JPDA (aka JDI) implementation?

1999-06-22 Thread Anonymous


Hi folks!

Sun has a new debugger architechture, JPDA (Java Platform Debug
Architechture), sometimes also referred to as JDI (Java Debug
Interface). It is currently in Beta release
(http://developer.java.sun.com/developer/earlyAccess/jpda/index.html).

Are there plans or acivities yet to port this to Linux?  If so, is
there an estimate when it may become available?

Thanks,

Michael


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: [off-topic] stop bashing!

1999-06-22 Thread Anonymous

1) I am not a member of java-linux-porting@blackdown. They are a private porting club.
2) I have been porting the Javasoft's non-commercial JDK source for a number of 
years.. But just for for alpha/linux.
3) I have also been a member of java-linux@blackdown for  a number of years.
4) the change of anything in '.java', including the lp ->lpr, would violate your 
notion, sorry , interpretation of the license requirements
5) In the land of alpha longs are 64bits as well as address pointers =>

.java
intlets_store_a_32_bit_address;
private nativeint wheres_my_window();
lets_store_a_32_bit_address = wheres_my_window();
..

Native .c
intwheres_my_window(blah*,blah ){
 longaddr;
addr = (Window) w;
return(addr);
}
This particular example results in the truncation of a returned 64 bit 'c' address, 
into a 32bit '.java' variable/storage. u can of course change the 'c' routine to 
return all 64bits, but that will be, eventually, truncated by the jvm to 32bits. In 
the Land of "OSF/true64" for the Dec Alpha, u can convince the compiler to produce 
32bit pointer arithemetic, where long == 32bits. But the currently available 
linux/alpha compilers a long==pointer==64bits.

6) Actually complaining, or presenting my viewpoints does do some good. The whole 
linux-java community is not just java-linux-porting@blackdown. There are other linux 
folks out there, of which Sun pays little attention to.  Cant imagine how the 
java-linux-porting folks got their commercial license, if it weren't for all of those 
complaining linux/Intel folks - in this forum, and in the Sun's bug-tracking system.

7) I didn't ask u to set a reasonable fee, I just threw back ur statement as to what 
would be a reasonable fee be to port java/linux/alpha. After all sun did say that they 
support the linux community. I'm just waiting for the support that they had advertised.

gat

Jeff Galyan wrote:

> If you're part of the Java-linux porting team, then the concessions Sun
> made to them should also extend to you, right?
>
> Now, a change of 'lp' to 'lpr' in a Runtime.exec call is *very*
> different than changing the size of an int. The license I have for the
> Java VM source says that none of the .java files are to be modified at
> all. I'm sure Apple makes some modifications, too, but again, this is
> what my license says.
>
> As for stating a reasonable fee, I'm not the person who would be setting
> the fees, so I cannot give you a quote of any kind. If the Blackdown
> distribution license doesn't extend to you, then you'd best contact Sun
> to find out why (since you say you're porting for Linux on Alpha) and
> what can be done to remedy that. Complaining on a mailing list that Sun
> isn't catering to you isn't going to do anyone any good.
>
> --Jeff




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



JDK 1.2.2 Performance

1999-06-22 Thread Anonymous

Hi,

I'm new to linux (but not java) so maybe I'm doing something

wrong and you guys can point me in the right direction.

I thought Swing was slow on Windows... I appreciate that JDK
1.2.2
is still beta, but i'm having serious performance problems
with my
apps on linux, where I don't with the other os.

A good example is the SwingSet demo. In the other os it
loads in
30 secs (40 secs with the JIT). In linux the best I've seen
is 60 secs,
this on an otherwise "empty" machine (i.e. all usual started
tasks).

The problem is storage - I have 32m ram. According to top,
the SwingSet
demo has a size of 22M (13M RSS) after loading - this with
-Xxm32m and no JIT. (With a 16m heap it quickly goes
OutOfMemory). The JIT adds about 3meg.

And it gets worse: Choose options|change L&F from the menu 5
times,
switching between metal & cde, size goes up to 34m and RSS
to 22m. Of course by now the system is thrashing wildly, and
each L+F change
takes a minute or two (Yes, same without top running). On
the other os
each L+F change takes about 5 secs !

(Incidently, UIManager.getInstalledLookAndFeels() offers
Windows L+F, but of course it aint there.)

I have exactly these problems, to a lesser degree, with my
own apps.

