eek, I'm going to ask Darin what happened
Juergen Kreileder wrote:
Marc St-Jean writes:
BTW, I went to verify how the "blackdown binary license" would be
different from the Sun license with regards to distribution with a
distro. However after downloading twice from the ftp.tux.org mirror
the LIC
Lots of difficult questions here. Joining the blackdown porting project
means you use the blackdown binary license etc.
If you want to talk to someone at Sun let me know and I'll forward your
questions on
regards
calvin
Marc St-Jean wrote:
>On Jan 7, 2005, at 12:58 PM, Dominic Duval wrote:
>
>
I believe you are seeing a generic javaws bug that is unfortunately a
change in behavior of glibc/waitid Juergen take a look at 6188963, Hui
believes you can just switch in waitpid instead of waitid. btw the 1.5
32bit JVM should also work on your machine too, but the javaws has the
same issue
I found a couple of bug reports you have logged before, most have been
filed to the netbeans group which is why we've probably never seen them.
I don't use Mandrake myself, I did run your program on 1.4.2 (as 2.1)
and saved and it loaded fine. I'm sure debian is fine too. If the
machine is rea
Does it ever resume. 1.5beta2 has more diagnostics in it and you may be
able to even attached with jdb (hey tim :*)
Without seeing the other threads its difficult to see what is happening,
however there was a bug fix in 1.5 to stop the cache growing out of
control. A quick test with 1.5 would h
Join me, Hui Huang and the one and only, Juergen Kreileder of
blackdown.org
in a JDC chat next week!
see below for more details
http://developer.java.sun.com/developer/community/chat/
regards
calvin
--
To UNSUBSCRIBE, email to
Yup native threads doing blocking IO on linux is expensive. Things start
looking progressively ugly from 50 connections upwards.
If you move to NIO you will see a significant improvement in both
stability/predictability and performance.
What would make it easier for you to try the multiplexing
No news is good news. Juergen has got some 1.4.1 beta builds in testing
regards
calvin
Vincent Touquet wrote:
>
> Is there still anyone from the blackdown project active on this list ? :)
>
> Only mails I've been getting so far , is "I'm away on holiday"
> and unsubscription emails :/
>
> W
There is a very important patch to Redhat 7.2 SMP kernel, you need the
2.4.9-31smp patch available under 'security patches' from redhat.com
Unpatched redhat 7.2 SMP will create duplicate process ids. That is not good!
The bug is also fixed in 2.4.18 kernels. Hui Huang tracked this down a while
I recently used JRE 1.4 with netscape 6.2.2 and that worked you can do the
same for mozilla too.
Download JRE 1.4 if you can (1.3.1 does work too but I've seen a couple
of annoying glitches with the default plugin they have)
Then do the following, if libjavaplugin_oji does not exist don't worry.
try one of the old gotchas
ulimit -s 2048
> Barnet Wagman wrote:
>
> With Blackdown j2sdk-1.3.1-02a-FCS-linux-i386, I get the following when I
> run 'java -version'
>
> java version "1.3.1"
> Java(TM) 2 Runtime Environment, Standard Edition (build
> Blackdown-1.3.1-02a-FCS)
>
blackdown has additional ppc/sparc/arm ports. JMF, java3d , customized java
web start, debian installer. java comm api.
Also Blackdown give the opportunity for developers to contribute to the java
on linux product. ie if you find a bug or want to change something its going
to be easier to get the
Redhat 7.0? If so either upgrade or get the rh 7.0 patches from
their website. Some large apps also need the C stack size to be reset
ulimit -s 2048
which is fixed in later glibc builds
regards
calvin
Kunal Bisla wrote:
>
> Hi,
> I'm trying to run weblogic600sp2_linux.bin in my
> Linux7.0
> I
It could be
http://developer.java.sun.com/developer/bugParade/bugs/4422213.html
which is a generic hotspot bug but was found on linux :*(
regards
calvin
Alexandre Saur wrote:
>
> Yes, I know... RedHat 7.0 is kind of old now...
> but my glibc is the latest for redhat 7.0 as far as I know...
> Hmm .. would increasing the heap size affect the stack size ?
It is very possible you will see a stack overflow if the heap is
exhausted. The options to change the stack in non-hotspot vms were -ss for c
stack -oss for Java stacks. These are for each thread and are not configurable
in the hotsp
I would recommend using JDK 1.3.1, it is way faster than 1.2.2.
