In some cases, you may find it useful to support plug-ins written for an architecture other than the one your application is running on at the time. You may need this simply for debugging purposes, but this approach may also be useful if you want your application to support existing plug-ins on newer architectures. For example, audio software manufacturers may find it easier to drive adoption of 64-bit versions of their application if they also support existing 32-bit audio unit plug-ins.
Designing a host application to load a given plug-in is a highly specialized task. This chapter provides an overview of common approaches to doing this. This chapter assumes that you have already written a dummy plug-in loader that loads the plug-in into memory (even if it doesn’t actually do anything with the plug-in yet).
In addition, this chapter describes several common interprocess communication APIs, explains how to pass large amounts of data between the host application and the plug-in helper host, and tells how to launch that host for a particular architecture.
Choosing a Helper Host Architecture Model
Using Interprocess Communication
Launching the Helper Host
Last updated: 2008-04-08