Outrageous Thoughts on Ada Compilers.

Outrageous Thoughts on Ada Compilers.

[Austin Obyrne]

1) Am I right in thinking that Ada compilers are written in the 'C' programming langauge.

2) Outside of 'Adacore' is it possible to access the Ada compiler itself source code.

If 2), then, I can encrypt  a customised version of any Ada compiler !*?

adacrypt

Re: Outrageous Thoughts on Ada Compilers.

[Pascal Obry]

Le mardi 23 février 2016 à 08:54 -0800, Austin Obyrne a écrit :
> 1) Am I right in thinking that Ada compilers are written in the 'C'
> programming langauge.

No GNAT is written in Ada. Now that depends what you count. As you
probably know GNAT is part of GCC which is written in C. So most
accurately GNAT frontend and middleend (expander) are written in Ada.

Now most of the GNAT runtime is written in Ada, but part are in C which
is unavoidable to get proper values from the OS (such as constants for
example).

As you see the answer is not simple.

-- 
  Pascal Obry /  Magny Les Hameaux (78)

  The best way to travel is by means of imagination

  http://v2p.fr.eu.org
  http://www.obry.net

  gpg --keyserver keys.gnupg.net --recv-key F949BD3B

Re: Outrageous Thoughts on Ada Compilers.

[Nasser M. Abbasi]

On 2/23/2016 11:03 AM, Pascal Obry wrote:

>
> No GNAT is written in Ada. Now that depends what you count. As you
> probably know GNAT is part of GCC which is written in C.

btw, gcc switched to c++, as of 4.8

https://lwn.net/Articles/542457/
"GCC's move to C++"

http://developers.slashdot.org/story/13/03/22/2126235/gcc-480-release-marks-completion-of-c-migration?sdsrc=rel
""GCC 4.8.0 has been released (download), and with it, the developers of
the GNU Compiler Collection have switched to C++ as the implementation language,"

etc...

Re: Outrageous Thoughts on Ada Compilers.

[Austin Obyrne]

On Tuesday, February 23, 2016 at 5:03:15 PM UTC, Pascal Obry wrote:
> Le mardi 23 février 2016 à 08:54 -0800, Austin Obyrne a écrit :
> > 1) Am I right in thinking that Ada compilers are written in the 'C'
> > programming langauge.
> 
> No GNAT is written in Ada. Now that depends what you count. As you
> probably know GNAT is part of GCC which is written in C. So most
> accurately GNAT frontend and middleend (expander) are written in Ada.
> 
> Now most of the GNAT runtime is written in Ada, but part are in C which
> is unavoidable to get proper values from the OS (such as constants for
> example).
> 
> As you see the answer is not simple.
> 
> -- 
>   Pascal Obry /  Magny Les Hameaux (78)
> 
>   The best way to travel is by means of imagination
> 
>   http://v2p.fr.eu.org
>   http://www.obry.net
> 
>   gpg --keyserver keys.gnupg.net --recv-key F949BD3B

Many thanks - your exposure is very useful to know - AOB

Re: Outrageous Thoughts on Ada Compilers.

[Niklas Holsti]

On 16-02-23 19:03 , Pascal Obry wrote:

> Now most of the GNAT runtime is written in Ada, but part are in C which
> is unavoidable to get proper values from the OS (such as constants for
> example).

Depends on the target, as you no doubt know. For example, the GNAT Pro 
Ravenscar Small Foot-Print bare-board runtime for the SPARC architecture 
has no C parts, all is in Ada except for a couple of assembly-language 
modules (crt0, context switch and such).

-- 
Niklas Holsti
Tidorum Ltd
niklas holsti tidorum fi
       .      @       .

Re: Outrageous Thoughts on Ada Compilers.

