Re: JAVA PROJECT! CODER WANTED!!
What, are you kidding? keith schmauss wrote: > > > > some of the project[s] [for fun!] are ==>> > 1] creating a very simple windowing manager coded in > JAVA! [mult-threading windows manager! ] 1) What are you going to use for your gui on this window manager? AWT is not thread safe, so you ultimately have to have all your events handled by one thread. Yes, you can distribute the work out to multiple threads, but the benefits of this are dubious. > > > > 2] porting the linux kernel from gnu c++ to java1.?? > [just think! write once Linux kernel run ANYWHERE!!] > [cool idea! comments please! :) ] 2) HAHAHAHA! That's really funny! > > > > 3] java and mmx + AMD k6 3d cpu >for games! > a gaming toolkit for linux writing in java or >game writing in linux java using mmx/k6 3d cpu mxx. > 3) I've done a fair bit of Java3D coding (the company I work for makes graphics libraries in Java & c++). Unless you've got good hardware acceleration (i.e. not simply MMX or K6 3D), everything gets done in software. Speed sucks. Now, if you've got good hardware acceleration, then performance is quite good. But it seems like you want to use java to reach more machines. Listen--just code it natively, and use one of the various gaming libraries out there. You'll do far better doing it this way. > > > > 4] a wsftp style gui ftp programme written in java for linux >so that someone can ftp in X gui. > 4) I think you can probably find one of these on one of the various java books out there. It seems like I've seen one or two, now that I think about it. With the new drag-and-drop stuff, these can be quite nice. However, this one is "been there, done that, what's next?" > > > > 5] a 3D login in XDM mgr in java . > 5) Why 3D? What do you want done in 3D? Why don't you just replace the gif in XDM with some cool 3D gif that you whip up in gimp? Quite frankly, I don't see the value added by a 3D login prompt, either in beauty or in functionality. > > > > if any one would like to help me with these ideas/project > please kindly email. > > If you have any cool java ideas/project share them with me/ > and post them here! > > Thanks ! > Keith Schmauss > Pob 704 > Laramie,Wyoming 82070 > > [EMAIL PROTECTED] > > MAXIMUM POWER!SPEED!3Dgraphics! = MAXIMUM-LINUX!! :) !! > Java is not nearly so slow as many of its pundits would have you believe (at least, not with TowerJ or a good JIT). However, it quite simply is *not* MAXIMUM POWER!SPEED! And quit yelling, for God's sake. At least you got the 3D graphics thing right. While Java3D is not near being complete yet, it's got a pretty darn good design. Oh, and get off the Microsoft gravy train. Stop using hotmail. Remember, this is a Linux newsgroup, and we Linux users don't much like Microsoft. Try iname if you want a decent web-based e-mail (and when you get a real e-mail address, they'll forward the mail to your new place for free, for life, whilst also filtering out spam).
Re: Compile Java to Native?
Yes, the tool is called TowerJ, from Tower Technology (http://www.twr.com/). TowerJ-compiled Java is faster than *any* JIT on *any* platform (as you might expect). Currently, this group has only ported TowerJ to HP's *nix and Linux (I just checked their homepage, and found out that they have ported it to Digital Unix for Alpha, as well!). Cygnus (I think) is also creating a front end to egcs (where new code is alpha/beta tested for gcc). This project is called gc1, I think. It is mentioned in the latest issue of Dr. Dobb's Journal. The benefit is that this project will capitalize on gcc's well-tested and robust optimization stages, as well as the diverse number of platforms under which gcc runs. My understanding of this project is that it is not yet finished, and may not be finished for a while. However, my sources for this understanding are dubious at best. Eugene Teh wrote: > Hi, > > I keep hearing people telling me that you can do this > But I have never found info about how to do this? > Is there such a utility - AFAIK Sun does not provide one? > Also what's the performance issue? How can it be tuned? > > Eugene Teh > Berkeley, CA >
Re: Compile Java to Native?
Correction to my previous post -- the Cygnus Java/gcc project is called jcl. You can read about it at: http://www.cygnus.com/news/whitepapers/evaluate.html It appears that it may be closer to completion than I was originally led to believe. -dan [EMAIL PROTECTED] [EMAIL PROTECTED] Eugene Teh wrote: > Hi, > > I keep hearing people telling me that you can do this > But I have never found info about how to do this? > Is there such a utility - AFAIK Sun does not provide one? > Also what's the performance issue? How can it be tuned? > > Eugene Teh > Berkeley, CA >
Re: Make depend for Java
Use jikes, from http://www.alphaworks.ibm.com. It also happens to be faster and more rigerous than javac. enjoy! -dan Bernd Kreimeier wrote: > The weird logic in which javac handles or ignores > dependencies (esp. source files more recent than > class files) never ceases to annoy me. Does anybody > have a decent solution to share that creates > dependencies from "import" and makes them work with > JDK? > > b. >
Re: fast multiple object creation ... can it be done!
I do believe the synchronization improvements are already shipping with JDK1.2beta4, which is available for Solaris and Win32. At any rate, synchronization suddenly got an order of magnitude faster from v1.1.6 to v1.2beta4. -dan [EMAIL PROTECTED] Paul Reavis wrote: > Steve Byrne wrote: > > > > Uncle George writes: > > > interesting, considering that 1.2 isn't out yet. U got a mole in the sun > > > organization? > > > > I used to work there until December of 97. I kept ensuring that 1.2 compiling > > cleanly with gcc, so the port to Linux wouldn't be too bad. Being on the > > inside, you get presentations from time to time about things related to the VM, > > and one of them talked about some of the improvements in 1.2. I think, though, > > the new synchronization mechanism is public knowledge, but I'm not certain. > > I've heard about it from various quarters; it's in the big bag of > HotSpot/1.2 improvements, and there are several white papers and > articles scattered around about those. Uncle George, you might dig > around on developer.javasoft.com and javaworld if you want more info. Or > just keysearch on +java +hotspot and see what crawls up. > > -- > > Paul Reavis [EMAIL PROTECTED] > Design Lead > Partner Software, Inc.http://www.partnersoft.com
Re: Write Once Run Anywhere?
Just a few points, but no real solution, I'm afraid.
Steve Cohen wrote:
>
> 2) Many wrote about the difficulties with copying files between the two
> systems. This is not relevant either, since the files actually copied are jar
> files containing the class files. This should be equivalent to zipping the
> files.
Actually, since a jar file is a binary file (i.e. it uses the 8th bit in each 8-bit
word), you still need to set the transfer mode to binary using FTP. That said,
copying the file on a disk should work.
> Now here is a detail that may be more relevant:
> There are 19 .java files in the package. The compilation under linux is making
> them into 19 class files. When compiling under Win95, 27 class files are
> created. Two of these are for non-public classes defined in other java files.
> The other six files have names like x$1.class where x.java is one of my
> project files (x.class is also created by the compilation). I don't know
> what these extra files are for, but the NoClassDefFoundError message is telling
> me that the class it cannot find is x$1. The extra files do not seem to be
> required under the linux environment, but they appear necessary under the Win95
> environment.
All of the "$1.class" files are actually just anonymous classes. For instance
if you say something like:
button.addActionListener(new ActionListener() {
public void actionPerformed(ActionEvent e) {
// handle the button press here!!!
System.exit(0);
}
}
you are actually creating an anonymous class, extended from the ActionListener
class. In order for Java to do the runtime lookup for this class, javac creates an
artificial class name for this anonymous class, as it does not have its own class
name (hence the label 'anonymous'). The name is created as 'x$#', where
'x' is the name of the class that contains the anonymous class, and '#' is the
#'th anonymous class found within the containing class.
As far as moving stuff back and forth between Win95, Linux, Solaris, and NT, I do
this all the time. I don't have any good suggestions for you, other than to try a
fresh installation of Java on your Windows95 system, and make sure all of the
examples work before you try your own code. I have had problems with Win95 & Java
when using native code, but these problems were not Java-related (native code
behaves markedly different under Win9x and WinNT, with errant code causing crashes
on Win9x, while running fine under NT).
-dan
[EMAIL PROTECTED]
[EMAIL PROTECTED]
Re: Write Once Run Anywhere? - MYSTERY SOLVED
Steve Cohen wrote: > > MYSTERY SOLVED! > Jar behaves rather oddly under Win95 and this caused me to become confused. > When I unjarred the file under Win95, I looked in the Windows Explorer for > the products. I immediately noticed under the directory where I did the > unjarring, a new path mirroring the path under linux where the files were > created, which contained a bunch of class files. I ALSO noticed some class > files in the directory where I did the unjarring, but I assumed that these > were from a previous compilation, deleted them, and copied in the ones from > the unix-like path. However, this set did NOT include any of the inner > classes, the $1, $2 etc. So I now had an incomplete set of files and the > program wouldn't run. > > When I just ran the files that jar unzipped, it all worked fine. Duh. > > I am wondering, however, why Jar makes this additional mirror of the > original directory structure and places in it an incomplete set of files. > It turns out that the wildcard expansion under Win95 does not include filenames containing '$'. Thus, if you say, "jar cvf fubar.jar *.class", you won't get any of your anonymous classes. Fixes: 1) Explicitly pass all .class files to the jar command line. 2) Try using a *real* shell. The dos shell, in a word, sucks. Try the bash shell, included in the Cygwin32 utilities, available from cygnus (http://www.cygnus.com). It behaves as expected, and makes the WinNT command line almost as comfortable to me as the Linux command line. -dan [EMAIL PROTECTED] [EMAIL PROTECTED]
Re: Kaffe
Yes, I agree completely. Additionally, Jikes also requires a more strict adherence to the Java standard than javac. The error messages are much more verbose and precise, as well, making debugging much easier. Finally, it has a command-line that allows you to output gcc-like error messages, allowing you to integrate jikes into the various Emacs versions (combine jikes with jacob & XEmacs, and you've got a pretty good IDE). -dan [EMAIL PROTECTED] [EMAIL PROTECTED] John Summerfield wrote: > > The compilers won't make much difference to runtime speed. The fastest > compiler I've seen is jikes (www.alphaworks.ibm.com) which is about an > order of magnitude faster than javac and the compiled code of the one > program I test it on is a little faster. > > jikes is now my preferred compiler. IBM has some other nice addons too > including a GUI debugger written in java. > > Cheers > John Summerfield > http://os2.ami.com.au/os2/ for OS/2 support. > Configuration, networking, combined IBM ftpsites index.
Re: jre failed
When using the '-cp' flag in jdk1.1.x, the classpath specified at runtime *replaces* the classpath specified in your environment. Make sure that you also specify the location of 'classes.zip' and 'swingall.jar' (if you are using swing, that is) in addition to 'testq.jar'. -dan [EMAIL PROTECTED] [EMAIL PROTECTED] Lukito Edi Nugroho wrote: > Hi, > > I tried to run an application packed in a JAR file: > $ jre -cp testq.jar QueueTest > but got the following message: > > Unable to initialize threads: cannot find class java/lang/Thread > Could not create Java VM > > I use jdk-1.1.6, RPM version (jdk-sbb-1.1.6-2.1.2glibc). > CLASSPATH, JAVA_HOME, and JDK_HOME are all > set appropriately (I think). > Am I missing something ? > > TIA, > Lukito
Re: jre/java troubles
Ian Wehrman wrote: > good day all, > > i'm sure this is a frequently asked question, but after looking around, > i haven't yet found anything. i have the JDK 1.1.6v4a for i386-glibc > linux, and compiling .java files with javac works fine, however whenever > i try to execute a class file with the 'java' command, i get: > > [iwehrman@r77h19 awt]$ java Main.class > Can't find class Main.class try "java Main" instead. the class name is 'Main', not 'Main.class'. -dan [EMAIL PROTECTED] [EMAIL PROTECTED]
Re: Free Java @ JavaOne 1999
Birds Of a Feather Chris Sommers wrote: > Kinf Folks, > What does "BOF" stand for? > > - chris sommers >
Re: RMI anyone?!
