No Subject
I am trying to compile traceroute in c, along with jni.h, and some java files but keep getting the following error message, anybody any ideas? Cheers. gcc -O2 -DHAVE_MALLOC_H=1 -DHAVE_STRERROR=1 -DHAVE_SETLINEBUF=1 -DSTDC_HEADE RS=1 -I. -Ilinux-include -c /home/murray/traceroute-1.4a4/traceroute.c In file included from /home/jdk117_v1a/include/oobj.h:25, from /home/jdk117_v1a/include/native.h:26, from /home/murray/traceroute-1.4a4/jumpDetails.h:2, from /home/murray/traceroute-1.4a4/traceroute.c:211: /home/jdk117_v1a/include/typedefs.h:18: typedefs_md.h: No such file or directory In file included from /home/jdk117_v1a/include/native.h:26, from /home/murray/traceroute-1.4a4/jumpDetails.h:2, from /home/murray/traceroute-1.4a4/traceroute.c:211: /home/jdk117_v1a/include/oobj.h:28: oobj_md.h: No such file or directory In file included from /home/jdk117_v1a/include/javaString.h:24, from /home/jdk117_v1a/include/native.h:29, from /home/murray/traceroute-1.4a4/jumpDetails.h:2, from /home/murray/traceroute-1.4a4/traceroute.c:211: /home/jdk117_v1a/include/java_lang_String.h:2: native.h: No such file or directo ry n-ary .. .. .. the uk java specialists http://www.n-ary.com [Servlets][JDBC][Applications] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
No Subject
am running a java method which includes a native method and am getting the following error messages, any ideas anyone? Cheers. Murray. SIGSEGV 11* segmentation violation stackbase=0xb9f0, stackpointer=0xb8f8 Full thread dump: "Finalizer thread" (TID:0x40655210, sys_thread_t:0x41399f04, state:R) prio=1 "Async Garbage Collector" (TID:0x40655258, sys_thread_t:0x41378f04, state:R) prio=1 "Idle thread" (TID:0x406552a0, sys_thread_t:0x41357f04, state:R) prio=0 "Clock" (TID:0x40655088, sys_thread_t:0x41336f04, state:CW) prio=12 "main" (TID:0x406550b0, sys_thread_t:0x818ddd8, state:R) prio=5: pending=jav a.lang.NoSuchFieldError *current thread* tracerouteJ.main(tracerouteJ.java:14) Monitor Cache Dump: Registered Monitor Dump: Thread queue lock: Name and type hash table lock: String intern lock: JNI pinning lock: JNI global reference lock: BinClass lock: Class loading lock: Java stack lock: Code rewrite lock: Heap lock: Has finalization queue lock: Finalize me queue lock: Dynamic loading lock: Monitor IO lock: Child death monitor: Event monitor: I/O monitor: Alarm monitor: Waiting to be notified: "Clock" (0x41336f04) Monitor registry: owner "main" (0x818ddd8, 1 entry) Thread Alarm Q: IOT trap/Abort (core dumped) n-ary .. .. .. the uk java specialists http://www.n-ary.com [Servlets][JDBC][Applications] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Don't download JDK1.2src! [Was: Re: JAVA2 source code]
On Fri, 26 Feb 1999, Cees de Groot wrote: > Oliver Fels wrote: > > > Just for your interest and in case noone has mentioned it yet: > > > > Sun is now making the source code to the Java® 2 platform > > available to the developer community. > > > > Yes, but *please* think twice before committing your soul to Sun! Isn't Sun worship an ancient tradition? ;) [snip] > You can see the openness involved with the SCSL: if they were serious about > openness, they'd have lifted the restrictions for the Blackdown porting team by > now. Methinks that with the JCK tests, Blackdown is at the debugging stage and > that can be perfectly paralellized - hundreds of people on the list would > gladly jump in. That they haven't done so, shows that Sun, as a corporation, is > not really committed to a Linux JDK or to openness. It's been my recent personal experience that Sun have some nasty internal communications problems, so I suspect it's more likely what we're seeing is due to sheer incompetence on Sun's part, not malevolence. Of course, in this day and age everyone suffers from some degree of paranoia, myself included. Scott -- = Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Slow loading on AMD Elan 486
On Wed, 3 Mar 1999 [EMAIL PROTECTED] wrote: > Running Blackdown 1.1.7 with linux 2.0.36 on an Elan SC400 development > board (486, 33 MHz, no floating point unit) results in extremely long load > times (up to an hour or more) for applets that load quickly on a desktop > linux system. After loading, the applet runs OK. > > Any ideas what is happening? > > One possibilty we have considered is the lack of a floating point unit, is > this plausible? The 486 has an integral FPU, it was the 386 that did not. Of course, to muddy the issue, the 486sx does not have one. >From my understanding of the class loading mechanism, poor floating point performance would have neglible impact on start up times. What kind of IO subsystem does your development board have? Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: glibc 2.1 binary
On Tue, 27 Apr 1999, Tom McMichael wrote: [snip!] > Good point Paul ... checked out jitter bug and according to the "DONE" > section the two choices for glibc 2.1 are: > 1) jdk 1.2 > 2) pre-pre-release of jdk117_v2 available at ... > > http://www.wisp.net/~kreilede/ > > I'm downloading it right now ... will post if I have success > Any comments on this pre-release ? It seems to fix the problem I (and others I think) had with Runtime.exec hanging sometimes when used with native threads. Which is good, as I was almost resigned to putting in some Linux specific code into the app I'm working on to avoid the problem under v1a. I don't want to sound ungrateful for the 1.2 effort (the initial results of which I'm using with great success), but is there an ETA for the release of jdk117_v2? Thanks, Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: HotSpot?
