[Caml-list] sources for ZINC?

[Caml-list] sources for ZINC?

[Nicolas Ojeda Bar]

Dear list,

I was wondering if the sources for ZINC, described in X. Leroy's thesis
are available anywhere...?

Thanks!
Nicolas

[Caml-list] Re: sources for ZINC?

[Hongbo Zhang]

On 7/12/13 3:59 PM, Nicolas Ojeda Bar wrote:
> Dear list,
>
> I was wondering if the sources for ZINC, described in X. Leroy's thesis
> are available anywhere...?
>
> Thanks!
> Nicolas
>
Hi Nicolas,
    basically current the compiler source tree (excluding asmrun and 
typechecking)

Re: [Caml-list] sources for ZINC?

[Xavier Leroy]

On 2013-07-12 21:59, Nicolas Ojeda Bar wrote:
> Dear list,
> 
> I was wondering if the sources for ZINC, described in X. Leroy's thesis
> are available anywhere...?

Caml Light is pretty close to what's described in the ZINC report,
esp. the abstract machine.  Available here:

http://caml.inria.fr/caml-light/release.html

- Xavier Leroy

Re: [Caml-list] sources for ZINC?

[Benedikt Meurer]

On Jul 16, 2013, at 17:33 , Xavier Leroy <Xavier.Leroy@inria.fr> wrote:

> On 2013-07-12 21:59, Nicolas Ojeda Bar wrote:
>> Dear list,
>> 
>> I was wondering if the sources for ZINC, described in X. Leroy's thesis
>> are available anywhere...?
> 
> Caml Light is pretty close to what's described in the ZINC report,
> esp. the abstract machine.  Available here:
> 
> http://caml.inria.fr/caml-light/release.html

http://caml.inria.fr/caml-light/release.en.html

Re: [Caml-list] sources for ZINC?

[Ashish Agarwal]

[-- Attachment #1: Type: text/plain, Size: 148 bytes --]

>
> > http://caml.inria.fr/caml-light/release.html
>
> http://caml.inria.fr/caml-light/release.en.html


http://ocaml.org/caml-light/releases/   :)

[-- Attachment #2: Type: text/html, Size: 574 bytes --]

[Caml-list] recursive mutexes in ocaml

[Chet Murthy]

Recently I'm writing some multi-threaded code, and ended up wishing I
had recursive mutexes.  Now .... I realize that this is a simple thing
to "get" -- just hack the code, maaaan.  But before (or, erm
.... whilst) I do that, I figured I'd ask -why- ocaml's mutexes aren't
recursive.  Or at least, why there isn't an option for recursive
mutexes?

I realize that at some level, you can -always- eschew recursive
mutexes by passing along extra parameters so that code can know
whether it's locked a particular mutex.  That said, it's (more than) a
bit of a pain, and surely complicates code ....

Is there some other -reason- that recursive mutexes aren't
implemented?  Or is it just a matter of taste?

Thanks,
--chet--

P.S. I found Markus' email about this:

>> I'd consider recursive lock acquisitions bad practice. There has
   never been a case in numerous complex bindings where I would have
   needed this feature. In mission-critical code I even prefer
   error-checking mutexes that prevent me from acquiring locks twice
   or releasing them once too often. As with everything multithreaded:
   the simpler the better. It's hard enough to reason about the simple
   case.

and this is about the ocaml master lock.  But it's the only instance I
find of somebody discussing recursive mutexes in ocaml.

Re: [Caml-list] recursive mutexes in ocaml

[Gerd Stolpmann]

Am Mittwoch, den 17.07.2013, 00:28 -0700 schrieb Chet Murthy:
> Recently I'm writing some multi-threaded code, and ended up wishing I
> had recursive mutexes.  Now .... I realize that this is a simple thing
> to "get" -- just hack the code, maaaan.  But before (or, erm
> .... whilst) I do that, I figured I'd ask -why- ocaml's mutexes aren't
> recursive.  Or at least, why there isn't an option for recursive
> mutexes?
> 
> I realize that at some level, you can -always- eschew recursive
> mutexes by passing along extra parameters so that code can know
> whether it's locked a particular mutex.  That said, it's (more than) a
> bit of a pain, and surely complicates code ....
> 
> Is there some other -reason- that recursive mutexes aren't
> implemented?  Or is it just a matter of taste?

The multi-threading feature in ocaml is quite old. I guess there was
some motivation to keep it as simple as possible, e.g. for maximizing
portability, so you can also build ocaml on an OS that does not support
such fancy features. Nowadays the OS landscape is quite different, of
course, and such frugality can be seen as limitation.

The demand for recursive mutexes is certainly low, because you can
easily have your own little wrapper around Mutex to get them (there is
normally an easy way to recognize the owner). It's just a counter after
all. But yes, having this directly in Mutex would make it simpler. Maybe
file a wish in the bugtracker?

Gerd


> 
> Thanks,
> --chet--
> 
> P.S. I found Markus' email about this:
> 
> >> I'd consider recursive lock acquisitions bad practice. There has
>    never been a case in numerous complex bindings where I would have
>    needed this feature. In mission-critical code I even prefer
>    error-checking mutexes that prevent me from acquiring locks twice
>    or releasing them once too often. As with everything multithreaded:
>    the simpler the better. It's hard enough to reason about the simple
>    case.
> 
> and this is about the ocaml master lock.  But it's the only instance I
> find of somebody discussing recursive mutexes in ocaml.
> 
> 
> 

-- 
------------------------------------------------------------
Gerd Stolpmann, Bad Nauheimer Str.3, 64289 Darmstadt,Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
Phone: +49-6151-153855                  Fax: +49-6151-997714
------------------------------------------------------------

Re: [Caml-list] recursive mutexes in ocaml

[Edgar Friendly]

[-- Attachment #1: Type: text/plain, Size: 3042 bytes --]

There was some back and forth on the batteries-devel mailing list about
getting rid of BatRMutex, batteries' implementation of recursive
mutexes(search the archives).  They're still in, if you want to use them.