JNI stands for Java Native Interface. It allows a Java program to access C & C++ methods. By using established C-to-Fortran bindings, you can also interface Java to Fortran using JNI. JNI will work for your purposes. Basically, you declare all of the C functions that you want to call as 'native' methods in Java. Then, you run 'javap', which parses your Java code and creates C "stubs" for all of the native methods (a stub is just a simple function in C that Java knows how to connect to). You then fill out these "stubs" to call the actual C routines that you intend to use. There are some things that need to be pointed out concerning the C stubs. Any data declared in Java that will be used in the C programs must be copied to local C variables (JNI provides functions that retrieve data from the JVM). To store values in the JVM from C code, you must likewise copy the local C variables to Java (again, using functions provided by JNI). The reason that variables need to be copied (as opposed to being used directly) is that Sun has declared in the JNI specification that the JVM can, at its will, move variables in memory (supposedly, the forthcoming HotSpot JVM will move memory in order to reduce memory contention between threads). If a C program were accessing data in the JVM directly, the pointer referring to the data would thus be out of date if the data had been moved by the JVM. This should get you the funtionality you require. JNI also specifies ways of throwing and catching Java exceptions, as well as ways of declaring new memory and garbage collecting old memory in the JVM. However, this functionality is rarely required, and adds significantly to the complexity of JNI programs. I hope this helps! -dan [EMAIL PROTECTED] Andrew McLaughlin wrote: > Hi Diego, > > I'm a little dense in this area, can you break down the components? I'm thinking >that I'll > have to write my own software on the server side to be able to select data as well >as do > the normal I/O. While UniData does have a C adapter so that I can access the data >files > from Unix I'm not sure what else that I would have to write to complete the chain. > > Is JNI (Java Network Interface)? Just guessing. That's a new one for me. > > Am I on the right track? > > Thanks, > Andrew (learning Java for one week so far) > > Diego Pons wrote: > > > Andrew McLaughlin wrote: > > > > > > Is anyone working with RMI so far? I'm launching into a project where I'll try to > > > connect Sanfrancisco to a legacy database (UniData). While UniData does provide >ODBC > > > it's too slow and, gulp, read-only. > > > > > > I welcome any and all commentary with open arms and squinted eyes!! > > > > > > > I'm using RMI in a project integrating a legacy RPC system with Java clients. > > It works very well and I do not see any significant performance degradation > > compared to regular RPC connections. > > > > If you decide to use JNI to interface directly to your legacy API, try very > > hard to centralize the API into a single JNI call (ioctl-like). It is very > > tedious and error-prone to develop JNI interface calls. > > It took +50 lines of the ugliest C code per each argument/call. > > > > -- > > Diego Pons Pharos Consulting LLC
Re: RMI anyone?!
I concur, in general, about the cryptic errors that occur with JNI coding. I tend not to put everything within a single function call, but I do follow the KISS principle (Keep It Simple, Stupid). May I also suggest the book, "essential JNI: Java Native Interface" (sic on capitalization), by Rob Gordon. This is an excellent (no, authoritative) book on the subject of JNI. It is very well written. Following the KISS principle, you'll probably only need to read the first 2 or 3 chapters. There are other native interfaces available, as well. Netscape's JVM has their own native interface, developed before JNI was available. However, Netscape has deprecated this interface in favor of JNI. Microsoft's JVM uses something called RNI (Runtime Native Interface), and does *not* support JNI (you can, however, use Sun's Java Plug-In, which encapsulates Sun's JVM inside an Active-X control). Microsoft will likely never support JNI. This is, in fact, the source of Sun's accusations against Microsoft for not supplying a JVM that is compliant with the Java standard. Why does Microsoft do this? WFC, Microsoft's Java interface to MFC, makes extensive use of native calls (as you might expect). In order to make WFC faster than it would otherwise be, RNI allows native code to get the address of Java variables directly (which, of course, would not work if data is moved in memory, as the JNI specification allows). To support JNI, Microsoft would have to cripple (in terms of speed, anyway) the WFC, something Microsoft is not likely to do. We have built a bridge between RNI and JNI for our internal use (we provide Java toolkits for data visualization). This way, our native code and our Java code remains intact. Our bridge maps the RNI stubs to call the JNI stubs, and provides the JNI data access functions in the event that the JNI is not available. However, we do not offer this as a product (yet). If there is sufficient interest, we could probably be coaxed into developing this into an actual product. Please e-mail me if you require this functionality. I hope this helps. Native interfaces have been the source of a lot of headaches for Java developers. I repeat my earlier advice: KISS. -dan [EMAIL PROTECTED] Diego Pons wrote: > > I'll try to be less terse this time. > First, RMI is a great solution, much simpler to code than say RPC's > for developing client/server interfaces. The only complication is that > to support legacy systems not written in Java, you will need to create > a (non-portable) bridge using JNI (see Dan Kirkpatrick's response to you). > > JNI allows you to interface with the C language, having thus access to > the underlying operating system or application interfaces in C. > As D.K. says, in JNI you have to map Java classes to C structures on > the output arguments, and C to Java on the input arguments, all the while > checking for errors at each step. This is a lot easier to say than to do. > Most errors will cause the JVM to dump a (most enigmatic) core. > > Given this difficulty, I suggest to minimize the number of JNI functions. > One way is to create a mega-call in JNI where the parameters are a command > and a related argument list as strings, in a way similar to ioctl() calls. > A typical API would then be handled in this way: > > my_jni_call(..., OPEN_CMD, arg, NULL) > my_jni_call(..., READ_CMD, arg, argsz) > my_jni_call(..., WRITE_CMD, arg, argsz) > etc, etc. > > While this goes against elegancy in design, it will add to the robustness > of the product by eliminating error-prone Java<->C mapping code. > Unfortunately, this came to me only after writing about 3000 lines of > the interface code. > > I suggest reading "Core Java, Volume II", by Horstmann and Cornell for > both RMI and JNI examples and references. > > -- > Diego Pons Pharos Consulting LLC
Re: HALLOWEEN
I'm not sure the "Halloween Document" belongs on this mailing list. Furthermore, this seems to me to be one more bit of propaganda that was intentionally "leaked" to the press. Look at the timing. This was released just 2 days before Bill Gates video testimony was made public in the DoJ vs. Microsoft trial. What better way to claim that they don't have a monopoly than the release of an "internal" memo that states that Microsoft is losing the battle against Open Source Software. I'd like to see some caution within the Open Source community before we embrace this as a victory for Our Side. It seems more plausible to me that this was a PR move designed to distract the DoJ from the trial. Just my $.02 worth. -dan [EMAIL PROTECTED] (home) [EMAIL PROTECTED] (work) p.s. - and JFK was assasinated by Cuban-Americans angry over the Bay of Pigs, Elvis is living in the Seychelles, and crack was an invention of the CIA to keep the minorities oppressed ;-)
Re: Linux Takes Lead in Server-Side Java Performance /IW November 23, 1998
I'm not one to normally defend Microsoft, but their JVM is a "true" JVM. The area that they depart from the Java standard is in their choice to leave out support for JNI (Java Native Interface). The reason they chose to do this is that JNI specifies that Java objects can be moved around in memory, at the whim of the JVM. This requires that java memory accesses from native code first explicitly request the value of the variable from the JVM, rather than simply checking the value pointed to by a pointer (after all, the object may have moved after the native pointer was set). Pure Java code that works other JVMs will work with Microsoft's. Pure Java code that works on Microsoft's JVM will work on other JVMs. Why did Microsoft do this? Because they were trying to optimize the performance of WFC calls (their Java interface to MFC). MFC is, of course, native code. If WFC required memory copies for every access to Java memory, the performance of WFC would suck. So Microsoft chose not to implement JNI. The answer to this, of course, is simply to avoid WFC like the plague. Besides, WFC is a really lousy UI. However, Microsoft's JVM is actually pretty decent. In the same way that I get upset when I hear FUD from Microsoft, I also get upset when I hear FUD directed *to* Microsoft. Oh, and for the record, their security model is also different than Sun's. But, then again, so is Netscape's. This isn't too bad, though. It simply means that you need to do a little more work in JavaScript if using signed applets. -dan [EMAIL PROTECTED] (home) [EMAIL PROTECTED] (work) "Matola, Tod" wrote: > I agree it is good to see (yeah Linux), but it is also a strange comparison. > > But my point was, is Micro$oft's sdk really a TRUE JVM? Didn't they buy performace > at the cost of portablity? Don't they play funny games with the low level calls > that get close to the OS? > > I have no first hand knowledge, but I question: is a High Performance compiler > that different from a vendor that doesn't follow the standard and regularly adjust > the OS to make an application perform (have you ever notice what DLL's get loaded > in the system directory during an install). > > Just my $.02. > > Cheers Tod... > > Bryce McKinlay wrote: > > > Although this is good to see, it is hardly fair to compare a native code > > compiler (a commercial one, at that) against a true JVM (which in Microsoft's > > case was only slightly slower) > > > > regards > > > > [ bryce ] > > > > Mario Camou wrote: > > > > > Hi all, > > > > > > Check this out (I'm not including the full text in the interest of conserving > > > bandwidth): > > > > > > http://www.internetworld.com/print/current/webdev/19981123-java.html > > > > > > Way to go!
Re: Linux Takes Lead in Server-Side Java Performance /IW November 23, 1998
Jauvane Cavalcante de Oliveira wrote: > > Why did Microsoft do this? Because they were trying to optimize the performance >of WFC > > calls (their Java interface to MFC). MFC is, of course, native code. If WFC >required > > memory copies for every access to Java memory, the performance of WFC would suck. >So > > Microsoft chose not to implement JNI. > > Most likely they tried to crush Java by creating a poluted version. This way many > developers could write code to their java language, hence being once again attached >to > Windows... Well, fortunately they lost the first instance in the lawsuit with Sun >and they > will have to comply with real Java (so, there is no such thing is MS JNI >implementation > anymore, I didn't even bother memorizing its name). This way Java code will have no >need > for any specific OS (such as Windows.. Did you get why they did that?). Yes, this did factor into Microsoft's reasoning. However, when the decision was originally made, it was made for performance reasons relating to WFC. The FUD aspect was an afterthought. This, of course, placed them in an alleged violation with their license with Sun (btw, the injuction is a preliminary decision to prevent further damage until the outcome of the trial is complete--the trial is not yet over, and Sun has not yet "won"). My point is that if you write pure Java code, it will work on the Microsoft JVM. Furthermore, the Microsoft JVM is pretty darn good, from a technical standpoint. I can't agree with Microsoft for desiring to split the Java camp. However, their decision not to support JNI was as much a technical decision as a FUD decision. The FUD is more related to Microsoft's inclusion of WFC in the first place, not in their decision not to support JNI. If you think I'm a MS suporter, or whatever, you simply need look at my posts on Slashdot (my user ID is "The Mayor"). I am *not* a suporter of MS. However, I work for a company that releases Java development tools, and have come in direct contact with VJ++ developers as a result. If they have me fooled as to their intentions, then so be it. I am certainly not immune to being fooled. But, as an "insider", I can confidently say that there are technical reasons for leaving JNI support out of the JVM (besides, JNI was not part of the standard when MS' JVM was first released). > > > I also get upset when I hear FUD directed *to* Microsoft. > > You will get upset a lot yet as the antitrust case goes on... I have no personal > complaints about MS, but their way of business, kinda "If I don't like, it won't >exist" > doesn't seem to be correct. They really gotta have some agency to hold it. I *do* have personal complaints about MS (code bloat, the price of their OSes to OEMs has quadrupled in the past 3 years, they've squashed WordPerfect, Dr. Dos, OS/2, and numerous others). Their decisions concerning Java were partly driven by their desire to split the Java camp. However, unlike their behavior wrt the aforementioned products, I don't really see how their decision concerning Java was anti-competitive. It was, IMO (and I'm no the judge presiding over the case, so it's merely an opinion), a violation of their license with Sun, and as such, will likely be reversed. But it was not anti-competitive (they didn't use an existing monopoly to create a new monopoly, or to extend their existing monopoly). Microsoft will get punished for their anti-competitive behavior. However, that is a separate subject from the Java issue. And, despite all the FUD on both sides, Microsoft's JVM is pretty darn good. -dan [EMAIL PROTECTED] (home) [EMAIL PROTECTED] (work)
Re: multi JVM's mem usage in linux?
Why not convert your app to an applet? Applets are designed to have multiple applets run in a single VM. Lotus' InfoBus takes advantage of this, in fact, to allow for data sharing between separate "programs" (actually, applets). This also encourages a more modular design--when a given applet (say, the spell checker) is finished with the data, the applet can exit & free up memory required by the applet. You'd also have the implicit advantage that your code would also run inside a browser, if you choose. -dan Maksim Lin wrote: > To get away from the torrent of 1.2 questions, here is one about > jdk1.1.7: > > I have a quick question about the jvm's memory usage in linux. For > instance if I have 3 jvm's running (all same version) will there be any > sharing of memory resources between the jvms? As I understand it, the > jdk is built using shared libs so should all of the running jvms be > sharing only 1 copy of these libs in memory ? > > Why I ask is that I have tried the above scenario and have found that > having 3 jvm's running uses up almost 3x the memory of having 1 runnning > - leading me to think there is no sharing and wondering how much memory > I'm going to need if I want to have a large number of small > (co-operating) java apps running similtaneously (I know an alternative > would be to use something like echidna which will let multiple apps run > inside 1 vm, but will this is a nice solution it still has a problem > since apps in the same vm are vulnerable to each other, ie. if one of > the apps dies it can take the whole vm and all the other apps in it with > it). > > TIA, > Maksim. > -- > Remember Darwin; building a better mousetrap merely results in smarter > mice. > > -- > 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: "can't find class file java/lang/Thread"
Yes, the poster, Kenny Freeman, is correct. To run a class HelloWorld, which
contains a static method 'main()', you need to type:
java HelloWorld
As long as the directory '.' is in your classpath, and the file
HelloWorld.class is in this directory, and HelloWorld is not a part of a
package, this will work. If HelloWorld is in a package (say, package
'com.fubar'), then you need to type:
java com.fubar.HelloWorld
Think of your classpath as a sort of LD_LIBRARY_PATH, with packages being
contained in subdirectories under directories contained within your CLASSPATH
(replacing the '.' with a '/', of course).
