Personal tools

ThreadScope Tour

From HaskellWiki

(Difference between revisions)
Jump to: navigation, search
m
m (Digging into a program with spark events: link to spark cycle)
Line 31: Line 31:
 
<ul>
 
<ul>
 
<li>[[ThreadScope_Tour/RTS|GHC RTS flags]]: a subset of flags relevant to ThreadScope</li>
 
<li>[[ThreadScope_Tour/RTS|GHC RTS flags]]: a subset of flags relevant to ThreadScope</li>
<li>[[ThreadScope_Tour/spark-lifecycle.png|Spark lifecycle]]: Lifecycle of a spark</li></ul>
+
<li>[[:Image:spark-lifecycle.png|Spark lifecycle]]: Lifecycle of a spark</li></ul>
 
</li>
 
</li>
 
<li><p>[[ThreadScope_Tour/Spark|Spark rates]]: study spark creation/conversion</p></li>
 
<li><p>[[ThreadScope_Tour/Spark|Spark rates]]: study spark creation/conversion</p></li>

Revision as of 16:08, 7 December 2011


A guided tour of ThreadScope

In this tutorial, we'll be working through concrete examples on using ThreadScope to debug the performance of parallel programs.

We aim to keep each module in the tutorial self-contained, so you can either follow the progression suggested or jump to just the sections we need.

Note that you will need:

  • ThreadScope 0.2.1 or higher
  • GHC 7.4 or higher. You could try earlier versions, but more advanced features such as spark events will not available

1 Getting Started

  1. Installation: install ThreadScope and run a sample trace
  2. Hello world: Run ThreadScope on a small test program

2 Basic skills

  1. Initial statistics: collect some simple statistics

  2. Profile: examine the profile for a real program

  3. Profile 2: examine the profile for an improved program

  4. Zoom: zooming and bookmarking

3 Digging into a program with spark events

  1. Spark overview

  2. Spark rates: study spark creation/conversion

  3. Spark rates 2: spark debugging continued

  4. Spark sizes: spark debugging continued

4 Reference

5 Planned

  1. Trace: custom events
  2. Consolidate: tease out the sequential parts of code