bit code should pay the penalty of larger pointer (reference)
and benefit (a little ?) from faster 64 bit integer arithmetic.
Hope you enjoy,
Kazuyuki Shudo[EMAIL PROTECTED] http://www.shudo.net/
--
To
oper/earlyAccess/pp4zaurus/
I am not sure this one work on your SL5600 but it will run very
likely. I could have it work even on iPAQ/Linux (with a trivial tweak).
The runtime runs faster than Blackdown's one.
Kazuyuki Shudo[EMAIL PROTECTED]
java.sun.com.
Blackdown Offers Java Support for the Upcoming AMD Opteron
and AMD Athlon Processors Based On Hammer Technology
http://www.blackdown.org/java-linux/java2-status/pr/Blackdown-20021029.html
I have just noticed this release now.
I assume that the adapted J2SE exploits 64-b
u mean Blackdown's 1.4.2 exploits 64-Bit Mode of AMD64?
Was the adaptation made by Sun or Blackdown?
I cannot find a version of J2SE exploiting the 64-Bit Modde on the
J2SE download page in java.sun.com.
Kazuyuki Shudo[EM
65 MFlops
Yes, Linpack benchmark is not good for benchmarking those runtimes
because flopating-point operations are performed by software emulating
FPU. What the benchmark measures is not clear. But, J2ME PP is much
faster I suppose.
Kazuyuki Shudo
ads provided -tss
option. When the option is given, threads are scheduled in a
time-slicing fashion. I guess even user-space threads can implement
time-slicing with kinds of signal and alarm(2). But, I'm not sure the
-tss option could work well ever on Linux.
Kazuyuki Shudo[EMAIL
j2re-1.3.1-02b-FCS-linux-i386.bin'
Note that JDK (JRE) 1.4 does not contain the Classic VM.
Thus we cannot use green threads with 1.4.
Kazuyuki Shudo[EMAIL PROTECTED] http://www.shudo.net/
--
To UNSUBSCRIBE, ema
f Java 2 SDK.
I could actually use Classic VM and green threads with Netscape
Navigator 4.X by modifying the source code. FreeBSD JDK 1.3.1p6
includes a Java Plug-in which uses green threads since the green
threads version of the their JDK is still stabler.
Kazuyuki Shudo
E for Linux/ARM:
http://www.handhelds.org/pipermail/ipaq/2001-June/007221.html
I can see Duke dancing on my iPAQ when type `java -jar IPAQDemo.jar'.
The demo takes about 75 seconds (!) to start up.
Can you see Duke?
Kazuyuki Shudo[EMAIL PROTECT
ides, I realized there are other changes to
Sun's raw 1.3.1 source code.
Will you get the patch available publicly?
Or will the changes be incorporated into Sun's code?
Kazuyuki Shudo[EMAIL PROTECTED]
-
r Linux/Itanium soon.
Of course, this is only my guess.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subj
web page of a small
web server implemented with /bin/sh and /bin/dd.
It has no relation to Java/Linux at all.
Kazuyuki SHUDO Happy Hacking!
Muraoka Laboratory, Waseda University
--
To UNSUBSCRIBE,
to-date
> version, although still without serious optimizations. But
> that codebase is still slushing, in prep for a release.)
Kazuyuki SHUDO Happy Hacking!
Muraoka Laboratory, Waseda University
--
www.tu-chemnitz.de/urz/java/news/00021.html
The results shown in this page are quite old.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE
st12-pre7
>glibc 2.2
>Pentium III UP
Both HotSpot Client VM and Server VM work on my machine, that has
Linux 2.4.0-test12-pre7, glibc 2.2 and Pentium III.
The Classic VM also works.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Gr
Kees Jan Koster wrote the following interesting article
to the FreeBSD-Java list.
It mentions the relationship between Sun and Blackdown.
(He permitted me forward this article to this list.)
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. &
> I just upgraded my laptop from RedHat 6.2 to RedHat 7.0 and found out that :
>
> - The Sun JDK 1.2.2-006 seg faults
I also found that Green threads does not work with glibc 2.1.93.
Both of Blackdown and Sun JDK 1.2.2 fall in SIGSEGV.
But they run if I use native threads.
Kazuy
nal scores also vary.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe&q
Hi,
Can we have the patch for JDK 1.2.2 FCS ?
What I can find now is only the patch for 1.2 pre-v2.
Or, doesn't Sun allow Blackdown team to release it?