On Thu, 29 Apr 1999, Dimitris Vyzovitis wrote: > Uncle George wrote: > > > What was more interesting about the proj is that they say there is little to do > > for any cpu architecture to get it to run. :- > > gat > > > > Which - to get it even furter - combined with the fact that "... will be available > for downloads free of charge for developers soon..." leads to the logical > conclusion that the final version might not even be ready yet (even for > windows/i386) ;-}} The Win32 HotSpot release has already been made available to partners for testing, so it's unlikely that it won't be released this week. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Java/C benchmark
On Mon, 3 May 1999, J.P.Lewis wrote: > > but I'm afraid I have to say your followup comparisons are unfair... > > hold on - this post was intended to be just a statement of fact, > not a complaint. I use blackdown and am very grateful for its existence. > I've timed tya across releases, and it's gotten steadily faster over the > last 6 months. The fact that it already keeps up with the sun-supplied > jdk12/linux jit is impressive. > > I primarily wanted to point out the curious fact that the > jit supplied by sun for Windows is much faster than the > jit supplied by sun for Linux, and to ask what this meant. It's quite likely that the Linux JDK 1.2 JIT is based on the one that Sun provides with the Solaris Reference JDK 1.2, which does not seem to provide as good of a performance boost as the Win32 JIT or the Solaris Production JDK JIT. All IMHO, of course. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: glibc 2.1 binary
On Tue, 4 May 1999, Steve Byrne wrote:
> Scott Murray writes:
[snip]
> > It seems to fix the problem I (and others I think) had with Runtime.exec
> > hanging sometimes when used with native threads. Which is good, as I
> > was almost resigned to putting in some Linux specific code into the app
> > I'm working on to avoid the problem under v1a.
>
> What kind of hanging? Were you waiting for the process to complete explicitly?
> I'm very interested to find out more.
>
> We're having that problem in 1.2 with native threads, which is one of the
> reasons that we don't say we pass JCK using native threads yet.
I'm attaching the source to my test program below as it's pretty short. It's
something I pasted together with code snippets from the Java Programmer's
FAQ as a proof-of-concept Solaris JDK 1.2 version checker, so it's not the
greatest code in the world. I do call Process.waitFor(), but it's probably
unnecessary, as I imagine that the exec'd Process is already dead when its
stderr is closed.
Under jdk117_v1a with native threads, the program hangs at the first
readLine call, but it works fine when run with the pre-release jdk117_v2 or
with jdk1.2pre-v1. That it works with jdk1.2pre-v1 makes me suspect that it
might not be of much use to you for debugging purposes, but you never know.