Now one of two things may have occured, either you really got a stack overflow
or stack overflow was incorrectly detected. If its the first case try
giving java some more heap space to use (java -Xmx128m ). If its the second
case you
I replied to this earlier but it didn't show up to the list. (I've
resubscribed
myself now)
you can remove the messages by running xprop -remove _MOTIF_DEFAULT_BINDINGS
which was set by another motif program
regards
calvin
"Jesus M. Salvo Jr." wrote:
>
> -BEGIN PGP SIGNED MESSAGE-
>
Yes this is a new bug (4458946), its caused by the daemon randomizer thread
not
shutting down in time.
The engineer who has looked into has this workaround for now
add
Runtime.getRuntime().exit(0);
to the end of main
regards
calvin
Timothy Reaves wrote:
>
> Hello.
>
> When I
I would recommend trying 1.3.1 not only does it have the fix to the wrapper
script below (/usr/bin/cut went to /bin/cut)
But it has many hotspot fixes too. If you fix the cut stuff you will also
need to set LD_ASSUME_KERNEL=2.2.5 for 1.3.0 releases
regards
calvin
Homayoun Yousefi'zadeh wrote:
I've seen a couple instances of this, almost all are attributed to the sound
card driver. If you have sb live cards then there is an open source driver
from opensouce.creative.com (driver called emu10k).
Otherwise check with the vendors site, until ALSA is complete, opensound.com
have probably
The 1.3.1 release has the 'cut' fix, the next release of 1.3.1 (rc2)
has the final floating stack code that was needed.
Alternatively set LD_ASSUME_KERNEL=2.2.5 when running Java and it will
use the backward compatability libraries with Redhat 7,1 , this will work
for 1.3.0_002 as well,
There
I know where the first part of the message came from
current locale is not supported in X11, locale is set to C
X locale modifiers are not supported, using default
The directory
/usr/X11R6/lib/X11/locale
is missing files that it needs.
even if you are displaying remote X to another DISPLAY yo
As of release 1.3.1 all platforms which includes Linux are released on the
same day. I know that only covers x86 but I would guess the ppc version
should be in the same ballpark
The 1.3.1 release is currently in rc form and has better debug support and
is faster again than 1.3.0
regards
calvin
Read the next section in the link I posted yesterday. there is a section
on memory issues with JNI, global references etc
http://developer.java.sun.com/developer/onlineTraining/Programming/JDCBook/jniref.html
regards
calvin
Zhihong Pan wrote:
>
> Hi,
>
> When I call my c function in java, I
The last element is the initializer for the array, I wrote an example of its
use here
http://developer.java.sun.com/developer/onlineTraining/Programming/JDCBook/jnistring.html
regards
calvin
Zhihong Pan wrote:
>
> Could anybody teach me hot to use NewObjectArray in jni ?
> jarray NewObjectArra
If this is a JNI program you may get more verification that it was inside
your own code or from the VM from running with -classic .
The new 1.3.1 release will have better hotspot diagnostics than the cryptic
message below. If you do 'man ascii' you can read off the file name below,
(its in hex) t
I've seen this before on redhat 7.0 and it is logged on the redhat
bug page as a known bug, currently they had fixed it in 7.1 but have
not issued a patch to 7.0
RedHat bug Id: 23758
location: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=23758
The classic VM in 1.3 does return the expect
There are some 1.3 hotspot related jdb bugs for example the one listed
below. If you happen to come across one I would recommend using the classic
VM (jdb -classic) until those bugs are fixed
regards
calvin
Joi Ellis wrote:
>
> On Sat, 10 Feb 2001, Jesus M. Salvo Jr. wrote:
>
> > I had more su
Redhat 6.1 was used to build the 1.2.2 source, later releases of gcc
refused to even compile the assembler files, and you need to remove the
comment lines if you are using a later release. One of the challenges
with developing on linux is how fast distributions change, the tools
all changed in a y
I've raised the priority of the bug in question, I don't think we have
a test case we can reproduce yet with this bug, so if anyone has
something
then please forward it on
regards
calvin
Joseph Shraibman wrote:
>
> Vote for:
> http://developer.java.sun.com/developer/bugParade/bugs/4372197.html
I don't know if anyone else can verify this, but the bug listed below
seems to be BSD socket related (ie OS related)
I hacked some code together from an online example and included what
Java does on the server side , compile cc -o serv serv.c and
test with the Broadcaster client from the bug rep
Hi
I've been at comdex this week so I'm just catching up on email, I hope we
can resolve redhats distribution issue soon. The fonts problem on linux
will improve with the availability of truetype rendering in Xfree, but we
should also continue to work on the fonts supplied with java
regards
cal
That particular (4102292) bug has been partially fixed in linux 1.3, the
window manager decides where the frame will go for the window managers I tried
sawfish, kwm etc and it will be placed as the window manager decides and
not stuck at 0,0 or worse -x, -y. That fix will also go into the 1.3.1
We are coming to the end of a great year for Java on linux,
hotspot has finally arrived on linux as well
as the optional packages JMF and Java 3D, additional chipset
ports, plugin support for mozilla and netscape to name a just
few achievements.
