On Wed, 2005-03-09 at 11:38 -0500, Thomas Fitzsimmons wrote:
On Wed, 2005-03-09 at 15:04 +0000, Andrew Haley wrote:
>I see. Of course, I have no idea why tools.jar is attempting to use
>the gcjlib:// class loader, or why it is so slow when it does. It
>really should not explicitly use this class loader.
I doesn't any more. We were doing that to avoid having to always add
jdtcore.jar to the classpath. I also have no idea why it's so slow but
I did notice a significant performance hit when compared to interpreted
mode.
Ziga, you can get the sources for java-gcj-compat from
sources.redhat.com:
Ah, much better, thanks. Here are the revised times (BTW, that's a
1.5GHz Pentium M):
HelloWorld
ecj | ecj-native | jikes
-------------------------------------------------------------
real 0m1.863s | real 0m1.614s | real 0m0.067s
user 0m1.758s | user 0m1.536s | user 0m0.050s
sys 0m0.103s | sys 0m0.076s | sys 0m0.012s
GNU Classpath (cd lib; make)
ecj | ecj-native | jikes
-------------------------------------------------------------
real 1m24.539s | real 0m24.552s | real 0m9.439s
user 1m23.157s | user 0m23.047s | user 0m7.486s
sys 0m1.142s | sys 0m1.139s | sys 0m0.771s
Note that classpath sources need a tiny hack to keep ecj from crashing.
But that looks like an upstream Eclipse bug, since I could reproduce it
running ecj with java-1.5.0-sun.
Andrew, I don't suppose you still need oprofile data (or are there other
places where gcjlib:// loading might be a problem)?
--
Ziga