native code

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

native code

Samar Yazdani
hello all,

i was wondering if there is any possibility in VW 7.x to gather native
microcode from Smalltalk specification...?


thanks in advance
Samar

Reply | Threaded
Open this post in threaded view
|

AW: native code

Georg Heeg
Samar,

To see the byte code, just open a browser, select a method and use the
inspect menu item. If you mean machine code from the JIT, there are some
debugging tools for VM engineers, but I do not know whether they are
available in general.

Georg

Georg Heeg eK, Dortmund und Köthen, HR Dortmund A 12812
Tel. +49-3496-214328, Fax +49-3496-214712

> -----Ursprüngliche Nachricht-----
> Von: Samar Yazdani [mailto:[hidden email]]
> Gesendet: Donnerstag, 3. Mai 2007 09:57
> An: vwnc
> Betreff: native code
>
> hello all,
>
> i was wondering if there is any possibility in VW 7.x to gather native
> microcode from Smalltalk specification...?
>
>
> thanks in advance
> Samar

Reply | Threaded
Open this post in threaded view
|

Re: AW: native code

Samar Yazdani
Thanks Georg for your prompt reply...yes, actually i was looking to
gather machine code from the JIT....

nebody who knows about such "debugging tools" that are available for free???

regards/thanks in advance
Samar

Georg Heeg wrote:

>Samar,
>
>To see the byte code, just open a browser, select a method and use the
>inspect menu item. If you mean machine code from the JIT, there are some
>debugging tools for VM engineers, but I do not know whether they are
>available in general.
>
>Georg
>
>Georg Heeg eK, Dortmund und Köthen, HR Dortmund A 12812
>Tel. +49-3496-214328, Fax +49-3496-214712
>
>  
>
>>-----Ursprüngliche Nachricht-----
>>Von: Samar Yazdani [mailto:[hidden email]]
>>Gesendet: Donnerstag, 3. Mai 2007 09:57
>>An: vwnc
>>Betreff: native code
>>
>>hello all,
>>
>>i was wondering if there is any possibility in VW 7.x to gather native
>>microcode from Smalltalk specification...?
>>
>>
>>thanks in advance
>>Samar
>>    
>>
>
>
>
>  
>

Reply | Threaded
Open this post in threaded view
|

Re: AW: native code

Reinout Heeck-2
In reply to this post by Georg Heeg
Georg Heeg wrote:
>  If you mean machine code from the JIT, there are some
> debugging tools for VM engineers, but I do not know whether they are
> available in general.
>  
The VM implements some secret primitives that will show a disassembly of
the JITed code.

To get access to these secrets you could enroll in the VW Developers
Program:
  http://parcplace.net



R
-


>> -----Ursprüngliche Nachricht-----
>> Von: Samar Yazdani [mailto:[hidden email]]
>> Gesendet: Donnerstag, 3. Mai 2007 09:57
>> An: vwnc
>> Betreff: native code
>>
>> hello all,
>>
>> i was wondering if there is any possibility in VW 7.x to gather native
>> microcode from Smalltalk specification...?
>>
>>
>> thanks in advance
>> Samar
>>    
>
>
>  

Reply | Threaded
Open this post in threaded view
|

Re: AW: native code

Reinout Heeck
>
> To get access to these secrets you could enroll in the VW  
> Developers Program:
>  http://parcplace.net

I forgot to mention:

   Yes it is free (gratis, not libre).


R
-

Reply | Threaded
Open this post in threaded view
|

Re: AW: native code

Eliot Miranda-2
In reply to this post by Reinout Heeck-2


On 5/3/07, Reinout Heeck <[hidden email]> wrote:
Georg Heeg wrote:
>  If you mean machine code from the JIT, there are some
> debugging tools for VM engineers, but I do not know whether they are
> available in general.
>
The VM implements some secret primitives that will show a disassembly of
the JITed code.


Actually they aren't secret anymore.  Load  OEprogrammingHacks and OEProfiler and you'll find  access to all he support that's there, e.g. inspecting a CompiledMethod will now include a translation selection showing native code.  But Reinout makes an excellent suggestion.  If as a University researcher you enroll in the vw-dev program you'll get access to the VM source code.  If you're interested in optmization ask Andres or myself about AOStA, a far-from-complete adaptive optimization framework for VW for which there is some support in the VM.

Enjoy :)

To get access to these secrets you could enroll in the VW Developers
Program:
  http://parcplace.net



R
-


>> -----Ursprüngliche Nachricht-----
>> Von: Samar Yazdani [mailto:[hidden email]]
>> Gesendet: Donnerstag, 3. Mai 2007 09:57
>> An: vwnc
>> Betreff: native code
>>
>> hello all,
>>
>> i was wondering if there is any possibility in VW 7.x to gather native
>> microcode from Smalltalk specification...?
>>
>>
>> thanks in advance
>> Samar
>>
>
>
>


Reply | Threaded
Open this post in threaded view
|

VW 64-bit issue wrt shared varible definitions

Ian Upright-2
In reply to this post by Reinout Heeck
Hello,

I'm trying out VW 7.5 running on Ubuntu 7.04 64-bit AMD.

One issue is it seems like defining shared variables in a namespace
takes an extremely long time.
Any code that defines shared variables, or loading packages out of Store
that has a large number of shared variables,
takes an extremely long time..

Using VW 7.5 with 32-bit windows doesn't seem to have this problem at all.

Has anyone else noticed this issue?

Thanks, Ian

Reply | Threaded
Open this post in threaded view
|

Re: VW 64-bit issue wrt shared varible definitions

Martin McClure
ian Upright wrote:

> Hello,
>
> I'm trying out VW 7.5 running on Ubuntu 7.04 64-bit AMD.
>
> One issue is it seems like defining shared variables in a namespace
> takes an extremely long time.
> Any code that defines shared variables, or loading packages out of Store
> that has a large number of shared variables,
> takes an extremely long time..
>
> Using VW 7.5 with 32-bit windows doesn't seem to have this problem at all.
>
> Has anyone else noticed this issue?

I haven't noticed that issue, but I just tried out VW 7.5 64-bit
yesterday for an hour or two.

Is 64-bit 7.5 stable for you on Ubuntu? I had trouble keeping it running
-- the VM would exit on parcel loads, use of Store, etc.  These were not
repeatable, but I don't think I ever kept the thing running for more
than five minutes at a time.

Regards,

-Martin