Re: config/mips/trampolines.c on SGI
> Is there some documentation as to what this functions is *expected* to >do or some other means by which it can be achieved? Given context, >I'm sure I can fix it, I just don't have time to track down that >context from the sources. If you want to know how the trampoline function works or why it is needed, please see http://www2.biglobe.ne.jp/~inaba/trampolines.html By the way, the introduction of JIT code to mips machine is new and I have not yet tested on my machine. Kiyo
Re: [Re: Errors encountered in compiling Java Tutorial example]
> Still no luck! I've downloaded the following classes.zip: > 11550429 May 11 12:26 jdk_1.1.7-v3-glibc-x86.tar.gz There's no swing in jdk 1.1.x, you should get the swing-1.1.1fcs package from JavaSofts website. -- Peter C. Mehlitz Tel: +1 510 704 1527 Transvirtual Technologies, Inc., Fax: +1 510 704 1893 Berkeley, CA, USA.Email:[EMAIL PROTECTED]
Re: [Re: Errors encountered in compiling Java Tutorial example]
"Peter C. Mehlitz" <[EMAIL PROTECTED]> wrote: > > > ..ApptoAppl.java:11: class JLabel not found in class ApptoAppl.. > > Do you have the (JavaSoft) swing classes installed? They are not part of our > Klasses.jar (but we can run them). > > -- > Peter C. Mehlitz Tel: +1 510 704 1527 > Transvirtual Technologies, Inc., Fax: +1 510 704 1893 > Berkeley, CA, USA.Email: [EMAIL PROTECTED] Still no luck! I've downloaded the following classes.zip: 8828109 May 10 06:07 /usr/local/share/kaffe/classes.zip from: 11550429 May 11 12:26 jdk_1.1.7-v3-glibc-x86.tar.gz And here's the same problem with the latest Kaffe: javac -verbose ApptoAppl.java [ START COMPILATION IN VERBOSE MODE ] [ 1307ms TO PARSE ApptoAppl.java ] [678ms TO CHECK INTERFACE OF ] ApptoAppl.java:11: Can''t find class "JLabel" §8 [250ms TO CHECK BODY OF ApptoAppl.java ] javac -version KOPI Java Compiler Version: 1.3C What am I missing? It's got to be a trivial config. problem !! Thanks for any help, [EMAIL PROTECTED] Get free email and a permanent address at http://www.netaddress.com/?N=1
Re: 'make build-classes' error for kaffe-1.0.5
Thank you for your answer, but I try the latest CVS and still have the same problem. Do I miss something? On Thu, 28 Oct 1999, Archie Cobbs wrote: > Ren-Song Ko writes: > > I couldn't make build-classes for those extension javalib in > > kaffe-1.0.5 on PC Linux and Sparc Solaris. The following are last couple of > > lines of message I had when I tried to rebuild rmi.jar, > > I believe Tim has fixed this since 1.0.5 .. try the latest CVS. > > -Archie > > ___ > Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com -- == Ren-Song Ko, Department of Computer Science, MSU [EMAIL PROTECTED], http://www.cps.msu.edu/~korenson !!! Error 3032 - Recursion error. See error 3032. !!! "There are only two things which are infinite, the universe and human stupidity,and I'm not really sure about the universe" Albert Einstein ==
About Kaffe
Hello Core Team. My name`s Dani, from UPC (Politechnical University of Catalunya, Spain). Let me tell you my problem directly. I`m working with Kaffe for an Alpha machine. I would like to use the JIT option, but it looks like it still doesn`t work. But I`m confused because in the README file it`s written that Kaffe forr Alpha supports JIT. The configure program don`t enable the option. And forcing it neither works. Will this feature be soon available? Or it`s me that I don`t know how to make it run properly. I don`t know. Could you help me, please? Thanks. Well, another question. I `have to` make a deep analysis of the Kaffe Virtual Machine, describing its techniques for each module, etc. I `have to` write a documentation about it to let more people work with it. To achieve it, I must understand all the code, and as it`s C, it`s never easy. Do you know if exist any documentation yet? Could you tell where I can get it? You would `save` me from losing a lot of time doing such `dirty` work, if it has been done yet, instead of using your marvellous piece of code ;) Now, seriously, could you help me? Lots of thanks to you. Bye. --Dani , the hardworker... :((
Re: 'make build-classes' error for kaffe-1.0.5
Ren-Song Ko writes: > I couldn't make build-classes for those extension javalib in > kaffe-1.0.5 on PC Linux and Sparc Solaris. The following are last couple of > lines of message I had when I tried to rebuild rmi.jar, I believe Tim has fixed this since 1.0.5 .. try the latest CVS. -Archie ___ Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com
'make build-classes' error for kaffe-1.0.5