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Wa
e attributes of both of the
HotSpot Client and Server VM, namely peak performance
and quick start-up.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
CM Java Grande 2000 Conference
http://www.extreme.indiana.edu/java00/
Java Grande Forum
http://www.javagrande.org/
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
---
contains only SPARC stuffs.
Will the HotSpot VM have a JIT compiler?
I know Sun migrates from the classic VM to the HotSpot VM.
One of my concerns is the effect of that migration
on developments of JIT compilers.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. Schoo
nd OpenJIT. They are helpful in
understanding the JIT API and the JDK source code may be
also helpful. And JIT developers on this list may answer
if you have specific questions.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. Sch
on the following pages:
ftp://gonzalez.cyberus.ca/pub/Linux/java/
http://www.dsv.su.se/~jens-and/tya/
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE,
printed paper.
It is listed on the publications page of one of the authors:
http://www.wi.leidenuniv.nl/home/ajcbik/pub.html
But a printable file isn't provided there.
Kazuyuki SHUDO Happy Hacking!
eloped by IBM Tokyo Research Lab.(TRL)
researchers. Some papers I've read state optimization
techniques which implemented to the JIT and effects of
each methods in detail.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., School of Sci. & Eng., Waseda Univ.
---
blicly
yet, but we can see some part of it in some header files
located at JDK_HOME/include/, oobj.h and interpreter.h.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
---
ify the following
line as you like:
DEFAULT_THREADS_FLAG=native
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subj
I guess, JDK
doesn't pass sigcontext to a signal handler because
Linux 2.0 and libc5 can't pass it to the handler. But
most OSes and JDKs can do it.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
Hi Blackdown team,
When can we get the linux patch to JDK 1.2.2?
I need a native thread part of the patch to debug my software.
Especially, I'm eager for threads_linux.c.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Wa
Paolo Ciccone wrote:
> On Thu, Dec 23, 1999 at 03:47:56PM +0900, SHUDO Kazuyuki wrote:
>
> > And I have more questions about JVMDI. Does Inprise JIT
> > (libjavacomp.so) support JVMDI? And how about Sun's
> > (libsunwjit.so)? I suppose Sun JIT does not and Inpris
d "On stack
replacement". The function can convert the execution
context in compiled code to the context in JVM. And IBM
JIT can also do it as far as I heard.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., School of Sci. & Eng., Waseda Univ.
---
ckdown
JDK 1.2. The reference implementation seems to have an
implementation of JVMDI as far as I check the source
code of it. It's no wonder JVMDI doesn't work properly
with Sun's JIT (sunwjit.so). Doesn't JVMDI in the
Blackdown JDK work even if JIT is disabled?
Kazuyuki SHUDO
s for exception handling don't work well with the RC3.
Inprise/Sun 1.2.2 RC1 doesn't have this problem.
It is unfortunate that the Blackdown team doesn't know
requirements of JIT well. It is more unfortunate that
Inprise gave me so bad impression today. :)
Kazuyuki SHUDO
gt;
>public static int divfunc(int x, int y) {
> return x/y;
>}
>
>public static void main(String args[]) {
>
> int i = 0x8000;
> int j = -1;
>
> int k = divfunc(i,j);
> System.out.println("Result is "+Integer.toHexSt
T off and
> > it will work.
> >
> > Juergen
> btw: the following is also wrong:
>
> int x=6;
> double d=7.7;
> System.out.println( (x>d) );
> > true
>
> with the JIT on and JDK1.2 .
> when will (will it?) this be fixed ?
Kazuyuki SHUDO
land.public.jbuilder.linux .
JAVA_COMPILER_VERBOSE is a debug level.
`0',`1',`2' and `3' are effective.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
> From: Bobby Hitt
> /usr/local/jdk1.2/bin/i386/native_threads/java: error in loading
> shared libraries
> /lib/libjvm.so: undefined symbol: __bzero
I guess you're using JDK for glibc 2.1 on a system based on glibc 2.0.
You must use the JDK corresponding to your libc
ent value has the same number of
significand bits as a normal double precision, it's 52
bit, not 64 bit.
Additionally, JDK 1.1.7 for Linux doesn't adjust
rounding precision. It is left as extended (80 bit)
precision. JDK 1.2 pre-release 2 does ajust.
Kazuyuki SHUDO
AccessError should be thrown when a substitution
to a final variable is tried out.
IncomatibleClassChangeError should be thrown when a
class doesn't implement the interface expected to be
implemened.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School o
Does JDK 1.2pre2 support Java Virtual Machine Debug
Interface (JVMDI)? My understanding is that it is a C
API described in JDK_ROOT/include-old/jvmdi.h.