BTW, if you ever decide to package your '.class' files into a jar file, the
jar file needs to be explicitly included in your class path. Let's assume you
package the HelloWorld.class into a jar file called 'fubar.jar', by using the
command:
jar cvf fubar.jar HelloWorld.class
Then, to run HelloWorld, you need to have fubar.jar in your classpath:
export CLASSPATH=${CLASSPATH}:./fubar.jar
Finally, you can now run the program (again, assuming that HelloWorld is not
part of a package) using:
java HelloWorld
I hope this helps!
-dan
[EMAIL PROTECTED]
Aaron Brick wrote:
> my JDK installation seems to be weird. i saw a previous post relevant to
> this problem in the archive but the responses i saw weren't helpful.
>
> when i do "java HelloWorld.class" it complains about not being able to
> find the class file - even if i pass the full path - until i do a
> "-classpath `pwd`", or its equivalent. is this normal behavior?
>
> then - this is the step i have not been able to solve - it complains about
> not being able to find a class file called "java/lang/Thread". well, i
> have tried a large number of things with the Thread.class file, including
> renaming it and putting its path in classpath, putting it in ad-hoc
> subdirs in the java/classes dir (where i'm trying all this), etc. i just
> can't get it to find the file.
>
> i mailed one member of this list and he told me to download some different
> classes.zip file - however, it didn't appear to be on the russian server
> he gave me. why the discrepancy, and what is the deal with this problem,
> anyone?
>
> i have:
>
> JDK 1.1.6v2, glibc, i386 on Debian 2.0 (2.0.34), libc6. i don't have my
> machine handy so i'm sorry i don't have the loader or X versions but i can
> certainly supply them later.
>
> i would say this problem is of fairly high severity because it prevents me
> from running any program at all! it may be that it is an isolated incident
> but i sure thought the install went fine and my system is not weirdly
> hacked or anything.
>
> thanks for your responses.
>
> aaron brick.
>
Re: jEdit/Blackdown 1.1.7a
The Sun JDK for NT comes with a JIT compiler. To get a better comparison of the two, try using the "-nojit" option under NT. I think you will find the performance to be comparable to Linux. There are several JIT compilers available for Linux, but it is my understanding that their performance still does not compare well against the Symantec JIT (included in Sun's JDK) or the Microsoft JIT (includes in IE & VisualJ++). With the growing acceptance of Linux in mainstream business, I imagine that will soon change. As for the stability, I cannot comment. I have experienced no serious problems with the Linux port. -dan Christopher Rowan wrote: > Howdy. > > I use jEdit on RH5.2/jdk 1.1.7a/glibc-2.0.7-29. > > Unfortunately, it is quite unstable and vaporizes frequently. Poof! > Just disappears :( > > So I tried something crazy for comparison purposes: I ran jEdit on > NT4.0sp3 with Sun's jdk1.1.7B. It didn't crash even once in 8 hours. > And the performance on a Pentium 200/64MB was much better (except for > startup time) than with Linux on a Pen II 300/196MB. > > I strongly prefer Linux over NT, but in this case NT makes Linux look > silly. > > Could there be another explanation for the performance and stability > difference, or is the JVM on NT just that much better? Someone please > say it isn't so! > > Feedback please! Save my trusty Pen II 300 from an NT install! > _ > Christopher Rowan [EMAIL PROTECTED] > the guy with too many [EMAIL PROTECTED] > e-mail addresses[EMAIL PROTECTED] > [EMAIL PROTECTED] > > -- > 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: Line numbers vs Compiled Code in stack traces
Try running with the "-nojit" option. Turning off the JIT allows the exception stack trace to display line numbers. Jason Proctor wrote: > Hi everyone, > > I've noticed that exception stack traces sometimes print helpful line > numbers and sometimes print "Compiled Code" instead of line numbers. The > latter is quite annoying when tracking down bugs and I was wondering what > determines whether line numbers are included or not. It seems to be random, > even in the code I'm writing myself! > > Does anyone know what determines this, and whether there's a way of always > getting line numbers? Thanks in advance. > > J > > -- > 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: Unidentified subject!
Sun's Java site has a pretty good set of tutorials and examples. Check them out at: http://java.sun.com/docs/books/tutorial/index.html For more extensive demos, check out Sun's Java Developer site. You have to register to view the page, but registration is free: http://developer.java.sun.com/developer/onlineTraining/index.html In general, these are among the best starting points for learning about almost any Java topic. -dan CURTIS David wrote: > Greetings, > > I am new to Linux and Java and would like some help. Can some one point me to some >examples and information with using Java and database? I would apprechiate this very >much. I have seen some literature on JDBC and am interested in this technology. > > -- > 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: (0.3 * 3) = 0,8999999999 ????
Actually, Java is supposed to have bitwise compatibility accross platforms (unlike
C). I know they (Sun) had severe performance problems in the early days
implementing Sun's floating point model under the Intel architecture, and that
changes were made to the VM as a result. Is this one of the consequences of these
changes? Are there ways to ensure bitwise compatibility?
-dan
Gerrit Cap wrote:
> At 10:04 12/05/99 +0200, Alessio Dragoni wrote:
> >Hi
> >
> > I try to run the code appended at the end of this message on v1.2pre1
> > on linux and in the jdk1.2 and jdk1.2.1 on Win and the result is
> > that 0.3*3 is equals to 0,899 and not simply to 0.9 !!!
> >
> > is this a bug or what?
> >
> > trying to run this one and you can see it!
> > Anyone knows some bug-fix to resolve this strange
> > error ?
> >
> > any comments are welcome!
> >
> > bye Alessio
> >
>
> Thats not a bug, thats just typical behaviour with floating point
> algorithm, try to compile and run this little C program on your linux machine:
>
> #include
> void main() {
> double d1 = 0.3, d2 = 3, d3;
>
> d3 = d1 * d2;
>
> printf("%20.20lf", d3);
> }
>
> Marble Consulting --
> Gerrit Cap http://www.ping.be/marble
> OO Solutions Engineer mailto:[EMAIL PROTECTED]
> Marble Consulting
> Blauwe Gaanweg, 53 tel : +32 75 72.94.36
> B-9150 Kruibeke-Bazel fax : +32 3 744.19.17
> Belgium
>
> --
> 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: Write a file from an applet
You've got a couple of choices.
You can use RMI or CORBA (IIOP), going outside the HTTP protocol, if the files
are big. In your case, this sounds like overkill. You can also encode the
items into POST variables using URLConnection or HttpURLConnection. To do this,
do something like:
URL url = new URL(myURL);
HttpURLConnection urlConn = (HttpURLConnection)url.openConnection();
urlConn.setDoOutput(true);
urlConn.setRequestMethod("POST");
After this, you must create your own writer for the output stream of the
URLConnection:
PrintWriter httpOutput = new PrintWriter(new
OutputStreamWriter(urlConn.getOutputStream()));
Then, for each item you wish to output (you can encode the entire file into one
item), you must write out the items like so:
httpOutput.print(URLEncoder.encode("VAR_NAME_1") + "=" +
URLEncoder.encode(var1) + "&");
httpOutput.print(URLEncoder.encode("VAR_NAME_2") + "=" +
URLEncoder.encode(var2));
Between each item, you must write out an "&". The encode() method encodes text
for the HTTP protocol. The text is automatically decoded on the other side.
All you need to do is write a CGI script or a servlet that accepts POST forms.
You can then look up the items as you have named them above (i.e. "VAR _NAME_1"
and "VAR_NAME_2").
I hope that helps.
-dan
Gustavo Medina del Rosario wrote:
> Hi:
>
> My name is Guss and I write from Spain, so sorry for my poor
> english.
>
> I am trying to write e file in a server from an applet. My applet
> ask some questions to the user (name, address, etc.) and proccess the
> information and I want it to write in the server (where the applet is
> stored) in a text file.
>
> Some people in spain emailed me that it can be made with the URL
> class but I can't.
>
> Thank you.
>
> Guss
>
> --
> 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: Java2D Servlets on Linux
Robbie Baldock wrote: > Nathan Meyers wrote: > > > No, this won't work. > > I didn't think it would! > > > Environment variables are associated with Unix processes and passed to > > child processes. You can't really set one "just for the servlet", you need > > to set it for the Unix process under which the servlet is running - > > meaning the JVM. > > But isn't this a bit crap? I mean, does this mean that all java > processes running on a Linux box (whether they be servlets or > applications) have to share the same DISPLAY settings?! As someone else noted, only the java apps running within a single instance of the JVM will be affected. When the web server that supports servlets is started, it starts a JVM for the servlets. Thus, only servlets started from that web server will run within that JVM. One solution you might use to get around any unusual servlet issues is to have your servlet use RMI to connect to separate application running under a different JVM. At my company, we have used a similar approach with servlets, albeit for a far different problem (no java2d). We use the session ID from the servlet as the RMI lookup name, thus providing persistance for the client (servlets are stateless, and die after the servlet manufactures the response). When the servlet first attempts to connect to the "server" process via RMI, the connection fails (the server has not yet been started). When this happens, we launch the server process from the servlet, passing the RMI lookup name as one of the arguments. Subsequent connections can connect to the server process using this RMI lookup name. I hope this helps... -dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
How to trace with invocation?
Kevin wrote: > Try invoking the java_g interpreter with the -tm (or was that -t) trace > method option and look for exceptions being throuwn during the class init. How do you do this with the invocation interface? I linked with -ljava_g, but I don't see anything documenting a trace method option in the jni docs. - Dan
Exception handling broken by jni?
I'm having a wee bit of trouble with Blackdown JDK for Linux
and exception handling inside code called from JNI.
It seems that try/catch blocks don't work if they are
contained in a method called directly or indirectly by JNI.
This is hitting me in the face because the jdbc driver I'm trying to use,
http://www.worldserver.com/~mmatthew/mysql/
while it does run better under Blackdown JDK than the last one I tried,
seems to rely on exception handling during normal operation
to detect the end of the query results:
catch (ArrayIndexOutOfBoundsException E) {
// Horrible hack
if (Driver.debug) {
System.out.println("* NULL Packet *");
}
return null;
}
Since try/catch seems broken, the driver doesn't return null here,
and instead my top level code gets the exception:
java.lang.ArrayIndexOutOfBoundsException: 2
at org.gjt.mm.mysql.Buffer.readLong(Buffer.java)
at org.gjt.mm.mysql.Buffer.readLength(Buffer.java)
at org.gjt.mm.mysql.MysqlIO.nextRow(MysqlIO.java)
at org.gjt.mm.mysql.MysqlIO.sqlQuery(MysqlIO.java)
at org.gjt.mm.mysql.Connection.execSQL(Connection.java)
at org.gjt.mm.mysql.Statement.executeQuery(Statement.java)
at score_65.scoreReceived(score_65.java:52)
Exception in thread "main"
Has anyone else had trouble with exception handling in code
called from JNI?
- Dan
RE: Java-Linux
http://www.blackdown.org/java-linux.html has the ports. Be sure to read http://www.blackdown.org/java-linux/docs/libraries.html and http://www.blackdown.org/java-linux/docs/faq/FAQ-java-linux.html first! - Dan > -Original Message- > From: Günter Zell [SMTP:[EMAIL PROTECTED]] > Sent: Wednesday, September 02, 1998 11:59 PM > To: [EMAIL PROTECTED] > Subject: Java-Linux > > Steve Byrne , > > I got your e-mail address from a Linux-Usergroup. > Iám looking for a JDK-Port for Linux. > Can you give me a hint > from where I can get one ? > > Best Regards, > Günter Zell, FvS School, Germany
Re: Optimal Directory Structure?
Wim Ceulemans wrote: > I personally use a little shell script which I name 'd' (for develop), and > that looks like: > > # script for setting up development environment > PATH=$PATH:/usr/local/pgsql/bin >... Come to think of it, I use that technique, too- but my little script sets up the environment absolutely, without appending to the old variables, so I can run it to fix up a corrupted environment or to switch JDK's. (This may be more important on WinNT/95/MS-DOS, though, where my day job is, because the environment is so fragile there.) - Dan
RE: and where is v4a?