[Austin Obyrne]

On Tuesday, February 23, 2016 at 10:10:33 PM UTC, Niklas Holsti wrote:
> On 16-02-23 19:03 , Pascal Obry wrote:
> 
> > Now most of the GNAT runtime is written in Ada, but part are in C which
> > is unavoidable to get proper values from the OS (such as constants for
> > example).
> 
> Depends on the target, as you no doubt know. For example, the GNAT Pro 
> Ravenscar Small Foot-Print bare-board runtime for the SPARC architecture 
> has no C parts, all is in Ada except for a couple of assembly-language 
> modules (crt0, context switch and such).
> 
> -- 
> Niklas Holsti
> Tidorum Ltd
> niklas holsti tidorum fi
>        .      @       .

A point I want to establish is that - reference the compiler - no matter how it is done and no matter what langauges are used to complete it all of these languages use 'ASCII' as their encryption domain - if I can then somehow access the compiler sourcecode I can customise a unique compiler for my own exclusive use by encrypting a private version that I can ahare with another entity having made a secure delivery to that person of the modified compiler.

Do you know how to access the compiler sourcecode?

Thanks - AOB

Re: Outrageous Thoughts on Ada Compilers.

[MM]

On Wednesday, 24 February 2016 08:36:14 UTC, Austin Obyrne  wrote:
> Do you know how to access the compiler source code?

You asked this question in sci.crypt, and it was answered there.

Here is a link to Ada compiler source (again):

http://www.dragonlace.net/

If you are having problems getting to exactly what you need, then please explain
what is going wrong so folks can guide you further.

M
-- 

Re: Outrageous Thoughts on Ada Compilers.

[Simon Wright]

Austin Obyrne <austin.obyrne@hotmail.com> writes:

> A point I want to establish is that - reference the compiler - no
> matter how it is done and no matter what langauges are used to
> complete it all of these languages use 'ASCII' as their encryption
> domain - if I can then somehow access the compiler sourcecode I can
> customise a unique compiler for my own exclusive use by encrypting a
> private version that I can ahare with another entity having made a
> secure delivery to that person of the modified compiler.

Months ago I provided you with source code demonstrating that your
woefully inefficient encryption could be used on any input, including
binary; so ASCII is irrelevant.

You could certainly produce a compiler for your exclusive use; it might
be a copyright violation (not in the case of GNAT), but you'd be very
unlikely to be found out. You could encrypt the source, and/or the built
compiler, and distribute the result; in the case of GNAT, the licensing
terms would oblige you to provide the other entity with the source as
well as the compiler. For other compilers, the legality would depend on
the copyright/licensing status of the source material.

> Do you know how to access the compiler sourcecode?

For GNAT, yes; otherwise, no.

Re: Outrageous Thoughts on Ada Compilers.

[Austin Obyrne]

On Thursday, February 25, 2016 at 9:34:04 AM UTC, Simon Wright wrote:
> Austin Obyrne <austin.obyrne@hotmail.com> writes:
> 
> > A point I want to establish is that - reference the compiler - no
> > matter how it is done and no matter what langauges are used to
> > complete it all of these languages use 'ASCII' as their encryption
> > domain - if I can then somehow access the compiler sourcecode I can
> > customise a unique compiler for my own exclusive use by encrypting a
> > private version that I can ahare with another entity having made a
> > secure delivery to that person of the modified compiler.
> 
> Months ago I provided you with source code demonstrating that your
> woefully inefficient encryption could be used on any input, including
> binary; so ASCII is irrelevant.
> 
> You could certainly produce a compiler for your exclusive use; it might
> be a copyright violation (not in the case of GNAT), but you'd be very
> unlikely to be found out. You could encrypt the source, and/or the built
> compiler, and distribute the result; in the case of GNAT, the licensing
> terms would oblige you to provide the other entity with the source as
> well as the compiler. For other compilers, the legality would depend on
> the copyright/licensing status of the source material.
> 
> > Do you know how to access the compiler sourcecode?
> 
> For GNAT, yes; otherwise, no.

Hi Simon,

I realise I am opening a massive can of worms here and I cannot even think of defending any thing that experienced Ada designers such as yourself bring up. 

But there is no need to do anything just yet.

I have a powerful encryption program called "ShuttlePads" that can encrypt anything that emanates from ASCII.

*Claim.
(In the present context)

Every compiler uses ASCII as its eventual output domain i.e. the eventual compiler source code is some combination of the 95 elements of ASCII.

*I can encrypt these to form a customised version of the parent compiler.

I can then manage this with another entity to create a unique loop for our sole use.

There is no need to go back to machine code??


<Do you know how to access the compiler sourcecode?
<For GNAT, yes; otherwise, no.

That alone may be sufficient to disable the compil;er in question.??

AOB 

 

Re: Outrageous Thoughts on Ada Compilers.

[MM]

On Thursday, 25 February 2016 16:37:19 UTC, Austin Obyrne  wrote:
> Every compiler uses ASCII as its eventual output domain i.e. the eventual compiler
> source code is some combination of the 95 elements of ASCII.

This is unbelievably confused.

1) The "source code" is *input* not output.
2) The output of a compiler is executable instructions.

> *I can encrypt these to form a customised version of the parent compiler.

Encrypted code is NOT customised code. Encrypted code will neither compile nor
execute unless you somehow build a seriously specialised tool to do this. I suggest
you learn how to do ROT-13 first.

> I can then manage this with another entity to create a unique loop for our sole use.

OK. Demonstrate this. I bet you can't.

> There is no need to go back to machine code??

How will your programs run, then?

> <Do you know how to access the compiler sourcecode?
> <For GNAT, yes; otherwise, no.
> 
> That alone may be sufficient to disable the compil;er in question.??

Hardly.

Simon knows where the compiler source that he is interested in is. Him not knowing
this disables nothing; somebody else will know.

M
-- 

Re: Outrageous Thoughts on Ada Compilers.

[Austin Obyrne]

On Thursday, February 25, 2016 at 9:34:04 AM UTC, Simon Wright wrote:
> Austin Obyrne <austin.obyrne@hotmail.com> writes:
> 
> > A point I want to establish is that - reference the compiler - no
> > matter how it is done and no matter what langauges are used to
> > complete it all of these languages use 'ASCII' as their encryption
> > domain - if I can then somehow access the compiler sourcecode I can
> > customise a unique compiler for my own exclusive use by encrypting a
> > private version that I can ahare with another entity having made a
> > secure delivery to that person of the modified compiler.
> 
> Months ago I provided you with source code demonstrating that your
> woefully inefficient encryption could be used on any input, including
> binary; so ASCII is irrelevant.
> 
> You could certainly produce a compiler for your exclusive use; it might
> be a copyright violation (not in the case of GNAT), but you'd be very
> unlikely to be found out. You could encrypt the source, and/or the built
> compiler, and distribute the result; in the case of GNAT, the licensing
> terms would oblige you to provide the other entity with the source as
> well as the compiler. For other compilers, the legality would depend on
> the copyright/licensing status of the source material.
> 
> > Do you know how to access the compiler sourcecode?
> 
> For GNAT, yes; otherwise, no.

< For GNAT, yes; otherwise, no.

Would you kindly publish a few lines of Gnat here just for me to show readers what I mean


Austin

Re: Outrageous Thoughts on Ada Compilers.

[MM]

On Thursday, 25 February 2016 18:13:31 UTC, Austin Obyrne  wrote:
> Would you kindly publish a few lines of Gnat here just for me to show readers what I mean

Are you having trouble doing this yourself?

M
-- 

Re: Outrageous Thoughts on Ada Compilers.

[Luke A. Guest]

Austin Obyrne <austin.obyrne@hotmail.com> wrote:
> 
> 1) Am I right in thinking that Ada compilers are written in the 'C' programming langauge.
> 
> 2) Outside of 'Adacore' is it possible to access the Ada compiler itself source code.
> 
> If 2), then, I can encrypt  a customised version of any Ada compiler !*?
> 
> adacrypt
> 

Nobody seems to be asking the most relevant question here. What's the
point?

Re: Outrageous Thoughts on Ada Compilers.

[MM]

On Sunday, 28 February 2016 02:20:44 UTC, Luke A. Guest  wrote:
> Nobody seems to be asking the most relevant question here. What's the
> point?

It's the OP not understanding the basics, and coming up with these
elaborate contraptions to compensate.

M
--