Bài giảng Computer Architecture: Chapter 2 - Prof. Jerry Breecher

Chia sẻ: Codon_03 Codon_03 | Ngày: | Loại File: PPT | Số trang:37

Thêm vào BST

Báo xấu

65
lượt xem 4
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Đến với "Bài giảng Computer Architecture: Chapter 2 - Instruction Sets do Prof. Jerry Breecher biên soạn các bạn sẽ được tìm hiểu các vấn đề chính về introduction; classifying Instruction Set Architectures; memory Addressing; Operations in the Instruction Set;...

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Bài giảng Computer Architecture: Chapter 2 - Prof. Jerry Breecher

Computer Architecture Chapter 2 Instruction Sets Prof. Jerry Breecher CSCI 240 Fall 2003
Introduction 2.1 Introduction 2.2 Classifying Instruction Set Architectures 2.3 Memory Addressing 2.4 Operations in the Instruction Set 2.5 Type and Size of Operands 2.6 Encoding and Instruction Set 2.7 The Role of Compilers 2.8 The MIPS Architecture Bonus Chap. 2 Instruction Sets 2
Introduction The Instruction Set Architecture is that portion of the machine visible to the assembly level programmer or to the compiler writer. software instruction set hardware 1. What are the advantages and disadvantages of various instruction set alternatives. 2. How do languages and compilers affect ISA. 3. Use the DLX architecture as an example of a RISC architecture. Chap. 2 Instruction Sets 3
Classifying Instruction Set 2.1 Introduction Architectures 2.2 Classifying Instruction Set Architectures 2.3 Memory Addressing 2.4 Operations in the Instruction Set Classifications can be by: 2.5 Type and Size of Operands 2.6 Encoding and Instruction Set 2.7 The Role of Compilers 1. Stack/accumulator/register 2.8 The DLX Architecture 2. Number of memory operands. 3. Number of total operands. Chap. 2 Instruction Sets 4
Instruction Set Basic ISA Architectures Classes Accumulator: 1 address add A acc acc + mem[A] 1+x address addx A acc acc + mem[A + x] Stack: 0 address add tos tos + next General Purpose Register: ALU Instructions 2 address add A B EA(A) EA(A) + EA(B) can have two or 3 address add A B C EA(A) EA(B) + EA(C) three operands. Load/Store: 0 Memory load R1, Mem1 ALU Instructions can load R2, Mem2 have 0, 1, 2, 3 operands. add R1, R2 Shown here are cases of 0 and 1. 1 Memory add R1, Mem2 Chap. 2 Instruction Sets 5
Instruction Set Basic ISA Architectures Classes The results of different address classes is easiest to see with the examples here, all of which implement the sequences for C = A + B. Stack Accumulator Register Register (Register-memory) (load-store) Push A Load A Load R1, A Load R1, A Push B Add B Add R1, B Load R2, B Add Store C Store C, R1 Add R3, R1, R2 Pop C Store C, R3 Registers are the class that won out. The more registers on the CPU, the better. Chap. 2 Instruction Sets 6
Instruction Set Intel 80x86 Architectures Integer Registers GPR0 EAX Accumulator GPR1 ECX Count register, string, loop GPR2 EDX Data Register; multiply, divide GPR3 EBX Base Address Register GPR4 ESP Stack Pointer GPR5 EBP Base Pointer – for base of stack seg. GPR6 ESI Index Register GPR7 EDI Index Register CS Code Segment Pointer SS Stack Segment Pointer DS Data Segment Pointer ES Extra Data Segment Pointer FS Data Seg. 2 GS Data Seg. 3 PC EIP Instruction Counter Eflags Condition Codes Chap. 2 Instruction Sets 7
Memory Addressing 2.1 Introduction 2.2 Classifying Instruction Set Architectures 2.3 Memory Addressing 2.4 Operations in the Instruction Set Sections Include: 2.5 Type and Size of Operands 2.6 Encoding and Instruction Set Interpreting Memory Addresses 2.7 The Role of Compilers 2.8 The DLX Architecture Addressing Modes Displacement Address Mode Immediate Address Mode Chap. 2 Instruction Sets 8
Memory Interpreting Memory Addressing Addresses What object is accessed as a function of the address and length? Objects have byte addresses – an address refers to the number of bytes counted from the beginning of memory. Little Endian – puts the byte whose address is xx00 at the least significant position in the word. Big Endian – puts the byte whose address is xx00 at the most significant position in the word. Alignment – data must be aligned on a boundary equal to its size. Misalignment typically results in an alignment fault that must be handled by the Operating System. Chap. 2 Instruction Sets 9
Memory Addressing Addressing Modes This table shows the most common modes. A more complete set is in Figure 2.6 Addressing Mode Example Meaning When Used Instruction Register Add R4, R3 R[R4]
Memory Displacement Addressing Addressing Mode How big should the displacement be? For addresses that do fit in displacement size: Add R4, 10000 (R0) For addresses that don’t fit in displacement size, the compiler must do the following: Load R1, address Add R4, 0 (R1) Depends on typical displaces as to how big this should be. On both IA32 and DLX, the space allocated is 16 bits. Chap. 2 Instruction Sets 11
Memory Immediate Address Addressing Mode Used where we want to get to a numerical value in an instruction. At high level: At Assembler level: a = b + 3; Load R2, 3 Add R0, R1, R2 if ( a > 17 ) Load R2, 17 CMPBGT R1, R2 goto Addr Load R1, Address Jump (R1) So how would you get a 32 bit value into a register? Chap. 2 Instruction Sets 12