Scott
--
=
Scott Murray email: [EMAIL PROTECTED]
http://www.interlog.com/~scottm ICQ#: 10602428
-
"Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness"
import java.io.*;
import java.util.*;
class CheckVersion {
public static void main(String[] args) {
boolean valid = false;
String path = "/usr/local/jdk1.2/bin/java";
if(args.length >= 1) {
path = args[0];
}
String command[] = { path, "-version" };
//String command = path + " -version";
System.out.println("executing command: " + path + " -version");
try {
System.out.println("Before exec");
Process p = Runtime.getRuntime().exec(command);
System.out.println("After exec");
BufferedReader pErr = new BufferedReader(new
InputStreamReader(p.getErrorStream()));
System.out.println("Reader created");
// Get stderr
String s = pErr.readLine();
System.out.println("Read output line: " + s);
String last = s;
while (s != null) {
last = s;
s = pErr.readLine();
System.out.println("Read output line: " + s);
}
//For testing: String last = null;
p.waitFor();
System.out.println("Command return code = " + p.exitValue());
if(last != null) {
System.out.println("Last line output: " + last);
StringTokenizer st = new StringTokenizer(last);
String type = null;
if(st.hasMoreTokens()) {
type = st.nextToken();
}
if(type != null) {
System.out.println("vm type = " + type);
if(type.equals("Classic")) {
valid = true;
}
}
}
}
catch (Exception e) { }
if(valid) {
System.out.println("Valid 1.2 VM version.");
}
else {
System.out.println("Invalid 1.2 VM version!");
}
}
}
Re: sunwjit slows down JDK
On Tue, 8 Jun 1999, Bob Cadenza wrote: [snip] > For images this is almost 5 times as slow in jdk1.2. You can even see > it, the benchmark even redraws itself noticably slow. You can chalk that difference up to the addition of Java2D in JDK 1.2. Check out the number 1 bug in Bug Parade, #4185726. Sadly, Sun seem to be dragging their heels on fixing this problem. Scott -- = Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: missing header files in JDK1.2
On Fri, 11 Jun 1999, Sumit Gupta wrote: > REPOSTING: > > I am running into trouble compiling some code that compiles and runs > on a solaris box (with jdk1.1.7 for solaris) but cannot find several > header files when compiled against jdk1.2 on linux. (for example, > oobj.h, sysmacros_md.h, threads.h etc) > > The include directory on the jdk1.2 seems to lack several required > header files. Some of them can be found in include-old but they > don't seem correct either. > > did anybody else face this problem? Where are the right include files > in the jdk1.2.? I would really appreciate the help. Nothing is missing; I believe the header files in include-old should allow you to compile older JNI code so that it works against 1.2. Make sure that you also include include-old/linux to get the platform specific stuff. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Version for GlibC
Andreas Jaeger <[EMAIL PROTECTED]> writes: > Don't mix the distributions: SuSE 6.0 (and we're discussing SuSE here) > is libc5 - RedHat 6.0 is glibc 2.1. I think you are mistaken. The SuSe 6.0 which I am running here came with glibc 2.07. SuSe 5.3 was libc5. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Corrupted JDK-1.2.2
On Mon, 29 Nov 1999, Albert Y.C. Lai wrote: > "John N. Alegre" <[EMAIL PROTECTED]> writes: > > > What gives? Why the shar file anyway? bz2 was ok, gz even better. > > shar files aren't even compressed > > I think the main purpose of the installer script is to make you read > and agree to the license before it unpacks the files, like all > software packages for DOS, Apple DOS, and Slowaris do these days. The Solaris Reference JDKs are indeed packaged in the same manner. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Select/poll for java-linux
On Mon, 29 Nov 1999, Artur Biesiadowski wrote: > Has anybody implemented it already ? Sun does not seem to try to solve > the subject, but I suppose that it should be possible to add select like > functionality on top of current jdk (including some native code of > course). > > I'm not thinking about being particulary portable here - as long as I > can run it on any 2.2.x linux kernel (no need of windows/solaris > compatibility). Sun have been working on some kind of poll extension, but AFAIK it is only available in some of their Solaris Production JDK's. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ#: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Sun and Inprise?