Help !

BTW, I have compared the sun & symantec JITs on a
compute-intensive
application (matrix diagonalisation). Both reduce elapsed
time from about
8 mins to less than 2 - pretty impressive. (The symantec JIT
is actually
35 sec faster, but all the difference is in one method, so
this isnt
very representative).

Nick



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



msqljava implementation

1999-06-22 Thread Anonymous

Hi all

I'm going to develop a jdbc cleint applet for a msql database in my company.
The lib of msqljava wants me to add it to the classpath (so I created one).
When I try to compile written code the "import msql.*;" is not accepted,
because it's not found. I tried the classpath extension of the javac but now
I'm able to import msql.* only. How to add foreign classes to the system.
Even if I know that I'm asking beginner questions I hope one of you'll send
me an answer. I tried 1.2pre and 1.1.7 v1a, but it's the same with both
packages.

Thanks for your help.

Jens


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: msqljava implementation

1999-06-22 Thread Anonymous

Hi Jens,

let me say that you made the typical beginner's error. I don't mean your
problem with java or linux - all of us have such problems from time to
time, even experts. The real error is to not give us enough information
to help you. Perhaps you just made a typo when setting the CLASSPATH: we
cannot find out if we do not know exactely what you did.

So please give the exact commands that you used: the working ones and
the not-working ones. Tell us what shell you are using. Give us the
exact error messages. And, check on your system that the user-id the
java VM is running under has the permissions to read the archive you
want to add. Check the jar file if it really contains the classes you
think it does.

Matthias


"Neumann, Jens" wrote:
> 
> Hi all
> 
> I'm going to develop a jdbc cleint applet for a msql database in my company.
> The lib of msqljava wants me to add it to the classpath (so I created one).
> When I try to compile written code the "import msql.*;" is not accepted,
> because it's not found. I tried the classpath extension of the javac but now
> I'm able to import msql.* only. How to add foreign classes to the system.
> Even if I know that I'm asking beginner questions I hope one of you'll send
> me an answer. I tried 1.2pre and 1.1.7 v1a, but it's the same with both
> packages.
> 
> Thanks for your help.
> 
> Jens
> 
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



RE: msqljava implementation

1999-06-22 Thread Anonymous

Hi Matthias

thanks for your quick response. I tried some things in between and the
problem moved. My Applet is connecting to a database using port 1114 and
using mslqjava 1.2.8.
I found the right place to locate the classes and comilation works fine. I
copied the Applet1.class file and another Applet1$SymAction.class on my web
server and made a test2.html.

I checked the rights (755) and it's nop problem to run the applet on some
machiones. Other computers with the same Browser fail with an security error
(IE says:" com.ms.security.SecurityExceptionEx(Apllet1.init)" and Mozilla
says:"Applet Applet1 can't start: class Applet1 got a security violation:
method verification error"

I don't see the reason for the different result with different machines? Do
you or somebody else know this problem?

Thanks for your help.

Jens


Jens Neumann
Senior Support Engineer

Digi International AG
(formerly ITK AG)
Joseph-von-Fraunhofer-Str. 23
44227 Dortmund
Germany

Tel.: +49-231-9747-250
Fax: +49-231-9747-777
Email: [EMAIL PROTECTED]


> -Original Message-
> From: Matthias Pfisterer
> [SMTP:[EMAIL PROTECTED]]
> Sent: Tuesday, June 22, 1999 4:15 PM
> To:   Neumann, Jens
> Cc:   '[EMAIL PROTECTED]'
> Subject:  Re: msqljava implementation
> 
> Hi Jens,
> 
> let me say that you made the typical beginner's error. I don't mean your
> problem with java or linux - all of us have such problems from time to
> time, even experts. The real error is to not give us enough information
> to help you. Perhaps you just made a typo when setting the CLASSPATH: we
> cannot find out if we do not know exactely what you did.
> 
> So please give the exact commands that you used: the working ones and
> the not-working ones. Tell us what shell you are using. Give us the
> exact error messages. And, check on your system that the user-id the
> java VM is running under has the permissions to read the archive you
> want to add. Check the jar file if it really contains the classes you
> think it does.
> 
> Matthias
> 
> 
> "Neumann, Jens" wrote:
> > 
> > Hi all
> > 
> > I'm going to develop a jdbc cleint applet for a msql database in my
> company.
> > The lib of msqljava wants me to add it to the classpath (so I created
> one).
> > When I try to compile written code the "import msql.*;" is not accepted,
> > because it's not found. I tried the classpath extension of the javac but
> now
> > I'm able to import msql.* only. How to add foreign classes to the
> system.
> > Even if I know that I'm asking beginner questions I hope one of you'll
> send
> > me an answer. I tried 1.2pre and 1.1.7 v1a, but it's the same with both
> > packages.
> > 
> > Thanks for your help.
> > 
> > Jens
> > 
> > --
> > To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> > with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Servlet / AWT question

