Transmeta’s New Processor Another way to design CPU By Wu Cheng

Slides:



Advertisements
Similar presentations
JUST-IN-TIME COMPILATION
Advertisements

Intro to the “c6x” VLIW processor
CPE 731 Advanced Computer Architecture ILP: Part V – Multiple Issue Dr. Gheith Abandah Adapted from the slides of Prof. David Patterson, University of.
Pipelining 5. Two Approaches for Multiple Issue Superscalar –Issue a variable number of instructions per clock –Instructions are scheduled either statically.
1 Advanced Computer Architecture Limits to ILP Lecture 3.
Computer Organization CS224 Fall 2012 Lesson 44. Virtual Memory  Use main memory as a “cache” for secondary (disk) storage l Managed jointly by CPU hardware.
Lecture 34: Chapter 5 Today’s topic –Virtual Memories 1.
Transmeta’s Crusoe Architecture Umran A. Khan Microprocessors.
Virtual Memory Operating Systems Lecture # 8. Multi-tasking OS OS Excel MS Word Outlook 0x0000 0x7000 0x4000 0x8000 0x9000.
Translation Buffers (TLB’s)
Choice for the rest of the semester New Plan –assembler and machine language –Operating systems Process scheduling Memory management File system Optimization.
State Machines Timing Computer Bus Computer Performance Instruction Set Architectures RISC / CISC Machines.
A Survey on Virtualization Technologies
A Survey on Virtualization Technologies. Virtualization is “HOT” Microsoft acquires Connectix Corp. EMC acquires VMware Veritas acquires Ejascent IBM,
IT253: Computer Organization Lecture 4: Instruction Set Architecture Tonga Institute of Higher Education.
INTRODUCTION Crusoe processor is 128 bit microprocessor which is build for mobile computing devices where low power consumption is required. Crusoe processor.
Anshul Kumar, CSE IITD CS718 : VLIW - Software Driven ILP Example Architectures 6th Apr, 2006.
Transmeta and Dynamic Code Optimization Ashwin Bharambe Mahim Mishra Matthew Rosencrantz.
Hardware Support for Compiler Speculation
Virtual Memory Expanding Memory Multiple Concurrent Processes.
Spring 2003CSE P5481 VLIW Processors VLIW (“very long instruction word”) processors instructions are scheduled by the compiler a fixed number of operations.
Virtual Memory Review Goal: give illusion of a large memory Allow many processes to share single memory Strategy Break physical memory up into blocks (pages)
Chapter 8 CPU and Memory: Design, Implementation, and Enhancement The Architecture of Computer Hardware and Systems Software: An Information Technology.
Implementing Precise Interrupts in Pipelined Processors James E. Smith Andrew R.Pleszkun Presented By: Ravikumar Source:
VAX-11/780 A VIRTUAL ADDRESS EXTENSION TO THE DEC PDP-11 FAMILY VAX-11/780 A VIRTUAL ADDRESS EXTENSION TO THE DEC PDP-11 FAMILY W.D.STRECKER W.D.STRECKER.
1 CSE451 Architectural Supports for Operating Systems Autumn 2002 Gary Kimura Lecture #2 October 2, 2002.
Next Generation ISA Itanium / IA-64. Operating Environments IA-32 Protected Mode/Real Mode/Virtual Mode - if supported by the OS IA-64 Instruction Set.
DSP Architectures Additional Slides Professor S. Srinivasan Electrical Engineering Department I.I.T.-Madras, Chennai –
Spring 2003CSE P5481 Precise Interrupts Precise interrupts preserve the model that instructions execute in program-generated order, one at a time If an.
Operating Systems ECE344 Ashvin Goel ECE University of Toronto Demand Paging.
CS2100 Computer Organisation Virtual Memory – Own reading only (AY2015/6) Semester 1.
Pentium Architecture Arithmetic/Logic Units (ALUs) : – There are two parallel integer instruction pipelines: u-pipeline and v-pipeline – The u-pipeline.
Virtual Memory Ch. 8 & 9 Silberschatz Operating Systems Book.
Virtual Memory Review Goal: give illusion of a large memory Allow many processes to share single memory Strategy Break physical memory up into blocks (pages)
1 Aphirak Jansang Thiranun Dumrongson
Crusoe Processor Seminar Guide: By: - Prof. H. S. Kulkarni Ashish.
MODERN OPERATING SYSTEMS Third Edition ANDREW S
Protection in Virtual Mode
15-740/ Computer Architecture Lecture 3: Performance
Chapter 8: Main Memory.
Visit for more Learning Resources
Morgan Kaufmann Publishers
Henk Corporaal TUEindhoven 2009
Chapter 8: Main Memory.
Morgan Kaufmann Publishers The Processor
From Address Translation to Demand Paging
Superscalar Processors & VLIW Processors
Superscalar Pipelines Part 2
A Survey on Virtualization Technologies
Lecture 23: Static Scheduling for High ILP
Codesigned Virtual Machine -Transmeta CRUSOE-
Henk Corporaal TUEindhoven 2011
Morgan Kaufmann Publishers Memory Hierarchy: Virtual Memory
CC423: Advanced Computer Architecture ILP: Part V – Multiple Issue
Translation Buffers (TLB’s)
TLB Performance Seung Ki Lee.
Prof. Onur Mutlu Carnegie Mellon University Fall 2011, 9/30/2011
CSE 451: Operating Systems Autumn 2003 Lecture 10 Paging & TLBs
CSE 451: Operating Systems Autumn 2003 Lecture 2 Architectural Support for Operating Systems Hank Levy 596 Allen Center 1.
Midterm 2 review Chapter
CSE 451: Operating Systems Autumn 2001 Lecture 2 Architectural Support for Operating Systems Brian Bershad 310 Sieg Hall 1.
Overview Prof. Eric Rotenberg
Translation Buffers (TLB’s)
CSE 451: Operating Systems Autumn 2003 Lecture 10 Paging & TLBs
CSE 451: Operating Systems Winter 2003 Lecture 2 Architectural Support for Operating Systems Hank Levy 412 Sieg Hall 1.
October 29 Review for 2nd Exam Ask Questions! 4/26/2019
Translation Lookaside Buffers
CPU Structure CPU must:
Translation Buffers (TLBs)
Review What are the advantages/disadvantages of pages versus segments?
Presentation transcript:

