sablevm-developer — SableVM developer mailing list

[Sablevm-developer] Bug#238768: sablevm should provide java2-runtime

[Etienne G. <gag...@uq...>]

Package: sablevm
Version: 1.1.0-5
Severity: normal

If you use dselect to install "ant", dselect will try to install
another JVM than sablevm (such as Sun's non-free JDK) to run it,
because sablevm does not "provide" java2-runtime.

Yet, sablevm should be sufficient to run ant and many other Java
applications.  SableVM does implement the "Java 2" specification, and
thus should provide java2-runtime, even though it is incomplete;
incompleteness is simply an incentive to contribute more to this Free
project.  Not "providing java2-runtime" excludes sablevm as a runtime
system for software such ant, which goes against Debian's objective to
promote Free software (such as SableVM) over non-free software (such
as Sun's JDK).

Etienne


-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.24
Locale: LANG=C, LC_CTYPE=C

Versions of packages sablevm depends on:
ii  java-common                 0.22         Base of all Java packages
ii  libc6                       2.3.2.ds1-11 GNU C Library: Shared libraries an
ii  libffi2                     1:3.3.3-3    Foreign Function Interface library
ii  libltdl3                    1.5.2-1      A system independent dlopen wrappe
ii  libpopt0                    1.7-4        lib for parsing cmdline parameters
ii  libsablevm1                 1.1.0-5      Free implementation of JVM second 
ii  unzip                       5.50-3       De-archiver for .zip files

-- no debconf information

[Sablevm-developer] Bug#238651: java-sablevm dequotes parameters

[Grzegorz B. P. <ga...@de...>]

Hi Ilguiz

You're absolutely right, these changes fix the problem (I tested it with
some other shell scripts).

This will go into 1.1.1-1 upload (hopefuly soon).

Thanks a lot,

				Grzegorz B. Prokopski

PS: I think we'd also get similar breakage if we ex. try to use a path
for CLASSPATH with spaces in the name, or if we try to set value of
a property to something with spaces or special chars... Yes, shell is
far from ideal here, but I'll try to test and fix some such problems in
the next version.

If it's not possible given shell syntax - I guess we're gonna switch to
perl, or even better - write a *binary* wrapper (in C).

Hmm... that's a nice idea! and not hard to do. any takers? 


On Thu, 2004-03-18 at 01:51, Ilguiz Latypov wrote:
> Package: sablevm
> Version: 1.1.0-5
> 
> The /usr/bin/java-sablevm shell script dequotes parameters submitted by
> the user's command line.  E.g., running
> 
>   java mypackage.MyApp "select * from table"
> 
> will yield only the word "select" in args[1] of the MyApp's main(String
> args[]) method.  
> 
> I believe the lines
> 
>     $echo $JVM -Y $CLASSPATH $OPTS $REST $@
>     $exec $JVM -Y $CLASSPATH $OPTS $REST $@
> 
> can be replaced with
> 
>     $echo $JVM -Y $CLASSPATH $OPTS $REST "$@"
>     $exec $JVM -Y $CLASSPATH $OPTS $REST "$@"
> 
> Also, the line
> 
>       EST="$REST $1"
> 
> should probably read
> 
>       REST="$REST $1"
> 
> Regards,

[Sablevm-developer] Bug#238651: java-sablevm dequotes parameters

[Ilguiz L. <ila...@in...>]

Package: sablevm
Version: 1.1.0-5

The /usr/bin/java-sablevm shell script dequotes parameters submitted by
the user's command line.  E.g., running

  java mypackage.MyApp "select * from table"

will yield only the word "select" in args[1] of the MyApp's main(String
args[]) method.  

I believe the lines

    $echo $JVM -Y $CLASSPATH $OPTS $REST $@
    $exec $JVM -Y $CLASSPATH $OPTS $REST $@

can be replaced with

    $echo $JVM -Y $CLASSPATH $OPTS $REST "$@"
    $exec $JVM -Y $CLASSPATH $OPTS $REST "$@"

Also, the line

      EST="$REST $1"

should probably read

      REST="$REST $1"

Regards,

-- 
Ilguiz Latypov

tel.  +1 (519) 569-8746
66 Wildlark Crescent
Kitchener, Ontario N2N 3E9

Re: [Sablevm-developer] boot.library.path vs. boot.class.path

[Archie C. <ar...@de...>]

Chris Pickett wrote:
> --property=sablevm.boot.class.path=/some-transformed-path/
> 
> and
> 
> --property=sablevm.boot.library.path=/some-transformed-path/
> 
> do, or at least tell me where to look?  In other words, what is the 
> difference between them?