It's there now; took a while to mirror. e.g. ftp://ftp.tux.org/pub/java/JDK-1.1.6/i386/glibc/v4a/ - Dan > -Original Message- > From: zembee [SMTP:[EMAIL PROTECTED]] > Sent: Monday, September 07, 1998 10:06 AM > To: [EMAIL PROTECTED] java; [EMAIL PROTECTED] > java > Subject: and where is v4a? > > I'm lost... > > With the announcement of JDK1.1.6v4a I cannot find this release. > I've been to these US ftp sites: > > ftp://ce.usu.edu/pub/mirror/linux/JDK > ftp://ftp.connectnet.com/pub/java > ftp://ftp.infomagic.com/pub/mirrors/linux/Java > ftp://ftp.tux.org/pub/java > > and to the appropriate directory (for RH 5.0): > > /JDK-1.1.6/i386/glibc > > and only find > > v1/ > v2/ > > I assume that it would be jdk1.1.6-v4a-glibc.tar.gz and that it would > be in a v4 directory. (The FAQ says the latest version is v2, but I > wouldn't expect the FAQ be updated). Did I miss something here? > > Thanks for the directions. > > > smb > > -- > Visit my home page at http://zembee.home.ml.org > [EMAIL PROTECTED] > >
RE: Questionable files
Perhaps the README in that FTP area should contain a link to the FAQ. http://www.blackdown.org/java-linux/docs/faq/FAQ-java-linux-2.html#ss2.3 should answer this question - but it doesn't seem to! FAQ author, take note... The answer is- just grab the JDK. You don't need the others. - Dan > -Original Message- > From: Jason William Jones [SMTP:[EMAIL PROTECTED]] > Sent: Monday, September 07, 1998 1:59 PM > To: [EMAIL PROTECTED] > Subject: Questionable files > > Hello, > > I am rather new to Linux Java and when downloading and installing > jdk1.1.6-v2-glibc.tar.gz I noticed the below files in the same ftp > directory: > > i18n1.1.6-v2-glibc.tar.gz > jre1.1.6-v2-glibc.tar.gz > rt1.1.6-v2-glibc.tar.gz > > Are these files included in jdk1.1.6-v2-glibc.tar.gz? > > And what are they? > > TIA > > Jason > [EMAIL PROTECTED] >
RE: Questionable files
Where is that file? I don't see it linked to from www.blackdown.org! http://www.68k.org/sysdocs/java/Java-Linux-1.1.1-HOWTO.txt is one of the only two .org hosts that has it... IMHO blackdown should put a link to the HOWTO right next to its link to the FAQ - either that, or merge the two... - Dan > -Original Message- > From: Marc Evelyn [SMTP:[EMAIL PROTECTED]] > Sent: Monday, September 07, 1998 3:54 PM > To: [EMAIL PROTECTED] > Subject: Re: Questionable files > > see the Java-Linux-1.1.1-HOWTO.txt in the /JDK/JDK-1.1.1 directory > > peace, Marc > > On Mon, 7 Sep 1998, Jason William Jones wrote: > > # Hello, > # > #I am rather new to Linux Java and when downloading and installing > # jdk1.1.6-v2-glibc.tar.gz I noticed the below files in the same ftp > directory: > # > # i18n1.1.6-v2-glibc.tar.gz > # jre1.1.6-v2-glibc.tar.gz > # rt1.1.6-v2-glibc.tar.gz > # > #Are these files included in jdk1.1.6-v2-glibc.tar.gz? > # > #And what are they? >
Native method problems, 1.1.5v7 glibc
I've got a project which I've been working on for several months which implements a Java GUI for a bunch of C++ DNA mapping algorithms. After I upgraded to RH 5.0, I had to go through hell to install a full set of libc5 development tools because at the time, there were no reliable glibc JDKs. Since then, I have tried some of the new JDK/glibc versions, but for some reason, my native methods cause the program to crash. In particular, I am running Steve Byrne's JDK 1.1.5v6 for libc5, and everything works fine as long as I link to libc5 when I compile the native methods. Yesterday, I thought I would try once again to use the latest glibc JDK, 1.1.5v7. I downloaded it, I grabbed the latest update of glibc off the RedHat site, and recompiled all of my native code, linking it to glibc. Even under this version, the program still crashes as soon as it tries to execute a native method, even using java_g (it runs fine up until that first native call, however). Just as a sanity check, I've run it again under the libc5 version, and also IBM's JDK 1.1.4 for AIX, which I have available on another machine. It works fine under both those conditions. Does anyone have any ideas on how to fix this problem, or if there is any way I can figure out what is going wrong? I can keep going with libc5 for now, but I would love to do away with the kludgy setup I need to compile with libc5 under RH 5.0. Dan
OS/2 JVM three times faster than others!?
Check out http://www.infoworld.com/cgi-bin/displayTC.pl?/980914analysis.htm It shows OS/2's latest JVM to be *three times* faster than any other. (They haven't tried Blackdown's v4a yet; they only used v2.) Perhaps the Blackdown team will be interested in getting a copy of the benchmark code they used; it looked easier to run than the Volanomark. - Dan
Re: SIGSEGV 11* segmentation violation
I had a similar problem with JDK v1.1.x under RedHat 5.0 (i.e., with glibc). It was solved by upgrading to gcc 2.8. libc5 linux systems don't seem to have this problem. Dan On Wed, 15 Jul 1998, Steffen Tacke wrote: > Hi ! > > Currently i have to develop an application that is based on a C++ Core. > For testing purpose i just called some C++-Functions that simply do some > cout stuff from within a selfwritten Java-GUI. The functions are called > succesfully but if i want proceed in the Java-Programm i ALWAYS get an: > SIGSEGV 11* segmentation violation > stackbase=0x413c5000, stackpointer=0x413c4a28 > > Full thread dump: > "Thread-5" (TID:0x4065e658, sys_thread_t:0x414fef04, state:MW) prio=5 > [...and so on...] > > The same kind of error happens also with a different C++ Library > but similar calls (it doesn't matter if the c++ calls are doing some cout > stuff ore anything else). > The C++ Calls are compiled with g++ 2.7.2 with the -shared switch. > The same type of Code (Java and C++) runs without any problems on a Sun > Solaris 2.5.1 System (also with gcc 2.7.2). > > So my conclusion is that there is a (are some) serious bug(s) in the Java > Native Interface of the JDK-Linux-Port... > Does anyone have the same kind of problem with jdk1.1.6v2 and JNI Calls > (calls of C++ Methods from Java) or does anyone have some further > information or hints for me what to do ? > > -- > mfg. > Steffen Tacke > > * > ** | *** Wilhelm-Schickardt-Insitut fuer Informatik ** ** > ** | *** Tuebingen ** ** > ** | *** Sand 13** ** > ** | ***email [EMAIL PROTECTED]** ** > ** | *** [EMAIL PROTECTED] ** ** > ** | ***Member #402 of TeamOS/2 Germany ** ** > ** | *** HOMEPAGE: http://www-ti.informatik.uni-tuebingen.de/~tacke ** ** > ***** > > > == Dan Fasulo | "That's life: trust and you're Dept. of Comp. Sci. & Eng. | betrayed; don't trust and you University of Washington | betray yourself." [EMAIL PROTECTED]| Knight of Shadows -- http://www.cs.washington.edu/homes/dfasulo ==
Re: Compilation hardware
Stefaan A Eeckels wrote: > I have a dual PII-266 (Intel Dakota, with 128Mb and SCSI disks > and I compile the Linux kernel in about 7 minutes, honest). How long does it take if you disable one CPU? What speed disks (7200RPM?) are you using? - Dan
Re: JDK problem with findClass in native code
Bernd Kreimeier wrote: > Not the invocation API, this time. I have finally given > up on that one. One last thing I tried is: > - a top level Java class, with one native method Main(), > and a loadLibrary call >... a native "init()" loads a couple > of other classes used only in native code. > > Again, JDK seems to fail me. The very same classpath > that works to load the main Java class (import/javac) > does not work when I call env->FindClass in the > native code. The same source works with Japhar > (unfortunately, Japhar has other problems). I'm about to try this kind of stuff today with v2. Where can I get later versions to test with if I run into trouble? And has anyone successfully used the invocation API with Blackdown's port yet? Please reply to [EMAIL PROTECTED] - Dan (Oddly, couldn't get any version of java to run on a RH5.0 machine yesterday. I know v2 runs on RH5.1 at home, so I'll install RH5.1 at work today.)
Re: JDK problem with findClass in native code
Bernd Kreimeier wrote: > ... The very same classpath > that works to load the main Java class (import/javac) > does not work when I call env->FindClass in the > native code. The same source works with Japhar > (unfortunately, Japhar has other problems). > > The error appears in 1.1.5 and 1.1.6v3. I rigged up > a test program, available on demand. I can load classes > that are not in packages, from the directories set > in the classpath (invocation API or bin/java). I can't > load a class that is in a package (below the root > directory), from inside a native method implementation. I tried the invocation API today; seemed to work ok. Where I ran into trouble was in loading classes. The same code seemed to work fine in Win95 with Sun JDK1.1.6, but failed identically every time with Blackdown JDK1.1.3v2 and 1.1.6v2. The code works fine if I invoke the problem code from a regular java app instead of from a C program that dynamically loads classes. Here's the sequence of calls: C main() initialize vm load init class load my classloader class call init class static method to open jdbc connection call my classloader class to load score_65 class from a jar file call score_65 class method, pass it the jdbc connection score_65 tries to retrieve record from database jdbc driver tries to load private class (it's on classpath) vm opens class file successfully vm reads bytes from class file vm reports "class not found" Here's what was printed out: java.lang.NoClassDefFoundError: twz1/jdbc/mysql/jdbcMysqlField at twz1.jdbc.mysql.jdbcMysqlResult.buildColumn(jdbcMysqlResult.java:392) at twz1.jdbc.mysql.jdbcMysqlResult.getResult(jdbcMysqlResult.java:258) at twz1.jdbc.mysql.jdbcMysqlStmt.mExecQuery(jdbcMysqlStmt.java:318) at twz1.jdbc.mysql.jdbcMysqlStmt.executeQuery(jdbcMysqlStmt.java:293) at score_65.scoreReceived(score_65.java:61) So I wonder if I'm running into class loader magic shit here? Note that the class file *is opened successfully* and bytes are read from it, yet it is reported as not really there. Could it be stronger bytecode verification is going on because this class is being loaded during the execution of a class that was loaded by my loader? The jdbc driver worked fine when I ran it from a normal app. Clearly, I should try doing my class-loading magic without involving JNI or the invocation API, to see which of the two is 'at fault'. My code is available to the maintainers if they want to see it. They probably want to wait until after I poke into this a bit more, though. - Dan
Re: Searching RPM for JDK 1.1.6 libc5
Dirk Vanhalle wrote: > Sorry if this is has been asked before. > Is there a RPM for JDK 1.1.6 (sbb), libc 5 version available ? > I tried RedHat, but there seems to be only a glibc version or a 1.1.5 RPM. The trick is to go to the redhat search *page*, then use the ftp site search for jdk. One thing it finds is ftp://ftp.redhat.com/pub/contrib/grouped/libc5/i386/Development/Languages/Java/jdk-1.1.5-8.i386.rpm I find it easy enough to just use the Blackdown tarballs, though. - Dan
RE: JNI/fileio problems (Re: JDK problem with findClass in native cod
Bernd,
thanks for the elaboration.
I have switched from using invocation interface to
calling a c_main native method from a java program per your suggestion.
I had to do this to turn on method tracing, since
I couldn't figure out how to do that with the invocation interface anyway.
My class-loading problem seems to persist even after I remove all
native code, though, so it might be in my funky class loader stuff.
- Dan
-- original message from bernd kreimeier follows ---
> [EMAIL PROTECTED] wrote:
>Bernd wrote:
>> The invocation API will get you in big trouble if you use any
>> system calls redirected by green threads.
> > I just threw together a quick JNI and fileio test program, and
> > didn't see any problems.
> > Could you give us more specific details about where you've seen
> > problems?
>
> I'll rephrase and elaborate. It is not file IO, it is *system calls*.
> E.g. an open("/dev/dsp") , read/write etc. In some USENET Java FAQ
> there is an entire list of it, I can dig it up in my backlogs at home.
>
> The invocation API as such works fine with small test programs, and
> larger apps, as long as you stay away from these calls. Even if you
> do use them, you might be fine - it all depends on the linkage order,
> the libraries used, and what calls are executed in which sequence.
>
> For example, the port of the Quake2 Game DLL to Java
> (http://www.openquake.org/jquake/q2java/) has not been ported
> to Linux successfully. The problems are:
>
> a) invocation API crashes with libdl/dlopen/dlsym calls
> (thus the original Win32 approach does not work)
> b) invocation API interferes with the main() C code that
> opens the sound device
> (thus the Solaris port that linked directly, avoiding
>the libdl use, can not be ported to Linux - the Solaris
>Q2 binary is a dedicated server)
> c) invocation API intereferes with the X11 libraries the
> Q2 binary is linked against on Linux (so even if you
> avoid the sound code, the green_threads changes to
> system calls break the invocation API with existing libs).
