follow, instead…

Advanced Vector Extension (announced 2008)

Longer SSE registers, more instructions
 If Intel didn’t extend with compatibility, its
competitors would!

Technical elegance ≠ market success

Chapter 2 — Instructions: Language of the Computer — 85

Basic x86 Registers

Chapter 2 — Instructions: Language of the Computer — 86

Basic x86 Addressing Modes
 Two operands per instruction
Source/dest operand Second source operand
Register Register
Register Immediate
Register Memory
Memory Register
Memory Immediate
 Memory addressing modes
 Address in register
 Address = Rbase + displacement
 Address = Rbase + 2scale × Rindex (scale = 0, 1, 2, or 3)
 Address = Rbase + 2scale × Rindex + displacement
Chapter 2 — Instructions: Language of the Computer — 87
x86 Instruction Encoding
 Variable length
encoding
 Postfix bytes specify
addressing mode
 Prefix bytes modify
operation

Operand length,
repetition, locking, …

Chapter 2 — Instructions: Language of the Computer — 88

Implementing IA-32
 Complex instruction set makes
implementation difficult
 Hardware translates instructions to simpler
microoperations

Simple instructions: 1–1

Complex instructions: 1–many
 Microengine similar to RISC
 Market share makes this economically viable
 Comparable performance to RISC
 Compilers avoid complex instructions
Chapter 2 — Instructions: Language of the Computer — 89
 §2.18 The Rest of the RISC-V Instruction Set
Other RISC-V Instructions
 Base integer instructions (RV64I)
 Those previously described, plus
 auipc rd, immed // rd = (imm<<12) + pc

follow by jalr (adds 12-bit immed) for long jump
 slt, sltu, slti, sltui: set less than (like MIPS)
 addw, subw, addiw: 32-bit add/sub
 sllw, srlw, srlw, slliw, srliw, sraiw: 32-bit shift
 32-bit variant: RV32I
 registers are 32-bits wide, 32-bit operations

Chapter 2 — Instructions: Language of the Computer — 90

Instruction Set Extensions
 M: integer multiply, divide, remainder
 A: atomic memory operations
 F: single-precision floating point
 D: double-precision floating point
 C: compressed instructions
 16-bit encoding for frequently used
instructions

Chapter 2 — Instructions: Language of the Computer — 91

 §2.19 Fallacies and Pitfalls
Fallacies
 Powerful instruction  higher performance
 Fewer instructions required
 But complex instructions are hard to implement

May slow down all instructions, including simple ones
 Compilers are good at making fast code from simple
instructions
 Use assembly code for high performance
 But modern compilers are better at dealing with
modern processors
 More lines of code  more errors and less productivity

Chapter 2 — Instructions: Language of the Computer — 92

Fallacies
 Backward compatibility  instruction set
doesn’t change
 But they do accrete more instructions

x86 instruction set

Chapter 2 — Instructions: Language of the Computer — 93

Pitfalls
 Sequential words are not at sequential
addresses
 Increment by 4, not by 1!
 Keeping a pointer to an automatic variable
after procedure returns
 e.g., passing pointer back via an argument
 Pointer becomes invalid when stack popped

Chapter 2 — Instructions: Language of the Computer — 94

 §2.20 Concluding Remarks
Concluding Remarks
 Design principles
1. Simplicity favors regularity
2. Smaller is faster
3. Good design demands good
compromises
 Make the common case fast
 Layers of software/hardware
 Compiler, assembler, hardware
 RISC-V: typical of RISC ISAs
 c.f. x86

Chapter 2 — Instructions: Language of the Computer — 95