1999-06-22 Thread Anonymous

I am using RedHat 6.0 with the 1.1.7 V3 JDK.  I have some charts that get
created from within servlets.  Does anybody know if I am required to have a
user logged in with X running for AWT to work?  It seems that the user who
starts the servlet engine, JRun, must stay logged in an have X up?  This
makes it impossible to have the system start JRun properly from rc.d.

Is there anyway around these problems?

Thanks in advance,
Rob


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Servlet / AWT question

1999-06-22 Thread Anonymous

> APP INF writes:

APP> I am using RedHat 6.0 with the 1.1.7 V3 JDK.  I have some
APP> charts that get created from within servlets.  Does anybody
APP> know if I am required to have a user logged in with X running
APP> for AWT to work?  It seems that the user who starts the
APP> servlet engine, JRun, must stay logged in an have X up?  This
APP> makes it impossible to have the system start JRun properly
APP> from rc.d.

Yes, AWT doesn't work without a connection to an Xserver.

APP> Is there anyway around these problems?

I haven't tried it but maybe Xvfb will help you:

 xvfb provides an X server that can run on machines with no display hardware
 and no physical input devices.  It emulates a dumb framebuffer using virtual
 memory.  The primary use of this server was intended to be server testing, but
 other novel uses for it have been found, including testing clients against
 unusual depths and screen configurations, doing batch processing with Xvfb as
 a background rendering engine, load testing, as an aid to porting the X server
 to a new platform, and providing an unobtrusive way to run applications that
 don't really need an X server but insist on having one anyway.



Juergen


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: JDK 1.2.2 Performance

1999-06-22 Thread Anonymous

--- Nick Lawson <[EMAIL PROTECTED]> wrote:

> I thought Swing was slow on Windows... I appreciate that JDK
> 1.2.2
> is still beta, but i'm having serious performance problems
> with my
> apps on linux, where I don't with the other os.

No, this is pretty much what I'd expect.  Windows has a lot more
resources to develop JVM's.  It has always been the fastest
platform to run Java on.

> The problem is storage - I have 32m ram. According to top,
> the SwingSet
> demo has a size of 22M (13M RSS) after loading - this with
> -Xxm32m and no JIT. (With a 16m heap it quickly goes
> OutOfMemory). The JIT adds about 3meg.

sunwjit, or are you running tya?  Tya is really very good, but I
haven't measured the increase in memory.

> Help !

Maybe IBM will release a JVM for 1.2 that's more efficient.  The
blackdown is a straight port from the solaris version, and I've
got a funny feeling they're not allowed (by the license) to
distribute modifications other than that required to do the port
itself.  (ie, no rewrites for performance enhancements).  Maybe
sun will release the hotspot sourcecode to us!

Paul
_
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Servlet / AWT question

1999-06-22 Thread Anonymous

"Schlackman, Robert (APP INF)" wrote:
> 
> I am using RedHat 6.0 with the 1.1.7 V3 JDK.  I have some charts that get
> created from within servlets.  Does anybody know if I am required to have a
> user logged in with X running for AWT to work?  It seems that the user who
> starts the servlet engine, JRun, must stay logged in an have X up?  This
> makes it impossible to have the system start JRun properly from rc.d.
> 
> Is there anyway around these problems?

You need the AWT but not any part of it that happens to use X. Sounds
like a long shot, but I wonder if running the version of the JDK1.1
executable that doesn't use X (set the NS_JAVA environment variable)
will work in this circumstance.

Failing that, Juergen Kreileder's suggestion on using Xvfb is a clever
kluge.

