Blame Passing for Analysis and Optimisation - PowerPoint PPT Presentation

About This Presentation
Title:

Blame Passing for Analysis and Optimisation

Description:

10/9/09. Async Forum. 1. Blame Passing for Analysis and Optimisation. Charlie Brej. APT Group. University of Manchester. 10/9/09. Group Talk. 2. Overview. Introduction ... – PowerPoint PPT presentation

Number of Views:47
Avg rating:3.0/5.0
Slides: 21
Provided by: brej
Learn more at: https://brej.org
Category:

less

Transcript and Presenter's Notes

Title: Blame Passing for Analysis and Optimisation


1
Blame Passing for Analysis and Optimisation
  • Charlie Brej
  • APT Group
  • University of Manchester

2
Overview
  • Introduction
  • Synchronous methods
  • Blame passing
  • Optimisations
  • Results

3
Why we need optimisation
  • High level entry languages
  • RTL
  • Gates
  • Transistors
  • Silicon
  • Large designs
  • More important things to worry about
  • Optimal designs often difficult to read
  • Difficult to determine inefficiencies

4
Critical path analysis
5
Retiming
6
Why we dont use the Critical Path
  • Assumes isochronic arrival of all inputs in each
    stage
  • Or predictable sequencing and timing of arrival
    of inputs (deterministic)
  • No cyclic circuits
  • No C-elements
  • Only looks at logic and not latches
  • No method of determining the interaction between
    different computation cycles
  • Determines and improves the worst (not average /
    mode) case performance

7
Slowest Path
  • Designed for asynchronous circuits
  • Allows cyclic circuits
  • Allows non-deterministic behavior
  • Uses a real simulation
  • Gathers average case data (not worst case)
  • Sort of like critical path of a full benchmark
  • Start point is the release of the reset
  • End point is the benchmark completion signal

8
Blame Game
2nd Place
Output
1st Place
Benchmark complete
9
Blame Game
10
Pass the blame
  • Blame passing forms a path from completion back
    to reset release
  • How to extract path
  • Static timing analysis (cyclic paths)
  • Simulation dump (lots of data)
  • Use a custom simulator
  • Small overhead (30)
  • Only remember the slowest path transitions

11
Decrementer
while (1) if (Reg ! 0) Reg Reg
1 else Reg Const
12
Graphical representation
13
Apply Optimisations
  • Find known paths through elements
  • Paths with positive effect
  • Paths with negative effect
  • Apply optimisation
  • Simulate again to determine the effect

14
Optimisation Rules
15
Optimisation Rules
16
Optimisation Rules
17
Optimisation Results (Dec)
18
Optimisation Results (GCD)
19
Optimisation Results (MIPS)
20
Conclusions
  • Novel analysis and optimisation technique
  • Applicable to many asynchronous systems
  • Not just circuits
  • Automatic reanalysis system
  • Future work
  • Read SDF files
  • Extension to other systems (Balsa etc)
  • Connection to a full design system (Biscotti)
Write a Comment
User Comments (0)
About PowerShow.com