[Bug libgcj/20135] New: Returning true from LinkedHashMap.removeEldestEntry does not remove entry

2005-02-21 Thread thhal at mailblocks dot com 
The LinkedHashMap fails to remove the eldest entry when returning true from the
removeEldestEntry method.

-- 
   Summary: Returning true from LinkedHashMap.removeEldestEntry does
not remove entry
   Product: gcc
   Version: 3.4.2
Status: UNCONFIRMED
  Severity: normal
  Priority: P1
 Component: libgcj
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: thhal at mailblocks dot com
CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20135

[Bug libgcj/20135] Returning true from LinkedHashMap.removeEldestEntry does not remove entry

2005-02-21 Thread thhal at mailblocks dot com 

--- Additional Comments From thhal at mailblocks dot com  2005-02-22 07:31 
---
Created an attachment (id=8251)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8251&action=view)
Code to reproduce the problem

Compile using gcj --main foo and run the result. Should print "ERROR, Size is
20". Compile and run using a standard Java, should print "OK, Size is 15".

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20135

[Bug libgcj/20136] New: Endless loop when using a SecurityManager

2005-02-21 Thread thhal at mailblocks dot com 
Installing a SecurityManager causes subsequent attempts to obtain a ClassLoader
to go into an endless loop.

-- 
   Summary: Endless loop when using a SecurityManager
   Product: gcc
   Version: 3.4.2
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: libgcj
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: thhal at mailblocks dot com
CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20136

[Bug libgcj/20136] Endless loop when using a SecurityManager

2005-02-21 Thread thhal at mailblocks dot com 

--- Additional Comments From thhal at mailblocks dot com  2005-02-22 07:46 
---
Created an attachment (id=8252)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8252&action=view)
Sample that reproduces the problem

Use gcj --main foo and run the resulting file. It never exits. Run it with a
standard java, it prints "OK, No endless loop"

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20136

[Bug libgcj/20193] New: Variable names incorrectly resolved in inner classes.

2005-02-24 Thread thhal at mailblocks dot com 
An inner class that references a static variable in an outer class will fail if
the super class of that outer class defines the same variable with a visibility
that makes it inaccessible to the inner class.

-- 
   Summary: Variable names incorrectly resolved in inner classes.
   Product: gcc
   Version: 4.0.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: libgcj
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: thhal at mailblocks dot com
CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20193

[Bug libgcj/20193] Variable names incorrectly resolved in inner classes.

2005-02-24 Thread thhal at mailblocks dot com 

--- Additional Comments From thhal at mailblocks dot com  2005-02-24 15:16 
---
Created an attachment (id=8271)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=8271&action=view)
Java class that illustrates the problem.

Try gcj -c ScopeProblem.java. It fails with:
ScopeProblem.java:12: error: Can't access package-private field ...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20193

[Bug libgcj/18893] New: null pointer exception in the Logger.getCallerStackFrame()

2004-12-08 Thread thhal at mailblocks dot com 
I have a problem using the java.util.logging.Logger from a Java function  that
originates from a JNI call. Apparently I get a null pointer exception in the
Logger.getCallerStackFrame() method. This is the stacktrace:

   at
_ZN4java4util7logging6Logger19getCallerStackFrameEv(/usr/lib64/libgcj.so.5.0.0)
   at
_ZN4java4util7logging6Logger3logEPNS1_5LevelEPNS_4lang6StringEP6JArrayIPNS5_6ObjectEE(/usr/lib64/libgcj.so.5.0.0)
   at
_ZN4java4util7logging6Logger3logEPNS1_5LevelEPNS_4lang6StringE(/usr/lib64/libgcj.so.5.0.0)
   at
_ZN4java4util7logging6Logger4infoEPNS_4lang6StringE(/usr/lib64/libgcj.so.5.0.0)
   at org.postgresql.pljava.example.Parameters.print(java.sql.Date)(Unknown 
Source)
   at
_ZN16_Jv_InterpMethod9run_classEP7ffi_cifPvP7ffi_rawS2_(/usr/lib64/libgcj.so.5.0.0)
   at
_Z18_Jv_CallAnyMethodAPN4java4lang6ObjectEPNS0_5ClassEP10_Jv_MethodbbP6JArrayIS4_EP6jvalueSB_b(/usr/lib64/libgcj.so.5.0.0)
E

Andrew Haley reports that this cannot be reproduced with 4.0-pre so perhaps this
has been fixed already.

-- 
   Summary: null pointer exception in the
Logger.getCallerStackFrame()
   Product: gcc
   Version: 3.4.3
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: libgcj
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: thhal at mailblocks dot com
CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
dot org
  GCC host triplet: Linux 2.4.21 on x86_64


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18893

[Bug libgcj/18893] null pointer exception in the Logger.getCallerStackFrame()

2004-12-08 Thread thhal at mailblocks dot com 

--- Additional Comments From thhal at mailblocks dot com  2004-12-08 21:13 
---
Created an attachment (id=7707)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7707&action=view)
test case

type make, then run the resulting executeable gcjtest

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18893