Some recent work involved generating, saving and loading large dictionaries, up to 1.9M entries. The performance of said dictionary natuarally became a concern.

Refactorings describes how to refactor code using the GUI. But if you want to programmatically refactor the API is either not intuitive (to me) or a little inconsistent.

I’m currently trying to understand how to commit code using Iceberg (programatically, the UI is fine).

Like most Smalltalk impelementations, Pharo supports ‘green threads’, meaning that multiple smalltalk threads can be active at once, but they are all run from a single operating system thread, i.e. only one CPU is active at a time.

If you want to get started in tuning Pharo’s memory management some mandatory reading is:

This post is an update on my ongoing GC crash investigation. For an introduction see GC Crash Summary.

This is a copy of an email available from the vm-dev mail archive.

This post is an update on my ongoing GC crash investigation. For an introduction see GC Crash Summary.

This blog post will be updated with the current status of my investigations and attempts to resolve this issue.