>
>
> In short, you should be fine as long as you start from scratch,
> avoid any libc wrapped system calls changed by green_threads
> in your app, and have full control of the linkage of all
> libs used. If you take an existing app in native code (e.g.
> of the complexity of the Q2 engine) and try to retrofit it
> with an embedded JVM invoked from that app, you might have a
> lot of problems. The same might apply for adding a Java service
> to some existing legacy server etc.
>
> This seems to be a non-issue with Win32 JDK, which does not
> use green_threads. I have no information about the Solaris
> native thread JDK. I do know that the Open Group JDK does not
> support the invocation API at all (doesn't even return proper
> error codes, because they didn't bother to catch or document
> this omission).
>
>
>
> > #define USER_CLASSPATH "." /* where Prog.class is */
>
> Changing (appending) to the Linux JDK supplied class path reportedly
> crashes on Solaris, crashed in some Linux JDKs for me, too. The
> default classpath returned by the invocation is completely bogus
> in Blackdown JDK, but seems to be different in OG JDK and Solaris.
> I am not aware of any specification.
>
>
> > (*jvm)->DestroyJavaVM(jvm);
>
>
> There is a documented and reported bug in all JDK 1.1.x, that,
> according to the Bug Parade evaluation, will not be fixed in 1.2.
> DestroyJavaVM always fails and returns an error code. As the
> SMI JNI specs do not specify any issues regarding multiple
> JVM use within a single process (serial or parallel), you might
> not want to destroy a JVM. You can't invoke a second one anyways,
> the Create call fails. Might be relevant for a server that is
> supposed to restart.
>
>
> Final remark: you might be better off to avoid the invocation
> API entirely. Write a top level Java control code and main(),
> and put your legacy or augmentation native code into some classes,
> using JNI in there. It seems that Sun does not put much effort
> in the invocation API, but JNI gets some more efforts (as long
> as javac and java run, nobody cares about invocation outside
> of what these do. JNI is required by all the dirty hacks in
> the core classes, so to say).
>
>
> b.
>
>
>
Re: Benchmark
Per Widerlund wrote: > Where can I find a widely used benchmark program? > > There are currently quite many JVM:s available for > Linux, and it would be nice to be able to compare > them. Search javaworld - they have some good stuff on the Volanomark: http://www.javaworld.com/javaworld/jw-00-1998/jw-00-volanomark.html http://www.javaworld.com/javaworld/jw-12-1997/jw-12-volanomark.html This is a good benchmark if you're doing networking, I think. - Dan
Re: ORB uder linux?
Paul V. Drobnich wrote: > Anybody knows anything about free CORBA2 ORB uder linux, See http://www.gnome.org/links.shtml for links to a few. - Dan
Re: Optimal Directory Structure?
Steve Cohen wrote: > I realize that there is not one answer to this question but I would like > some of you veteran java/linux developers to suggest what the optimal > directory structure and setup of the CLASSPATH variable is. >... > I know it isn't rocket science, but it is annoying. What I am looking > for is a strategy. I want to be able to set it one way and forget about > it, at least until the next project. For me, the best way to deal with it is explicitly. I use Sun's JDK, and everything I do looks kinda like this: every compile or execution is driven by a shell script (or batch file) that starts off with something like MYJDK=/usr/local/jdk1.1.6 MYCLASSPATH=.:$MYJDK/lib/classes.zip and then java -classpath $MYCLASSPATH ... In other words, the classpath is completely specified each time I run java or javac. No more mysteries. - Dan
Re: [Fwd: [Fwd: Distribution of runtime portions]]
Rohit Kaila wrote: > I had requested the following information some time back. > I will be most obliged if you can give me an idea about > the following : > > > > The SUN documentation says that while JRE can be distributed > > > freely with the Java Applications/Applets, the JDK cannot be > > > distributed. I wanted to distribute the runtime features of > > > the JDK ported to linux with my Java application, can I do so > > > freely ?? You *can* distribute the JRE. That's exactly what you need. - Dan
Re: Where is 1.1.6v4 published?
Chuck Huber wrote: > I've looked at serveral mirror sites across North America, eastern > Europe, and Japan. The best all of them have is 1.1.6v2. Your page at > http://www.blackdown.org/java-linux/ports.html shows that v4 is > available at mirror sites. Oddly enough, the only one I've found is ftp://ftp.tux.org/pub/java/JDK-1.1.6/i386/glibc/v4a/ tux.org is a bit slow at the moment- maybe everyone's downloading the jdk from there... - Dan
Java benchmark article
Hi Mr. Young, I enjoyed reading your article, http://www.infoworld.com/cgi-bin/displayTC.pl?/980914analysis.htm I really liked your technique of factoring out everything but the Java VM in question. Good job! I understand that Blackdown's JDK1.1.6v4a has removed a bottleneck that impacted its Volanomark scores. I'm cc'ing the Java-Linux list so they can correct me if I'm wrong. You might want to repeat the test with that version in your copious spare time :-) The raw version is at ftp://ftp.tux.org/pub/java/JDK-1.1.6/i386/glibc/v4a/jdk1.1.6v4a-i386-glibc.tar.gz The RPM version can be installed with rpm -Uvh ftp://ftp.redhat.com/pub/contrib/hurricane/i386/jdk-sbb-1.1.6-4a.glibc.i386.rpm or so I'm told. Thanks, Dan
Re: OS/2 JVM three times faster than others!?
Dan Kegel wrote: > > Check out > http://www.infoworld.com/cgi-bin/displayTC.pl?/980914analysis.htm > > It shows OS/2's latest JVM to be *three times* faster than > any other. (They haven't tried Blackdown's v4a yet; they only > used v2.) I should mention - the test was run on dual-processor machines (except for Linux, which they couldn't get running in SMP mode). So perhaps OS/2's JVM is only 1.5x times faster on a single processor, and has much better native threading than the others. - Dan
Re: RXTX, SerialImp.c
[EMAIL PROTECTED] wrote: > The major thing I have done with ioctls() is to do a nonblocking read > FIONREAD which also returns the number of bytes available for reading. I > could not find a similar function using termios. I can't think of a way > to avoid this. (I would actually like to go even lower and use > interrupts for notification as the polling is wasteful of resources.) Would select() be applicable? (This is the greatest functions for doing I/O efficiently; you can sleep the current thread until any of a number of file descriptors has i/o ready for you. Java suffers greatly from its absence. It's the key to implementing truly monsterously high performance web servers and the like, where the number of clients far exceeds the number of available threads.) - Dan
Re: Java interface to MySQL
Corey Shelton wrote: > I was told that there's a JAVA web interface for MySQL that works much > better than the one currently in Perl. Could someone point me in the right > direction? Here's my setup (from a post in comp.databases): I needed to do some web database stuff on Linux. I got up and running fairly quickly, so I thought I'd post a note about exactly what packages I installed. The packages I picked were MySQL, Apache, PHP3, JDK1.1.6, JDBC, and Red Hat Linux 5.1. In particular, the RPM's ftp://ftp.redhat.com/pub/contrib/hurricane/i386/apache_php3-1.3b6-1.i386.rpm ftp://ftp.redhat.com/pub/contrib/hurricane/i386/libgd-1.3-2.i386.rpm http://mysql.he.net/Downloads/MySQL-3.21/RPMS/i386/mysql-client-3.21.32a-rh50.1.i386.rpm (You may need to retrieve the RPM's with plain old FTP, not with a web browser. Web browsers tend to get confused by the extension.) plus Java support (these aren't RPM's) ftp://ftp.tux.org/pub/java/JDK-1.1.6/i386/glibc/v4a/jdk1.1.6v4a-i386-glibc.tar.gz http://www.worldserver.com/~mmatthew/mysql/dist/mm.mysql.jdbc-1.1.zip Here's example PHP2 code (pretty easy to convert to PHP3; PHP3 code available several places I can't recall just now): http://www.webtechniques.com/features/1998/01/note/note.shtml All these things seem to be playing together nicely for me. For more info on PHP3, see http://www.php.net/ For more info on MySQL, see http://mysql.he.net/ For more info on Java under Linux, see http://www.blackdown.org/java-linux.html For more info on a jdbc driver for mysql, see http://www.worldserver.com/~mmatthew/mysql/ Be sure to read the MySQL documentation and try out a few examples. The FAQ in the doc really does get you through most common questions. - Dan
Re: urgent help needed please.......Rý??²?ü¢??*sü^£?³ý¬???ý´?¶¹¥
satya seshu kumar dammu wrote: > %gcc -g -shared -lc_nonshared -I/usr/include > -I/usr/share/jdk1.1.6/include > -I/usr/share/jdk1.1.6/include/genunix -L/lib -L/usr/lib > -L/usr/share/jdk1.1.6/lib > -L/usr/share/jdk1.1.6/lib/i686/green_threads hello.c > -o libhello.so > > why is the libhello.so still having "lstat" as undefined (i mean U) > even though i am providing "gcc" the -l/c_nonshared option Why are you using -lc_nonshared? I'm not familliar with that... lstat is surely defined in the standard c shared library... - Dan
Re: Using Native methods from an applet
Stephen Martin wrote: > > I'm trying to access a native method from an applet under netscape > 4.06 (and 4.5b2) and I have managed to get the thing to almost > work. It finds and loads that shared library, it invokes the > native method which returns a String, but if I use the String > in any context such as trying to print it, it blows up the > broweser. Native methods that return ints work fine. > > Does anyone have any idea what this might be?? No idea... perhaps you could post the code that generates the return value? In the meantime, here's a link that goes thru the native method for Netscape stuff, for those who haven't already seen it: http://www.javaworld.com/jw-10-1998/jw-10-apptowin32.html?092198txt - Dan
Re: JDK 1.1.6 Version 4a and VolanoMark 2.0
John Neffenger wrote: > Linux JDK 1.1.6 v4a 317, 321, 319 320 (+ 37%) Thanks John for updating us with the new results! Great, we're now neck and neck with that speed demon, MRJ 2.0 :-) Does the the Volano benchmark run with the TYA JIT (http://www.tya.home.ml.org)? Or for that matter with Kaffe (http://www.kaffe.org)? Haven't seen any Volano results posted for those VM's... - Dan
Re: Netscape 4.05 support Java 1.1??
Chi-Ming Yang wrote: >I got netscape 4.05 that came with RedHat 5.1. >Does it support Java 1.1? ... No. You need netscape 4.5pr2, I think. >I have JDK1.1.6 installed. Will that help netscape? No. Netscape 4.x doesn't have a good way to use an external JRE (unless you count the Java Plugin). - Dan
Re: Install Problems!!
Lonnie Cumberland wrote: > I'm new to Java for Linux, and am having some problems getting the > JDK1.1.6v4a installed on my Linux Redhat 5.1 system. It seams that java > wants me so set some library path variable, but I do not know which one. > > I downloaded the jdk1.1.6v4a for glibc and untarred it just like the > instructions said. I then set the PATH and CLASSPATH variable just like > I was told to do. After that I just tried to run the appletviewer with > a demo, but it says that "no library path set". This also occurs when I > just try to use javac and .java. Don't set CLASSPATH. Set PATH to include just the bin directory. e.g. I unzip the file to /usr/local/jdk116a, make a symbolic link with cd /usr/local; ln -s jdk116a jdk and put /usr/local/jdk/bin in my PATH. (That way I can switch between JDK's by changing a single symlink.) It may help to uninstall older JRE's on your machine, too. - Dan
Re: Netscape 4.05 support Java 1.1??