Anyone from the Blackdown team care to comment on: http://www.computerworld.com/home/news.nsf/all/9912035javalinux wherein a "source close to Inprise" claims that Inprise and Sun are going to announce that they are "porting the Java 2 development kit to the Linux operating system". No mention of the Blackdown team's current porting effort is made in the article. If this is true, I'm going to be seriously PO'd at Sun. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Sun and Inprise Java 2 announcement
Check out: http://www.newsalert.com/bin/story?StoryId=CoeYuubWbu0zuvteXmW Their J2SE 1.2.2 RC1 is available at: http://developer.java.sun.com/developer/earlyAccess/j2sdk122/ and here's their "System Requirements": This version of the Java 2 SDK is supported on Intel Pentium platform running the Linux kernel v 2.2.5 and GLibC v 2.1, 32 megabytes RAM minimum. Recommended 48 megabytes of RAM, 16-bit color mode, KDE and KWM window managers. You should have 65 megabytes of free disk space before attempting to install the Java 2 SDK software. If you also install the separate documentation download bundle, you need an additional 90 megabytes of free disk space. We do not support or recommend running Java 2 SDK on SMP kernels. Also the only thread model supported is green threads. I'm praying that the Blackdown team continues it's work if this is what Sun consider a useful release. I'm really curious as to why the hell Sun and Inprise went off on their own when the Blackdown port is available... Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun and Inprise Java 2 announcement
On Tue, 7 Dec 1999, Paolo Ciccone wrote: > On Tue, Dec 07, 1999 at 09:19:28AM -0500, Scott Murray wrote: > > I'm praying that the Blackdown team continues it's work if this is what Sun > > consider a useful release. I'm really curious as to why the hell Sun and > > Inprise went off on their own when the Blackdown port is available... > > Besides extensive testing (full JCK) and performace tuning, this version > includes JPDA and several Swing bugs that we found while developing JBuilder > Solaris edition. Without native threads support, this release is IMO useless for running any kind of serious Java applications on Linux. While I will admit that I myself have been sometimes frustrated by the speed of the Blackdown porting team, they have been doing a good job quality-wise. I still don't understand why Sun would start their own Linux JDK port and not use the Blackdown team's work as a base. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Native vs. green threads
On Tue, 7 Dec 1999, Matt Welsh wrote: > > [EMAIL PROTECTED] (Nelson Minar) writes: > > > > However, I disagree that native threads are required for serious > > applications. Green threads work surprisingly well for many > > applications. In some, they work better. I recently wrote a spider > > program that was invoking another program in a subprocess and blocking > > on IO to it. When I had 20 threads (and 20 subprocesses) the green > > threads did much better than native threads. That's not entirely > > surprising, either. > > This is all fine and good until you need to exploit an SMP system! To clarify my original statement, this is mostly what I meant when I said "serious" java applications. From the customer support issues we receive at my day job for the product I work on, it seems that a reasonable percentage of the people running Java servers out in the commercial world are using SMP boxes. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun and Inprise Java 2 announcement
On Tue, 7 Dec 1999, Paolo Ciccone wrote: > On Tue, Dec 07, 1999 at 04:39:07PM -0500, Derek Glidden wrote: [snip] > > It looks like either someone at Inprise or at Sun isn't playing fair. > > We actually are. We didn't say that this is a clean port, in fact we > are giving credit to Blackdown for the port. This has been done with > the press at the Java Expo in New York and I personally posted here > and on /. about our use of the early patches from Blackdown. The story > is simple, we took the patches for 1.2.1 and started working on > 1.2.2. From that early effort, fundamental to start the port, we > actually added a lot of other stuff. The AWT/Motif for example has > been completely replaced and a lot of other fixes have been added. > This is truly a collective achievement, starting froim the early > effort of Blackdown and then accelerated by Sun and Inprise and > brought to a level of stability and performance that we feel is > acceptable before the release of 1.3/Hotspot. > > My previous post, that you quoted, was about the RC1/RC3 naming, the > two ports are indeed different and I wanted to clarify some of the > confusion. I don't really mean to be a bastard here, but here's an excerpt from one of your postings to this list earlier today: The Sun/Inprise port is not the same of Blackdown's, that's why the release number is different. The Sun/Inprise port is actually the latest codebase from Sun and includes several fixes to Swing that were found in the stock 1.2.2 release. What were we supposed to think? As well, I've read three different press articles on this today, and none of them contained the word "Blackdown". If Inprise/Sun wanted to give credit to the Blackdown team for providing a base for their further work, I would have expected something, especially since the press releases for this were delayed several days. It's not like Sun doesn't know about the Blackdown team, they made them a licensee last year with great fanfare. Scott -- = Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun and Inprise Java 2 announcement
On Tue, 7 Dec 1999, Scott Murray wrote: [snip] > What were we supposed to think? As well, I've read three different press > articles on this today, and none of them contained the word "Blackdown". > If Inprise/Sun wanted to give credit to the Blackdown team for providing > a base for their further work, I would have expected something, especially > since the press releases for this were delayed several days. It's not like > Sun doesn't know about the Blackdown team, they made them a licensee last > year with great fanfare. I know it's bad form to reply to your own post, but I just found a press article linked from LinuxToday that mentions the Blackdown team in association with Sun's 1.2.2 RC1 release. Check out: http://news.cnet.com/news/0-1003-200-1484634.html?tag=st.ne.1002. Oddly, it does not mention Inprise at all, but has lots of detail on how IBM think Sun should give up control of Java. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Native vs. green threads
On Tue, 7 Dec 1999, Weiqi Gao wrote: > Scott Murray wrote: > > > > On Tue, 7 Dec 1999, Matt Welsh wrote: > > > > > > [EMAIL PROTECTED] (Nelson Minar) writes: > > > > > > > > However, I disagree that native threads are required for serious > > > > applications. Green threads work surprisingly well for many > > > > applications. In some, they work better. I recently wrote a spider > > > > program that was invoking another program in a subprocess and blocking > > > > on IO to it. When I had 20 threads (and 20 subprocesses) the green > > > > threads did much better than native threads. That's not entirely > > > > surprising, either. > > > > > > This is all fine and good until you need to exploit an SMP system! > > > > To clarify my original statement, this is mostly what I meant when I > > said "serious" java applications. From the customer support issues > > we receive at my day job for the product I work on, it seems that a > > reasonable percentage of the people running Java servers out in the > > commercial world are using SMP boxes. > > And how are they exploiting the SMP system with green threads? On Solaris, they use the native thread support in either the Reference or Production VM. On Linux, they use either the Blackdown or IBM version of the JDK. That was kind of my point, when I started this thread by saying: Without native threads support, this release is IMO useless for running any kind of serious Java applications on Linux. Of course, it is true that my definition of "serious" does not necessarily coincide with everyone else's, but I did kind of mean SMP scalability for running application servers when I wrote that sentence this morning. Scott -- = Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
JDK 1.1.8-v1 files on mirrors
It seems that the Intel version of the Blackdown JDK 1.1.8 has been released, as I noticed an i386 directory in JDK-1.1.8 tonight. I've been unable to run it just yet here at home, as I'm still running a stock RedHat 6.0, but I've noticed the following two glitches so far: - the files are .bz instead of .bz2, so I had to rename them before bunzip2 would uncompress them. - the jre118_v1 package seems to be missing the native_threads directories, although the jdk118_v1 package has them. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Using JPDA with Blackdown JDK?
On Wed, 22 Dec 1999, SHUDO Kazuyuki wrote: [snip] > Only JVM itself can provide JVMDI functions. I > understood it is the reason why Inprise guys decideed to > develop their JDK. > > It's a wonder that JVMDI can't work with the Blackdown > JDK 1.2. The reference implementation seems to have an > implementation of JVMDI as far as I check the source > code of it. It's no wonder JVMDI doesn't work properly > with Sun's JIT (sunwjit.so). Doesn't JVMDI in the > Blackdown JDK work even if JIT is disabled? >From what I've been able to determine, it looks like JVMDI works with the Blackdown JDK 1.2.2 release candidates, even with the JIT turned on. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Using JPDA with Blackdown JDK?
On Wed, 22 Dec 1999, Scott Murray wrote: [snip] > >From what I've been able to determine, it looks like JVMDI works with > the Blackdown JDK 1.2.2 release candidates, even with the JIT turned on. Oops, I have to correct myself here. A teammate at work reminded me that the wrapper for our JVMDI-using program always disables the JIT due to some problems discovered on Win32 JDKs. So, the best I can say right now is that JVMDI works on the Blackdown 1.2.2 JDKs with the JIT disabled. I'll probably investigate this issue after my Christmas holidays just to pin down the situation better. Scott -- ===== Scott Murray email: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Problem with exec()
> Dimitris Terzis wrote:
>
> Hi guys...
>
> I am trying to launch an external application from my Java program,
> using exec().
>
> I do Runtime.getCurrentRuntime() and then try to exec() the
> application (not internal command), providing a properly created path.
> My application has a GUI which is never displayed. However, if I type
> "ps -aux" I can see that the process is there, which is reasonable
> since, in all cases, the Process object returned from
> Runtime.getCurrentRuntime().exec(...) is valid and no IOException (or
> whatever other error) occurs.
>
> I am running JDK 1.2.2-RC2 from Sun, on a RedHat 6.1 box.
>
> Any ideas?
I've noticed that sometimes the program being exec'd will seem to
block if you do not read its standard output. I encountered this
while working on the JProbe Solaris 2.5 installer.
