IA32 Processors 15 213 Totally Dominate Computer Market The course that gives CMU its Zip Machine Level Programming I Introduction Sept 09 2006 Topics Evolutionary Design Starting in 1978 with 8086 Added more features as time goes on Still support old features although obsolete Complex Instruction Set Computer CISC Assembly Programmer s Execution Model Many different instructions with many different formats Hard to match performance of Reduced Instruction Set Computers RISC But Intel has done just that z But only small subset encountered with Linux programs Accessing Information z Registers z Memory Arithmetic operations class04 ppt 15 213 F 06 x86 Evolution Programmer s View Abbreviated Name 8086 Date Transistors 1978 29K 16 bit processor Basis for IBM PC DOS Limited to 1MB address space DOS only gives you 640K 1985 x86 Evolution Programmer s View Machine Evolution 386 15 213 F 06 2 275K Extended to 32 bits Added flat addressing Capable of running Unix Referred to as IA32 32 bit Linux gcc uses no instructions introduced in later models 486 1989 1 9M Pentium 1993 3 1M Pentium MMX 1997 4 5M PentiumPro 1995 6 5M Pentium III 1999 8 2M Pentium 4 2001 42M Added Features Instructions to support multimedia operations Instructions to enable more efficient conditional operations z Parallel operations on 1 2 and 4 byte data both integer FP Linux GCC Evolution 3 15 213 F 06 4 None 15 213 F 06 New Species IA64 X86 Evolution Clones Name Date Transistors Advanced Micro Devices AMD Itanium 2001 10M Historically z AMD has followed just behind Intel Extends to IA64 a 64 bit architecture Radically new instruction set designed for high performance Can run existing IA32 programs z A little bit slower a lot cheaper Recently z Recruited top circuit designers from Digital Equipment Corp and z On board x86 engine Joint project with Hewlett Packard Itanium 2 other downward trending companies z Exploited fact that Intel distracted by IA64 2002 z Now are close competitors to Intel 221M Developed x86 64 its own extension to 64 bits Big performance boost Itanium 2 Dual Core z Started eating into Intel s high end server market 2006 1 7B Itanium has not taken off in marketplace Lack of backward compatibility 15 213 F 06 5 15 213 F 06 6 Intel s 64 Bit Dilemma Our Coverage Intel Attempted Radical Shift from IA32 to IA64 IA32 Totally different architecture Executes IA32 code only as legacy Performance disappointing x86 64 AMD Stepped in with Evolutionary Solution Hard to admit mistake or that AMD is better 2004 Intel Announces EM64T extension to IA32 7 The emerging standard Presentation x86 64 now called AMD64 Intel Felt Obligated to Focus on IA64 The traditional x86 Book has IA32 Handout has x86 64 Lecture will cover both Labs Extended Memory 64 bit Technology Lab 2 x86 64 Almost identical to x86 64 Lab 3 IA32 Our Saltwater fish machines 15 213 F 06 8 15 213 F 06 Assembly Programmer s View CPU Turning C into Object Code Memory Addresses Registers P C PC Compile with command gcc O p1 c p2 c o p z Use optimizations O z Put resulting binary in file p Instructions C program p1 c p2 c text Compiler gcc S Programmer Visible State Code in files p1 c p2 c Object Code Program Data OS Data Data Condition Codes Stack Program Counter text z Address of next instruction Asm program p1 s p2 s z Called EIP IA32 or RIP x86 Assembler gcc or as 64 Register File z Heavily used program data Condition Codes z Store status information about most recent arithmetic operation z Used for conditional branching binary Memory z Byte addressable array z Includes stack used to support Compiling Into Assembly C Code int sum int x int y int t x y return t binary procedures 15 213 F 06 Static libraries a Linker gcc or ld z Code user data some OS data 9 Object program p1 o p2 o Executable program p 15 213 F 06 10 Assembly Characteristics Generated IA32 Assembly sum pushl ebp movl esp ebp movl 12 ebp eax addl 8 ebp eax movl ebp esp popl ebp ret Minimal Data Types Integer data of 1 2 or 4 bytes z Data values z Addresses untyped pointers Floating point data of 4 8 or 10 bytes No aggregate types such as arrays or structures z Just contiguously allocated bytes in memory Primitive Operations Obtain with command gcc O S code c Perform arithmetic function on register or memory data Transfer data between memory and register z Load data from memory into register Produces file code s z Store register data into memory Transfer control z Unconditional jumps to from procedures z Conditional branches 11 15 213 F 06 12 15 213 F 06 Object Code Code for sum Machine Instruction Example Assembler C Code int t x y Translates s into o 0x401040 sum Binary encoding of each instruction 0x55 Total of 13 0x89 Nearly complete image of executable bytes 0xe5 code Each 0x8b instruction 1 Missing linkages between code in 0x45 2 or 3 bytes different files 0x0c Starts at 0x03 address Linker 0x45 0x401040 0x08 Resolves references between files 0x89 Combines with static run time libraries 0xec z E g code for malloc printf 0x5d 0xc3 Some libraries are dynamically linked z Linking occurs when program begins execution Assembly addl 8 ebp eax z Same instruction whether signed or unsigned x y Or int ebp eax ebp 2 0x401046 03 45 08 Object Code 0x401040 0x55 0x89 0xe5 0x8b 0x45 0x0c 0x03 0x45 0x08 0x89 0xec 0x5d 0xc3 ebp esp ebp 0xc ebp eax 0x8 ebp eax ebp esp ebp 0x0 esi esi objdump d p 15 15 213 F 06 Disassembled Object Disassembler Useful tool for examining object code Analyzes bit pattern of series of instructions Produces approximate rendition of assembly code Can be run on either a out complete executable or o file 3 byte instruction Stored at address 0x401046 Alternate Disassembly Disassembled push mov mov add mov pop ret lea Register eax Memory M ebp 8 Register eax Return function value in eax 14 Disassembling Object Code 00401040 sum 0 55 1 89 e5 3 8b 45 0c 6 03 45 08 9 89 ec b 5d c c3 d 8d 76 00 Operands x y t int eax 15 213 F 06 Add 2 4 byte integers z Long words in GCC parlance Similar to expression 13 Add two signed integers 15 213 F 06 0x401040 0x401041 0x401043 0x401046 0x401049 0x40104b 0x40104c 0x40104d sum sum 1 sum 3 sum 6 sum 9 sum 11 sum 12 sum 13 push mov mov add mov pop ret lea ebp esp ebp 0xc ebp eax 0x8 ebp eax ebp esp ebp 0x0 esi esi Within gdb Debugger gdb p disassemble sum Disassemble procedure x 13b sum 16 Examine the 13 bytes starting at sum 15 213 F 06 What Can be Disassembled Moving Data IA32 eax edx objdump d WINWORD EXE WINWORD EXE Moving Data movl
View Full Document
Unlocking...