I am looking for a good (standard) free implementation of Common Lisp on
the web. I know that there are several out there but I would like to have
some suggestions from Seasoned Lispers as to which one is most popular.
Thanks,
Dipa.
"A mind once stretched by a new idea never regains its original dimension."
===============================================================================
Dipanwita Sarkar Office: LH415
Graduate Student
Computer Science Dept. 150 S Woodlawn Avenue
Indiana University Bloomington IN 47405-7104
Phone : 855-3608
===============================================================================
From: Kent M Pitman
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <sfwvgpjep73.fsf@world.std.com>
Dipanwita Sarkar <·······@cs.indiana.edu> writes:
> I am looking for a good (standard) free implementation of Common Lisp on
> the web. I know that there are several out there but I would like to have
> some suggestions from Seasoned Lispers as to which one is most popular.
Free like no cost or free like the GNU people want you to think free means?
>>>>> Kent M Pitman writes:
Kent> Dipanwita Sarkar <·······@cs.indiana.edu> writes:
>> I am looking for a good (standard) free implementation of Common Lisp on
>> the web. I know that there are several out there but I would like to have
>> some suggestions from Seasoned Lispers as to which one is most popular.
Kent> Free like no cost or free like the GNU people want you to think free means?
For a free one like the GNU people want you to think free means,
which is, IMO, an excellent definition of free (maybe french as
a mother tongue influences me there), there's CLISP and CMUCL.
For a _gratis_, but definitely not _free_ version, others will surely
recommend the demo versions which both Franz and Xanalys offer.
Also in this category, if you're on windoze, if memory serves, Corman
Lisp can be downloaded and used (the compiler only, not the whole
toolkit).
--
It would be difficult to construe Larry Wall, in article
this as a feature. <·····················@netlabs.com>
Alain Picard <·······@optushome.com.au> writes:
>For a free one like the GNU people want you to think free means,
>which is, IMO, an excellent definition of free (maybe french as
>a mother tongue influences me there), there's CLISP and CMUCL.
For the record, CMUCL is not under the GPL, although "free enough" for
the GNU people. In fact, personally I hate the GPL.
The original poster didn't even say which OS and machine he wants a CL
for.
Martin
--
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <········@bik-gmbh.de> http://www.bik-gmbh.de/~cracauer/
FreeBSD - where you want to go. Today. http://www.freebsd.org/
From: David Bakhash
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <m3pufqvmqc.fsf@alum.mit.edu>
········@counter.bik-gmbh.de (Martin Cracauer) writes:
> For the record, CMUCL is not under the GPL, although "free enough"
> for the GNU people. In fact, personally I hate the GPL.
I'm not so crazy about it either. It impedes business in many ways.
I find the LGPL to be a reasonable compromise, though, and wonder how
people feel about it relative the the GPL. I like it in that it
doesn't affect code around it the way the GPL does, and that code is
(in my opinion) separate enough to be closed-source, if that's how
developers want to keep it.
dave
········@counter.bik-gmbh.de (Martin Cracauer) writes:
> For the record, CMUCL is not under the GPL, although "free enough"
> for the GNU people.
Where on the gnu.org web pages is freedom (with respect to software)
defined in terms of the GPL?
The definition of freedom I found here:
<URL:http://www.gnu.org/philosophy/free-sw.html>
doesn't even mention the GPL.
Colin Walters <·········@cis.ohio-state.edu> writes:
>········@counter.bik-gmbh.de (Martin Cracauer) writes:
>> For the record, CMUCL is not under the GPL, although "free enough"
>> for the GNU people.
>Where on the gnu.org web pages is freedom (with respect to software)
>defined in terms of the GPL?
>The definition of freedom I found here:
><URL:http://www.gnu.org/philosophy/free-sw.html>
>doesn't even mention the GPL.
That's what I said, the definitition of free is not really coupled to
the GPL, not even for the GNU people.
Martin
--
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <········@bik-gmbh.de> http://www.bik-gmbh.de/~cracauer/
FreeBSD - where you want to go. Today. http://www.freebsd.org/
From: Hannah Schroeter
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <98laqr$cpf$1@c3po.schlund.de>
Hello!