A class path is a search path for Java class files, while a library
path is a search path for shared library files (e.g., ELF files), such
as may be needed by classes that rely on external native code.

-Archie

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com

[Sablevm-developer] boot.library.path vs. boot.class.path

[Chris P. <chr...@ma...>]

Hi!

Can somebody explain what specifying each of

--property=sablevm.boot.class.path=/some-transformed-path/

and

--property=sablevm.boot.library.path=/some-transformed-path/

do, or at least tell me where to look?  In other words, what is the 
difference between them?

Cheers,
Chris

[Sablevm-developer] Bug#235838: marked as done (sablevm: error while loading shared libraries)

[<ow...@bu...>]

Your message dated Mon, 15 Mar 2004 14:14:19 -0500
with message-id <1079378059.1164.7.camel@glue>
and subject line Closing: error while loading share libraries
has caused the attached Bug report to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere.  Please contact me immediately.)

Debian bug tracking system administrator
(administrator, Debian Bugs database)

--------------------------------------
Received: (at submit) by bugs.debian.org; 2 Mar 2004 18:59:45 +0000
>From la...@ao... Tue Mar 02 10:59:45 2004
Return-path: <la...@ao...>
Received: from aopsysadsl.net1.nerim.net (stan.aopsys.com) [62.212.101.94] 
	by spohr.debian.org with esmtp (Exim 3.35 1 (Debian))
	id 1AyF7I-00010Z-00; Tue, 02 Mar 2004 10:59:44 -0800
Received: from laurent by stan.aopsys.com with local (Exim 3.36 #1 (Debian))
	id 1AyEvs-0005Mv-00; Tue, 02 Mar 2004 19:47:56 +0100
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: Laurent Martelli <la...@be...>
To: Debian Bug Tracking System <su...@bu...>
Subject: sablevm: error while loading shared libraries
X-Mailer: reportbug 2.48
Date: Tue, 02 Mar 2004 19:47:56 +0100
Message-Id: <E1A...@st...>
Sender: Laurent Martelli <la...@ao...>
Delivered-To: su...@bu...
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_01 
	(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Status: No, hits=-5.0 required=4.0 tests=HAS_PACKAGE autolearn=no 
	version=2.60-bugs.debian.org_2004_03_01
X-Spam-Level: 

Package: sablevm
Version: 1.0.9-3
Severity: grave
Justification: renders package unusable

laurent@stan:~$ sablevm 
sablevm: error while loading shared libraries: libsablevm.so.1: cannot open shared object file: No such file or directory

Make /usr/lib/libsablevm.so.1 a symbolic link to
/usr/lib/libsablevm.so.0 seems to fix the problem.

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (990, 'testing'), (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.4.22-pre8
Locale: LANG=fr_FR@euro, LC_CTYPE=fr_FR@euro (ignored: LC_ALL set to fr_FR@euro)

Versions of packages sablevm depends on:
ii  java-common                0.22          Base of all Java packages
ii  libc6                      2.3.2.ds1-10  GNU C Library: Shared libraries an
ii  libffi2                    1:3.3.2-0pre5 Foreign Function Interface library
ii  libltdl3                   1.5.2-1       A system independent dlopen wrappe
ii  libpopt0                   1.7-2         lib for parsing cmdline parameters
ii  libsablevm1                1.1.0-2       Free implementation of JVM second 
ii  unzip                      5.50-3        De-archiver for .zip files

