11 - 380C Last Time Today Interactionsofscheduling...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
  Lecture 11            CS 380C                1 380C Last Time Interactions of scheduling  and register usage Today Interactions of scheduling  and instruction level  parallelism
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
  Lecture 11            Shape of Expressions Proebsting & Fischer assume a fixed expression  tree  Hunt et al. reorganize commutative and  associative operations in expression trees to  Increase ILP Decrease critical path length Group constants
Background image of page 2
  Lecture 11            Motivation Long pipelines and fine grain parallel processors  (e.g., SuperScalar RISC, VLIW & EDGE)  benefit from instruction  level parallelism. Decreasing critical path length improves loop  performance Grouping constants improves constant  propagation.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
  Lecture 11            Example Let M denote intermediate values we  need to preserve.
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 03/09/2012 for the course CS 380 taught by Professor Shmat during the Fall '08 term at University of Texas at Austin.

Page1 / 18

11 - 380C Last Time Today Interactionsofscheduling...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online