In moving forward I would be very interested in
Minor corrections, you can legally
a) share source with other SCSL licensees for free, for example if
you wanted to work in a porting team or use it for your research project.
b) you can ship a binary if it passes the TCK test, there are
no royalties but the test itself is not free.
regards
c
Yes this is a horrible side effect of a fix for key press/key release
events requested by an ISVs
http://developer.java.sun.com/developer/bugParade/bugs/4371923.html
It is committed to be fixed
regards
calvin
Juergen Kreileder wrote:
>
> > "Joi" == Joi Ellis <[EMAIL PROTECTED]> writes:
>
tid is the system thread pointer, you can usually use it to cross
reference in the
monitor dump, unfortunately the hotspot stack trace isn't as full as the
classic
trace yet. The nid is the native thread id, on linux this translates to
the unix process
number.
Monitor entry means that it is tryin
I think Nigel had a different question (about JMF)
by default it will go into /usr/java, you can run rpm --query -l -p
on any rpm file to find out where it will put files.
You can install the JDK files to other places (although not advised) by
installing
as --badreloc --relocate /usr/java=/usr
Very likely a monitor contention leak bug that has been fixed in
hotspot, I think in
beta refresh. btw the more details you can give a group like blackdown
or Sun etc the
better, java -version for example :*) I've been using SMP for a while,
the only issue
I've seen on linux is that the later ve
This is a hotspot problem with SUSE, not a blackdown issue really, but there
is a workaround and it will be fixed in the next suse release
see the release notes :*)
http://developer.java.sun.com/developer/earlyAccess/j2sdk13/relnotes-linux.html
regards
calvin
>Hello,
>
>when I start java using
There used to be issues with some window managers with the Sun JDK, but
Blackdown and
ourselves have fixed most if not all the issues.
In the next release of the 1.3 Sun JDK most of the window managers we
have tried
sawfish, enlightenment, fvwm, kwm, twm behave correctly (if not log a
bug)
regar
For jdb
the command use should work for most class files I just tried
1.2.2 fcs and it
worked ok (perhaps run strace to see why it is not working)
I managed to use oldjdb with 1.3 and use worked ok there for simple
.class files too
For gdb you'll need to handle SIGUSR1 (the following works for
>This might not be a blackdown problem, but I need help now, and the IBM
>newsgroups seems down? and sun, well I have never gotten a response from
>them.
Well there are apparently 2.5 million developers out there!
Don't forget that Java 2 has a finer granularity security manager. I would
sugge
Submit it as a bug and/or send us the short example, it could very well be
another linuxthreads bug that we will have to work around or someone has
already discovered.
Java is one of the best stress tests of the linux threads library :*)
regards
calvin
>
>I've been unable to get blackdown na
There is a known bug list where this is listed, in a nutshell src.jar
is actually a gzip archive, run unzip or gunzip to extract the files
from that archive for this release.
regards
calvin
>Did anyone else get an exception when unjaring this file?
>
>
>--
Yes it is a bug with the way that the port handles locales, basically
it tries to interogate the locale data for en, which is probably incomplete
It was doing this to enable japanese support.
The workaround is to set LANG to POSIX or a full locale, also the
blackdown port does not have this bug .
Depending on what type of embedding you are doing, but you may find J2SE
works. I don't know of the release date of J2ME on linux so I've forwarded
your email to those engineers
regards
calvin
>Hi
>
>I try to embed Linux in a device and I also want to run java (including
>GUI). How many librari
One suggestion, not sure if this will work. I have created a patch
file to move from jdk1.2.1 to jdk1.2.2 on the scsl site. You may be able
to reverse patch that jdk1.2.2 release to create a jdk1.2.1 release
which should work with the blackdown diffs.
Although there are over 1000 fixes I am hopi
Louis-David I will show you the code! Hop on over to
http://developer.java.sun.com/developer/products/java2cs
you can get all the Java 2 source code for no fee. Then make a trip
to blackdown.org, you can get Java 2 releases. blackdown has the diffs
that I will be linking to make a Java 2 linux s
49 matches
Mail list logo