Operations In The 2.1 Introduction 2.2 Classifying Instruction Set Architectures 2.3 Memory Addressing Instruction Set 2.4 Operations in the Instruction Set 2.5 Type and Size of Operands Sections Include: 2.6 Encoding and Instruction Set 2.7 The Role of Compilers Detailed information about types 2.8 The DLX Architecture of instructions. Instructions for Control Flow (conditional branches, jumps) Chap. 2 Instruction Sets 13
Operations In The Operator Types Instruction Set Arithmetic and logical - and, add Data transfer - move, load Control - branch, jump, call System - system call, traps Floating point - add, mul, div, sqrt Decimal - add, convert String - move, compare Multimedia - 2D, 3D? e.g., Intel MMX and Sun VIS Chap. 2 Instruction Sets 14
Operations In The Control Instructions Instruction Set Conditional branches are 20% of all instructions!! Control Instructions Issues: • taken or not • where is the target • link return address • save or restore Instructions that change the PC: • (conditional) branches, (unconditional) jumps • function calls, function returns • system calls, system returns Chap. 2 Instruction Sets 15
Operations In The Control Instructions Instruction Set There are numerous tradeoffs: There are numerous tradeoffs: Compare and branch condition in general-purpose register + no extra compare, no state passed + no special state but uses up a register between instructions -- branch condition separate from branch logic in pipeline -- requires ALU op, restricts code some data for MIPS scheduling opportunities > 80% branches use immediate data, > Implicitly set condition codes - Z, N, V, C 80% of those zero + can be set ``for free'' 50% branches use == 0 or 0 -- constrains code reordering, extra compromise in MIPS state to save/restore branch==0, branch0 Explicitly set condition codes compare instructions for all other + can be set ``for free'', decouples compares branch/fetch from pipeline -- extra state to save/restore Chap. 2 Instruction Sets 16
Operations In The Control Instructions Instruction Set Link Return Address: Save or restore state: implicit register - many recent What state? architectures use this function calls: registers system calls: registers, flags, PC, PSW, etc + fast, simple -- s/w save register before next call, Hardware need not save registers surprise traps? Caller can save registers in use Callee save registers it will use explicit register + may avoid saving register Hardware register save IBM STM, VAX CALLS -- register must be specified Faster? processor stack Many recent architectures do no register + recursion direct saving -- complex instructions Or do implicit register saving with register windows (SPARC) Chap. 2 Instruction Sets 17
Type And Size of Operands 2.1 Introduction The type of the operand is usually encoded in the Opcode – a LDW 2.2 Classifying Instruction Set Architectures implies loading of a word. 2.3 Memory Addressing Common sizes are: 2.4 Operations in the Instruction Set Character (1 byte) Half word (16 bits) 2.5 Type and Size of Operands Word (32 bits) 2.6 Encoding and Instruction Set Single Precision Floating Point (1 Word) 2.7 The Role of Compilers Double Precision Floating Point (2 Words) 2.8 The DLX Architecture Integers are two’s complement binary. Floating point is IEEE 754. Some languages (like COBOL) use packed decimal. Chap. 2 Instruction Sets 18
Encoding And Instruction Set 2.1 Introduction This section has to do with how an 2.2 Classifying Instruction Set Architectures assembly level instruction is 2.3 Memory Addressing encoded into binary. 2.4 Operations in the Instruction Set 2.5 Type and Size of Operands Ultimately, it’s the binary that is 2.6 Encoding and Instruction Set read and interpreted by the 2.7 The Role of Compilers machine. 2.8 The DLX Architecture We will be using the Intel instruction set which is defined at: http://developer.intel.com/design/Pentium4/manuals. Volume 2 has the instruction set. Chap. 2 Instruction Sets 19
Encoding And 80x86 Instruction Instruction Set Encoding for ( index = 0; index < iterations; index++ ) Here’s some 0040D3AF C7 45 F0 00 00 00 00 mov dword ptr [ebp-10h],0 sample code that’s 0040D3B6 EB 09 jmp main+0D1h (0040d3c1) been disassembled. 0040D3B8 8B 4D F0 mov ecx,dword ptr [ebp-10h] It was compiled 0040D3BB 83 C1 01 add ecx,1 with the debugger 0040D3BE 89 4D F0 mov dword ptr [ebp-10h],ecx option so is not 0040D3C1 8B 55 F0 mov edx,dword ptr [ebp-10h] optimized. 0040D3C4 3B 55 F8 cmp edx,dword ptr [ebp-8] 0040D3C7 7D 15 jge main+0EEh (0040d3de) long_temp = (*alignment + long_temp) % 47; This code 0040D3C9 8B 45 F4 mov eax,dword ptr [ebp-0Ch] was 0040D3CC 8B 00 mov eax,dword ptr [eax] produced 0040D3CE 03 45 EC add eax,dword ptr [ebp-14h] using Visual 0040D3D1 99 cdq Studio 0040D3D2 B9 2F 00 00 00 mov ecx,2Fh 0040D3D7 F7 F9 idiv eax,ecx 0040D3D9 89 55 EC mov dword ptr [ebp-14h],edx 0040D3DC EB DA jmp main+0C8h (0040d3b8) Chap. 2 Instruction Sets 20