Re: java/2916: gcj 2.97 produces wrong code for concatenation and final

Re: java/2916: gcj 2.97 produces wrong code for concatenation and final

[tromey]

The following reply was made to PR java/2916; it has been noted by GNATS.

From: tromey@gcc.gnu.org
To: gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org, gcc-prs@gcc.gnu.org,
  iglewski@solidum.com, java-prs@gcc.gnu.org, nobody@gcc.gnu.org,
  tromey@gcc.gnu.org
Cc:  
Subject: Re: java/2916: gcj 2.97 produces wrong code for concatenation and final
Date: 14 Nov 2001 19:14:19 -0000

 Synopsis: gcj 2.97 produces wrong code for concatenation and final
 
 Responsible-Changed-From-To: unassigned->tromey
 Responsible-Changed-By: tromey
 Responsible-Changed-When: Wed Nov 14 11:14:18 2001
 Responsible-Changed-Why:
     I'm handling this.
 State-Changed-From-To: analyzed->closed
 State-Changed-By: tromey
 State-Changed-When: Wed Nov 14 11:14:18 2001
 State-Changed-Why:
     This appears to have been fixed on the trunk.
     The fix will appear in gcj 3.1.
     If you try the trunk and it doesn't work for you,
     tell me and I will reopen the PR.
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=2916&database=gcc

Re: java/2916: gcj 2.97 produces wrong code for concatenation and final

[tromey]

Synopsis: gcj 2.97 produces wrong code for concatenation and final

Responsible-Changed-From-To: unassigned->tromey
Responsible-Changed-By: tromey
Responsible-Changed-When: Wed Nov 14 11:14:18 2001
Responsible-Changed-Why:
    I'm handling this.
State-Changed-From-To: analyzed->closed
State-Changed-By: tromey
State-Changed-When: Wed Nov 14 11:14:18 2001
State-Changed-Why:
    This appears to have been fixed on the trunk.
    The fix will appear in gcj 3.1.
    If you try the trunk and it doesn't work for you,
    tell me and I will reopen the PR.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=2916&database=gcc

Re: java/2916: gcj 2.97 produces wrong code for concatenation and final

[green]

Synopsis: gcj 2.97 produces wrong code for concatenation and final

State-Changed-From-To: open->analyzed
State-Changed-By: green
State-Changed-When: Sat Jul 14 08:16:35 2001
State-Changed-Why:
    This problem still exists in the trunk.  It's definitely a
    bytecode compilation problem.  Compiling directly to native
    code works fine.  Also, Sun's javac will produce proper .class
    for gij or gcj.  gcj -C is at fault somehow.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=2916&database=gcc

java/2916: gcj 2.97 produces wrong code for concatenation and final

[iglewski]

>Number:         2916
>Category:       java
>Synopsis:       gcj 2.97 produces wrong code for concatenation and final
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed May 23 10:56:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Michal Iglewski <michal@solidum.com>
>Release:        unknown-1.0
>Organization:
>Environment:
linux
>Description:
The concatenation operator produces, in some (?) situations,
wrong code if the first arg is a final string.
Removing 'final' in the example below solves
the problem.
Replacing '+' by concat() also solves the problem.

public class ERROR {
  static public void error(String msg){
    System.err.println("Error: " + msg);
  }
  public static final String sth = "sth";
}

public class test {
  public static void main(String args[]){
		System.out.println(ERROR.sth+"aaaa");
		ERROR.error(ERROR.sth+"bbbb");
	}
}

$ gcj -C ERROR.java
$ gcj -C test.java
$ jar cf t.jar ERROR.class test.class
$ gcj -c -g  t.jar
$ gcj -o t t.o --main=test
$ ./t
s
Error: s
>How-To-Repeat:

>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: