java/3602: Problems with bytes

java/3602: Problems with bytes

[Anthony Green]

>Number:         3602
>Category:       java
>Synopsis:       Problems with bytes
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Sat Jul 07 02:56:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Anthony Green
>Release:        3.1 20010705 (experimental)
>Organization:
>Environment:
System: Linux fencer.cygnus.com 2.2.12-20smp #1 SMP Mon Sep 27 10:34:45 EDT 1999 i686 unknown
Architecture: i686

	
host: i686-pc-linux-gnu
build: i686-pc-linux-gnu
target: i686-pc-linux-gnu
configured with: ../gcc/configure --enable-language=c,c++,java --enable-threads --prefix=/horton/green/net/i : (reconfigured) 
>Description:
	gcj is having problems with bytes.  See the attached program.  I'm seeing..
OK
Bad
-73

	That last number should be 0xB7.  -73 is 0xFFFFFFB7. :-(

	I don't think there's any work-around.
>How-To-Repeat:
	Compile the following and run...

public class Bytes {

  public static void foo (byte b)
  {
    if (b == (byte)0xB7)
      {
	System.out.println ("OK");
      }
    
    switch (b)
      {
      case INVOKESPECIAL:
	System.out.println ("Match");
      default:
	System.out.println ("Bad");
      }

    System.out.println (b);
  }

  public static void main (String args[])
  {
    foo (INVOKESPECIAL);
  }

  public static final byte
    PUTSTATIC = (byte)0xB3,
    GETFIELD = (byte)0xB4,
    PUTFIELD = (byte)0xB5,
    INVOKEVIRTUAL = (byte)0xB6,
    INVOKESPECIAL = (byte)0xB7,
    INVOKESTATIC = (byte)0xB8,
    INVOKEINTERFACE = (byte)0xB9,
    XXXUNUSEDXXX = (byte)0xBA,
    NEW = (byte)0xBB,
    NEWARRAY = (byte)0xBC,
    ANEWARRAY = (byte)0xBD,
    ARRAYLENGTH = (byte)0xBE,
    ATHROW = (byte)0xBF;
}


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

Re: java/3602: Problems with bytes

[green]

Synopsis: Problems with bytes

State-Changed-From-To: feedback->closed
State-Changed-By: green
State-Changed-When: Fri Jul 13 20:06:34 2001
State-Changed-Why:
    Fixed - thanks!
    
    rhino (the Mozilla javascript interpreter) appears to work a little bit now...
    
    fencer[132] ./RunScript 123+2354234/2345+34
    1160.9377398720683
    
    It operates somewhat like kava (compiling to bytecode and interpreting).

http://sources.redhat.com/cgi-bin/gnatsweb.pl?cmd=view&pr=3602&database=gcc

Re: java/3602: Problems with bytes

[apbianco]

Synopsis: Problems with bytes

Responsible-Changed-From-To: unassigned->apbianco
Responsible-Changed-By: apbianco
Responsible-Changed-When: Fri Jul 13 15:27:37 2001
Responsible-Changed-Why:
    Mine.
State-Changed-From-To: open->feedback
State-Changed-By: apbianco
State-Changed-When: Fri Jul 13 15:27:37 2001
State-Changed-Why:
    Here's a patch I'm testing at the moment.

http://sources.redhat.com/cgi-bin/gnatsweb.pl?cmd=view&pr=3602&database=gcc

Re: java/3602: Problems with bytes

[Alexandre Petit-Bianco]

The following reply was made to PR java/3602; it has been noted by GNATS.

From: Alexandre Petit-Bianco <apbianco@cygnus.com>
To: Anthony Green <green@cygnus.com>
Cc: gcc-gnats@gcc.gnu.org
Subject: Re: java/3602: Problems with bytes
Date: Mon, 9 Jul 2001 11:11:00 -0700 (PDT)

 Anthony Green writes:
 
 > OK
 > Bad
 > -73
 
 I see we're not handling the switch statement very well with negative
 byte values, but I think the last println is fine (b = 0xB7 = -73.)
 
 I'll look into the switch problem.
 
 ./A