I'm under review in using JVMDI to reconstruct a thread
migration system I'v developed.
And how about JVMPI?
Kazu
the trial seems to succeed.
This feature also works with JDK 1.1.7 not only JDK 1.2.
Now, with strictfp modifiers, shuJIT produces the same
results as what SPARC processor derives. An attached
test code shows:
% java StrictfpTest
shuJIT for Sun JVM/IA-32 Copyright 1998,199
But now
those routines are entirely written in Java. I guess
that the purpose of the rewrite was improvement of
portability of the class library.
If we decide to develop a JVM or a
Java-to-another-language translator, we must implement
native methods in a class library that we adopt. It is a
very
the application you are
> running.
>
> -Anand.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
d treat an `irem' bytecode instruction not only `idiv'.
I've fixed my JIT (http://www.shudo.net/jit/)
with SIGFPE as a suggestion by Matt.
Thanks to Matt.
Kazuyuki SHUDO Happy Hacking!
irectory in the linux jdk (for instance
> threads.h, sys_api.h, oobj.h, etc.) . some of these header
> files i could find in a directory called "include-old" in the linux
> JDK but including those leads to further missing header files.
Did you see an include-ol
tions. The JIT with JDK for Win32
(symcjit.dll) is developed by Symantec and the JIT with
JDK 1.2 for Linux (libsunwjit.so) is based on one for
Solaris developed by Sun.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci.
that kind of Java compiler doesn't affect
the results.
>byte int long
> 117a arraycopy 109 122 121
> 117a a[i] = b[i] 311 21
> 12pre arraycopy 123 117 120
> 12pre a[i] = b[i] 20 60 83
Kazuyuki SHUDO
ath. And on Win32, you have to specify a property
`env.class.path' to make javac find soruce codes.
System.setProperty("env.class.path", System.getProperty("java.class.path"));
Kazuyuki SHUDO Happy
des sunwjit but the 1.2 JDK isn't ready yet.
> We get sunwjit as a binary and when 1.2 is ready both the JDK and sunwjit
> will be released at the same time and in one package.
The JIT may show a bit less performance than a JIT for
Solaris/x86. I don't think Sun optim
pted to shuJIT better.
> and I'd be curious to hear about it's ease of use.
> (http://www.shudo.net/jit/)
Its usage is the same as other JITs.
You can use it by the JVM option `-Djava.compiler=shujit'
or setting an environmental variable `JAVA_COMPILER=shujit'.
Kazuyuki
tion work.
You can compile shuJIT which support -prof.
Add -DENABLE_PROFILING to CDEBUGFLAGS or CFLAGS line in GNUmakefile
and make it.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
2.0/gblic2/x86
ftp://ftp.netscape.com/pub/communicator/4.5/english/unix/unsupported/linux20_glibc2/navigator_standalone/navigator-v45-export.x86-unknown-linux2.0glibc.tar.gz
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
defined (__PPC__) \
|| (defined(__mips__) && (defined(MIPSEB) || defined (__MIPSEB__)))
# define BIG_ENDIAN
#else
# error architecture not supported by the Linux C library
#endif
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
libc 2.0.7 have a symbol _dl_symbol_value but 2.0.110
don't have it. I need JDK adapted to recent glibc.
Are there any plans?
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
_1'
even if LANG is `ja_JP.ujis' or `ja_JP.sjis' or
`ja_JP.iso-2022-jp'.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
criptors a process
and select(2) can handle. You may be able to get more
descriptors by increasing these numbers and re-compiling
your kernel.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
Are there any places to have a discussion referring to
source code of JDK? I'd like to an i18n problem I've
found with JDK source code.
Kazuyuki SHUDO Happy Hacking!
be archived with some tools such as smtpfeed
(http://www.kyoto.wide.ad.jp/smtpfeed/).
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
ased with Java 2 SDK (JDK
1.2) for Linux/PowerPC. Kevin B. Hendricks mentioned this here.
> Subject: Sun and Linux
> Message-Id: <v03010d01b2a351a29655@[209.96.178.126]>
> Date: Sun, 20 Dec 1998 20:28:37 -0500
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab.,
ll
native code is written in JNI.
Kazuyuki SHUDO Happy Hacking!
Muraoka Lab., Grad. School of Sci. & Eng., Waseda Univ.
E-mail: [EMAIL PROTECTED]
Address: Muraoka Lab., Waseda Univ.,
3-4-1 Okubo Shinjuku-ku Tokyo 169-8555 JAPAN
TEL: +81-3-3209-5198FAX: +81-3-3209-5198
61 matches
Mail list logo