Hi, I couldn't make build-classes for those extension javalib in kaffe-1.0.5 on PC Linux and Sparc Solaris. The following are last couple of lines of message I had when I tried to rebuild rmi.jar, kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/net/Socket" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/net/InetAddres s" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/net/UnknownHos tException" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/lang/reflect/C onstructor" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/lang/reflect/M ethod" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/lang/reflect/I nvocationTargetException" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/lang/Thread" § 8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/lang/Exception " §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/io/IOException " §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/io/DataInputSt ream" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/io/DataOutputS tream" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/io/ObjectInput Stream" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "java/io/ObjectOutpu tStream" §8 kaffe/rmi/server/UnicastServerRef.java:1: Can''t find class "kaffe/util/LongHash table" §8 kaffe/rmi/server/UnicastServerRef.java:78: Can''t find class "Object" §8 gmake: *** [lib/stamp] Error 1 == Ren-Song Ko, Department of Computer Science, MSU [EMAIL PROTECTED], http://www.cps.msu.edu/~korenson !!! Error 3032 - Recursion error. See error 3032. !!! "There are only two things which are infinite, the universe and human stupidity,and I'm not really sure about the universe" Albert Einstein ==
Re: patch to java.io.FileOutputStream
[EMAIL PROTECTED] writes: > I show a short patch to FileOutputStream.java. > > According to JLS, the constructers of FileOutputStream issue > FileNotFoundException rather than IOException. Thanks for the patch.. though according to http://java.sun.com/products/jdk/1.2/docs/api/java/io/FileOutputStream.html only the constructors taking a String argument throw FileNotFoundException. Will commit. -Archie ___ Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com
config/mips/trampolines.c on SGI
In the process of re-adapting kaffe for SGI Irix 6.x I've run into a problem with the trampoline definition. It seems that the lines: sw $31, 0($sp) \n l1: .word " C_FUNC_NAME(soft_fixup_trampoline) " \n .globl mips_do_fixup_trampoline\n .entmips_do_fixup_trampoline\n mips_do_fixup_trampoline: \n cause a problem with the SGI assembler in which it refuses to compile with the following error: /bin/sh ../../../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../../../config -I../../../include -I../../../kaffe/kaffevm/systems/unix-jthreads -I.. -I./.. -DTRANSLATOR -DJIT3 -DKVER='"1.0.5"' -I/home/canada2/leei/src/gcc/kaffe-cvs/./kaffe/kaffevm -I/home/canada2/leei/src/gcc/kaffe-cvs/./kaffe/kaffevm/systems/unix-jthreads -I../../../config -I../../../include -I/usr/local/include -g -O2 -Wall -Wstrict-prototypes -c -o machine.lo machine.c gcc -DHAVE_CONFIG_H -I. -I. -I../../../config -I../../../include -I../../../kaffe/kaffevm/systems/unix-jthreads -I.. -I./.. -DTRANSLATOR -DJIT3 -DKVER=\"1.0.5\" -I/home/canada2/leei/src/gcc/kaffe-cvs/./kaffe/kaffevm -I/home/canada2/leei/src/gcc/kaffe-cvs/./kaffe/kaffevm/systems/unix-jthreads -I../../../config -I../../../include -I/usr/local/include -g -O2 -Wall -Wstrict-prototypes -c machine.c -DPIC -Wp,-MD,.deps/machine.TPlo -o machine.lo as: Error: /var/tmp/cc0Bxeo8.s, line 12706: Cannot use data-generating directives in the .text section.: .word as: Error: /var/tmp/cc0Bxeo8.s, line 12706: Cannot use data-generating directives in the .text section.: .word make[3]: *** [machine.lo] Error 1 Is there some documentation as to what this functions is *expected* to do or some other means by which it can be achieved? Given context, I'm sure I can fix it, I just don't have time to track down that context from the sources. --- Lee Iverson SRI International [EMAIL PROTECTED] 333 Ravenswood Ave., Menlo Park CA 94025 http://www.ai.sri.com/~leei/(650) 859-3307
patch to java.io.FileOutputStream
Hello,
I show a short patch to FileOutputStream.java.
According to JLS, the constructers of FileOutputStream issue
FileNotFoundException rather than IOException.
Also, I'm afraid to less null-check for FileDescriptor and
SecurityManager... But I don't add any modifying about this probrem
yet.
Thanks,
Jiro
Index: FileOutputStream.java
===
RCS file: /home/cvspublic/kaffe/libraries/javalib/java/io/FileOutputStream.java,v
retrieving revision 1.4
diff -r1.4 FileOutputStream.java
24c24
< public FileOutputStream(File file) throws IOException
---
> public FileOutputStream(File file) throws FileNotFoundException
30c30
< {
---
> {
35c35
< public FileOutputStream(String name) throws IOException
---
> public FileOutputStream(String name) throws FileNotFoundException
40c40
< public FileOutputStream(String name, boolean append) throws IOException
---
> public FileOutputStream(String name, boolean append) throws FileNotFoundException
Kaffe 1.0.5 profiling patch
We have a patch for the new jit3 system in Kaffe 1.0.5 to generate profiling information for the x86 processors. The code has been tested on Pentium II and Pentium III processors. The code has the same limitation with multithreaded code and exceptions as earlier versions of the instrumentation. The patch can be obtained from: http://www.eb.uah.edu/~crash/javanauts/tools/prof.1.0.5.patch Because we are looking at the code generated, the debugging statements in the jit3-i386.def have been reordered to provide a better idea of instruction locations in the generated code. Output has also been added so it is easier to find where the code for a specific method starts. -Will Cohen