In article <·············@counter.bik-gmbh.de>,
Martin Cracauer <········@counter.bik-gmbh.de> wrote:
>[...]
>For the record, CMUCL is not under the GPL, although "free enough" for
>the GNU people. In fact, personally I hate the GPL.
Oh, I'd be very interested in running cmucl, however, I don't know
how to get it over to my OS (OpenBSD/x86). It seems much more
complicated to compile/cross-compile, even when I have access e.g.
to a Linux box, too.
And cmucl seems impossible to bootstrap with other Lisps (such as
sbcl, which exists for OpenBSD, but lacks many of the extensions
from cmucl which I *am* interested in, too).
Kind regards,
Hannah.
······@schlund.de (Hannah Schroeter) writes:
> Hello!
>
> In article <·············@counter.bik-gmbh.de>,
> Martin Cracauer <········@counter.bik-gmbh.de> wrote:
> >[...]
>
> >For the record, CMUCL is not under the GPL, although "free enough" for
> >the GNU people. In fact, personally I hate the GPL.
>
> Oh, I'd be very interested in running cmucl, however, I don't know
> how to get it over to my OS (OpenBSD/x86). It seems much more
> complicated to compile/cross-compile, even when I have access e.g.
> to a Linux box, too.
>
> And cmucl seems impossible to bootstrap with other Lisps (such as
> sbcl, which exists for OpenBSD, but lacks many of the extensions
> from cmucl which I *am* interested in, too).
Adding extensions to SBCL is probably(?) going to be less work
than getting CMUCL built for OpenBSD.
--
Raymond Wiker
·············@fast.no
From: Tim Moore
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <98lmru$3hd$0@216.39.145.192>
On 13 Mar 2001, Raymond Wiker wrote:
> ······@schlund.de (Hannah Schroeter) writes:
>
> > Hello!
> >
> > In article <·············@counter.bik-gmbh.de>,
> > Martin Cracauer <········@counter.bik-gmbh.de> wrote:
> > >[...]
> >
> > >For the record, CMUCL is not under the GPL, although "free enough" for
> > >the GNU people. In fact, personally I hate the GPL.
> >
> > Oh, I'd be very interested in running cmucl, however, I don't know
> > how to get it over to my OS (OpenBSD/x86). It seems much more
> > complicated to compile/cross-compile, even when I have access e.g.
> > to a Linux box, too.
You might try asking on the ··········@cons.org or ·········@cons.org
mailing list. FreeBSD is very well supported in cmucl so the changes to
support OpenBSD should be small.
> > And cmucl seems impossible to bootstrap with other Lisps (such as
> > sbcl, which exists for OpenBSD, but lacks many of the extensions
> > from cmucl which I *am* interested in, too).
That's a noble goal of sbcl; has sbcl been built using a Lisp other than
itself or cmucl? I'm not trying to be snide, but I'm wondering about the
realities of having another suitable Lisp laying around on a potential
target. Last I heard CLISP wasn't complete enough to host the sbcl
compiler.
> Adding extensions to SBCL is probably(?) going to be less work
> than getting CMUCL built for OpenBSD.
My feeling is the opposite :) That is, if the work has been done to get
sbcl running in OpenBSD, then the changes caused by differences between
FreeBSD and OpenBSD (signals, possibly some system call differences)
should be easy to backport to cmucl.
Tim
From: Hannah Schroeter
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <99aha2$kru$1@c3po.schlund.de>
Hello!
In article <··············@raw.grenland.fast.no>,
Raymond Wiker <·············@fast.no> wrote:
>[...]
> Adding extensions to SBCL is probably(?) going to be less work
>than getting CMUCL built for OpenBSD.
Might be, however, I still don't know how to do this.
Kind regards,
Hannah.
From: Oliver Bandel
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <99bclt$fs@first.in-berlin.de>
Hannah Schroeter <······@schlund.de> wrote:
[...]
