Method/Block Invocation Optimization

[dreamhead]

Hi!

Do you remember the proposal for method/block invocation?
http://groups.google.com/group/xruby-devel/browse_thread/thread/d1ce1790497e2771

Now, I've finished it for 0 or 1 argument method/block. The
performance gets a little improved. You can view the attached file to
know test results. Ruby version is 1.8.6.

I've noticed that bm_app_raise.rb and bm_app_strconcat.rb are faster
in C Ruby than XRuby now. Maybe it's optimization in C Ruby 1.8.6.

Ye Zheng

--
Everything is simple!

[Xue Yong Zhi]

benchmark/bm_loop_times.rb takes 24.78 secondsto finish with xruby 0.2.0,
now only 6.62 seconds with trunk. Great improvement.

>
>Now, I've finished it for 0 or 1 argument method/block. The
>performance gets a little improved.

_________________________________________________________________
Get a preview of Live Earth, the hottest event this summer - only on MSN
http://liveearth.msn.com?source=msntaglineliveearthhm

[Michael Chen]

Sorry may be a bad news for you: I updated the code just several
minutes ago and run the test/ruby/test_io.rb, which has been passed
but now has following exception trace:

Exception in thread "main" java.lang.VerifyError: (class:
optparse/warn$291, method: run signature:
(Lcom/xruby/runtime/lang/RubyValue;Lcom/xruby/runtime/value/RubyArray;Lcom/xruby/runtime/lang/RubyBlock;)Lcom/xruby/runtime/lang/RubyValue;)
Incompatible argument to function
...

I think it may be related to some issue during the byte code
generation. When I running other tests, it reports me the same error.

--
Michael Chen
--------------------------------
Blog: http://michael.nona.name
MSN: jzch...@hotmail.com

[dreamhead]

Michael, you're right! It's a bug. I've fixed it.

But the test doesn't still passed. You can view it.
<"ÿ"> expected but was<"��xyz">.

Ye Zheng

2007/6/11, Michael Chen <mech...@gmail.com>:

--
Everything is simple!

[Michael Chen]

Strange behaviour...It passed on my box, Mac BookPro, a unix based environment.

Any body can test it on other environment to make sure that it's an issue?

[Haofei]

On 11/06/07, Michael Chen <mech...@gmail.com> wrote:

Strange behaviour...It passed on my box, Mac BookPro, a unix based environment.

Any body can test it on other environment to make sure that it's an issue?

same error as dreamhead on Ubuntu 7.04

[dreamhead]

I also use Ubuntu 7.04.

2007/6/11, Haofei <yue...@gmail.com>:

>
>
> On 11/06/07, Michael Chen <mech...@gmail.com> wrote:
> > Strange behaviour...It passed on my box, Mac BookPro, a unix based
> environment.
> >
> > Any body can test it on other environment to make sure that it's an issue?
>
> same error as dreamhead on Ubuntu 7.04
>
> > On 6/11/07, dreamhead <dreamh...@gmail.com > wrote:
> > > Michael, you're right! It's a bug. I've fixed it.
> > >
> > > But the test doesn't still passed. You can view it.
> > > <"ÿ"> expected but was<"��xyz">.
> > >
> > > Ye Zheng
> > >

--
Everything is simple!