Nathan


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



ANNOUNCE: Manager/Factory utility in java using CORBA

1999-06-22 Thread Anonymous

Generic Manager/Factory implementation

needs: JacORB, >jdk1.1.x

Making a distributed system i had the need of having somehow something that
would allow to start on a network of machines distributed process. I was told
to use the factory paradigm, which means having on each machine of the network
a daemon that would instantiate the needed objects.

Trying to achieve this i hit into the limitation of the nameserver to allow
only unique bindings to names.

I looked at a GPL'd trader in java, but it needed dynamic classes, and as it
seems i have not the faintest idea of the purpose of a trader, further i am
using JacORB as an orb implementation, and it doesn't support the dynamic
functionalities requested by this trader.

So i searched further but finding nothing, implemented my-self a system which
acts like a modifyed nameserver, letting record service names and service
providers bound to them, letting ask some object to this manager to deliver an
object bound to a particular service (with a minimalistic loadbalancing).

Together with that manager are delivered some (hopefully) generic factories
able to instantiate objects which are directly derived from
org.omg.CORBA.Object or objects using the delegation-pattern and accessible
through a tie-class.

Hoping that this could be usable for others i make them available under the
GPL, look at http://erm1.u-strasbg.fr/~bboett/jtrader/ for further
information. 

-- 
ciao bboett
==
[EMAIL PROTECTED]
http://erm6.u-strasbg.fr/~bboett
===
the total amount of intelligence on earth is constant.
human population is growing


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: msqljava implementation

1999-06-22 Thread Anonymous

Hi Jens,

the reason for the failures are clear: The sandbox strikes back.
Different browsers impose different restrictions on the execution of
unsave operations. But again, there is not enough info to tell exactely
what's happening. To give concrete advice, I need to know more: Which in
the the successful case broser? How are the additional classes (mysql.*)
loadded (locally or via the network)?

There are only two hints I can give you blindly:
1) with appletviewer, it is possible to change the security
restrictions. Try to make them more restrictive untill you get a simular
error as in the browsers. Expecially try to forbid to open network
connections to hosts other than the one the applet was loaded from.
2) With Netscape Navigator, open the Java console
(Communicator/Tools/Java Console) before executing the applet. The stack
dump from the Exception will show in it. This way, you can find out
about the line number where the error occurs.


Matthias


"Neumann, Jens" wrote:
> 
> Hi Matthias
> 
> thanks for your quick response. I tried some things in between and the
> problem moved. My Applet is connecting to a database using port 1114 and
> using mslqjava 1.2.8.
> I found the right place to locate the classes and comilation works fine. I
> copied the Applet1.class file and another Applet1$SymAction.class on my web
> server and made a test2.html.
> 
> I checked the rights (755) and it's nop problem to run the applet on some
> machiones. Other computers with the same Browser fail with an security error
> (IE says:" com.ms.security.SecurityExceptionEx(Apllet1.init)" and Mozilla
> says:"Applet Applet1 can't start: class Applet1 got a security violation:
> method verification error"
> 
> I don't see the reason for the different result with different machines? Do
> you or somebody else know this problem?
> 
> Thanks for your help.
> 
> Jens
> 
> Jens Neumann
> Senior Support Engineer
> 
> Digi International AG
> (formerly ITK AG)
> Joseph-von-Fraunhofer-Str. 23
> 44227 Dortmund
> Germany
> 
> Tel.: +49-231-9747-250
> Fax: +49-231-9747-777
> Email: [EMAIL PROTECTED]
> 
> > -Original Message-
> > From: Matthias Pfisterer
> > [SMTP:[EMAIL PROTECTED]]
> > Sent: Tuesday, June 22, 1999 4:15 PM
> > To:   Neumann, Jens
> > Cc:   '[EMAIL PROTECTED]'
> > Subject:  Re: msqljava implementation
> >
> > Hi Jens,
> >
> > let me say that you made the typical beginner's error. I don't mean your
> > problem with java or linux - all of us have such problems from time to
> > time, even experts. The real error is to not give us enough information
> > to help you. Perhaps you just made a typo when setting the CLASSPATH: we
> > cannot find out if we do not know exactely what you did.
> >
> > So please give the exact commands that you used: the working ones and
> > the not-working ones. Tell us what shell you are using. Give us the
> > exact error messages. And, check on your system that the user-id the
> > java VM is running under has the permissions to read the archive you
> > want to add. Check the jar file if it really contains the classes you
> > think it does.
> >
> > Matthias
> >
> >
> > "Neumann, Jens" wrote:
> > >
> > > Hi all
> > >
> > > I'm going to develop a jdbc cleint applet for a msql database in my
> > company.
> > > The lib of msqljava wants me to add it to the classpath (so I created
> > one).
> > > When I try to compile written code the "import msql.*;" is not accepted,
> > > because it's not found. I tried the classpath extension of the javac but
> > now
> > > I'm able to import msql.* only. How to add foreign classes to the
> > system.
> > > Even if I know that I'm asking beginner questions I hope one of you'll
> > send
> > > me an answer. I tried 1.2pre and 1.1.7 v1a, but it's the same with both
> > > packages.
> > >
> > > Thanks for your help.
> > >
> > > Jens
> > >
> > > --
> > > To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> > > with a subject of "unsubscribe". Trouble? Contact
> > [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Java 1.2 pre2 and SMP?