> Oh, I'd be very interested in running cmucl, however, I don't know
> how to get it over to my OS (OpenBSD/x86). It seems much more
> complicated to compile/cross-compile, even when I have access e.g.
> to a Linux box, too.
i heard about CMUC, that it's not completely written in C.
It's only the innerst part of CMUCL in C, the rest is in
LISP itself.
Therefore you need a running CMUCL to compile CMUCL.
That means: You have to start with a binary.
Ciao,
Oliver
From: Hannah Schroeter
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <9ahpmu$1he$1@c3po.schlund.de>
Hello!
In article <·········@first.in-berlin.de>,
Oliver Bandel <······@first.in-berlin.de> wrote:
>Hannah Schroeter <······@schlund.de> wrote:
>[...]
>> Oh, I'd be very interested in running cmucl, however, I don't know
>> how to get it over to my OS (OpenBSD/x86). It seems much more
>> complicated to compile/cross-compile, even when I have access e.g.
>> to a Linux box, too.
>i heard about CMUC, that it's not completely written in C.
>It's only the innerst part of CMUCL in C, the rest is in
>LISP itself.
>Therefore you need a running CMUCL to compile CMUCL.
>That means: You have to start with a binary.
Yes. And that means, about no way to port it over to OpenBSD.
I wonder how the ports e.g. from Linux to FreeBSD have been done.
>Ciao,
> Oliver
Kind regards,
Hannah.
······@schlund.de (Hannah Schroeter) writes:
> Hello!
>
> In article <·········@first.in-berlin.de>,
> Oliver Bandel <······@first.in-berlin.de> wrote:
>
> >Hannah Schroeter <······@schlund.de> wrote:
> >[...]
> >> Oh, I'd be very interested in running cmucl, however, I don't know
> >> how to get it over to my OS (OpenBSD/x86). It seems much more
> >> complicated to compile/cross-compile, even when I have access e.g.
> >> to a Linux box, too.
>
> >i heard about CMUC, that it's not completely written in C.
> >It's only the innerst part of CMUCL in C, the rest is in
> >LISP itself.
> >Therefore you need a running CMUCL to compile CMUCL.
> >That means: You have to start with a binary.
>
> Yes. And that means, about no way to port it over to OpenBSD.
> I wonder how the ports e.g. from Linux to FreeBSD have been done.
Well, it happened tha other way around. The first x86 port of CMUCL
was for FreeBSD by Paul Werkowski. Then Peter Van Eynde did the Linux
port.
Cheers
--
Marco Antoniotti ========================================================
NYU Courant Bioinformatics Group tel. +1 - 212 - 998 3488
719 Broadway 12th Floor fax +1 - 212 - 995 4122
New York, NY 10003, USA http://bioinformatics.cat.nyu.edu
"Hello New York! We'll do what we can!"
Bill Murray in `Ghostbusters'.
>>>>> "Marco" == Marco Antoniotti <·······@cs.nyu.edu> writes:
Marco> ······@schlund.de (Hannah Schroeter) writes:
>> Yes. And that means, about no way to port it over to OpenBSD.
>> I wonder how the ports e.g. from Linux to FreeBSD have been done.
Marco> Well, it happened tha other way around. The first x86 port of CMUCL
Marco> was for FreeBSD by Paul Werkowski. Then Peter Van Eynde did the Linux
Marco> port.
Yes. And Paul used the Alpha port to cross-compile to the FreeBSD
port.
From a historical view, it would be interesting to know what the first
port was and how the other ports came about.
Ray
From: Tim Bradshaw
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <nkjelv7z2rx.fsf@tfeb.org>
Raymond Toy <···@rtp.ericsson.se> writes:
>
> From a historical view, it would be interesting to know what the first
> port was and how the other ports came about.
>
I'm not sure what the `first port' means, but as far as I know the
first port to a commodity OS was to SPARC/SunOS 4.x (probably 4.1, but
might have been 4.0.3 or something) in 1992 or so. It ran on
SPARC/Mach before that (as well as other processors), so it was mostly
getting it to talk to the OS not actually getting the compiler to
generate code that caused pain. I remember mmap being a horrible
nightmare (I think SunOS's mmap didn't really work right).
