Squeak image format documentation

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

Squeak image format documentation

Michael Haupt-3
Dear all,

where can I find detailed documentation on the Squeak image format and
loading process?

Thanks in advance,

Michael Haupt

Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

timrowledge

On 2-May-06, at 9:29 AM, Michael Haupt wrote:

> Dear all,
>
> where can I find detailed documentation on the Squeak image format

You will find some of that in the various chapters of the nuBlue book  
(online copies of chapters at http://www.iam.unibe.ch/~ducasse/ 
FreeBooks/CollectiveNBlueBook/) and in comments and methods in the  
ObjectMemory class that is a part of the VMMaker package on  
SqueakMap. The book chapters are somewhat outdated in a few details  
and I expect the code comments could do with some improving too.


> and
> loading process?
>

The image loading process is implemented in  
Interpreter>readImageFromFile:HeapSize:StartingAt: and related  
methods. All it does is open the relevant image file, read it into  
memory, possibly run through to flip endianness, possibly run through  
to update pointers and some othe minor housekeeping. Nothing really  
clever.

What's the reason for your interest?


tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
For every action, there is an equal and opposite criticism.



Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

Michael Haupt-3
Hi Tim,

many thanks for your quick response. I tried to find information via
the Swiki at minnow.cc.gatech.edu, but that server seems to be down
(?!); at least, it refuses connections.

On 5/2/06, tim Rowledge <[hidden email]> wrote:
> You will find some of that in the various chapters of the nuBlue book
> (online copies of chapters at http://www.iam.unibe.ch/~ducasse/
> FreeBooks/CollectiveNBlueBook/)

Many thanks - would that be your "A Tour of the Squeak Object Engine"
chapter? Are there any others I should look at?

> and in comments and methods in the
> ObjectMemory class that is a part of the VMMaker package on
> SqueakMap.

I was afraid of that. ;-) I wanted to avoid looking at too much source
code... a concise specification of the format would have been great.
Well, I'll see what I can do.

> The image loading process is implemented in
> Interpreter>readImageFromFile:HeapSize:StartingAt: and related
> methods. ...

Thanks once more.

> What's the reason for your interest?

On the one hand, sheer interest in VMs generally and Smalltalk VMs
particularly. On the other hand, I'm doing some teaching on VMs this
summer semester, and this Thursday's topic, amongst others, is how
applications (or entire "snapshot" worlds) are made available to VMs.

Best,

Michael

Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

stéphane ducasse-2
I would really to see your material if possible?

> On the one hand, sheer interest in VMs generally and Smalltalk VMs
> particularly. On the other hand, I'm doing some teaching on VMs this
> summer semester, and this Thursday's topic, amongst others, is how
> applications (or entire "snapshot" worlds) are made available to VMs.


Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

Laurence Rozier
In reply to this post by Michael Haupt-3


On 5/2/06, Michael Haupt <[hidden email]> wrote:
Hi Tim,

many thanks for your quick response. I tried to find information via
the Swiki at minnow.cc.gatech.edu, but that server seems to be down
(?!); at least, it refuses connections.

On 5/2/06, tim Rowledge <[hidden email]> wrote:
> You will find some of that in the various chapters of the nuBlue book
> (online copies of chapters at http://www.iam.unibe.ch/~ducasse/
> FreeBooks/CollectiveNBlueBook/)

Many thanks - would that be your "A Tour of the Squeak Object Engine"
chapter? Are there any others I should look at?

All of the Part 2 is relevant but there is an excellent description of how image segments are generated in in the Object Memory section of  "Back to the Future Once More".

Regards,
Laurence

> and in comments and methods in the
> ObjectMemory class that is a part of the VMMaker package on
> SqueakMap.

I was afraid of that. ;-) I wanted to avoid looking at too much source
code... a concise specification of the format would have been great.
Well, I'll see what I can do.

> The image loading process is implemented in
> Interpreter>readImageFromFile:HeapSize:StartingAt: and related
> methods. ...

Thanks once more.

> What's the reason for your interest?