-- no debconf information


---------------------------------------
Received: (at 235838-done) by bugs.debian.org; 15 Mar 2004 21:46:32 +0000
>From ga...@de... Mon Mar 15 13:46:32 2004
Return-path: <ga...@de...>
Received: from anis.telecom.uqam.ca [132.208.250.6] 
	by spohr.debian.org with esmtp (Exim 3.35 1 (Debian))
	id 1B2zuq-0005gs-00; Mon, 15 Mar 2004 13:46:32 -0800
Received: from anis4.telecom.uqam.ca (anis4.telecom.uqam.ca [132.208.250.236])
	by sortant.uqam.ca (8.12.10/8.12.1) with SMTP id i2FLfBLF014704
	for <235...@bu...>; Mon, 15 Mar 2004 16:41:58 -0500 (EST)
Received: from antivirus.uqam.ca ([132.208.250.6])
 by anis4.telecom.uqam.ca (SAVSMTP 3.1.1.32) with SMTP id M2004031516430520725
 for <235...@bu...>; Mon, 15 Mar 2004 16:43:05 -0500
Received: from glue.info.uqam.ca (glue.info.uqam.ca [132.208.135.95])
	by intrant.uqam.ca (8.12.10/8.12.2/uqam-filtres) with ESMTP id i2FLd5o7004341
	for <235...@bu...>; Mon, 15 Mar 2004 16:39:07 -0500 (EST)
X-Spam-Filter: Filtre-Uqam re:  ab...@uq...
Subject: Closing: error while loading share libraries
From: "Grzegorz B. Prokopski" <ga...@de...>
To: 235...@bu...
Content-Type: text/plain
Message-Id: <1079378059.1164.7.camel@glue>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.4.5 
Date: Mon, 15 Mar 2004 14:14:19 -0500
Content-Transfer-Encoding: 7bit
Delivered-To: 235...@bu...
X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_12 
	(1.212-2003-09-23-exp) on spohr.debian.org
X-Spam-Status: No, hits=-2.0 required=4.0 tests=BAYES_00 autolearn=no 
	version=2.60-bugs.debian.org_2004_03_12
X-Spam-Level: 

This bug has been fixed since when SableVM 1.1.0 Debian packages
migrated to testing.

				GBP

Re: [Sablevm-developer] Deserialization of arrays of objects.

[David <db...@cs...>]

On Sun, Mar 14, 2004 at 04:40:29PM -0500, James Damour wrote:
>=20
> The source code in the attached file, when compiled with Sun's JDK
> v1.3.1 throws "java.lang.ClassNotFoundException: Tryit$Inne
>    at gnu.java.lang.SystemClassLoader.findClass
> (SystemClassLoader.java:79)" when run under SableVM v1.1.0 (Debian
> testing).
>=20

Hi James,

This bug as well as a few others concerning reflection on array class
types have been fixed in SableVM/staging and will probably made their
way into next SableVM release 1.1.1 that will be available some time
this week.

I can run your sample code and the output is below.  Note that I get
warning about some files.  I don't know much about these missing files
warnings.  I guess SableVM would need to provide these some day.

--------------------
Before serialization
Value #0 : 0
Value #1 : 1
Value #2 : 2
Value #3 : 3
Value #4 : 4
Value #5 : 5
Value #6 : 6
Value #7 : 7
Value #8 : 8
Value #9 : 9
WARNING: could not properly read security provider files:
         file:///home/david/local/sablevm-jit/lib/security/SableVM.securi=
ty
         file:///home/david/local/sablevm-jit/lib/security/classpath.secu=
rity
         Falling back to standard GNU security provider
After serialization
Value #0 : 0
Value #1 : 1
Value #2 : 2
Value #3 : 3
Value #4 : 4
Value #5 : 5
Value #6 : 6
Value #7 : 7
Value #8 : 8
Value #9 : 9
--------------


David

---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Java-gnome-developer] Re: [Sablevm-developer] Re: [kaffe] Ja va-Gnome: jni or cni

[Archie C. <ar...@de...>]

Jeffrey Morgan wrote:
> > Actually, I reckon that the same argument applies to java-gnome in
> > general.  Anyone who uses the java-gnome libraries in a GUI-based Java
> > application is risking tying that application into GNOME.  By 
> > contrast,
> > if they use Swing or SWT, they can easily port the 
> > application across a
> > wide range of supported platforms.
> 
> But this is the point of the whole project.  Our goal is to target
> gnome and gtk environments.  java-gnome is for writing applications
> that target the gnome desktop environment.  As a result, we have
> many classes that are designed to provide the developer the type
> of integration that they would expect on a platform if they were
> using c or c++.

If you choose CNI, then please change the name of your project
(as suggested by someone else) to "gcj-gnome" from "java-gnome"
to avoid confusion.

Thanks,
-Archie

__________________________________________________________________________
Archie Cobbs      *        CTO, Awarix        *      http://www.awarix.com

Re: [Sablevm-developer] Deserialization of arrays of objects.