Transmeta’s New Processor Another way to design CPU By Wu Cheng

What’s Crusoe CPU X86 ISA Compatible Simple VLIW Hardware Software/Hardware Mixed Architecture Low Power Self-Turning of Clock & Voltage

Structure

X86 Applications X86 BIOS Code Morphing Software X86 Operating System VLIW Processor

VLIW Hardware 64/128 bits molecule 2 Integer, 1 Float, 1 LD/ST, 1 Branch Optimized for simple, fast and low-power [F]ADDBRCCLDADD 128 bit molecule

Code Morphing Layer Started When Boot Translate/Optimize ISA to Native ISA Insert Instrument Code to … –Execute path analysis –Detect CPU need for efficient power consuming –Any other analysis..

Code Morphing Layer 8-16M Physical Memory on main-board Code Morph itself And Translated Code Cache Virtual Physical Memory BIOS and OS can see

Translation Step  Interpretation  Translate one instruction at a time  Simple-minded  Optimize Code that Compilers do  Highly Optimized  Reorder the instructions and make parallelism

Special Issues for X86 Generate Same Condition Code Operates same 80-bit floating point TLB has same protection bits TLB address mapping as X86 processors

Precise Interrupt –Shadow Registers (int / fp) –Gated Store Buffer –Commit Operation –Rollback Operation

Alias Hardware Alias Hardware (Memory) –Reorder limited by load/store pairs –Load-and-protect Save address and size of loaded data –Store-Under-Alias-Mask Check alias hardware Raise exception when overwrite

Example of Alias ld %r30, [%x] // first load from location x … st %data, [%y] // might overwrite location x ld %r31, [%x] // this accesses location X again use %31 ldp %r30, [%x] // load from location x and protect it … stam %data, [%y] // traps if it writes x use %30 // may use data from first load

Self-Modifying Code Keep a “translated” bit in MMU When a protected page is written –Invalidate the affected translation(s)

Low Power!!! Pentium!!!Crusoe Reduced Power Consumption mainly by reduced x86 decode logic