Here's the relevant snippet from JProbe's Java Installshield
registration extension, which prints any output as it runs the
JProbe product at the end of the install:
// Run the product
System.out.println("Running " + product_fullname + " for registration
purposes");
try {
Process p = Runtime.getRuntime().exec("/bin/sh " + product_exe +
options);
// Get stdout
BufferedReader pOut =
new BufferedReader(new InputStreamReader(p.getInputStream()));
String s = pOut.readLine();
while (s != null) {
System.out.println(s);
s = pOut.readLine();
}
p.waitFor();
}
catch(Exception e) {
e.printStackTrace();
}
Hope this helps,
Scott
--
Scott MurrayKL Group Inc.
JProbe development 260 King St. E.
Tel: (416) 594-1026, ext. 355 Toronto, Ontario
Fax: (416) 594-1919 Canada, M5A 4L5
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Mount Table File per Linux Distributon
On Fri, 31 Mar 2000, David Marshall wrote: > 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" Ummm, no, that's where you specify what you want mounted, not what is currently mounted. /etc/mtab should exist on all recent Linux distributions, because it is created by mount itself. From the mount(8) man page: The programs mount and umount maintain a list of currently mounted file systems in the file /etc/mtab. If no arguments are given to mount, this list is printed. When the proc filesystem is mounted (say at /proc), the files /etc/mtab and /proc/mounts have very similar contents. So, I'd say that both /etc/mtab and /proc/mounts are reasonably cross-distribution ways to look for the currently mounted filesystems. Scott -- = Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SMP ok?
On Sun, 2 Apr 2000, Matt Welsh wrote: > > Dimitrios Vyzovitis <[EMAIL PROTECTED]> writes: > > > > I have a 2xPIII-500 box and it kicks butt (both ibm and blackdown jdk). > > Go for it! > > We have had a lot of problems with native threads on SMP machines, for > all JDK's tested (Sun 1.1.x, Sun 1.2.x, and IBM). This includes JVM crashes > and mysterious behavior which we can't attribute to bugs in our own code. Just curious, what distribution/glibc are you using? I've found that upgrading to glibc 2.1.3 greatly improved the stability of IBM's 1.1.8 on my SMP box with RH 6.1. As a counter-example, I've seen greater instability when running my application under Mandrake 7.0, probably due to their "optimized" libraries. > We are running fairly complicated applications with many threads and > "interesting" interactions between them. A lot of people may never run into > these problems because most multithreaded applications are not so complex. > > The only way we've been able to achieve stability is to use Sun JDK 1.1.7v3 > with green threads, which clearly defeats the purpose of using SMP systems. > > Caveat emptor. Have you tried Blackdown's 1.2.2-RC4? I ask because it's not explicitly in your list, and my experiments have shown it to be quite stable with native threads, albeit a bit slow compared to IBM's 1.1.8. AFAIK, Blackdown's 1.2.2 VM has the most mature native threads implementation of all the VMs available for Linux. As well, worse comes to worse, if you can reproduce threading problems with it, IMO the Blackdown team are probably more inclined to fix them compared to either Sun or IBM. Scott -- = Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What is fcs?
On Tue, 18 Apr 2000, Ganesh Sivaraman wrote: > Hi, > > Do anyone know what it means by fcs? I have just downloaded Swing1.1.1fcs > version. But the ironical situation is there are many fcs, when searched > but no where it is explained. What could it mean? For ex. I am using Kaffe > and it says that Swing 1.1 cannot work but 1.1.1 fcs does. So what is this > fcsall about > > And my willd guess saysit is Foundation classes..:), which > isnt. Grrr...these Acronyms.. > > Any other wild guess are welcome..:). and ofcourse the serious and the > right one. No WAG required, as I'm pretty sure. At Sun, FCS means "First Customer Ship", which basically means the release version of a product. They probably shouldn't be labelling the downloads with FCS, as it's not really meaningful to anyone besides Sun and their licensees. Scott -- ====== Scott Murray KL Group Inc. JProbe development 260 King St. E. Tel: (416) 594-1026, ext. 355 Toronto, Ontario Fax: (416) 594-1919Canada, M5A 4L5 "I'm sick and tired of people saying that we put out 11 albums that sound exactly the same. In fact, we've put out 12 albums that sound exactly the same." -- AC/DC guitarist Angus Young -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: What is fcs?