--tim
>>>>> "Tim" == Tim Bradshaw <···@tfeb.org> writes:
Tim> Raymond Toy <···@rtp.ericsson.se> writes:
>>
>> From a historical view, it would be interesting to know what the first
>> port was and how the other ports came about.
>>
Tim> I'm not sure what the `first port' means, but as far as I know the
Tim> first port to a commodity OS was to SPARC/SunOS 4.x (probably 4.1, but
Tim> might have been 4.0.3 or something) in 1992 or so. It ran on
Tim> SPARC/Mach before that (as well as other processors), so it was mostly
Tim> getting it to talk to the OS not actually getting the compiler to
Tim> generate code that caused pain. I remember mmap being a horrible
Tim> nightmare (I think SunOS's mmap didn't really work right).
Basically 'first port' was the first machine to run CMUCL (or was it
Spice Lisp then?). What other architectures did it run on, and in
what order ports to other arch's and OS's happened?
I understand that a full recompile used to take some 15 hours in the
early days of the Python compiler. (I think). It would also be
interesting to know what capabilities various versions of the compiler
had. When was native code generation added or was it always there?
How much was done in C and how much in Lisp? When did it get into
it's current form where almost nothing is in C except for GC and
everything is in Lisp?
"Inquiring minds want to know" :-)
I find this kind of history quite interesting and often left out in
classes.[1]
Ray
Footnotes:
[1] Back in college, I did a thought experiment: With a good
telescope, could I, from first principles, measure the size of the
universe? I think I got as far as the stars within a parsec or two,
but never could went farther than that. Never had the chance to look
up the literature or go much farther than that. I knew about the
variable stars with fixed intrinsic brightness (Seifert variables?),
but didn't know if I could measure their distance with a telescope.
That would have extended my reach to the Milky Way, perhaps.
From: Tim Moore
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <9aiffa$eog$0@216.39.145.192>
On 5 Apr 2001, Raymond Toy wrote:
> >>>>> "Tim" == Tim Bradshaw <···@tfeb.org> writes:
>
> Tim> Raymond Toy <···@rtp.ericsson.se> writes:
> >>
> >> From a historical view, it would be interesting to know what the first
> >> port was and how the other ports came about.
> >>
>
> Tim> I'm not sure what the `first port' means, but as far as I know the
> Tim> first port to a commodity OS was to SPARC/SunOS 4.x (probably 4.1, but
> Tim> might have been 4.0.3 or something) in 1992 or so. It ran on
> Tim> SPARC/Mach before that (as well as other processors), so it was mostly
> Tim> getting it to talk to the OS not actually getting the compiler to
> Tim> generate code that caused pain. I remember mmap being a horrible
> Tim> nightmare (I think SunOS's mmap didn't really work right).
>
> Basically 'first port' was the first machine to run CMUCL (or was it
> Spice Lisp then?). What other architectures did it run on, and in
> what order ports to other arch's and OS's happened?
I recall seeing a historical note from ram, but I can't find it now.
Spice Lisp was bootstrapped from MacLisp on a PDP-10, targeted at the
now-obscure Perq. At some point the compiler became too "Common Lispy" to
run in MacLisp. I think the IBM RT was the next port; around that time
(mid-80s?) Spice Lisp mutated into CMUCL and the new Python compiler was
written.
Tim
> I understand that a full recompile used to take some 15 hours in the
> early days of the Python compiler. (I think). It would also be
Heh. Now it takes less than 15 minutes to build compiler and then
recompile everything, including the compiler.
> interesting to know what capabilities various versions of the compiler
> had. When was native code generation added or was it always there?
> How much was done in C and how much in Lisp? When did it get into
> it's current form where almost nothing is in C except for GC and
> everything is in Lisp?
Probably as much as possible was done in Lisp from the beginning. The
"genesis" technique for crafting a Lisp image inside another Lisp is very
old.