On the one hand, sheer interest in VMs generally and Smalltalk VMs
particularly. On the other hand, I'm doing some teaching on VMs this
summer semester, and this Thursday's topic, amongst others, is how
applications (or entire "snapshot" worlds) are made available to VMs.

Best,

Michael




Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

timrowledge
In reply to this post by Michael Haupt-3

On 2-May-06, at 10:10 AM, Michael Haupt wrote:
>
> On 5/2/06, tim Rowledge <[hidden email]> wrote:
>> You will find some of that in the various chapters of the nuBlue book
>> (online copies of chapters at http://www.iam.unibe.ch/~ducasse/
>> FreeBooks/CollectiveNBlueBook/)
>
> Many thanks - would that be your "A Tour of the Squeak Object Engine"
> chapter? Are there any others I should look at?

The back to the future and porting chapters would probably be helpful  
too
>
>> What's the reason for your interest?
>
> On the one hand, sheer interest in VMs generally and Smalltalk VMs
> particularly. On the other hand, I'm doing some teaching on VMs this
> summer semester, and this Thursday's topic, amongst others, is how
> applications (or entire "snapshot" worlds) are made available to VMs.

The key point is that the snapshot is merely a memory dump of the  
object memory. No coding, decoding or other messing around.

tim
--
tim Rowledge; [hidden email]; http://www.rowledge.org/tim
Strange OpCodes: CPP: Crush Plotter Pen



Reply | Threaded
Open this post in threaded view
|

Swiki status?

Chris Muller
In reply to this post by Michael Haupt-3
Does anyone know how we will restore Swiki?
 
 I've only heard reports of it being down, that it may be moving.
 
 Great timing, I guess, for the beginners list since this great resource is now unavailable!
 
 I do hope its absence is brief.
 

----- Original Message ----
From: Michael Haupt <[hidden email]>
To: The general-purpose Squeak developers list <[hidden email]>
Sent: Tuesday, May 2, 2006 12:11:14 PM
Subject: Re: Squeak image format documentation

Hi Tim,

many thanks for your quick response. I tried to find information via
the Swiki at minnow.cc.gatech.edu, but that server seems to be down
(?!); at least, it refuses connections.

On 5/2/06, tim Rowledge <[hidden email]> wrote:
> You will find some of that in the various chapters of the nuBlue book
> (online copies of chapters at http://www.iam.unibe.ch/~ducasse/
> FreeBooks/CollectiveNBlueBook/)

Many thanks - would that be your "A Tour of the Squeak Object Engine"
chapter? Are there any others I should look at?

> and in comments and methods in the
> ObjectMemory class that is a part of the VMMaker package on
> SqueakMap.

I was afraid of that. ;-) I wanted to avoid looking at too much source
code... a concise specification of the format would have been great.
Well, I'll see what I can do.

> The image loading process is implemented in
> Interpreter>readImageFromFile:HeapSize:StartingAt: and related
> methods. ...

Thanks once more.

> What's the reason for your interest?

On the one hand, sheer interest in VMs generally and Smalltalk VMs
particularly. On the other hand, I'm doing some teaching on VMs this
summer semester, and this Thursday's topic, amongst others, is how
applications (or entire "snapshot" worlds) are made available to VMs.

Best,

Michael






Reply | Threaded
Open this post in threaded view
|

Re: Swiki status?

ccrraaiigg

 > Does anyone know how we will restore Swiki?

        As I recall, there were people (Marcus?) who were working on making a
clone of it available through our own servers.

        At the very least, we can salvage the text from Google's cache.


-C

--
Craig Latta
improvisational musical informaticist
www.netjam.org
Smalltalkers do: [:it | All with: Class, (And love: it)]



Reply | Threaded
Open this post in threaded view
|

Re: Swiki status?

Chris Muller
Salvage??
 
 You mean to say, the bits that resided on a drive at GA Tech are gone?  Surely, they keep a backup!  Who knows who at GA tech can get in touch with someone to obtain the latest state of the Swiki?
 
 (I think my internal furnace just kicked on, time to relax, count to ten..)
 

