Status update: I have identified some of the current bugs while building
OpenJDK7.
Here is the list:
http://svn.cs.lth.se/trac/jastadd-trac/ticket/16
http://svn.cs.lth.se/trac/jastadd-trac/ticket/18
http://svn.cs.lth.se/trac/jastadd-trac/ticket/19
http://svn.cs.lth.se/trac/jastadd-trac/ticket/22
http://svn.cs.lth.se/trac/jastadd-trac/ticket/23
As of yet I have only made progress with the method type signature bug.
/Jesper
On 11/28/2011 12:33 PM, Jesper Öqvist wrote:
> On 11/27/2011 09:07 PM, Na, Hyunik wrote:
>> Hello Jesper,
>>
>>
>>
>>
>>> I have seen the second error message. I solved that by removing all
>>> *.class files and compiling all *.java files at once using the _at_<file>
>>> option to provide the list of *.java files.
>>>
>> Probably, you meant 'detoured around' rather than 'solved'.
>>
>
> Yes, that would be more correct :) Although it did solve my current
> problem at that time!
>>
>>> I do not know if there is an error in the bytecode generation, or only the bytecode parsing of JastAddJ.
>>> I added a ticket for this error when I noticed it and continued without
>>> more investigation.
>>>
>>>
>> At least, bytecode generation seems to have some problems.
>>
>> While I use javac in compiling the 7638 java files in the "one-by-one" fashion and with -verbose option,
>> if I put the class files produced by Java7Backend (revision 9178)
>> in the foremost part of the boot class path, then I encounter many error messages from javac, such as
>>
>> -=-=-=
>>
>> ### compiling ./org/jcp/xml/dsig/internal/DigesterOutputStream.java
>> [parsing started ./org/jcp/xml/dsig/internal/DigesterOutputStream.java]
>> [parsing completed 122ms]
>> [search path for source files: .]
>> [search path for class files: ../byjastaddj/build,.,./build,/home/hina/myWork/research/openjdk-1.6.0.0.hina/build/linux-i586/classes,.]
>> [loading ../byjastaddj/build/java/io/ByteArrayInputStream.class]
>> [loading ../byjastaddj/build/java/io/InputStream.class]
>> [loading ../byjastaddj/build/java/io/OutputStream.class]
>> [loading ../byjastaddj/build/java/security/MessageDigest.class]
>> [loading ../byjastaddj/build/java/util/logging/Logger.class]
>> [loading ./java/util/logging/Level.java]
>> [parsing started ./java/util/logging/Level.java]
>> [parsing completed 14ms]
>> [loading ./com/sun/org/apache/xml/internal/security/utils/UnsyncByteArrayOutputStream.java]
>> [parsing started ./com/sun/org/apache/xml/internal/security/utils/UnsyncByteArrayOutputStream.java]
>> [parsing completed 3ms]
>> [loading ../byjastaddj/build/java/io/Closeable.class]
>> [loading ../byjastaddj/build/java/lang/Object.class]
>> [loading ../byjastaddj/build/java/io/Flushable.class]
>> [loading ../byjastaddj/build/java/io/ByteArrayOutputStream.class]
>> [loading ../byjastaddj/build/java/util/ResourceBundle.class]
>> [loading ../byjastaddj/build/java/io/Serializable.class]
>> [loading ../byjastaddj/build/java/util/ArrayList.class]
>> ./java/util/logging/Level.java:62: cannot access java.util.ArrayList
>> bad class file: ../byjastaddj/build/java/util/ArrayList.class
>> bad signature: ^H^_at_´^A^_at_^H, S
>> Please remove or make sure it appears in the correct subdirectory of the classpath.
>> private static java.util.ArrayList<Level> known = new java.util.ArrayList<Level>();
>> ^
>> ### result = 1, compiling ./org/jcp/xml/dsig/internal/DigesterOutputStream.java
>>
>> -=-=-=
>>
>> and
>>
>> -=-=-=
>>
>> ### compiling ./sun/rmi/server/MarshalInputStream.java
>> [parsing started ./sun/rmi/server/MarshalInputStream.java]
>> [parsing completed 125ms]
>> [search path for source files: .]
>> [search path for class files: ../byjastaddj/build,.,./build,/home/hina/myWork/research/openjdk-1.6.0.0.hina/build/linux-i586/classes,.]
>> [loading ../byjastaddj/build/java/io/IOException.class]
>> [loading ../byjastaddj/build/java/io/InputStream.class]
>> [loading ../byjastaddj/build/java/io/ObjectInputStream.class]
>> [loading ../byjastaddj/build/java/io/ObjectStreamClass.class]
>> [loading ../byjastaddj/build/java/io/StreamCorruptedException.class]
>> [loading ../byjastaddj/build/java/net/URL.class]
>> [loading ../byjastaddj/build/java/security/AccessControlException.class]
>> [loading ../byjastaddj/build/java/security/Permission.class]
>> [loading ./java/rmi/server/RMIClassLoader.java]
>> [parsing started ./java/rmi/server/RMIClassLoader.java]
>> [parsing completed 13ms]
>> [loading ../byjastaddj/build/java/io/ObjectInput.class]
>> [loading ../byjastaddj/build/java/io/DataInput.class]
>> [loading ../byjastaddj/build/java/lang/Object.class]
>> [loading ../byjastaddj/build/java/io/ObjectStreamConstants.class]
>> [loading ../byjastaddj/build/java/io/Closeable.class]
>> [loading ../byjastaddj/build/java/util/Map.class]
>> [loading ../byjastaddj/build/java/lang/Runnable.class]
>> [loading ../byjastaddj/build/java/lang/Class.class]
>> An exception has occurred in the compiler (1.6.0_20). Please file a bug at the Java Developer Connection (http://java.sun.com/webapps/bugreport) after checking the Bug Parade for duplicates. Include your program and the following diagnostic in your report. Thank you.
>> java.lang.AssertionError
>> at com.sun.tools.javac.code.Type.asMethodType(Type.java:365)
>> at com.sun.tools.javac.code.Type$ForAll.asMethodType(Type.java:1059)
>> at com.sun.tools.javac.jvm.ClassReader.readMemberAttr(ClassReader.java:865)
>> at com.sun.tools.javac.jvm.ClassReader.readMemberAttrs(ClassReader.java:1027)
>> ...
>> ...
>>
>> -=-=-=
>>
>> That is, javac fails to load some jastaddj-produced class files.
>> ( I attach the first half of the whole messages (gzipped) to this posting. )
>>
>> On the contrary, while I use Java7Backend compiler,
>> if I remove jastaddj-produced class files from the boot class path and
>> put the OpenJDK-produced class files there,
>> then I don't see any error messages regarding class loading.
>>
>>
>> -- Hyunik
>
> Thank you, your testing is very valuable! I was only interested in
> measuring compilation time with JastAddJ for my thesis, so I did not
> try to load any of the compiled class files in any JVM - mainly due to
> lack of time. I think you are correct that there is an error in the
> bytecode generation.
>
> On wednesday I might be able to debug some of this.
>
> /Jesper
>>
>>
>> _______________________________________________
>> JastAdd mailing list
>> JastAdd_at_cs.lth.se
>> https://mail1.cs.lth.se/cgi-bin/mailman/listinfo/jastadd
>
>
> _______________________________________________
> JastAdd mailing list
> JastAdd_at_cs.lth.se
> https://mail1.cs.lth.se/cgi-bin/mailman/listinfo/jastadd
>
Received on Wed Dec 07 2011 - 15:05:35 CET