[EMAIL PROTECTED](Nelson Minar) said:
> I found a link to another JIT for Linux, shuJIT, at
> http://www.shudo.net/jit/
Thanks. I'm the author of the JIT.
I've forgotten to announce releace of my JIT compiler.
> Executive summary: also works with FreeBSD,
The JIT is being developing on Linux and ported to
FreeBSD by myself. Porting work requires some knowledge
of FreeBSD (2.X), but is not too difficult. I suppose
that it is desirable to port TYA to some platforms other
than Linux.
> says performance is about like TYA,
ShuJIT shows better performance than TYA with simple
programs such as Linpack or Eratosthenes sieve
benchmark.
But maybe TYA is superior to shuJIT with usual
applications. The current weakest point of shuJIT is
performance of method invocation.
If we pursuit real good performance, I think that we
should deploy the intermediate representation such as
GCC's RTL and perform more optimizations like C compiler
and register allocation rather than direct translation
like TYA and shuJIT. Because my primary goal is not
performance, I don't plan to construct a JIT in this
manner.
> interested in working eventually with thread migration.
Primary purpose of my development of JIT is having our
thread migration system worked with JIT. Thread
migration requires being able to take execution context
out in machine independent form. Common JIT's don't
allow us that.
To 16:00 Nov 12, we have a booth in research exhibits of
SC98 international conference at Orlando
(http://www.supercomp.org/sc98/). We'll show
demonstration of our system if you can visit our booth,
parallel and distributed computing environment, R119.
> The author is in a research group working on "OpenJIT",
Sorry for misleading my writings, I'm not a member of
OpenJIT project. I respect the leader of the project,
Prof. Matsuoka and I'm very interesting in the project.
> a project to make
> the JIT itself be controllable by Java to make it easier to tune.
> Looks quite interesting.
Performance of the prototype implementation is as good
as Sun JIT. Rather shows a bit better performance than
Sun JIT. To my regret current implementation of OpenJIT
is not open to public.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., School of Sci. Eng., Waseda Univ.
E-mail: [EMAIL PROTECTED]
Address: Muraoka Lab., Waseda Univ.,
3-4-1 Okubo Shinjuku-ku Tokyo 165-8555 JAPAN
TEL: +81-3-3209-5198 FAX: +81-3-3209-5198