1999-06-22 Thread Anonymous

I recently (this weekend) attempted an upgrade to Java 2 on our web
servers, and came across a few problems someone might be able to help
with. Briefly:

Java 1.2 green threads won't start on one system: It just hangs taking
all processor time and memory. This is a Dual PII machine with SMP
running kernel 2.2.9 and I've reported this as a bug to Blackdown. It's
not my software: it happens even if I just run "java" on its own.

Java 1.2 *native* threads seems to work fine, and really fast too (no
JIT even) on the same machine, but software problems exist - those I
think are mine I've fixed, but something weird is happening wrt Postgres
6.5 - more detail later.

Java 1.2 green threads works perfectly on the develop machine (1 x
AMDK6/2)

Java 1.2 native threads works the same on this machine as on the SMP
one, same problems too.

I'm doing all this because I need a particular TimeZone bug fixed, which
was fixed in 1.1.8 or later, but there isn't a 1.1.8 for Linux.

OK, more detail:

Both systems are SuSE Linux 6.0 with some updates: Apache 1.3.6,
Postgres 6.5, Live Software's JRunPro 2.3.1 build 145.  The develop
system has a single AMD K6/2 processor, is running the SuSE kernel
2.0.36, whereas the live server has twin Pentium IIs and is running a
home-cooked SMP 2.2.9 (with the ICMP DoS fix applied).

The problem with green threads on the SMP system is the big one at the
moment, as if that was fixed, I could deploy instantly.

The native thread problem I'm having seems to be down to software. The
JVM seems to work fine (odd seeing about 26 little Java processes
sitting there awaiting hits) and once I fixed my own code (mostly
involved removing statics and improving design so I didn't need them -
shouldn't have been there anyway) so it doesn't keep opening and never
closing Postgres connections, that seems to work fine - unless I do two
identical searches via the website at the same time, at which point the
two Postgres tasks suddenly take a long time, eat up all processor time,
then give up leaving Postgres in an unusable state. As the same problem
doesn't occur when running under Green threads, or when concurrent
searches are not looking for the same things, I don't know now whether
the problem is in my code, the JDBC driver, or the JVM.

-- 
Rachel



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Java 1.2 pre2 and SMP?

1999-06-22 Thread Anonymous

Rachel Greenham wrote:
> 
> I recently (this weekend) attempted an upgrade to Java 2 on our web
> servers, and came across a few problems someone might be able to help
> with. Briefly:
> 
> Java 1.2 green threads won't start on one system: It just hangs taking
> all processor time and memory. This is a Dual PII machine with SMP
> running kernel 2.2.9 and I've reported this as a bug to Blackdown. It's
> not my software: it happens even if I just run "java" on its own.

Last I knew (unless it's been fixed in preV2) the JIT compiler was
incompatible with green threads. Try running with -Djava.compiler= in
the command-line to disable JIT.

Nathan


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Java 1.2 pre2 and SMP?

1999-06-22 Thread Anonymous

Nathan Meyers wrote:
> 
> Rachel Greenham wrote:
> >
> > I recently (this weekend) attempted an upgrade to Java 2 on our web
> > servers, and came across a few problems someone might be able to help
> > with. Briefly:
> >
> > Java 1.2 green threads won't start on one system: It just hangs taking
> > all processor time and memory. This is a Dual PII machine with SMP
> > running kernel 2.2.9 and I've reported this as a bug to Blackdown. It's
> > not my software: it happens even if I just run "java" on its own.