[David <db...@cs...>]

On Sun, Mar 14, 2004 at 04:40:29PM -0500, James Damour wrote:
> I ran into a bug in SableVM v1.0.9 on February second with
> deserialization of objects in arrays.  The bug is still in v1.1.0, so I
> thought I'd resend my proposed fix, along with a simple testcase to sho=
w
> the bug.
>=20
> Basically, the deserialization code makes an assumption about the lengt=
h
> of the classname in the bytestream that isn't correct for arrays of
> objects.
>=20
> The source code in the attached file, when compiled with Sun's JDK
> v1.3.1 throws "java.lang.ClassNotFoundException: Tryit$Inne
>    at gnu.java.lang.SystemClassLoader.findClass
> (SystemClassLoader.java:79)" when run under SableVM v1.1.0 (Debian
> testing).
>=20
> I haven't tested the following "obvious" diff (I don't have the sources
> installed), but it shows what I *think* is the offending line of code.=20
> I'm sorry that I can't submit a fully tested patch, but I'm tainted
> (I've read Sun's source), so this is the closest I dare come.  I'd be
> happy to take whatever steps are suggested to help debug this issue,
> however.
>=20
> --- ClassLoader.java    Fri Feb  6 22:45:56 2004
> +++ ClassLoader.java-patched    Fri Feb  6 22:46:09 2004
> @@ -365,7 +365,7 @@
>            =20
>           case 'L':
>             {
> -             elementName =3D elementName.substring (1,
> elementName.length() -
> 1);
> +             elementName =3D elementName.substring (1,
> elementName.length());
>               element =3D loadClass (elementName, resolve);
>             }
>             break;
>=20
> James Damour
>=20

Hi,

Thanks for your time of resending this bug report.

This bug is actually not located in this class, the -1 is necessary
to remove the ending ';' for the element array name "[Tryit$Inner;".

However, due to one bug in the reflection code I wrote, the ; was
getting removed for array types and this method was receiving
incorrectly "[Tryit$Inner" from time to time.

I am currently testing a fix for this bug.  I found a few more bugs
with your example.

I will look into it and report later.


David

---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Sablevm-developer] Crash on System.arraycopy()

[David <db...@cs...>]

On Sun, Mar 14, 2004 at 09:51:04PM -0500, Etienne Gagnon wrote:
> Hi David,
>=20
> I don't know if we have yet implemented the simple unsigned comparison
> for array bounds check.
>=20

Hi Etienne,

The interpreter uses an unsigned comparisons, at least at some places.
Never checked if everywhere.

> I think that, again, unsigned comparison would save the day, here.
>=20
> (_svmt_u32) src_position + (_svmt_u32) length > (_svmt_u32)=20
> src_instance->size
>=20

Yes, this works.  I checked in this solution to the bug fix in staging.


David

---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Sablevm-developer] Crash on System.arraycopy()

[Etienne G. <gag...@uq...>]

Hi David,

I don't know if we have yet implemented the simple unsigned comparison
for array bounds check.

I think that, again, unsigned comparison would save the day, here.

See below.

David B=E9langer wrote:
> SableVM code to check bounds:
>     if (src_position < 0 ||
> 	dst_position < 0 ||
> 	length < 0 ||
> 	src_position + length > src_instance->size ||
> 	dst_position + length > dst_instance->size)
>...=20
> So basically: src_position + length overflows and become negative.

(_svmt_u32) src_position + (_svmt_u32) length > (_svmt_u32) src_instance-=
>size

Etienne

--=20
Etienne M. Gagnon, Ph.D.             http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/

Re: [Sablevm-developer] request for log messages

[David <db...@cs...>]

On Sun, Mar 14, 2004 at 05:22:15PM -0500, Chris Pickett wrote:
> Hi,
>=20
> I think it would be good if all log messages were written to stderr.=20
> This means you can run benchmarks, and use the stdout to check that the=
y=20
> executed properly, and stderr to analyse your changes.  If this was=20
> done, I think it means that there would be zero occurrences of stdout i=
n=20
> the SableVM tree, save for the license / help information ... but=20
> perhaps even that could be changed.
>=20
> Anyway, it's something that I'm doing now, just thought I'd mention it=20
> in case there are any big problems.
>=20

Hello,

I assume you are talking about the verbose mode and stats.
I will not comment on what is better since I'm never sure myself
what to use.  :)

But I guess one option could be to use a variable for the FILE* to use
for these "log" messages.  The FILE* would default to either stdout or
stderr but it could also be set to other (ex: a disk file) if stdout
and stderr are used for other purposes.