----- Original Message ----
From: Craig Latta <[hidden email]>
To: [hidden email]
Sent: Tuesday, May 2, 2006 2:34:12 PM
Subject: Re: Swiki status?


 > Does anyone know how we will restore Swiki?

    As I recall, there were people (Marcus?) who were working on making a
clone of it available through our own servers.

    At the very least, we can salvage the text from Google's cache.


-C

--
Craig Latta
improvisational musical informaticist
www.netjam.org
Smalltalkers do: [:it | All with: Class, (And love: it)]








Reply | Threaded
Open this post in threaded view
|

Re: Swiki status?

Michael Rueger-6
Chris Muller wrote:
> Salvage??
>  
>  You mean to say, the bits that resided on a drive at GA Tech are gone?  Surely, they keep a backup!  Who knows who at GA tech can get in touch with someone to obtain the latest state of the Swiki?

An intern at Impara, Thomas Froeb, was working on porting the swiki data
to his persistency version of SmallWiki and Pier. So we actually might
have a system ready to go. Thomas?

Michael

Reply | Threaded
Open this post in threaded view
|

Re: Swiki status?

ccrraaiigg
In reply to this post by Chris Muller

 > > At the very least, we can salvage the text from Google's cache.
 >
 > Salvage??
 >
 > You mean to say, the bits that resided on a drive at GA Tech are gone?

        Certainly not... that's why I used the words "at the very least". :)

        I do suspect, though, that one of the clones of that data (e.g., the
one Michael Rueger just mentioned) will turn out to present an easier
path to reinstatement than dealing with Georgia Tech directly (and that
is indeed rather sad).


-C

--
Craig Latta
improvisational musical informaticist
www.netjam.org
Smalltalkers do: [:it | All with: Class, (And love: it)]



Reply | Threaded
Open this post in threaded view
|

Re: Swiki status?

Guzdial, Mark
Folks,  I'm reading the Squeak Swiki right now.  There seems to be a DNS
hiccup, and people are looking into it.  The server is running fine, no
bits are lost.

Mark

Craig Latta wrote:

>
> > > At the very least, we can salvage the text from Google's cache.
> >
> > Salvage??
> >
> > You mean to say, the bits that resided on a drive at GA Tech are gone?
>
>     Certainly not... that's why I used the words "at the very least". :)
>
>     I do suspect, though, that one of the clones of that data (e.g.,
> the one Michael Rueger just mentioned) will turn out to present an
> easier path to reinstatement than dealing with Georgia Tech directly
> (and that is indeed rather sad).
>
>
> -C
>


Reply | Threaded
Open this post in threaded view
|

Re: Squeak image format documentation

Michael Haupt-3
In reply to this post by timrowledge
Hi,

thank you all very much for your help!

Best,

Michael

Reply | Threaded
Open this post in threaded view
|

RE: Swiki status?

Thomas Koenig-2
In reply to this post by Guzdial, Mark
Yes.  Thanks much.  Swiki is a great resource

> -----Original Message-----
> From: [hidden email] [mailto:squeak-dev-
> [hidden email]] On Behalf Of Mark Guzdial
> Sent: Tuesday, May 02, 2006 4:25 PM
> To: The general-purpose Squeak developers list
> Subject: Re: Swiki status?
>
> Folks,  I'm reading the Squeak Swiki right now.  There seems to be a DNS
> hiccup, and people are looking into it.  The server is running fine, no
> bits are lost.
>
> Mark
>
> Craig Latta wrote:
> >
> > > > At the very least, we can salvage the text from Google's cache.
> > >
> > > Salvage??
> > >
> > > You mean to say, the bits that resided on a drive at GA Tech are gone?
> >
> >     Certainly not... that's why I used the words "at the very least". :)
> >
> >     I do suspect, though, that one of the clones of that data (e.g.,
> > the one Michael Rueger just mentioned) will turn out to present an
> > easier path to reinstatement than dealing with Georgia Tech directly
> > (and that is indeed rather sad).
> >
> >
> > -C
> >
>