Re: gcj

Re: gcj

[Andrew Haley]

LAMOME Julien CS-SI wrote:
> Hello.
> I read this page : http://gcc.gnu.org/java/ and see no news until about one year. The most links (status, done) have the same old update.
> Gcj project do it abandon or stop ? Or just stop web maintain ?
> I'm interest by advance of integration of java 1.5 into gcj. When a new release ? etc.

gcj is released with gcc, on the same schedule with the same
version numbers.

gcj has supported Java language version 1.5 for some time now,
and is in Fedora, Debian, etc.

Andrew.

Re: gcj

[Tom Tromey]

>>>>> ">" == Jayakrishnan M <jak_76_in@yahoo.com> writes:

>> 1. Does a program compiled into native executable
>> using gcj still require the presence of a jvm. If in
>> my program, I doesn't use awt or swing, can I
>> completely eliminate the big jre when distributing the
>> program ?

A program compiled by gcj requires libgcj.so to be available.  This
may or may not include the whole JVM (depending on platform), but it
always includes compiled forms of the standard classes.

>> 2. Does gcj supports IBM swt on win and lin ?

Yes.

>> 3. Does the gcc runtime system has its own jvm ? Or
>> does I have to bundle a jvm with my app in order to
>> run it on the linux platform ?
   
It includes its own.

Tom

gcj

[Jayakrishnan M]

Dear Friend,

   Please answer my following questions.

1. Does a program compiled into native executable
using gcj still require the presence of a jvm. If in
my program, I doesn't use awt or swing, can I
completely eliminate the big jre when distributing the
program ?

2. Does gcj supports IBM swt on win and lin ?

3. Does the gcc runtime system has its own jvm ? Or
does I have to bundle a jvm with my app in order to
run it on the linux platform ?
   
Best Regards
Jayakrishnan

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

Re: GCJ

[Tom Tromey]

>>>>> "Vivin" == Vivin Suresh Paliath <computer_maniac@hotmail.com> writes:

Vivin> How can I build the libgcj that comes with gcc-2.95.3 under
Vivin> Cygwin? I'm told that it is not possible to currently build it
Vivin> because you can't build libffi. Is it possible to edit the
Vivin> configure.in file so that it _tries_ to build libgcj anyway? By
Vivin> editing noconfigdirs?

It is possible, sure.  But why do you want to do this?  The Windows
code isn't really actively maintained.

Vivin> I was also told that I could use some sort of reverse patch on
Vivin> GCC3.0 or GCC3.1 to try to build libgcj.

Maybe, but that is likely to be hard.  You might get farther by trying
to get the gcc 3.0 snapshot and see how far it gets.  However, even
that won't fully work on Windows.  It might get further into the build
process.

Tom

GCJ

[Vivin Suresh Paliath]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 634 bytes --]

How can I build the libgcj that comes with gcc-2.95.3 under 
Cygwin? I'm told that it is not possible to currently build it because you can't 
build libffi. Is it possible to edit the configure.in file so that it _tries_ to 
build libgcj anyway? By editing noconfigdirs? 
 
I was also told that I could use some sort of reverse patch on 
GCC3.0 or GCC3.1 to try to build libgcj. I'm not well versed as far as these 
kind of things go so I would really appreciate it if someone told me how to 
build libgcj - either by editing configure.in or applying the reverse 
patch.
 
Thanks!
 
Vivin Suresh Paliath
 
e^(i * pi) = -1

RE: gcj

[Anthony Green]

Tom wrote:
> Offhand I don't know.  gcj partially supports the ARM, but I don't
> know how the ARM 7 relates to this support.  My understanding is that
> we have ARM support for everything except the interpreter.
>
> "ARM 7" isn't very specific.  libgcj requires knowledge not only of
> the processor but also of the operating system.

You can also run on bare metal.  Xscale is part of the ARM family, and with
a bit of tweaking I got gcj compiled code and the GC running on the Xscale
simulator (although, with no threads or networking support).

AG

Re: gcj

[Alexandre Petit-Bianco]

Gary Mueller writes:

> I am not sure this is the right email group.  If not, please point
> me to the right group.

It's the right one.

> I have an ARM 7 processor and would like to use gcj.  I would like
> to "complie" java code and run it on the ARM 7 processor.  Is the
> ARM7 processor supported by gcj?


Gcj can be configured to generate arm code, either as a native or as a
cross compiler. This shouldn't be too hard to achieve. In order to run
application, you'll need a port of the run-time, which besides
building it with a toolchain that generates arm code, will require
some porting (GC, threading, files, signal -- I don't know what's the
minimal amount of work necessary to run Hello World.)