Chi-Ming Yang wrote: > Is there a RPM for the latest Netscape? (4.06?) The latest is 4.07, I think. You're better off downloading it from Netscape directly; it's pretty easy to install even without the RPM. Just install it to /usr/local/nn4.07/ so you can uninstall it easily... > Why did Red Hat 5.1 give me an old Netscape? Because RH5.1 was put together many months ago, and they only use things that have been stable for a while. > It seems that Netscape is not taking Java seriously. Too bad. I can't really complain. Mozilla can now run using Japhar as its JVM - it looks like Netscape is pretty serious about providing a choice of JVM's in NN5.0. - Dan
Re: Java app without X installed
Steve Byrne wrote: > John Summerfield writes: > > > Aw, come on, man! RTFM! I covered this in great detail; you should take the > > Well I for one don't know which frigging document you're talking about. > I guess README.linux *could* be named more obviously :-) >... > > I can see no mention of this in any of the documentation included with the > > package jre1.1.6-v4a-i386-libc5.tar.gz. > > Look again in README.linux. The doc could be made a little more accessible. Did you know there's no link to README.linux on http://www.blackdown.org/java-linux.html ? It ought to be right there next to the FAQ. It should also be linked to from inside the FAQ, e.g. under a question "What documentation is there for the Linux port of the JDK?"; I didn't see it when I checked briefly just now. - Dan
Re: Java app without X installed
Steve Byrne wrote: > > Dan Kegel writes: > > > Look again in README.linux. > > > > The doc could be made a little more accessible. Did you know there's > > no link to README.linux on http://www.blackdown.org/java-linux.html ? > > It ought to be right there next to the FAQ. > > It should also be linked to from inside the FAQ, e.g. under > > a question "What documentation is there for the Linux port of the JDK?"; > > I didn't see it when I checked briefly just now. > > But, Dan, it *is there: look in the ports file. It's *there* as the release > notes. Granted it doesn't say "Here lies README.linux", but it *is* > there and in a pretty accessible place, though it should also be right next to > the FAQ (as well as being covered in the FAQ). Aha. OK, guess I fell into the "expected the README to be right next to the file I was downloading" trap. Or the "can't read the screen" trap. Now that I think about it, the FTP sites suffer from the same problem: the README is there, but you have to read lots of stuff (a whole page!-) and look around for it. I suspect that people who know they need to download a file get into a mode where all they notice on a page are the files for download; likewise, people looking for a particular link only notice links. That's why it's important to put the README link right next to the things these users are looking for. My final suggestion (I promise) is: * link to the README directory from the FAQ and from the top web page right next to the FAQ link * include a symbolic link to the README right next to the actual download files in the FTP area; don't make people navigate over to the doc directory and last but least, * use a .txt extension on the README for the benefit of web browsers This should make it easier for caffeine-crazed, MTV-paced people to actually accidentally notice the README as they download the JDK or JRE. - Dan
Re: Java Awt/Swing is fatally flawed
Dan Kegel wrote: > It seems to me that the Swing architects came to > a conclusion about how to write thread-safe programs: > thread safety is difficult. The best way to achieve > it is to carefully segment programs such that the > interfaces between threads are particularly well understood. A page that goes into more detail on this is http://www.javaworld.com/javaworld/jw-10-1998/jw-10-assure.html - Dan
Re: Swing and Threads
Dave Postill wrote:
> On 7 Oct 1998 16:40:34 GMT, [EMAIL PROTECTED] (Mohd) wrote:
> | I have a swing application which needs to start two threads. The threads use
> | Runnable classes. When I start the threads from inside my Swing application,
> | The application hangs completely. I know the problem is not with threads as
> | they work fine from an AWT based application. Anybody knows whats goin on?
> |
> | Can I have a multi-threaded application with Swing? If yes how do I go about
> | it?
>
> <http://java.sun.com/products/jfc/tsc/swingdoc-archive/threads.html>
> <http://java.sun.com/products/jfc/tsc/swingdoc-archive/swing_worker.html>
A similar question was asked on java-linux yesterday. I'm redirecting
that
discussion here:
Michael Emmel wrote:
> Dan Kegel wrote:
> > It seems to me that the Swing architects came to
> > a conclusion about how to write thread-safe programs: ...
> > they expect you to segment your program so that only
> > one thread manages the user interface, and communicates
> > with the rest of your program through a carefully
> > designed thread-safe interface (e.g. a queue or two). ...
A page that goes into more detail on this in general, not just for
Swing, is
http://www.javaworld.com/javaworld/jw-10-1998/jw-10-assure.html
> Well that illustrates my point. A developer should not have to do that.
> The Swing team should provide the queue interface it should not have to be
> cobbled together by every developer that wants to write a complex graphics
> program.
> I never had any problems with Display Postscript under Nextstep. The
> interface was very easy to use and multi threaded drawing was much
> easier to implement.
> Just fill a message buffer and send it. The locking is at the message level.
I think Display Postscript is analogous to AWT's Graphics, not to Swing.
Even if Swing was based on Display Postscript rather than awt's
Graphics, the comments about thread safety would still apply; you
really want just one thread manipulating the Swing widgets. I don't
think there's an easy way around it, unless you want all Swing calls
to be queued and nonblocking, and to return results to you via a queue,
which wouldn't be very intuitive for most programmers.
> Plus translation and current paint mode should be exposed
> and thus reversible. A simple reset to default for graphics
> contexts would be quit useful. ...
> The SwingGraphics state does not match the underlying native graphics state
> so I finally gave up and decided to spend my time doing something
> more useful like bitching about Java's drawing model. ...
> Also the decision to not make SwingGraphics public should be reviewed.
> Drawing should not be this hard IMHO. ...
> For me I dont trigger a bug until sevberal object try to share a Native
> context.
> public void paint( Graphics g ) {
> super.paint(g);
> g.setPaintMode(); // makes the context un reusable.
> g.dispose();
> }
> This code should have no effect at all except to cause the graphics to really
> dispose; Now maybe I don't know what I'm doing but it seems to blow the
> Swing up when it gets several objects sharing a Native graphics context.
I don't know enough about Java graphics to deal with this question, but
it sounds like you have discovered that Java only has one native
graphics
context that all AWT Graphics objects share, and that you are unhappy
with
this. Can anyone knowledgable comment on this?
- Dan
Re: Java Awt/Swing is fatally flawed
> Michael Emmel wrote: > > > <http://java.sun.com/products/jfc/tsc/swingdoc-archive/threads.html> > > > <http://java.sun.com/products/jfc/tsc/swingdoc-archive/swing_worker.html> > > Thanks I read these once befor and went searching for them but could not find the > key words. ... > invokelater clears up the problems. I neede to do this becuase > of possible async calls from the awt. > I cannot control this. > It also means I dont share a graphics context between the Awt and Swing so I can > no longer get my graphics bug using my current example. > > Thanks for your help. So it looks like reading the doc at the Swing Connection regarding threads helped Michael a bunch. End of thread (I hope!). - Dan
Re: Java Awt/Swing is fatally flawed
Michael Emmel wrote: > Hello I've been working for a while on and all java Windowing system. > My current version is written on top of the Swing toolkit. Which is > practically all Java. > Well it turns out even Swing is flawed. > > The problem with Swing is the attempt to reuse Graphics context and not > being thread safe. ... Could we pause for a moment here, and have you explain the problem in greater detail? In particular, what is the problem with not being thread-safe? It seems to me that the Swing architects came to a conclusion about how to write thread-safe programs: thread safety is difficult. The best way to achieve it is to carefully segment programs such that the interfaces between threads are particularly well understood. For instance, rather than allowing lots of threads to call the many methods of a Graphics or other complex object, and attempt to have all these methods lock all the resources they reference in a way that avoids deadlock, only allow simple objects like queues to be used by multiple threads. That's why Swing is not thread-safe: they expect you to segment your program so that only one thread manages the user interface, and communicates with the rest of your program through a carefully designed thread-safe interface (e.g. a queue or two). I happen to have helped fix a bunch of deadlocks in a complicated Java user interface by doing just that, so I trust their judgement. You go on to write: > The NT thread model is flawed it allows race conditions. I thought NT implemented threads fairly well- and that's why Java code that runs on other Java ports sometimes has race conditions on NT Java, because those other platforms don't do preemptive multithreading, and therefore hide race conditions. This is an interesting discussion, but it probably deserves to be split up into a few pieces- the Swing discussion should probably move to comp.language.java.gui. - Dan
Re: Java Awt/Swing is fatally flawed
Michael Emmel wrote:
> > It seems to me that the Swing architects came to
> > a conclusion about how to write thread-safe programs: ...
> > they expect you to segment your program so that only
> > one thread manages the user interface, and communicates
> > with the rest of your program through a carefully
> > designed thread-safe interface (e.g. a queue or two). ...
> > A page that goes into more detail on this is
> > http://www.javaworld.com/javaworld/jw-10-1998/jw-10-assure.html
> Well that illustrates my point. A developer should not have to do that.
> The Swing team should provide the queue interface it should not have to be
> cobbled together by every developer that wants to write a complex graphics
> program.
> I never had any problems with Display Postscript under Nextstep. The
> interface was very easy to use and multi threaded drawing was much
> easier to implement.
> Just fill a message buffer and send it. The locking is at the message level.
I think Display Postscript is analogous to AWT's Graphics, not to Swing.
Even if Swing was based on Display Postscript rather than awt's
Graphics,
the comments about thread safety would still apply; you really want
just one thread manipulating the Swing widgets. I don't think there's
an easy way around it, unless you want all Swing calls to be queued and
nonblocking, and to return results to you via a queue, which wouldn't
be very intuitive for most programmers.
> Plus translation and current paint mode should be exposed
> and thus reversible. A simple reset to default for graphics
> contexts would be quit useful. ...
> The SwingGraphics state does not match the underlying native graphics state
> so I finally gave up and decided to spend my time doing something
> more useful like bitching about Java's drawing model. ...
> Also the decision to not make SwingGraphics public should be reviewed.
> Drawing should not be this hard IMHO. ...
> For me I dont trigger a bug until sevberal object try to share a Native
> context.
> public void paint( Graphics g ) {
> super.paint(g);
> g.setPaintMode(); // makes the context un reusable.
> g.dispose();
> }
> This code should have no effect at all except to cause the graphics to really
> dispose; Now maybe I don't know what I'm doing but it seems to blow the
> Swing up when it gets several objects sharing a Native graphics context.
This is interesting, but out of the scope of java-linux. I am reposting
this
to Usenet's comp.lang.java.gui. Please redirect all discussion on this
thread there.
- Dan
Re: Threads / CLASSPATH?
Michael Sinz wrote: > Hmmm... This tends to happen when the Java classes.zip file is not found > by the JVM. > ... > I personally think that no user should every point at classes.zip and > that the system should automatically make sure it is at the end of > the classpath so that things work. Maybe a good compromise would be for the JVM to print out a special, helpful, error message right at startup if java.lang.Thread can't be found? We might want to assume that it should be stored in a file classes.zip, and if classes.zip is not on the classpath, append it and retry the operation. If that fails, it could print out the following message showing where it tried to load classes.zip from: Error: class java.lang.Thread not found. Classes.zip not in classpath. Cannot find file /usr/java/lib/classes.zip. You may need to reinstall Java. (OK, OK, I should just add the code myself... but I'm too lame right now...) - Dan
Re: HP wrong, Linux right... right?