On Wed, 19 Apr 2000, Paolo Ciccone wrote: > On Tue, Apr 18, 2000 at 11:02:41AM -0400, Scott Murray wrote: > > No WAG required, as I'm pretty sure. At Sun, FCS means "First Customer > > Ship", which basically means the release version of a product. They > > probably shouldn't be labelling the downloads with FCS, as it's not really > > meaningful to anyone besides Sun and their licensees. > > Scott, I disagree with that since I saw the term used in several companies > in the industry. While I agree that other companies probably use the same acronym or similiar acronyms (e.g. IBM's GA), I personally don't think Joe Q. Public should have to pull out his magic decoder ring to determine what version a product is. Scott -- ====== Scott Murray KL Group Inc. JProbe development 260 King St. E. Tel: (416) 594-1026, ext. 355 Toronto, Ontario Fax: (416) 594-1919Canada, M5A 4L5 "I'm sick and tired of people saying that we put out 11 albums that sound exactly the same. In fact, we've put out 12 albums that sound exactly the same." -- AC/DC guitarist Angus Young -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: IBM has posted 1.3 pre-release
On Thu, 4 May 2000, SHUDO Kazuyuki wrote: > Juergen Kreileder wrote: > > > IBM has its own JVM. Our 1.3 will have Sun's HotSpot VM. > > Is the forthcoming Blackdown JDK 1.3 based on the > source code of HotSpot VM available under SCSL ? > I infer the answer is no. My understanding from reading the status page and talking with several people at LinuxWorld in February is that that is correct. However, I believe Sun have said that they will make the 1.3 source available under the SCSL at some point, probably after they get around to shipping the FCS version. > The archive containing the sourcecode (hotspot1_0_1-src.tar.gz) > seems to contains only SPARC stuffs. Yes that's the original server side HotSpot VM. 1.3 includes the so-called client side HotSpot VM, which is supposed to have better performance with interactive applications. > Will the HotSpot VM have a JIT compiler? I'm pretty sure that the Classic VM included with the 1.3 release candidates for Windows does not include a JIT. Since the default VM is HotSpot, which has their dynamic compilation technology in it, that won't be a problem for most people. However, if you hit a bug in HotSpot that requires you to switch to the Classic VM, you're screwed, as it has no JIT (unless you install one). :( > I know Sun migrates from the classic VM to the HotSpot VM. > One of my concerns is the effect of that migration > on developments of JIT compilers. >From what I've seen of IBM's 1.3, with its improvements on their already fast 1.1.8 VM, I think you have more to fear from it in this regard. Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: idltojava
On Fri, 21 Jul 2000, Brett Smith wrote: > I tried to ask this question yesterday, but apparently was suffering > from some form of dyslexia (javatoidl). > > Here is my 2nd attempt: Is there a version of idltojava for Linux? I > am using 1.2.2, rc4. Unfortunately, Sun don't seem to have given the 1.2.x idltojava compiler source to anyone to port, as neither their 1.2.2 J2SE release or the Blackdown 1.2.2 RC4 have it. A quick search in Google and some digging on eGroups turned up: http://www.egroups.com/message/java-linux/13108?&start=10760 where someone describes running the Windows idltojava under Wine. If that's not palatable to you, both IBM and Sun's JDK 1.3 releases include a new pure Java IDL compiler, idlj. One would hope that it generates stubs that work under JDK 1.2.2, this being Java and all... Scott -- ====== Scott Murray KL Group Inc. JProbe development 260 King St. E. Tel: (416) 594-1026, ext. 355 Toronto, Ontario Fax: (416) 594-1919Canada, M5A 4L5 "I'm sick and tired of people saying that we put out 11 albums that sound exactly the same. In fact, we've put out 12 albums that sound exactly the same." -- AC/DC guitarist Angus Young -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Does "HotSpot" always mean no "green threads"?
On Sun, 19 Nov 2000, Volker wrote: > Hi, > > does HotSpot only support "native threads" but not "green ones"? Yes, that seems to be the case with all of the the existing HotSpot VMs. Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
JDK1.3 only classic works
I have just downloaded the Blackdown 1.3-FCS and only the "classic" VM works. If I try client (the default) or server there is no error shown - it just hangs. It does this even with 'java -version'. >From the command line, I can either use java -classic or change jvm.cfg, but my main reason for upgrading was to get java support working in Mozilla (daily updates) but this "hardwires" the 'client' jvm. I am using Linux 2.4.0-test12-pre7 glibc 2.2 Pentium III UP -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: building 1.3? building anything? hello?