Correction, the latter seems to sometimes work. But starting the servlet
engine (JRunPro) always dies, though it starts up fine on my develop
machine, and on native threads.

> Last I knew (unless it's been fixed in preV2) the JIT compiler was
> incompatible with green threads. Try running with -Djava.compiler= in
> the command-line to disable JIT.

Hmm. Well, it's been saying it can't find the JIT anyway. Actually, I'll
just check that's the case on the SMP box, it definitely is here, I get,
when I run Java:

Warning: JIT compiler "sunwjit" not found. Will use interpreter. 

Yes, also on the SMP box.

$ java -version
Warning: JIT compiler "sunwjit" not found. Will use interpreter.
java version "1.2"
Classic VM (build Linux_JDK_1.2_pre-release-v2, green threads, nojit)

This is from the JDK downloaded from Blackdown (actually the UK sunsite
mirror) yesterday, and all I've done is change the default thread type
to green.

-- 
Rachel


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: JDK 1.2.2 Performance

1999-06-22 Thread Anonymous

Paul Mclachlan wrote:

> --- Nick Lawson <[EMAIL PROTECTED]> wrote:
>
> > I thought Swing was slow on Windows... I appreciate that JDK
> > 1.2.2
> > is still beta, but i'm having serious performance problems
> > with my
> > apps on linux, where I don't with the other os.
>
> No, this is pretty much what I'd expect.  Windows has a lot more
> resources to develop JVM's.  It has always been the fastest
> platform to run Java on.

Well, I hope not ! I've seen various (unsupported) claims for the
superior
performance of linux - and as I mentioned I'm new to linux, so
that's
what I would like to see. (BTW, this is Suns jdk under Windows,
not MS).

> > The problem is storage - I have 32m ram. According to top,
> > the SwingSet
> > demo has a size of 22M (13M RSS) after loading - this with
> > -Xxm32m and no JIT. (With a 16m heap it quickly goes
> > OutOfMemory). The JIT adds about 3meg.
>
> sunwjit, or are you running tya?  Tya is really very good, but I
> haven't measured the increase in memory.

I dont think JIT is an issue here - loading should be (and is)
faster
without a JIT, but still slow because of paging. My impression from
watching top is that other processes dont get swapped out very fast
- a linux rather than a java vm issue.

However there is a point here, which surprised me - OutOfMemory
occurs sooner without a JIT than with a JIT (also in Windows).
Can't imagine why.

There is a jvm aspect which worries me, and that is the rate at
which
it chews up the heap when you change L&F - about 2.5m per shot!
Then it runs out of memory - why? With the JIT, its a lot less per
shot,
but still runs out of memory eventually.

> > Help !
>
> Maybe IBM will release a JVM for 1.2 that's more efficient.  The
> blackdown is a straight port from the solaris version, and I've
> got a funny feeling they're not allowed (by the license) to
> distribute modifications other than that required to do the port
> itself.  (ie, no rewrites for performance enhancements).  Maybe
> sun will release the hotspot sourcecode to us!

We wish!

Nick



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: JDK 1.2.2 Performance

1999-06-22 Thread Anonymous

Paul Mclachlan wrote:

> --- Nick Lawson <[EMAIL PROTECTED]> wrote:
>
> > I thought Swing was slow on Windows... I appreciate that JDK
> > 1.2.2
> > is still beta, but i'm having serious performance problems
> > with my
> > apps on linux, where I don't with the other os.
>
> No, this is pretty much what I'd expect.  Windows has a lot more
> resources to develop JVM's.  It has always been the fastest
> platform to run Java on.

Well, I hope not ! I've seen various (unsupported) claims
for the
superior
performance of linux - and as I mentioned I'm new to linux,
so
that's
what I would like to see. (BTW, this is Suns jdk under
Windows,
not MS).

> > The problem is storage - I have 32m ram. According to top,
> > the SwingSet
> > demo has a size of 22M (13M RSS) after loading - this with
> > -Xxm32m and no JIT. (With a 16m heap it quickly goes
> > OutOfMemory). The JIT adds about 3meg.
>
> sunwjit, or are you running tya?  Tya is really very good, but I
> haven't measured the increase in memory.

I dont think JIT is an issue here - loading should be (and
is)
faster
without a JIT, but still slow because of paging. My
impression from
watching top is that other processes dont get swapped out
very fast
- a linux rather than a java vm issue.