Here's a short discussion that took place recently on the topic of arm
support:

  http://gcc.gnu.org/ml/java/2001-01/threads.html#00600

The Java section of gcc's website feature a short `porting the gcj
runtime' section:

  http://gcc.gnu.org/java/docs.html

./A

Re: gcj

[Tom Tromey]

>>>>> "Gary" == Gary Mueller <garymueller@uswest.net> writes:

Gary> I am not sure this is the right email group.  If not, please
Gary> point me to the right group.

This is right.

Gary> I have an ARM 7 processor and would like to use gcj.  I would
Gary> like to "complie" java code and run it on the ARM 7 processor.
Gary> Is the ARM7 processor supported by gcj?

Offhand I don't know.  gcj partially supports the ARM, but I don't
know how the ARM 7 relates to this support.  My understanding is that
we have ARM support for everything except the interpreter.

"ARM 7" isn't very specific.  libgcj requires knowledge not only of
the processor but also of the operating system.

Tom

gcj

[Gary Mueller]

I am not sure this is the right email group.  If not, please point me to
the right group.

I have an ARM 7 processor and would like to use gcj.  I would like to
"complie"
java code and run it on the ARM 7 processor.  Is the ARM7 processor
supported
by gcj?


                                    Thanks,

                                    Gary Mueller

Re: Gcj

[Anthony Green]

A few months ago I wrote:
> Another simple trick I'd like to see implemented looks like this...
>
> In a method M, for each class C requiring initialization, have the
> compiler declare a new local called Ci.  Instead of unconditional
> calls to the initialization routine, emit them as:
> 
>         if (! Ci)
>           {
>             _Jv_InitClass (C);
>             Ci = 1;
>           }
> 
> The compiler will then magically eliminate redundant calls to
> _Jv_InitClass within methods via simple constant propagation.  This
> shows up a lot.

I just submitted a patch to do this...

http://gcc.gnu.org/ml/gcc-patches/2000-02/msg00410.html

Once committed -O0 code will be slightly larger and slower, but -O1
will be better.

AG

-- 
Anthony Green                                               Cygnus Solutions
                                                       Sunnyvale, California

Re: Gcj

[Anthony Green]

Kresten wrote:
> Since I started using gcj for this verifier I'm writing, it has
> started to annoy me that it is so expensive to use static methods.  

> But maybe it would be worth it to compile
> static method calls using an indirection, and then flip that
> indirection pointer the first time it is used?

> And it would also bring gcj closer to spec; the initialization is
> supposed to happen *at* the first real use.

Hmmm.. I'm pretty sure we implement correct semantics - but your
quality of implementation issues are valid.

The indirect call scheme might be reasonable in some cases.

For many classes, however, the main function of initialization is to
intern string constants.  There are some linker tricks we could
implement that might eliminate this step.  Another simple trick I'd
like to see implemented looks like this...

In a method M, for each class C requiring initialization, have the
compiler declare a new local called Ci.  Instead of unconditional
calls to the initialization routine, emit them as:

        if (! Ci)
          {
            _Jv_InitClass (C);
            Ci = 1;
          }

The compiler will then magically eliminate redundant calls to
_Jv_InitClass within methods via simple constant propagation.  This
shows up a lot.

AG

-- 
Anthony Green                                               Cygnus Solutions
                                                       Sunnyvale, California

Re: Gcj

[Kresten Krab Thorup]

Since I started using gcj for this verifier I'm writing, it has
started to annoy me that it is so expensive to use static methods.  

And well, this must be something that you have discussed a lot, so
there is no easy solution.  But maybe it would be worth it to compile
static method calls using an indirection, and then flip that
indirection pointer the first time it is used?  This is no worse than
calling a shared library function (on most modern OS'es)..  And it
would also bring gcj closer to spec; the initialization is supposed to
happen *at* the first real use.  Then there's just static field
initializaiton left...

Just 2 cents, ...

-- Kresten

 Kresten Krab Thorup, Ph.D. Candidate
 c/o Yonezawa Laboratory
 Department of Information Science   
 The University of Tokyo             
 7-3-1 Hongo, Bunkyo-ku, Tokyo 113 Japan
 Fax: +81-(0)3-5689-4365	 
 Phone: +81-(0)3-3812-2111 ext 4118
 Mobile: +81-(0)90-3693-5715