Charles Forsythe wrote: > I think they have a bug involving blocking I/O (which was a problem with > Linux JDK at one point). Is this really a bug, or is this acceptable > behaviour. Someone please tell me it's a bug. They want to charge me > $250/hr if they investigate it and it's not a bug. Does HP implement native threads? If so, you may have wildly different results running with the Blackdown JDK, which uses green threads. Try running it on Windows NT using either Sun or MS Java; I think they support native threads. If it fails there, too, it may be your code. - Dan
Re: JMEDIAFRAMEWORK
Patrick Lenders wrote: > is there any plan to port java media framework to linux? I haven't heard of any. It'd probably be a fresh implementation rather than a port, given the nature of JMF. There have been other implementations- check out Whipped Butter for MacOS - so it's not impossible. Hardest part is, Linux lacks much codec support. You might be able to leverage xanim's codebase, though. - Dan
Linux Red Hat ethernet driver
Hi- Sorry if this is the wrong list, but... I'm installing Red Hat Linux 5.2 on a Compaq. The ethernet driver, tlan.o, for this machine is copied to the file system. But Red Hat's install program doesn't list it amongst its drivers from which to select. I have the file, but don't know how to "install" it manually. Its name is tlan.o and is in the directory with all the other ethernet drivers. This is probably really easy, but I'm stumped, Dan
Trouble with '<' in awt text components
Blackdown-Linux developers- I'm having trouble figuring out which layer of code (JDK, Linux, XFree86, etc) is responsible for some anomalous behavior. I'm having trouble with java.awt.TextFields (and TextAreas) that don't capture a less-than character typed in by the user (it appears as a greater-than character). AWT KeyEvents do, however, see the less-than character properly. Test code and ldconfig -D output are attached. I've scoured Jitterbug and the mail list archives and not found any help. I'm not seeing the symptoms which are supposedly fixed by LD_PRELOAD=libBrokenLocale.so, but I have tried this, and I just get a seg fault. I would have filed a new bug but I'm still not sure this is actually a jdk bug. Any help you could provide would be greatly appreciated. Thanks. -Dan [EMAIL PROTECTED] jdk: 1.1.6v5 (also reproduces on 1.1.7v1a) arch: libc5-x86 linux: slack 3.5.0 ldconfig: version 1.9.9 /usr/local/lib: libxbase.so.1 => libxbase.so.1.0.7 libcscore.so => libcscore.so.3.49.0d libcsound.so => libcsound.so.3.49.0d libtya.so => libtya.so libpng.so.1 => libpng.so.1.0.89c libjpeg.so.6a => libjpeg.so.6a.0.0 libz.so.1.0.4 => libz.so.1.0.4 libgtk.so.1 => libgtk.so.1.0.0 libgdk.so.1 => libgdk.so.1.0.0 libglib.so.1 => libglib.so.1.0.0 /usr/X11R6/lib: libgtk.so.1 => libgtk.so.1.0.2 libglib.so.1 => libglib.so.1.0.2 libgdk.so.1 => libgdk.so.1.0.2 libgimpui.so.1 => libgimpui.so.1.0.0 libgimp.so.1 => libgimp.so.1.0.0 libgck.so.1 => libgck.so.1.0.0 libXpm.so.4 => libXpm.so.4.11 libXaw3d.so.6 => libXaw3d.so.6.1 libXtst.so.6 => libXtst.so.6.1 libXt.so.6 => libXt.so.6.0 libXp.so.6 => libXp.so.6.2 libXmu.so.6 => libXmu.so.6.0 libXi.so.6 => libXi.so.6.0 libXext.so.6 => libXext.so.6.3 libXaw.so.6 => libXaw.so.6.1 libXIE.so.6 => libXIE.so.6.0 libX11.so.6 => libX11.so.6.1 libSM.so.6 => libSM.so.6.0 libPEX5.so.6 => libPEX5.so.6.0 libICE.so.6 => libICE.so.6.3 libXm.so.1 => libXm.so.1.2.0 libMrm.so.1 => libMrm.so.1.2.0 /usr/ix86-linux/lib: /usr/ix86-linuxaout/lib: libPEX5.so.6 => libPEX5.so.6.0 libXpm.so.4 => libXpm.so.4.3 libXt.so.6 => libXt.so.6.0 libXaw.so.6 => libXaw.so.6.0 libXIE.so.6 => libXIE.so.6.0 libX11.so.6 => libX11.so.6.0 libXt.so.3 => libXt.so.3.1.0 libXaw.so.3 => libXaw.so.3.1.0 libX11.so.3 => libX11.so.3.1.0 libdb.so.1 => libdb.so.1.85.1 libvga.so.1 => libvga.so.1.2.9 /usr/openwin/lib: libxview.so.3 => libxview.so.3.2.2 libolgx.so.3 => libolgx.so.3.2.2 libsspkg.so.1 => libsspkg.so.1.0.0 /usr/i586-pc-linux-gnulibc2/lib: ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libc-2.0.7.so has inconsistent soname (libc.so.6) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libdb-2.0.7.so has inconsistent soname (libdb.so.2) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libm-2.0.7.so has inconsistent soname (libm.so.6) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libBrokenLocale-2.0.7.so has inconsistent soname (libBrokenLocale.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libutil-2.0.7.so has inconsistent soname (libutil.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libcrypt-2.0.7.so has inconsistent soname (libcrypt.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnss_files-2.0.7.so has inconsistent soname (libnss_files.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnss_db-2.0.7.so has inconsistent soname (libnss_db.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libresolv-2.0.7.so has inconsistent soname (libresolv.so.2) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnss_dns-2.0.7.so has inconsistent soname (libnss_dns.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnss_nis-2.0.7.so has inconsistent soname (libnss_nis.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnsl-2.0.7.so has inconsistent soname (libnsl.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libnss_compat-2.0.7.so has inconsistent soname (libnss_compat.so.1) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libpthread-0.7.so has inconsistent soname (libpthread.so.0) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/libdl-2.0.7.so has inconsistent soname (libdl.so.2) ldconfig: warning: /usr/i586-pc-linux-gnulibc2/lib/ld-2.0.7.so has inconsistent soname (ld-linux.so.2) ld-linux.so.2 => ld-2.0.7.so libdl.so.2 => libdl-2.0.7.so libpthread.so.0 => libpthread-0.7.so libnss_compat.so.1 => libnss_compat-2.0.7.so libnsl.so.1
Where did the Java Activator (plug-in) go?
Hi, I'd like to install the Java activator for Netscape but it has disappeared from your site with no explanation. There are dozens of current links to www.blackdown.org/activator ... from various sites (including java.sun). It would sure be a big help if you would put something at that link. It occurred to me that the Linux Java Activator may have been an unsupported binary that was withdrawn as Netscape and Java moved on. If that's the case, a few words to that effect would be greatly appreciated! Daniel Dunay [EMAIL PROTECTED]
Helping out with JDK1.2?
Judging from the web-page you've had a lot of people bugging you about JDK1.2. I was wondering if there's anything I can do to speed the process of delivery/porting up (other than stop emailing the porting team! :) ? Cheers, Dan. -- +--------+ | Dan Creswell Email: [EMAIL PROTECTED] | | Senior Analyst Programmer Tel: 01483 447828 | | M.A.C. UK Ltd. | | PGP Finger Print: 5B 86 BC 5C 21 4D 04 4E 2D 05 21 F0 B0 AE 7E 9B | ++
Sun's Java Web Server
Hi- I'm using Sun's Java Web Server with Red Hat 5.2 and the Backdown 1.1.7A jdk. Works great! Now, I want to move the web server to port 80 and I need to recompile some "native" libraries that come with the web server. (Sun's native libraries seems to have some Solaris dependencies, as one might expect.) If anyone has the modifications needed, that would really be appreciated! Thanks, Dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Appletviewer under 1.1.5(glibc version)
Hello,
I installed the jdk1.1.5 on my linux machine. I do have one
problem with the appletviewer. For some reason when the appletviewer
will not dispaly strings to the screen after it is first initialized.
However, after the applet is resized it works without error. This
also includes the status string that states "applet initialized" or
"applet started".
I know that when you resize the applet it has to redraw the
applet, so something the appletviewer uses initially is not working.
Any ideas? Please e-mail me if you do, I am not that familiar with
the way Xwindows works, so I can't make a valid guess as to why this
error occurs.
Thank You,
Dan Hughes
P.S. Has anyone else had this problem?
_
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com
2.0.37ac9?
Just wondering - the status page says the multicast sockets issues are fixed in 2.0.37ac9 - the latest patch I can see over @ Alan's site is ac8 - is there a later patch or is it a typo? Cheers, Dan. -- +----+ | Dan Creswell Email: [EMAIL PROTECTED] | | Senior Analyst Programmer Tel: 01483 447828 | | M.A.C. UK Ltd. | | PGP Finger Print: 5B 86 BC 5C 21 4D 04 4E 2D 05 21 F0 B0 AE 7E 9B | ++ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Another Question concerning the 1.2 port
In a similar vein, the status page also mentions that one requires the 2.0.37ac9 path - last patch I saw from Alan was ac8 which was announced and available for ftp - is there an ac9 or has the status page got a typo? Cheers, Dan. Nathan Meyers wrote: > > It means that it's now finally *in* the beta stages. You can download a > prerelease from the Blackdown site. > > Nathan > > Samuel A Mullen wrote: > > > > Blackdowners, > > > > I've been checking out the site on a pretty regular basis to see if 1.2 is ready > > for the big time and I just saw that the row for i386 is all green. Does that > > mean that it's out of the beta stages and ready??? > > > > If it's not ready yet, is there an expected date? > > > > Thanks, > > Samuel Mullen > > > > -- > > 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] -- ++ | Dan Creswell Email: [EMAIL PROTECTED] | | Senior Analyst Programmer Tel: 01483 447828 | | M.A.C. UK Ltd. | | PGP Finger Print: 5B 86 BC 5C 21 4D 04 4E 2D 05 21 F0 B0 AE 7E 9B | ++ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Running AOLim with JDK1.1.6 on Linux!
Hi, When I was still using Win95, I'd installed the Java version of AOL's Instant Messenger. I have my old Win95 disks mounted under Red Hat Linux 5.1. Tonight I tried running AOLim under Linux with Blackdown's JDK 1.1.6-2. All I had to do was: %cd '/mnt/c/Program Files/AOLim' %jre -cp AIM.jar main.Oscar Running AOL Instant Messenger (TM) version 1.1.19 for Java built 03/24/98 ... It worked! I used it today for real. A few wierdnesses: The Buddy List tree control is messed up visually a tiny bit. Also, the first time I tried to send a message, keystrokes were doubled, but it worked fine after that. Don't let anyone tell you WORA is just a dream! - Dan p.s. It took me a while to figure out how to get the JDK running (it was simple, but I didn't see the doc.) I ended up with this in my .bash_profile: JAVA_HOME=/usr/local/java/jdk1.1.6 PATH=$JAVA_HOME/bin:$PATH CLASSPATH=/usr/local/java/jdk1.1.6/lib/classes.zip #following not needed? LD_LIBRARY_PATH=$JAVA_HOME/lib/i586/green_threads:$LD_LIBRARY_PATH Seems like there ought to be an RH RPM for the JDK that sets this stuff up systemwide so users don't need to modify their personal .bash_profile's... if no one at Blackdown is doing that, maybe I'll try...
Re: can't find class ijvmasm
I believe the problem is specific to the software package (mic1) you are trying to run. With some of the earlier releases of the package, ijvmasm.class was missing from the classes.zip file. Try downloading the current development release from www.ontko.com/mic1 and reinstall. Or, you can compile source/ijvmasm.java and add that directory to your CLASSPATH. Dan TC Lin wrote: > > Hello, Could anyone figure out what the problem is ? > > Thanks. > > -Todd > > 1. > problem > > in ~hyu/mic1 directory I type in > $java ijvmasm echo.jas echo.ijvm > > I got the following ERROR message : > > Can't find class ijvmasm > > 2. > files in the misc1 directory > > COPYING.TXT echo.jas mal.html mic1sim.txt > README.TXT faq.html mic1.tar relnotes.html > add.ijvm ijvm.confmic1asm.txt source/ > add.jas ijvmasm.txt mic1ijvm.mal user_guide.html > classes.zip ijvmtest.jas mic1ijvm.mic1 > echo.ijvmjas.html mic1sim.gif > > 3. > CLASSPATH > > hyu:~/mic1$ echo $CLASSPATH > /home/hyu/mic1/classes.zip > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Jini1.0 and JDK1.1.7 compatiblity.
Jini requires Java 2 Dan "Nagaraj S.B" wrote: > Hi all, > I have > I would like to know, > Whether Jini1.0 works on JDK1.1.7 loaded on Linux(Slackware3.5, > libc5.4.44,Kernel2.0.34) machine ?.Does Jini1.0 requires JDK1.2 only?. > Thanks in advance, > Nagaraj S.B. > Bells Softech Ltd,Bells House,1036, > 26th Main,4th 'T' Block, > Jayanagar, Banglore - 560 041.Ph.No.:6650084/33. > > -- > 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 on RedHat 6.1
RedHat has just released the 6.1 version of Linux. I am curious if anyone has had any experience with the 1.1.7, 1.1.8 or 1.2 JDK and JRE on RedHat 6.1. I have some rather large applications to port, and I would like to know if there are any known pitfall introduced by the new kernel. I believe 6.1 is based on the 2.2.12 kernel. Thanks in advance, Dan Iuster [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RE: java-linux-digest Digest V99 #49
How do I unsubscribe from the java-linux-digest mailing list? [EMAIL PROTECTED] -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]Sent: Friday, October 15, 1999 7:06 AMTo: [EMAIL PROTECTED]Subject: java-linux-digest Digest V99 #49
Re: Mount Table File per Linux Distributon
David Marshall said, about Mount Table File per Linux Distributon > Peter Pilgrim wrote: > >> Hi >> >> I need some information about linux distributions and where >> they have the system mount table. >> >> For SuSE Linux this file is ``/etc/mtab'' >> >> I need to know what it is for Red Hat Linux, Debian, Slackware, TurboLinux >> and Linux Mandrake. THis is for a Java Linux portable JNI project > > For Redhat it's "/etc/fstab" /etc/fstab is where you have the things that you WILL or WANT to mount. /etc/mtab is where it keeps what IS mounted. just running the command 'mount' with no switches in most *NIX is a way to get what is mounted. Dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: JRE -and- JDK
Trieu Truong said, about JRE -and- JDK > A newbie question (this seems too obvious even for the FAQs): I need to > install the JRE (I'm going to use the one at blackdown.org) so that I > can run Oracle 8i on my Linux box. Does this either preclude or hinder > my ability to install and use Blackdown's JDK so that I may learn and > develop with Java on the same box? No. I've got JDK 1.1.8, JDK 1.1.7, JDK 1.2.2, JRE 1.2.2, and a couple others running around. Make sure you set your JAVA_HOME and CLASS_PATH in a shell script that runs your program and make sure that they point to the right one. Dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
high performance I/O?