However there is a point here, which surprised me -
OutOfMemory
occurs sooner without a JIT than with a JIT (also in
Windows).
Can't imagine why.

There is a jvm aspect which worries me, and that is the rate
at
which
it chews up the heap when you change L&F - about 2.5m per
shot!
Then it runs out of memory - why? With the JIT, its a lot
less per
shot,
but still runs out of memory eventually.

> > Help !
>
> Maybe IBM will release a JVM for 1.2 that's more efficient.  The
> blackdown is a straight port from the solaris version, and I've
> got a funny feeling they're not allowed (by the license) to
> distribute modifications other than that required to do the port
> itself.  (ie, no rewrites for performance enhancements).  Maybe
> sun will release the hotspot sourcecode to us!

We wish!

Nick


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Java 1.2 pre2 and SMP?

1999-06-22 Thread Anonymous

Rachel Greenham wrote:
> I get, when I run Java:
> 
> Warning: JIT compiler "sunwjit" not found. Will use interpreter.
> 
> Yes, also on the SMP box.
> 
> $ java -version
> Warning: JIT compiler "sunwjit" not found. Will use interpreter.
> java version "1.2"
> Classic VM (build Linux_JDK_1.2_pre-release-v2, green threads, nojit)
> 
> This is from the JDK downloaded from Blackdown (actually the UK sunsite
> mirror) yesterday, and all I've done is change the default thread type
> to green.

Three questions:

1) Do you have a jre/lib/i386/libsunwjit.so in your JDK installation?

2) Which JDK are you using, one for glibc2.0 or glibc2.1 (they came from
different subdirectories on the mirror)?

3) Which version of glibc are you using? Do you have a
/lib/libc-2.0. and/or a /lib/libc-2.1.?


What I'm getting at is the possibility that you're running a glibc2.1
system (reasonably likely if you're running a 2.2 kernel) but downloaded
the JDK for a glibc2.0 system.


Nathan


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: [off-topic] stop bashing!

1999-06-22 Thread Anonymous

Uncle George wrote:
> 
> 1) I am not a member of java-linux-porting@blackdown. They are a private porting 
>club.

Then you should be talking to Sun, not java-linux.

> 2) I have been porting the Javasoft's non-commercial JDK source for a number of 
>years.. But just for for alpha/linux.

And your point is...?

> 3) I have also been a member of java-linux@blackdown for  a number of years.

And your point is...?

> 4) the change of anything in '.java', including the lp ->lpr, would violate your 
>notion, sorry , interpretation of the license requirements

Okay, after Juergen's post on this, I can see how this could be argued
as "platform-dependent". The size of an int, though, is specified by the
Java Language Specification and therefore changing it would, as Cees
deGroot pointed out, result in a version of Java which would not be
compatible with other platforms' implementations.

> 5) In the land of alpha longs are 64bits as well as address pointers =>
> 
> .java
> intlets_store_a_32_bit_address;
> private nativeint wheres_my_window();
> lets_store_a_32_bit_address = wheres_my_window();
> ..
> 
> Native .c
> intwheres_my_window(blah*,blah ){
>  longaddr;
> addr = (Window) w;
> return(addr);
> }
> This particular example results in the truncation of a returned 64 bit 'c' address, 
>into a 32bit '.java' variable/storage. u can of course change the 'c' routine to 
>return all 64bits, but that will be, eventually, truncated by the jvm to 32bits. In 
>the Land of "OSF/true64" for the Dec Alpha, u can convince the compiler to produce 
>32bit pointer arithemetic, where long == 32bits. But the currently available 
>linux/alpha compilers a long==pointer==64bits.

Your example is flawed in that you assume you should return an int from
wheres_my_window(blah*, blah). If you return a long, everything is fine,
since in the land of Java, long is defined as a 64-bit data type.


> 
> 6) Actually complaining, or presenting my viewpoints does do some good. The whole 
>linux-java community is not just java-linux-porting@blackdown. There are other linux 
>folks out there, of which Sun pays little attention to.  Cant imagine how the 
>java-linux-porting folks got their commercial license, if it weren't for all of those 
>complaining linux/Intel folks - in this forum, and in the Sun's bug-tracking system.
> 

So does this mean you *are* attempting some communication with Sun?


