I have started to put together some documentation on Kaffe. It is not much but I would like to pass it along to the list. I hope to expand it soon. I started with some stuff off of the website and added information I have cleamed from other locations (such as sablevm and amazon).
Comments welcome. Michael Kaffe ===== This is an introduction to Kaffe. Kaffe is a clean room implementation of the Java virtual machine, plus the associated class libraries needed to provide a Java runtime environment. The Kaffe virtual machine is free software, licensed under the terms of the GNU General Public License. Kaffe has three different execution engines. A pure interepter (intrp) and two different just-in-time compilers, JIT and JIT3. Each execution engine follows a common architecture and are easily swapable (at compile time only). Implementation -------------- Kaffe, like other byte code interpreters is a stack based interpreter. It follows the standard method of reading in the byte code and then interpreting each opcode in an infinate loop. Execution only exits the loop when **** something here **** The definition of the opcodes (Java has 255 codes, for some reason opcodes 186 is not used) is in the kaffe.def file. This expands to elements of a switch statement that contain the code to execute for each JVM opcode. _______________________________________________ kaffe mailing list [email protected] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
