The bug hunting is progressing well, I've found a number of bugs since yesterday, and the standard build process goes further and further every day. At the moment, the build process builds miniparrot, parrot, libparrot, and bunch of libraries. I build PGE.pbc successfully, but the build fails at building compilers/pge/PGE/builtins_get.pir. I've been trying to trace through this build step manually, and the errors I'm getting are very strange, very cryptic, and very hard to debug.
Unfortunately, the "very hard" part is leading me to be a little bit burnt out today. Instead of running myself into the ground, I'm going to take the rest of the day off and pursue other things. There are some other bugs, patches, and features I can work on instead for Parrot. Plus, I want to get some work done, on Wikibooks and in other places. A little bit of a break will help me get my mind right, and I'll feel refreshed when I get back to work tomorrow or monday (I'm away for the long weekend).
Parrot builds, like I said above, and is passing the majority of the test suite. I have to run the tests manually, make test doesn't work because make doesn't complete yet. Based on a rough count, I think I'm passing all but 700 tests from the suite of over 7700. Of the tests that fail, many of them produce correct output but segfault during program cleanup and termination. This gives me some ideas about where to look, but isn't conclusive or particularly accurate. I need to scrutinize the failing tests for clues about what causes a segfault and what does not.
Anyway, I'm off to get some work done and clear my mind a bit. I'll post again either tomorrow or monday with a progress report.