David

---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Sablevm-developer] SableVM / Classpath 0.08 - now in staging

[Grzegorz B. P. <ga...@de...>]

W li=B6cie z nie, 14-03-2004, godz. 19:01, David B=E9langer pisze:=20
> This is to announce that SableVM Classpath has been updated to
> Classpath 0.08 in the staging area.

Great timing, David!

Thanks a lot,

				Grzegorz B. Prokopski

PS: So far it goes like that:
* friday(night) - GNU Classpath 0.08 release
* saturday - Imported into SVN
* sunday - merged into "staging"

I hope nobody expects me to make the 1.1.1 release tomorrow? ;-)

--=20
Grzegorz B. Prokopski <ga...@de...>
Debian GNU/Linux      http://www.debian.org
SableVM - LGPLed JVM  http://www.sablevm.org
Why SableVM ?!?       http://devel.sablevm.org/wiki/WhySableVM

Re: [Sablevm-developer] Crash on System.arraycopy()

[David <db...@cs...>]

On Sun, Mar 14, 2004 at 11:56:04PM +0100, Mark Wielaard wrote:
> Hi,
>=20
> The following crashes sablevm. It should give a
> ArrayIndexOutOfBoundsException. (Similar test is in Mauve btw.)
>=20

Hi,

Thanks for reporting the bug.


SableVM code to check bounds:
    if (src_position < 0 ||
	dst_position < 0 ||
	length < 0 ||
	src_position + length > src_instance->size ||
	dst_position + length > dst_instance->size)
      {
	_svmf_error_ArrayIndexOutOfBoundsException (env);
	goto end;
      }

So basically: src_position + length overflows and become negative.


What about moving the length the other side of > :

src_position > src_instance->size - length

because of the other tests, we know that:
src_position is >=3D 0
length is 0 <=3D length <=3D MAX_INT

and
src_instance->size is >=3D 0

So I guess it solves the overflow problem.

Any comments?
I will check this fix in staging.


David



---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

[Sablevm-developer] SableVM / Classpath 0.08 - now in staging

[David <db...@cs...>]

Hi,

This is to announce that SableVM Classpath has been updated to
Classpath 0.08 in the staging area.

Note that the SableVM/CP interface remained unchange.

Note that there do not seem to have additional build requirements.

Build it as usual.

Instructions are in file INSTALL-DEVEL located in SableVM source
directory.

Note to people who update infrequently, from revision 1648 (Feb 29),
the Java class files were moved from .../lib/... to .../share/...
in case you use customized installation scripts.

Have fun!

David


---

David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226

Web page:   http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt

Re: [Sablevm-developer] request for log messages

[Chris P. <chr...@ma...>]

David Bélanger wrote:
> On Sun, Mar 14, 2004 at 05:22:15PM -0500, Chris Pickett wrote:
> 
>>Hi,
>>
>>I think it would be good if all log messages were written to stderr. 
>>This means you can run benchmarks, and use the stdout to check that they 
>>executed properly, and stderr to analyse your changes.  If this was 
>>done, I think it means that there would be zero occurrences of stdout in 
>>the SableVM tree, save for the license / help information ... but 
>>perhaps even that could be changed.
>>
>>Anyway, it's something that I'm doing now, just thought I'd mention it 
>>in case there are any big problems.
>>
> 
> 
> Hello,
> 
> I assume you are talking about the verbose mode and stats.

Well, there's not much else :) -- I did introduce some verbose modes of 
my own though.

Side note: I have come to learn that gathering statistics via trace 
files generated with a given verbose mode and bash is not practical for 
normal-sized benchmarks (even if all they output, for example, is the 
signature of a method on each invoke).  A lot of the problem is that gnu 
sort is inefficient (yes, I tried splitting the outputs, sorting them, 
and doing 'sort -m').  It is still possible, if you really need temporal 
information, but I think it is better to wait for / work on some kind of 
*J integration.  At the moment I think one should use / change the stats 
stuff directly inside SableVM instead.