Tim
Raymond Toy <···@rtp.ericsson.se> writes:
> [1] Back in college, I did a thought experiment: With a good
> telescope, could I, from first principles, measure the size of the
> universe? I think I got as far as the stars within a parsec or two,
> but never could went farther than that. Never had the chance to look
> up the literature or go much farther than that. I knew about the
> variable stars with fixed intrinsic brightness (Seifert variables?),
Cepheid.
> but didn't know if I could measure their distance with a telescope.
> That would have extended my reach to the Milky Way, perhaps.
Using Cepheid variables gets you out of our galaxy, I think (I should
know; this is almost my field...); these days good telescopes can
resolve stars in quite far-away galaxies.
I believe that the concept of measuring distances is the `cosmological
distance ladder', which can be googled for.
Christophe
--
Jesus College, Cambridge, CB5 8BL +44 1223 524 842
(FORMAT T "(·@{~w ········@{~w~^ ~})" 'FORMAT T "(·@{~w ········@{~w~^ ~})")
>>>>> "Christophe" == Christophe Rhodes <·····@cam.ac.uk> writes:
Christophe> Cepheid.
>> but didn't know if I could measure their distance with a telescope.
>> That would have extended my reach to the Milky Way, perhaps.
Christophe> Using Cepheid variables gets you out of our galaxy, I think (I should
Christophe> know; this is almost my field...); these days good telescopes can
Christophe> resolve stars in quite far-away galaxies.
Yes, but you have to measure the distance to enough Cepheid variables
to figure out the intrinsic brightness was fixed function of the
period. Otherwise, you would know what the instrinsic brightness
was. :-)
Christophe> I believe that the concept of measuring distances is the `cosmological
Christophe> distance ladder', which can be googled for.
Thanks!!! I'll definitely look that up.
Ray
From: Will Deakin
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <3ACD7CDB.2020202@pindar.com>
Raymond Toy wrote:
> With a good telescope, could I, from first principles, measure the
> size of the universe?
Hmmm. This would depend by what you mean by a `good telescope' --
at certain distances the redshift on objects at extreme distance
is so large[1] that you would not physically be able to observe
the light from the objects due to absorption of light by the
glass making up the lens/mirror.
:) will
[1] As I am sure you know, there is a correlation between
redshift and distance. There is then a measurement called the Z
number which relates the redshift of the wavelength of light to
distance, where a Z of two corresponds to a halving of the
wavelength. IIRC object with Z's of three or more have been
observed and would be awkward to see using an optical telescope.
>>>>> "Will" == Will Deakin <········@pindar.com> writes:
Will> [1] As I am sure you know, there is a correlation between
Will> redshift and distance. There is then a measurement called
Will> the Z number which relates the redshift of the wavelength of
Will> light to distance, where a Z of two corresponds to a halving
Will> of the wavelength. IIRC object with Z's of three or more
Will> have been observed and would be awkward to see using an
Will> optical telescope.
But you have to measure the distance to those objects somehow to
figure out that there is a correlation. The question is how do you do
that. I guess Cepheid variables in other galaxies will get you that
correlation, once you can measure the distance to a few Cepheid
variables.
Anyway, this is way off-topic, so I think I'll go to the astronomy
groups if I want more info. :-)
Ray
From: Tim Bradshaw
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <nkjr8z6bby5.fsf@tfeb.org>
Raymond Toy <···@rtp.ericsson.se> writes:
>
> Basically 'first port' was the first machine to run CMUCL (or was it
> Spice Lisp then?). What other architectures did it run on, and in
> what order ports to other arch's and OS's happened?
I think you'd have to ask someone at CMU. It ran on RS/6000 and MIPS
at least, before, and I think it ran on the PC-RT before that. I
guess originally it ran on the perq, but I think that was pre-python.
> I understand that a full recompile used to take some 15 hours in the
> early days of the Python compiler. (I think). It would also be
> interesting to know what capabilities various versions of the compiler
> had. When was native code generation added or was it always there?
> How much was done in C and how much in Lisp? When did it get into
> it's current form where almost nothing is in C except for GC and
> everything is in Lisp?
I'd *guess* it has always been native-code, and always almost all in
Lisp. Given that I guess you could argue that the first version was
at Stanford in 1958 because you can likely trace back a chain to there
(:-).