On Sun, 17 Dec 2000, Brian Craft wrote: > Is there anyone out there that knows how to build 1.3 or 1.2.2 on linux? Not on Linux, but I will admit to having built 1.2.2 on Solaris. > I've been researching this for a week & can't find anyone who will admit > to being able to do it. Is this some blackdown secret? > > One person suggested I get the 1.3 source from sun, but the build is just > as broken on 1.3 as on 1.2.2. Looks like 1.3 only builds with an old > libc. It particular, it messes with the ucontext in a way that doesn't work. Well, I cannot seem to find a Blackdown source release, so you're probably going to require some help from someone on the team at Sun to get their source release to build. My experiences with building 1.1.x and 1.2.x in the past suggest that you'd better be prepared for some frustration before getting a usable build. Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 2.4 kernel and threads
On Thu, 4 Jan 2001, Joseph Shraibman wrote: > >From README.linux to the 1.2.2 distrib: > * If you get OutOfMemory errors when you try create more than xxx > threads you'll have to increase the number of tasks supported by > kernel (the default is 256 per user). Change NR_TASKS in > /usr/src/linux/include/linux/tasks.h and recompile the kernel, or > upgrade to a 2.4 kernel. > > > So can I assume a 2.4 kernel has no limits on threads? 2.4 allows you to set the maximum number of threads/processes via the /proc file /proc/sys/kernel/threads-max. From poking around in the kernel source, the default limit is such that the maximum amount of memory consumed by all of the in-kernel thread structures would be half of physical memory (not virtual). As an example, on my 192M x86 machine that translates into a default maximum of 12288 threads (12288 * 8K/thread = 96M). Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: address for ibm jdk bugs?
On Sun, 7 Jan 2001, noisebrain wrote: > Hello, > > does anyone know an email address at ibm that would be interested > seeing bug reports in their jdk? I have one that happens in theirs > but not in blackdown or sun. No email address that I could find, but the "Reporting Problems" section in the file IBMJava2-13/docs/README.SDK.HTML suggests posting to the newsgroup ibm.software.java.linux on news://news.software.ibm.com. I used to check the newgroup regularly when I worked on JProbe (which was up until four months ago), and the developers did answer questions and take the occasional bug report. It helps a lot if you can give them a simple test program to reproduce the bug; hard to reproduce problems don't tend to stay on their radar unless you're somewhat persistent. Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.interlog.com/~scottm ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 2.4 kernel and threads
On Wed, 10 Jan 2001, Joseph Shraibman wrote: > On a related note, what is the difference between "buffered" and > "cached" memory? I've noticed that with 2.4.0 there is barely any > buffered. Nowadays the buffers are purely used for buffering access to the block devices (e.g. disks). Pretty much everything else is now cached in the page cache. > How have other people been finding 2.4.0? I've noticed that some things > like resizing windows are much slower. On most benchmarks I've seen, 2.4.0 is faster than 2.2.x. I've been running the test kernels for quite a while, and the VM tweaks that went in in the last few before 2.4.0 improved interactive response quite a bit. Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.spiteful.org (coming soon) ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: 2.4 kernel and threads
On Thu, 11 Jan 2001, Joseph Shraibman wrote: > Scott Murray wrote: > > > > Nowadays the buffers are purely used for buffering access to the block > > devices (e.g. disks). Pretty much everything else is now cached in the > > page cache. > > That still doesn't answer my question. If everything is treated as a > file, what is the difference between buffer and cache? Oops, that really wasn't clear, was it. More accurately, filesystem block meta-data (e.g. inodes) is cached in the buffer cache, and the logical file pages are cached in the page cache. Google found a reasonably good explanation (but for Solaris) with some diagrams at: http://sunsite.uakom.sk/sunworldonline/swol-07-1999/swol-07-filesystem3.html Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.spiteful.org (coming soon) ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Object sizes
On Mon, 28 May 2001, Miloslaw Smyk wrote: > Hi! > > Just a quick question: how can I calculate amount of memory that is used by > an object running under JVM on Linux? I don't work on JProbe anymore (I quit Sitraka about 9 months ago), but AFAIK both it and Optimize It! can tell you this. You can download free demos of both of them: JProbe - http://www.sitraka.com/software/jprobe/ Optimize It! - http://www.optimizeit.com/productinfo.html Good luck, Scott -- ===== Scott Murrayemail: [EMAIL PROTECTED] http://www.spiteful.org (coming soon) ICQ: 10602428 - "Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
