Older blog entries for saugart (starting at number 12)

14 Jan 2005 (updated 14 Jan 2005 at 21:33 UTC) »

I libelled the Jikes Java-source-to-byte-code compiler in my diary entry of January 12th, 2005. Jikes seems to have done a fine job of building GNU Classpath; my troubles were almost certainly due to other causes, ones which I have still not fully tracked down. I apologize to the members of the Jikes project and retract that diary entry.

13 Jan 2005 (updated 14 Jan 2005 at 21:31 UTC) »
I retract my unfair accusations against the Jikes compiler. The problems I was experiencing were almost certainly due to other causes.

I have retroactively edited this diary entry to prefix it with this retraction, but am leaving the diary entry untouched.

--14 January 2005

Trying to get GNU Classpath 0.13+CVS and Jikes RVM to work together. Sigh; the Jikes compiler is not behaving very well these days. Jikes 0.21 [Here I meant 1.21 --14 Jan 2005] seems to be building a bogus version of GNU Classpath; ouch.

It's the first snow of the year here in New York. Whee.

I'm glad to see that Advogato is back up; I can post blog entries again now.

One of my software patents has been granted. Time to contact some sort of free patent foundation.

I've been sad for the past week, and not very interested in software. Nancy's pregnancy didn't take. Dimitri may end up being an only child.

I read Tom Tromey's blog entry on Planet Classpath, where he discusses concerns about large companies acquiring software patents and using them to stifle innovation.

I have three software patents pending, which I've filed for on my own. It is expensive -- several thousand dollars of legal fees each. I was working on them yesterday, wondering why I still bother, since the software startup that I started the applications for has long since folded. But I will grant a perpetual royalty-free license to them for any software that meets the Debian Free Software Guidelines.

Just uploaded my Debian Jikes RVM package to my personal FTP site for my Debian Mentor (Greg Prokopski, a.k.a. gadek) to look at.

I was disappointed to see that Debian's glibc apparently does not use the GS register to store per-thread data. This means that I had to compile a version of Jikes RVM that won't be capable of exploiting SMP processors. The correct fix to the problem would be to get rid of Jikes RVM's m:n threading. I wonder how much work that would be. It would solve other problems as well, since otherwise any JNI calls we do on such a platform may hang. However, a much better bet would be to finish the self-bootstrapping work that I paused in order to build the Debian packages before Sarge gets frozen.

I found out why I was so unenthusiastic about driving Dimitri up to Boston. I have a nasty cold of some sort -- been coughing up green mucus.

I spent much of the weekend resting.

Am still working on packing up Jikes RVM for Debian. My application to be a Debian Developer is currently holding, waiting for an AM.

I took my son, Dimitri, up to Boston for Easter and to visit with Dylan, his best friend from our old home in Los Angeles.

We might be able to go up and see Dylan again tomorrow or Saturday. But do I really want to drive another four hours each way (8 hours round trip) just so that he can see his friend? Do I care more about my convenience and my enthusiasm for hacking on Jikes RVM, or about my son's friendship? And how much will this matter a few months from now?

I certainly don't need to put in the hours at work; I've been putting in plenty of overtime.

Kaffe ran the Jikes RVM optimizing compiler for two and a half hours, and produced a "development" image, which is a "production" image with the assertions still in place. Good for Kaffe. The image it built runs my simple test cases.

I built Jikes RVM with its optimizing compiler and adaptive system last night, using entirely free tools. Hooray, another milestone. I use a two-stage build process; Kaffe OpenVM makes a Jikes RVM BaseBase* build (one with the baseline compiler only). Then we make an *Adaptive* build using the just-built BaseBase Jikes RVM to expand the templates and using Kaffe to do the Host VM's other work.

I've already merged the code to make optimizing builds into the CVS trunk, including what I hope is full documentation on the process. (I would like to get feedback on the documentation.)

We plan to cut the 2.3.2 release, including these features, next week. (We missed cutting the release last week, due to server trouble.)

There are several next steps. One of them is to add the free builds to the current Jikes RVM nightly regression test suite. Another is to adapt that regression test suite so that it can be run by a remote developer who doesn't have any of the unfree benchmarks (SPECjbb, etc.), and to further adapt it so that it doesn't require you to have all of the the free benchmarks installed. My goal is to let anyone build Jikes RVM using free tools and immediately run regression tests against the just-built VM.

Once the free builds pass all the tests consistently, I hope to work with Debian or another distribution so that we can ship a free working JDK and JRE that use Jikes RVM; SuSE Linux has included a kaffe package for years (at least since SuSE 6.1, which I installed in May of 1999).

3 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!