I'm sure it took 15 hours - everything used to take 15 hours.
I agree with you that this history is fascinating and almost always
forgotten (not just for CMUCL but for everything).
--tim
>>>>> "Tim" == Tim Bradshaw <···@tfeb.org> writes:
Tim> I'm sure it took 15 hours - everything used to take 15 hours.
I think earlier versions didn't take nearly that long because the
compiler wasn't doing as much work. There's some message in the CMUCL
mailing list from Rob McLachlan about this.
Ray
From: Rolf Marvin B�e Lindgren
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <lbzofua47br.fsf@aqualene.uio.no>
[Raymond Toy]
| With a good telescope, could I, from first principles, measure the
| size of the universe?
as far as I understand relativity, you'd never see the limits; due to
the force field you'd be creating as you approached the edge (or the
part you looked at, anyway) you'd be creating ever more of it.
--
Rolf Lindgren http://www.roffe.com/
·····@tag.uio.no
On 05 Apr 2001 14:35:23 -0400, Raymond Toy <···@rtp.ericsson.se> wrote:
> [1] Back in college, I did a thought experiment: With a good
> telescope, could I, from first principles, measure the size of the
> universe? I think I got as far as the stars within a parsec or two,
By the way, the great XVII century astronomer Christiaan Huygens estimated
the distance of the star Sirius with just the naked eye and a disk with a
tiny hole in it. His estimate was good, considering that he didn't know
about the difference in intrinsic brightness between Sirius and the Sun.
Paolo
--
EncyCMUCLopedia * Extensive collection of CMU Common Lisp documentation
http://cvs2.cons.org:8000/cmucl/doc/EncyCMUCLopedia/
From: Hannah Schroeter
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <9bmtsm$5vd$1@c3po.schlund.de>
Hello!
In article <···············@octagon.mrl.nyu.edu>,
Marco Antoniotti <·······@cs.nyu.edu> wrote:
>[...]
>> Yes. And that means, about no way to port it over to OpenBSD.
>> I wonder how the ports e.g. from Linux to FreeBSD have been done.
>Well, it happened tha other way around. The first x86 port of CMUCL
>was for FreeBSD by Paul Werkowski. Then Peter Van Eynde did the Linux
>port.
Okay, thanks for the correction.
However, the question *how* the port from FreeBSD to Linux was done.
Kind regards,
Hannah.
>>>>> "Hannah" == Hannah Schroeter <······@schlund.de> writes:
Hannah> Hello!
Hannah> In article <···············@octagon.mrl.nyu.edu>,
Hannah> Marco Antoniotti <·······@cs.nyu.edu> wrote:
>> [...]
>>> Yes. And that means, about no way to port it over to OpenBSD.
>>> I wonder how the ports e.g. from Linux to FreeBSD have been done.
>> Well, it happened tha other way around. The first x86 port of CMUCL
>> was for FreeBSD by Paul Werkowski. Then Peter Van Eynde did the Linux
>> port.
Hannah> Okay, thanks for the correction.
Hannah> However, the question *how* the port from FreeBSD to Linux was done.
Paul used an existing Alpha port to cross-compile to FreeBSD. Took
him a year, part-time, I think.
Ray
······@schlund.de (Hannah Schroeter) writes:
> Hello!
>
> In article <···············@octagon.mrl.nyu.edu>,
> Marco Antoniotti <·······@cs.nyu.edu> wrote:
>
> >[...]
>
> >> Yes. And that means, about no way to port it over to OpenBSD.
> >> I wonder how the ports e.g. from Linux to FreeBSD have been done.
>
> >Well, it happened tha other way around. The first x86 port of CMUCL
> >was for FreeBSD by Paul Werkowski. Then Peter Van Eynde did the Linux
> >port.
>
> Okay, thanks for the correction.
>
> However, the question *how* the port from FreeBSD to Linux was done.
By cross-compiling from a FreeBSD installation, or maybe from some
other OS, I don't remember exactly what the cross-compilation host
(which can differ from the "ported-from" OS) was.