> I will not comment on what is better since I'm never sure myself
> what to use.  :)
> 
> But I guess one option could be to use a variable for the FILE* to use
> for these "log" messages.  The FILE* would default to either stdout or
> stderr but it could also be set to other (ex: a disk file) if stdout
> and stderr are used for other purposes.

Actually, yes -- since Java allows you to write to stdout and stderr, it 
would be better if SableVM's messages could be sent elsewhere if the 
user was having problems (imagine a benchmark that generated output on 
stdout as well as a log of what it was doing on stderr).  For now stderr 
is fine for my purposes (SPECjvm98 and JOlden), but when I prepare this 
change I will send log / debugging messages to vm->stderr and the help / 
license information to vm->stdout.  The user will be able to specify 
'-o' for the stdout and '-e' for the stderr (with a clear explanation 
that this does not apply to the Java stdout / stderr).  Sound good?

Chris

[Sablevm-developer] Crash on System.arraycopy()

[Mark W. <ma...@kl...>]

Hi,

The following crashes sablevm. It should give a
ArrayIndexOutOfBoundsException. (Similar test is in Mauve btw.)

Cheers,

Mark

[Sablevm-developer] request for log messages

[Chris P. <chr...@ma...>]

Hi,

I think it would be good if all log messages were written to stderr. 
This means you can run benchmarks, and use the stdout to check that they 
executed properly, and stderr to analyse your changes.  If this was 
done, I think it means that there would be zero occurrences of stdout in 
the SableVM tree, save for the license / help information ... but 
perhaps even that could be changed.

Anyway, it's something that I'm doing now, just thought I'd mention it 
in case there are any big problems.

Cheers,
Chris

[Sablevm-developer] Deserialization of arrays of objects.

[James D. <jd...@ny...>]

I ran into a bug in SableVM v1.0.9 on February second with
deserialization of objects in arrays.  The bug is still in v1.1.0, so I
thought I'd resend my proposed fix, along with a simple testcase to show
the bug.

Basically, the deserialization code makes an assumption about the length
of the classname in the bytestream that isn't correct for arrays of
objects.

The source code in the attached file, when compiled with Sun's JDK
v1.3.1 throws "java.lang.ClassNotFoundException: Tryit$Inne
   at gnu.java.lang.SystemClassLoader.findClass
(SystemClassLoader.java:79)" when run under SableVM v1.1.0 (Debian
testing).

I haven't tested the following "obvious" diff (I don't have the sources
installed), but it shows what I *think* is the offending line of code. 
I'm sorry that I can't submit a fully tested patch, but I'm tainted
(I've read Sun's source), so this is the closest I dare come.  I'd be
happy to take whatever steps are suggested to help debug this issue,
however.

--- ClassLoader.java    Fri Feb  6 22:45:56 2004
+++ ClassLoader.java-patched    Fri Feb  6 22:46:09 2004
@@ -365,7 +365,7 @@
            
          case 'L':
            {
-             elementName = elementName.substring (1,
elementName.length() -
1);
+             elementName = elementName.substring (1,
elementName.length());
              element = loadClass (elementName, resolve);
            }
            break;

James Damour

Re: [Sablevm-developer] Errno usage considered harmful?!

[Grzegorz B. P. <ga...@de...>]

W li=B6cie z =B6ro, 10-03-2004, godz. 08:26, Clark VERBRUGGE pisze:=20
> I have some guesses, below.
> Try just using the include and removing the "extern int errno;"
> statement.

Works, indeed.

> > 2. In interrupt() implementation I use a pointer to pthread_cond_t:
> >   volatile pthread_cond_t *sleeping_on_cond;
> > I also keep getting this warning:
> >   java_lang_VMThread.c:342: warning: passing arg 1 of
> >   `pthread_cond_broadcast' discards qualifiers from pointer target
> type
> > which is such code:
> >   pthread_cond_broadcast (env_other->thread.sleeping_on_cond);
>=20
> The prototype for cond_broadcast doesn't expect a volatile type for
>  the argument, so that qualifier is being discarded when you call the
>  function.
> An explicit cast may eliminate the warning, but the problem is that
>  the cond_broadcast code was not compiled assuming volatile input
>  arguments. =20
>=20
> If that's a problem then you'll need to re-examine why you need
>  volatile in sleeping_on_cond.

