JVM Usenix paper got rejected

[Bryce McKinlay]

Jeff Sturm wrote:

> It was many individual runs.  But javac's performance may have been
> limited by runtime factors anyway... has anybody noticed that libgcj's
> classloader is far slower than Sun's when reading compressed .jar files?

Doesn't surprise me. There are certainly plenty of opportunities for
optimizations and cleanups in the classloader stuff. Was the slowness you saw
actually specific to zip files (and if so, specific to compressed zip files?) or
was it just slow classloader in general?

regards

  [ bryce ]