Given that SBCL has been ported to OpenBSD, I'll look into adding an
OpenBSD port to CMU CL, if I can find the time. Though I don't know
if I can find enough time to do continued maintenance on such a
port (i.e. building current binaries, etc.). Maybe you would be
interested in taking over such maintenance yourself?
Regs, Pierre.
--
Pierre R. Mai <····@acm.org> http://www.pmsf.de/pmai/
The most likely way for the world to be destroyed, most experts agree,
is by accident. That's where we come in; we're computer professionals.
We cause accidents. -- Nathaniel Borenstein
From: Hannah Schroeter
Subject: Re: Good and Free implementation of Common Lisp
Date:
Message-ID: <9c4ctg$6qi$1@c3po.schlund.de>
Hello!
In article <··············@orion.bln.pmsf.de>,
Pierre R. Mai <····@acm.org> wrote:
>[...]
>Given that SBCL has been ported to OpenBSD, I'll look into adding an
>OpenBSD port to CMU CL, if I can find the time. Though I don't know
>if I can find enough time to do continued maintenance on such a
>port (i.e. building current binaries, etc.). Maybe you would be
>interested in taking over such maintenance yourself?
Depends on how much work it is and how easy/complicated it is.
Perhaps it'd be managable if I get one set of cmucl binaries for
OpenBSD once, and then should do a bootstrap to newer versions
in some periods like e.g. monthly; *if* the bootstrap process is
somehow understandable for someone like me who doesn't know cmucl's
implementation interna too much. Ideally it'd be calling one
script like sbcl's make.sh (preceded by clean.sh), or just a few
step that I can e.g. let run on one box, and then do other work
while the compilation processes run.
>Regs, Pierre.
Kind regards,
Hannah.
······@schlund.de (Hannah Schroeter) writes:
> In article <··············@orion.bln.pmsf.de>,
> Pierre R. Mai <····@acm.org> wrote:
> >[...]
>
> >Given that SBCL has been ported to OpenBSD, I'll look into adding an
> >OpenBSD port to CMU CL, if I can find the time. Though I don't know
> >if I can find enough time to do continued maintenance on such a
> >port (i.e. building current binaries, etc.). Maybe you would be
> >interested in taking over such maintenance yourself?
>
> Depends on how much work it is and how easy/complicated it is.
>
> Perhaps it'd be managable if I get one set of cmucl binaries for
> OpenBSD once, and then should do a bootstrap to newer versions
> in some periods like e.g. monthly; *if* the bootstrap process is
> somehow understandable for someone like me who doesn't know cmucl's
> implementation interna too much. Ideally it'd be calling one
> script like sbcl's make.sh (preceded by clean.sh), or just a few
> step that I can e.g. let run on one box, and then do other work
> while the compilation processes run.
Yes, that's the idea. In my current setup (which once started as the
setup that Peter Van Eynde uses in his debian packaging, though it has
undergone increasing changes through time), recompiling CMU CL is just
as easy as typing make. The idea is that with that framework, you can
fetch/update the source directory from CVS, plug some recent binaries
of CMU CL into current (needed once only), put possibly needed
bootstrapping code into a bootstrap.lisp file, and off you go.
Since most bootstrapping issues are cross-platform, and nearly all are
independent of the OS, I'd be able to indicate known-good combinations
of binaries and bootstrapping.
I'll keep you posted on the status of the OpenBSD port...
Regs, Pierre.
--
Pierre R. Mai <····@acm.org> http://www.pmsf.de/pmai/
The most likely way for the world to be destroyed, most experts agree,
is by accident. That's where we come in; we're computer professionals.
We cause accidents. -- Nathaniel Borenstein
In article <············@c3po.schlund.de>, Hannah Schroeter wrote:
> ...
>>Therefore you need a running CMUCL to compile CMUCL.
>>That means: You have to start with a binary.
>
>Yes. And that means, about no way to port it over to OpenBSD.
>I wonder how the ports e.g. from Linux to FreeBSD have been done.
i think it was ported to linux from freebsd
hs