1 / 6

Correlation Between State and Control Signals in Out-of-order Issue Logic

This update highlights the data collected for individual instruction consistency, PC Fetch vs Issue Window consistency, register renaming, and instruction repetition. It also discusses challenges faced in collecting large pipetraces and rewriting tools for larger pipetraces.

jdominguez
Télécharger la présentation

Correlation Between State and Control Signals in Out-of-order Issue Logic

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Correlation Between State and Control Signals in Out-of-order Issue Logic Checkpoint 2: 11/21/00 Kenneth Barr Kenneth Conley Serhii Zhak

  2. Status Update • Highlights • Data for individual instruction consistency • Data for PC_Fetch vs. Issue Window consistency • Data for register renaming • Data for instruction repetition • Lowlights • Hard to collect large pipetraces • Spent a lot of time rewriting tools to support larger pipetraces • Repetition of research

  3. Renaming Results 0.90 0.80 0.70 Mapped to 0.60 Original Mapping 0.50 Fraction 0.40 Could Map to 0.30 Original Mapping 0.20 0.10 0.00 li go cc1 m88 perl vortex ijpeg compress Benchmark

  4. Individual Instruction Consistency

  5. Fetch_PC vs. Issue Window Consistency

  6. The Road to the Finish Line • Issue Oracle • Verify functionality • Collect data • Compare results against expected results (consistency data) • Test different oracle prediction rules • Renaming • Collect data for forced renaming • Consistency Data • Greater variety of traces

More Related