Re: c++/1804: c++ -c -I. -I.. -I../liboctave -I../src -I../libcruft/misc -I../glob -I../glob -DHAVE_CONFIG_H -fno-rtti -fno-exceptions -fno-implicit-templates -g -O2 -Wall sighandlers.cc

[Neil Booth <neil@daikokuya.demon.co.uk>]

The following reply was made to PR c++/1804; it has been noted by GNATS.

From: Neil Booth <neil@daikokuya.demon.co.uk>
To: A.Abbing@marin.nl
Cc: gcc-gnats@gcc.gnu.org
Subject: Re: c++/1804: c++ -c  -I. -I.. -I../liboctave -I../src -I../libcruft/misc  -I../glob -I../glob -DHAVE_CONFIG_H -fno-rtti -fno-exceptions -fno-implicit-templates -g -O2 -Wall sighandlers.cc
Date: Mon, 29 Jan 2001 19:11:55 +0000

 A.Abbing@marin.nl wrote:-
 > 
 > Compilation of the GNU octave source file octave-2.0.16/src/sighandlers.cc produces
 > an internal compiler error.
 
 *Please* provide preprocessed source.  You cannot expect bug-fixers to
 trawl the web looking for source code, then download and configure it
 and try to compile it.  See
 
 http://gcc.gnu.org/bugs.html
 
 Neil.
>From praun@inf.ethz.ch Sun Apr 01 00:00:00 2001
From: praun@inf.ethz.ch
To: gcc-gnats@gcc.gnu.org
Cc: praun@inf.ethz.ch
Subject: java/2280: gcj -C does not translate Java 'finally' correctly 
Date: Sun, 01 Apr 2001 00:00:00 -0000
Message-id: <200103132049.f2DKnMB10765@femme.inf.ethz.ch>
X-SW-Source: 2001-q1/msg02298.html
Content-length: 2623

>Number:         2280
>Category:       java
>Synopsis:       gcj -C does not translate Java 'finally' correctly
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Tue Mar 13 12:56:00 PST 2001
>Closed-Date:
>Last-Modified:
>Originator:     
>Release:        3.0 20010305 (prerelease)
>Organization:
ETH Zurich	
>Environment:
System: Linux femme 2.2.16-SMP #1 SMP Wed Sep 27 18:18:28 GMT 2000 i686 unknown
Architecture: i686

	
host: i686-pc-linux-gnu
build: i686-pc-linux-gnu
target: i686-pc-linux-gnu
configured with: /project/gcc-3.0/configure --enable-languages=java,c++,c --prefix=/project/gcc-3.0-install --enable-threads=posix --srcdir=/project/gcc-3.0
>Description:
	Problem occurred during compilation of java/io/ObjectOutputStream::writeObject
        ... java source of writeObject:
        catch (IOException e)
    {
      realOutput.writeByte (TC_EXCEPTION);
      reset (true);

      try
      {
        writeObject (e);
      }
      catch (IOException ioe)
      {
        throw new StreamCorruptedException ("Exception " + ioe + " thrown while 
exception was being written to stream.");
      }

      reset (true);
    }
    finally
    {
      isSerializing = was_serializing;

      if (! was_serializing)
        setBlockDataMode (true);
    }
  }

  Resulting Byteocde with gcj -C:

   830 invokespecial #289 <Method java.io.StreamCorruptedException(java.lang.Strin
g)>
 833 athrow
 834 aload_0
 835 iconst_1
 836 invokespecial #269 <Method void reset(boolean)>
 839 jsr 850
 842 astore 4    <---------------  THERE should be a GOTO
 844 jsr 850
 847 aload 4
 849 athrow
 850 astore_3
 851 aload_0
 852 iload_2
 853 putfield #90 <Field boolean isSerializing>
 856 iload_2
 857 ifne 865
 860 aload_0
 861 iconst_1
 862 invokespecial #65 <Method void setBlockDataMode(boolean)>
 865 ret 3
 867 return

  Resulting Bytecode with javac/jikes:

 831 invokespecial #88 <Method java.io.StreamCorruptedException(java.lang.String
)>
 834 athrow
 835 aload_0
 836 iconst_1
 837 invokespecial #79 <Method void reset(boolean)>
 840 jsr 854
 843 goto 872   <------------------------
 846 astore 11
 848 jsr 854
 851 aload 11
 853 athrow
 854 astore 12
 856 aload_0
 857 iload_2
 858 putfield #16 <Field boolean isSerializing>
 861 iload_2
 862 ifne 870
 865 aload_0
 866 iconst_1
 867 invokespecial #1 <Method void setBlockDataMode(boolean)>
 870 ret 12
 872 return

   

>
>How-To-Repeat:
	gcj -C ObjectOutputStream.java
>Fix:
	unknown.
>Release-Note:
>Audit-Trail:
>Unformatted: