[M3devel] Problems building CM3 from trunk.

Fri Jun 10 21:46:29 CEST 2011

Hi all:
and here it is (work in progress) in Modula-3 a JVM target compiler (maybe later another VM since UNSAFE is nop in JVM) to create another language (see for the later:
http://xerxys.de/index.php?title=Discussion

) :
http://www.xerxys.org/index.php?title=Xm3

So the all dream about low level code in Modula-3 is nice, is just closer each time I guess.

I wouldn't be surprised they even know the ARX OS, since there they coded everything in Modula-2+, so it's believable (more than SPIN) was (they always run in protected mode, so, I guess this would prove that safe codes needs not to live necessarily in kernel, though SPIN was, but to protect itself from others clients). Anyway, ARX coded even C runtime, so, it's not that surprising they did that yet it's wonderful.

Thanks in advance

--- El vie, 10/6/11, Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es> escribió:

> De: Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es>
> Asunto: Re: [M3devel] Problems building CM3 from trunk.
> Para: m3devel at elegosoft.com, "Hendrik Boom" <hendrik at topoi.pooq.com>
> Fecha: viernes, 10 de junio, 2011 10:42
> Hi all:
> besides, that the existence of linker, intermediate and low
> level cross assemblers from 16032 to ARM and a low level
> debugger in Acorn, though did live there until don't know
> when (even a PC emulator written back circa 1986, 
> available for their internal use).
> Some information had appeared in some of the manuals, I
> refer FYI one:
> http://mdfs.net/Docs/Books/ARMCoPro/ARMAssem.rtf 
> Similar to others that may be available
> 
> Ok, besides that, I would want to ask you please if you may
> tell me whether m3cg currently generates private structures
> or generated gcc compiler tree-lang or both of them, this is
> because if the former or the later we then have a lot of
> common with an UW-Milwaukee Professor John Boyland's PhD
> thesis [1] (a copy is larger than the one of the below url,
> by 134 pages, sorry there is not m3cg available man page):
> 
> http://www.eecs.berkeley.edu/Pubs/TechRpts/1996/CSD-96-916.pdf
> 
> Main difference I believe is that it targeted Oberon-2 (and
> doesn't generate M3CG µ-code), so it circumvents that, but
> formalizes through a description layer, but given compiler
> comparison available documentation in [2], we could be able
> to re target it for Modula-3, in fact his thesis proposes it
> (besides other Algol-derived languages plus another Module
> implementing intermediate tree-lang to C backend). In fact
> XDS Oberon-2/Modula-2 compiler that generates both target
> machine and C language, and it's based on Modula-3 compiler
> (so it might generate M3CG if so), see:
> http://wrapl.sourceforge.net/links.html#Programming_Languages
> 
> Yet another example of this Tycoon's TML which generates
> C:
> http://wwwmatthes.in.tum.de/file/Publications/1992/Math92/codegen.ps.gz
> 
> or indirectly through VM level (after the above, like
> Boyland thesis):
> http://wwwmatthes.in.tum.de/file/Publications/1994/GaMa94/paper.pdf
> 
> Yet before they tried Quest Cardelli's language machine and
> adding some persistence capabilities (a functional language
> stack based virtual machine):
> http://wwwmatthes.in.tum.de/file/Publications/1991/Matt91/P-Quest-Manual.ps.gz
> 
> 
> Thanks in advance
> 
> [1] J. T. Boyland, “Descriptional composition of compiler
> components,” University of California, Berkeley, 1996.
> 
> [2]Computer abstracts. 1993. p152
> 
> 
> 
> --- El jue, 9/6/11, Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es>
> escribió:
> 
> > De: Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es>
> > Asunto: Re: [M3devel] Problems building CM3 from
> trunk.
> > Para: m3devel at elegosoft.com,
> "Hendrik Boom" <hendrik at topoi.pooq.com>
> > Fecha: jueves, 9 de junio, 2011 11:04
> > Hi all:
> > it is that there is an actual compiler that using
> Modula-3
> > M3CG code generates C (why not may even C--) to object
> code
> > generation inside a Scale Java infrastructure and
> mostly
> > LGPL available (it makes type inference BTW). So the
> wish of
> > C is just out there :)
> > Ok, perhaps it's good to think how to generate
> additional
> > target and have fun:
> > http://www-ali.cs.umass.edu/Scale/images/dataflow.gif
> > 
> > Plusses are the optional optimization, inlining and
> > debugging of C/C-- lcc, ldb, mld (NJMCTK) tool chain
> > infrastructure.
> > 
> > ftp://ftp.cs.umass.edu/pub/techrept/techreport/1997/UM-CS-1997-058.ps
> > 
> > http://www.cs.tufts.edu/~nr/pubs/tk-usenix.html
> > 
> > I fell this sounds good for any given native JVM code
> > generate and/or cross assembling in CM3 (perhaps the
> hardest
> > part?)
> > 
> > If you may please take a look, thanks in advance
> > 
> > --- El dom, 5/6/11, Daniel Alejandro Benavides D.
> <dabenavidesd at yahoo.es>
> > escribió:
> > 
> > > De: Daniel Alejandro Benavides D. <dabenavidesd at yahoo.es>
> > > Asunto: Re: [M3devel] Problems building CM3 from
> > trunk.
> > > Para: m3devel at elegosoft.com,
> > "Hendrik Boom" <hendrik at topoi.pooq.com>
> > > Fecha: domingo, 5 de junio, 2011 21:40
> > > Hi all:
> > > well I think we would need a good C hacker (very
> > good), I
> > > guess all experience is welcomed, perhaps is good
> idea
> > to
> > > think in mobility I believe many others are
> working
> > towards
> > > that, perhaps is time to us to do it.
> > > 
> > > http://www.mobily.com.sa/wps/portal/apps-award
> > > 
> > > Modula-3 web apps and custom browsers I think
> idea is
> > > recycled here, every word I heard about mobile
> web is
> > > client, client for that, ...
> > > 
> > > If so perhaps that's not good enough so we may
> be
> > start to
> > > think what is better to start with, anyway I have
> not
> > heard
> > > of something like mobile web development
> environment
> > and so
> > > if they like idea why not? The biggest effort
> would be
> > in
> > > Arabic characters, I guess would need to fix that
> part
> > of
> > > current WIDECHAR libraries again not in m3core,
> > fortunately
> > > there is a lady who worked in DEC SRC and
> Professor in
> > an
> > > University who worked with BIll Kalsow, she's
> expert
> > in
> > > arabic language (see:
> > > http://www.goodreads.com/author/show/3504134.Violetta_Cavalli_Sforza
> > > ), it might that she can give us some help or a
> clue
> > for
> > > interfaces. This idea is not bad at all since
> you
> > surely
> > > know that internet has support now for that kind
> of
> > url, we
> > > may have a hand on it. I have some indian friends
> as
> > well,
> > > hopefully they may get some interest if so. My
> feeling
> > is
> > > that they sort of like the language.
> > > 
> > > Thanks in advance for all your comments on that
> > > 
> > > --- El dom, 5/6/11, Hendrik Boom <hendrik at topoi.pooq.com>
> > > escribió:
> > > 
> > > > De: Hendrik Boom <hendrik at topoi.pooq.com>
> > > > Asunto: Re: [M3devel] Problems building CM3
> from
> > > trunk.
> > > > Para: m3devel at elegosoft.com
> > > > Fecha: domingo, 5 de junio, 2011 17:01
> > > > On Sun, Jun 05, 2011 at 07:23:38PM
> > > > +0100, Daniel Alejandro Benavides D. wrote:
> > > > > Hi all:
> > > > > Yes, it would be nice, but you would
> need
> > to
> > > > understand that we don't want the RT to
> depend
> > on
> > > specific
> > > > implementations, but to be portable really
> and
> > truly,
> > > having
> > > > C underneath makes just another thing makes
> us
> > to
> > > depend on
> > > > C RT (that is less likely to be the best
> idea),
> > > instead of
> > > > that, a clean low level but truly
> independent
> > platform
> > > low
> > > > level C-- is good to start with, you might
> see
> > > another
> > > > implementation of this idea in portable
> Cedar
> > > (Modula-3 was
> > > > ported to it), just that their idea was to
> make
> > this
> > > even
> > > > further to interact with that language, us
> here
> > just
> > > need
> > > > the portable low level interfaces same way
> (I
> > don't
> > > deny
> > > > Modula-3 C runtime, but making it safe
> ideally
> > is
> > > C--),
> > > > something I don't know it's quite platform
> > dependent
> > > in
> > > > inner C currently if I may say so (who can
> say
> > that
> > > BTW).
> > > > Even SPIN did  transfered their ideal
> C
> > idiom to
> > > SPIN
> > > > Modula-3 (just what it missed, later did the
> same
> > in
> > > a
> > > > distributed JVM an nowadays I think in C#)
> > > > 
> > > > C-- is my favourite object code language,
> but it
> > > lacks
> > > > support.  As far as I know, 
> > > > it has a code generator for the 32-bit x86
> > processoe
> > > and
> > > > that's about it.  Several 
> > > > others have been mentioned as being nearly
> ready,
> > but
> > > that
> > > > isn't really good enough.  
> > > > There hasn't been much activity on the C--
> > newsgroup
> > > on
> > > > gmane.  THe project looks 
> > > > moribund.
> > > > 
> > > > I think it would be worth breathing life
> into the
> > C--
> > > > project, but I'm really not 
> > > > sure how to do it.
> > > > 
> > > > It would probably be a lot of work.
> > > > 
> > > > --  hendrik
> > > > 
> > > >
> > >
> >
> 