Anyone here interested in getting Java to handle heavy I/O nicely? I'm personally interested in insane things like trying to write an ftp daemon that can handle 5000 simultaneous clients in Java, but there are probably more sane examples. I have a few notes on the subject at http://www.kegel.com/c10k.html#java and am musing about how one might support /dev/poll ( http://www.kegel.com/c10k.html#/dev/poll ) in Java under Linux. Sun is trying to deal with these issues, too; see http://java.sun.com/aboutJava/communityprocess/jsr/jsr_051_ioapis.html - Dan -- Entia non sunt multiplicanda praeter necessitatem. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: high performance I/O?
Matt Welsh wrote: > Exactly the topic of my research :-) You should check out Jaguar, > http://www.cs.berkeley.edu/~mdw/proj/jaguar OK, I now link to it from http://www.kegel.com/c10k.html#java > We are also working on building large-scale Internet services in Java. > Many of the problems you talk about on your 'c10k' web page (which is > very good, by the way) are things we are addressing. Thanks. The table of contents I just added ought to make it more readable. > You also might want to read a paper we wrote recently about engineering > systems for high throughput. What we have in mind here is Java, > but it applies to any language and O/S: > http://www.cs.berkeley.edu/~mdw/papers/events.pdf Looks like a good read. I now link to it. (Wish my copy of Ghostscript didn't choke on half the pages!) > I'm also on the Expert Group for JSR 51 (the new I/O APIs for the Java > platform), so hopefully good things will happen there! Yes, it should be interesting! > My personal feeling is that there is a lot that will need to happen at > both the Java and the O/S level to get great I/O performance. I am not > sure I agree with many of the discussions in the linux-kernel list > that the right way to get high I/O bandwidth is just to use some > bastardization of signals; I think that the folks at Rice are a lot > closer to the mark with their novel event-delivery mechanism (by > this I mean http://www.cs.rice.edu/~druschel/usenix99event.ps.gz). The folks at the Linux Scalability Project seem to agree that the realtime signal approach isn't the top contender at the moment. The /dev/poll they implemented for Linux beats the realtime signal approach in their tests so far. See http://www.citi.umich.edu/techreports/reports/citi-tr-00-4.pdf > So ... it's a very interesting space to work in right now. Demanding > I/O applications place a lot of new demands on Java (and operating systems > as well). I would be interested in having more discussion with people on > this list about their experiences! Ditto. And if anyone interested in this happens to be in LA, drop in on the lajug javaklatch at Equator in old Pasadena at 1pm this Saturday and let's talk Java over java :-) - Dan -- Entia non sunt multiplicanda praeter necessitatem. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: J2ME/KVM for Linux?
Thomas, Check out the patch available at http://www.kvmworld.com - it will allow you to cross compile the Solaris source for the linux platform. Dan Hushon Pervasive Java Technology Practice Sun Microsystems, Inc. +1.703.981.3330 (GSM) Thomas Bonk wrote: > Hello, > > does there exist a Linux port of J2ME/KVM? > > Thanks in advance. > bye, Thomas > -- > "Es gibt zwei Dinge, die unendlich sind: Der > Weltraum und die menschliche Dummheit. Bei > dem Weltraum bin ich mir allerding noch nicht > ganz sicher." (Albert Einstein) > > -- > 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]
System.exit() value not being observed with 1.3.0 FCS?
Hi,
looks like System.exit(1) doesn't cause the java executable to
terminate with an exit status of 1 with the Blackdown 1.3.0.
It works fine with the IBM 1.3.0, though. Here's my test program:
import java.io.*;
public class foo {
public static void main(String args[]) {
System.out.println("Failing...");
System.exit(1);
}
}
And here's the shell script I run it with:
#!/bin/sh
set -x
set -e
java foo
echo Hmm: exit status was $?
With Blackdown 1.3.0, I see
$ sh foo.sh
+ set -e
+ java foo
Failing...
+ echo Hmm: exit status was 0
Hmm: exit status was 0
$
which is incorrect, but with IBM, I see the expected
$ sh foo.sh
+ set -e
+ java foo
Failing...
$
I suppose this is an old known problem... any workaround?
- Dan
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Brutal X-Windows performance of Swing apps on slow links
"Martin, Stephen" wrote: > I've been trying to display a Wwing application through X-Windows from one > machine to another across an ISDN line. Performace is horrible, it takes a > long time to open windows, display menus and react to mouse events. Other > X applications including emacs and Mozilla are quite usable over the same > link. What is it about Swing and/or AWT that causes this horrible > performance. Is there anything that can be done about it. See the suggestions in http://developer.java.sun.com/developer/bugParade/bugs/4204845.html Do any of those help? - Dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Brutal X-Windows performance of Swing apps on slow links
"Martin, Stephen" wrote: > > > See the suggestions in > > http://developer.java.sun.com/developer/bugParade/bugs/4204845.html > > Do any of those help? > > - Dan > > This and all the rest of the replies have been excellent. Clearly Sun is > aware of the problem and from reading the posts attached to the bug > it is clear that the user community wants them to do something about it. I dunno - seems like the only thing Sun could do is detect the slow connection, and automatically apply some of those workarounds. They're not hard. I do hope you try them. - Dan > I will try some of the suggestions and figure out what works best for me, > unfortunatley it looks like copying over the source and compiling and > running locally is going to be the way to go. :( -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: connecting netscape ?
"V.Suresh" wrote: > > At present, I am doing a stand alone application project, and I want > an event to open up netscape with a URL. How do I do this? Try http://www.tolstoy.com/samizdat/jconfig.html - Dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
JVM and threads
I just joined the list today so my apologies if this has already been covered but hasn't made it to the archive yet. >From lkml: first NPT vs. NGPT vs. LinuxThreads benchmark results at http://www.uwsg.iu.edu/hypermail/linux/kernel/0209.2/1581.html Note as well that there is an NPT mailing list. archives at https://listman.redhat.com/mailman/listinfo/phil-list best, dan -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
jdk1.5 and sparc
Hi, im trying to get answers about this, after reading the 1.5 status http://www.blackdown.org/java-linux/java2-status/j2se1.5-status.html it states there is jdk1.5 support for sparc among other platforms. However when i go into all the mirrors i can only see 1.4 available. Im wondering if i was able to get info if blackdown will of course run jdk1.5 in sparc linux. Unfortunately there is no sparc java port for freebsd, so after finding this info i tried to install ubuntu on this, my last resort is of course solaris 10. Let me know thanks.
Re: jdk1.5 and sparc
Stefaan A Eeckels wrote: On Mon, 04 Sep 2006 22:18:07 +1000 Dan Rossi <[EMAIL PROTECTED]> wrote: my last resort is of course solaris 10. In my experience, on SPARC Solaris 10 is the superior OS. Ok but in terms of admining the upkeep of a developer server obviouslly debian linux and freebsd are far superior in terms of keeping the server source and packages up to date, solaris, especially the patches is a nightmare, however i see they have a update manager in 10 now, so far behind the times. Im not a solaris admin so prefer to ignore that part altogether. It still doesnt answer my question what this page means then i dont even see a jdk1.5 package available, so what does it mean :) Im currently downloading solaris10 atm, however it was my last resort even though it may look like i have to go down that path lol. thanks.
Re: jdk1.5 and sparc
Klaus Wagner wrote: On Mon, 2006-09-04 at 22:18 +1000, Dan Rossi wrote: Hi, im trying to get answers about this, after reading the 1.5 status http://www.blackdown.org/java-linux/java2-status/j2se1.5-status.html it states there is jdk1.5 support for sparc among other platforms. However when i go into all the mirrors i can only see 1.4 available. Im wondering if i was able to get info if blackdown will of course run jdk1.5 in sparc linux. Unfortunately there is no sparc java port for freebsd, so after finding this info i tried to install ubuntu on this, my last resort is of course solaris 10. Let me know thanks. I think its true for solaris sparc. I once took a look at suns original java sources and figured out that there are mayor design differences between linux and solaris. As blackdown jdk is based on these sources it would be quite difficult to port it to linux sparc. linux sources use x86 centristic asm code while sparc code heavyly use solaris internals. So either way (starting from linux x86 or starting from solaris sparc) would be hard to go. And it is quite a neverending story because sun will never support these sources. I don't know if anyone intended to do this for jdk 1.5 but I gave up tracking it about a year ago. You might want to try solaris express or opensolaris as sun tries to move towards opensource in many ways. regards klaus Right i dont understand why that jdk1.5 status page is up then, if sparc support isnt possible. I also saw somewhere that a netbsd port supports sparc jdk 1.5 buts its vague. I dont really have the money to move to an amd system really just so i can run it on something other than intel linux preferably bsd. thanks,
Re: jdk1.5 and sparc
[EMAIL PROTECTED] wrote: According to Jurgen Kreileder, there is one bug left that prevents a release. At this time, no JVM is running on the sparc port because of some changes in the glibc that cause the 1.4.1 (only 1.4 jdm released) to crash on non trivial load. Don't ask me, I don't know how to fix it. The status release is that sparc-linux jvm is dead, as far as I know. Seb Ok thanks for your troubles, an upgrade to a dual opteron is at hand then at a later date, and ill come back to this :) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: jdk1.5 and sparc
[EMAIL PROTECTED] wrote: On Mon, Sep 04, 2006 at 08:49:43PM +0200, Johan Vos wrote: Hi, I did the port to Linux/SPARC for JDK 1.2, but I don't have a performant Linux/Sparc machine anymore and 1.5 is very different from 1.2. Apart from that, I don't think there is a huge interest in Linux/SPARC anymore, as Solaris indeed moved to Open Source. - Johan I have a sparc as my main gateway. I have linux on it and plans to keep it like this. I've been bitten by the java JVM 1.4.1b not in shape. I've tried to recompile from JVM 1.2, which went almost to the end (there is some problem with a float to text conversion wich makes the build choke). This one is almost ok. But almost is not ok. I don't know if there are more bugs lurking behing the surface. I've been able to build it because the sources were available. For 1.3, 1.4 and 1.5, the diffs against the source are not even available. Don't know why. Licence didn't changed, and diffs of previous JVM were released, leading me to the conclusion that it was ok to release diffs against the SUN source. The hotspot client and server VM has already been ported on the linux/sparc architecture but noone seems to maintains it anymore. Some people released a beta or some earlier work, but no final version were ever released for arch other than x86 and x86-64. There seems to be lacking some continuity in the blackdown's efforts. However, I'm confident that the efforts needed to reactivate the java port should be minimal. Maybe a refresh build for 1.4.1b to make it work with the new glibc should be enough to have a working port again. At least enough to use J2EE and tomcat (which is the most needed apps on sparc/linux). Look at the BSD effort, they keep the diffs available against all java JVM. This is IMHO the way to go. If we only could have the diffs of the impressive work the blackdown team did, it would be a great step toward a working JVM for sparc. Oh well, it would have been really nice, ive gone back to installing solaris 10, however it doesnt seem its progressed in any way at all. I managed to get both freebsd and ubuntu installed in under 20 mins, with solaris 10 ive spent a while day already trying to research why it finds a drive corrupt and unusuable even though both the others were absolutely fine with this, on top of solaris only seeing 130GB of one of the drives where the others see the entire size fine, makes you wonder how primitive it is lol .
Re: jdk1.5 and sparc
Stefaan A Eeckels wrote: On Wed, 06 Sep 2006 00:57:22 +1000 Dan Rossi <[EMAIL PROTECTED]> wrote: Are you installing Solaris 10 on a SPARC based box? The by now very old Ultra 5 and 10 had a IDE controller that was limited to 128GB, but that applies to all OSes you'd install on them. Ok getting off topic here, nope ubuntu and freebsd partitions 160GB , solaris only manages to see 130GB to partition on top of this the original drive that was in my fire v100 rack is now not usable in solaris as it was partitioned from the ubuntu installer , even deleting the partition solaris cant use it :) Habve to work out what bootable partition software can prepare the drive for solaris as its tools are shotty. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Java on a Cobalt RAQ
I have a Cobalt RAQ Microserver running Linux 2.0. It says that it operates on a 64 bit superscalar processor. Are you aware of any JREs for this environment? Thanks in advance. Dan -- Dan Coates, Vice President Millward Brown Interactive [EMAIL PROTECTED] San Francisco: 415.538.0923 Cincinnati:513.984.8811 Fax: 513.984.8388
Re: JDK1.2
Don't ask ??? Why? Don't you think it is a relevant question to ask for JDK1.2 for Linux on this mailing list ??? /Dan >DAMMIT!!! Why don't we just make a join message that says >"The damn JDK1.2 (Java 2) isn't. Don't ask." > > > > >- > >On Tue, 12 Jan 1999 [EMAIL PROTECTED] wrote: > >> >> >> Hi guys, >> >> Just wondering about the status of JDK1.2 (a.k.a Java 2) for Linux. Can we >> expect a release by Feb. 1, 1999 ? >> >> - Farley Caesar >> >> P.S. Great work on the JDK1.1.x series for Linux by the way. >> >> >>