Good hint. The problem was, that I wanted volatile _pointer_ to
pthread_cond_t, not a pointer to volatile pthread_cond_t.
I changed the declaration to sth. like:

	pthread_cond_t *volatile sleeping_on_cond;

and the warnings are all gone.

Thanks a lot,

				Grzegorz B. Prokopski

--=20
Grzegorz B. Prokopski <ga...@de...>
Debian GNU/Linux      http://www.debian.org
SableVM - LGPLed JVM  http://www.sablevm.org
Why SableVM ?!?       http://devel.sablevm.org/wiki/WhySableVM

Re: [Sablevm-developer] Classpath 0.08 import request

[Etienne G. <gag...@uq...>]

Etienne Gagnon wrote:
> Revision 1763 of:
>  svn://svn.sablevm.org/sablevm-classpath/vendor/current
> does contain Classpath 0.08.

Please use the head revision; 1763 had some missing file properties.

Etienne

-- 
Etienne M. Gagnon, Ph.D.             http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/

[Sablevm-developer] Re: Java-Gnome: jni or cni

[Thomas A. <ae...@gr...>]

On Sat, 2004-03-13 at 20:33, Momchil Velikov wrote:
>   Bullshit, sorry. 

Thanks for your politeness. I don't think that anything coming after
such a sentence is worth discussing.

Best regards,
Tom
----------------------------------------------------------------------------
Thomas Aeby, Kirchweg 40, 1735 Giffers, Switzerland, Voice : (+41)26
4180040
Internet: ae...@gr...                           PGP public key
available
----------------------------------------------------------------------------
Neurotic     (n.) : One who builds castles in the air.
Psychotic    (n.) : One who lives in them.
Psychiatrist (n.) : The one who collects the rent.

[Sablevm-developer] Re: Java-Gnome: jni or cni

[Momchil V. <ve...@fa...>]

>>>>> "Thomas" == Thomas Aeby <ae...@gr...> writes:

    Thomas> On Thu, 2004-03-11 at 18:44, Mark Howard wrote:
    >> The big question is: should we switch to CNI?

    Thomas> Actually you already know the answer. Will you rename the
    Thomas> "java-gnome" project to "gcj-gnome"? If not, stay with
    Thomas> JNI, if yes, do this step with all consequences - you are
    Thomas> talking about a library, then, which will not be available
    Thomas> to generic Java programs/developers. One will then have to
    Thomas> decide

    Thomas>   1) do I write a platform-independent Java program
    Thomas>      obeying Java standards and do not use gcj-gnome

    Thomas> or

    Thomas>   2) do I write a GCC-program in Java language and make
    Thomas>      use of gcj-gnome

  Bullshit, sorry.  The choice is already made -- to write a GNOME
program.  It is unlikely an environement, which has GNOME to not have
gcj.

    Thomas> (note the difference in writing a Java program and writing
    Thomas> a program in Java language)

  And you note the differencees between:

  a) a Java program and a GNOME program written in the Java language.

  b) a Java/GNOME API and its implementation.

  Nothing precludes you from using an alternative implementation of
the Java GNOME binding.

Re: [Sablevm-developer] Classpath 0.08 import request

[Etienne G. <gag...@uq...>]

Grzegorz B. Prokopski wrote:
> It'd be great to have it imported and merged into "staging" rather
> sooner than later.

Revision 1763 of:
  svn://svn.sablevm.org/sablevm-classpath/vendor/current
does contain Classpath 0.08.

In fact, this revision does not contain some erroneously resuscitated
files in the Classpath CVS classpath-0_08-release tag.

Etienne
-- 
Etienne M. Gagnon, Ph.D.             http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/

Re: [Sablevm-developer] [Java-gnome-developer] Re: Java-Gnome: jni or cni

[Etienne G. <gag...@uq...>]

Per Bothner wrote:
> ...
> then a pre-processor can generate the JNI or CNI headers.
>  ...

My main worry is that the programmers using these macros might not take
the good approach to JNI programming (which consists of "caching" method
and field ID's) if some macro hides method/field accesses.

Also, from a readability point of view, any Java programmer that uses JNI
regularly will not recognize JNI code [but this is not really critical ;-)].

Etienne
-- 
Etienne M. Gagnon, Ph.D.             http://www.info.uqam.ca/~egagnon/
SableVM:                                       http://www.sablevm.org/
SableCC:                                       http://www.sablecc.org/