> 7) I didn't ask u to set a reasonable fee, I just threw back ur statement as to what 
>would be a reasonable fee be to port java/linux/alpha. After all sun did say that 
>they support the linux community. I'm just waiting for the support that they had 
>advertised.
> 

The support that Sun "advertised" is there. All Sun said in the press
release from back in November was that they would provide some support
to the Blackdown efforts - not linux in general. Again, you should be
talking to Sun if you really have issues with this.


-- 
Jeff Galyan
http://www.anamorphic.com
http://www.sun.com
jeffrey dot galyan at sun dot com
talisman at anamorphic dot com
Sun Certified Java(TM) Programmer
==
Linus Torvalds on Microsoft and software development:
"... if it's a hobby for me and a job for you, why are you doing such a
shoddy job of it?"

The views expressed herein do not necessarily reflect those of my
employer.

Sun Microsystems, Inc., has no connection to my involvement with the
Mozilla Organization.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Java 1.2 pre2 and SMP?

1999-06-22 Thread Anonymous

Nathan Meyers wrote:
> 
> Rachel Greenham wrote:
> > I get, when I run Java:
> >
> > Warning: JIT compiler "sunwjit" not found. Will use interpreter.
> >
> > Yes, also on the SMP box.
> >
> > $ java -version
> > Warning: JIT compiler "sunwjit" not found. Will use interpreter.
> > java version "1.2"
> > Classic VM (build Linux_JDK_1.2_pre-release-v2, green threads, nojit)
> >
> > This is from the JDK downloaded from Blackdown (actually the UK sunsite
> > mirror) yesterday, and all I've done is change the default thread type
> > to green.
> 
> Three questions:
> 
> 1) Do you have a jre/lib/i386/libsunwjit.so in your JDK installation?

Yes.

> 2) Which JDK are you using, one for glibc2.0 or glibc2.1 (they came from
> different subdirectories on the mirror)?

glibc2.0

> 3) Which version of glibc are you using? Do you have a
> /lib/libc-2.0. and/or a /lib/libc-2.1.?

SuSE 6.0 is glibc 2.0 I'm given to understand, though it also attempts
libc5 binary support. Actually none of the above files exist as named in
my /lib though that might be a SuSE oddity.

> What I'm getting at is the possibility that you're running a glibc2.1
> system (reasonably likely if you're running a 2.2 kernel) but downloaded
> the JDK for a glibc2.0 system.

Good thought, but I don't think so. :-)

-- 
Rachel


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: JDK 1.2.2 Performance

1999-06-22 Thread Anonymous

Nick Lawson wrote:
> 
> Paul Mclachlan wrote:
> 
> > --- Nick Lawson <[EMAIL PROTECTED]> wrote:
> >
> > > I thought Swing was slow on Windows... I appreciate that JDK
> > > 1.2.2
> > > is still beta, but i'm having serious performance problems
> > > with my
> > > apps on linux, where I don't with the other os.
> >
> > No, this is pretty much what I'd expect.  Windows has a lot more
> > resources to develop JVM's.  It has always been the fastest
> > platform to run Java on.
> 
> Well, I hope not ! I've seen various (unsupported) claims for the
> superior
> performance of linux - and as I mentioned I'm new to linux, so
> that's
> what I would like to see. (BTW, this is Suns jdk under Windows,
> not MS).

The claims are valid.

But the reality of Java is that it's an extremely difficult port with
many platform dependencies, and the volunteer effort behind it cannot
match the efforts of large organizations that have thrown lots of money
behind their porting efforts.

Things will improve -- they've already started to improve -- but not as
fast as you'd like. The reality of the Linux revolution is that manning
the barricades gives you the responsibility to help make things better
and forfeits you the right to bitch and moan about them. That ethos has
led to a stunningly good operating system.

Nathan


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Bug in setResizable() ?

1999-06-22 Thread Anonymous


It seems that issuing setResizable(false) on a JFrame() causes the GUI
to flash for a moment but disappear:

JFrame jFrame = new JFrame("setResizable bug?");
Dimension d = new Dimension(400, 300);
jFrame.setSize(d);
jFrame.getContentPane().setLayout(new BorderLayout());
jFrame.addWindowListener(
new WindowAdapter() {
public void windowClosing(WindowEvent e) {
System.exit(0);
}
}
);
jFrame.setVisible(true);


This code appears to work fine on Windows and Solaris, but is breaking
on the linux 1.2.1 VM.

Marty


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]