E.


On Wed, Jul 17, 2013 at 6:54 AM, Gerd Stolpmann <info@gerd-stolpmann.de>wrote:

> Am Mittwoch, den 17.07.2013, 00:28 -0700 schrieb Chet Murthy:
> > Recently I'm writing some multi-threaded code, and ended up wishing I
> > had recursive mutexes.  Now .... I realize that this is a simple thing
> > to "get" -- just hack the code, maaaan.  But before (or, erm
> > .... whilst) I do that, I figured I'd ask -why- ocaml's mutexes aren't
> > recursive.  Or at least, why there isn't an option for recursive
> > mutexes?
> >
> > I realize that at some level, you can -always- eschew recursive
> > mutexes by passing along extra parameters so that code can know
> > whether it's locked a particular mutex.  That said, it's (more than) a
> > bit of a pain, and surely complicates code ....
> >
> > Is there some other -reason- that recursive mutexes aren't
> > implemented?  Or is it just a matter of taste?
>
> The multi-threading feature in ocaml is quite old. I guess there was
> some motivation to keep it as simple as possible, e.g. for maximizing
> portability, so you can also build ocaml on an OS that does not support
> such fancy features. Nowadays the OS landscape is quite different, of
> course, and such frugality can be seen as limitation.
>
> The demand for recursive mutexes is certainly low, because you can
> easily have your own little wrapper around Mutex to get them (there is
> normally an easy way to recognize the owner). It's just a counter after
> all. But yes, having this directly in Mutex would make it simpler. Maybe
> file a wish in the bugtracker?
>
> Gerd
>
>
> >
> > Thanks,
> > --chet--
> >
> > P.S. I found Markus' email about this:
> >
> > >> I'd consider recursive lock acquisitions bad practice. There has
> >    never been a case in numerous complex bindings where I would have
> >    needed this feature. In mission-critical code I even prefer
> >    error-checking mutexes that prevent me from acquiring locks twice
> >    or releasing them once too often. As with everything multithreaded:
> >    the simpler the better. It's hard enough to reason about the simple
> >    case.
> >
> > and this is about the ocaml master lock.  But it's the only instance I
> > find of somebody discussing recursive mutexes in ocaml.
> >
> >
> >
>
> --
> ------------------------------------------------------------
> Gerd Stolpmann, Bad Nauheimer Str.3, 64289 Darmstadt,Germany
> gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
> Phone: +49-6151-153855                  Fax: +49-6151-997714
> ------------------------------------------------------------
>
>
>
> --
> Caml-list mailing list.  Subscription management and archives:
> https://sympa.inria.fr/sympa/arc/caml-list
> Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
> Bug reports: http://caml.inria.fr/bin/caml-bugs
>

[-- Attachment #2: Type: text/html, Size: 4199 bytes --]

Re: [Caml-list] recursive mutexes in ocaml

[Xavier Leroy]

Hi Chet,

> I figured I'd ask -why- ocaml's mutexes aren't recursive.  Or at
> least, why there isn't an option for recursive mutexes?

Dave Butenhof, one of the main authors of POSIX threads, makes rather
strong points against recursive mutexes:
http://tinyurl.com/butenhof-recursive-mutexes

The killer issue in my opinion is how to make recursive mutexes play
well with pthread_cond_wait() / Condition.wait: POSIX's
pthread_cond_wait() only releases the mutex once, which leads to all
sorts of deadlock if it's been acquired N>1 times.  I believe Java
has a more useful behavior, whereas the mutex is released N times,
then reacquired N times.  But this cannot be implemented easily on top
of POSIX threads.  And don't get me started on Win32 threads...

It might make sense to have recursive mutexes as a separate type,
different from ordinary mutexes, so that it's not usable in
conjunction with Condition.wait.  Apparently, Batteries has/had that.

- Xavier Leroy

Re: [Caml-list] recursive mutexes in ocaml

[Chet Murthy]

Xavier,

Wow, that's excellent.  Made my morning!  I've implemented Java's
monitors, so yeah, they're as you describe, and I cannot imagine doing
it any other way.  What were they -thinkin'- in the POSIX committee?

Ah, well.

OK.  I'll just add the (few) lines of code needed to get basic
recursive mutexes for locking purposes, and leave it at that.  Given
Butenhof's analysis, I -fully- agree with leavin' 'em out of ocaml.
Yikes!

--chet--

On Wednesday, July 17, 2013 07:33:41 PM Xavier Leroy wrote:
> Hi Chet,
> 
> > I figured I'd ask -why- ocaml's mutexes aren't recursive.  Or at
> > least, why there isn't an option for recursive mutexes?
> 
> Dave Butenhof, one of the main authors of POSIX threads, makes rather
> strong points against recursive mutexes:
> http://tinyurl.com/butenhof-recursive-mutexes
> 
> The killer issue in my opinion is how to make recursive mutexes play
> well with pthread_cond_wait() / Condition.wait: POSIX's
> pthread_cond_wait() only releases the mutex once, which leads to all
> sorts of deadlock if it's been acquired N>1 times.  I believe Java
> has a more useful behavior, whereas the mutex is released N times,
> then reacquired N times.  But this cannot be implemented easily on top
> of POSIX threads.  And don't get me started on Win32 threads...
> 
> It might make sense to have recursive mutexes as a separate type,
> different from ordinary mutexes, so that it's not usable in
> conjunction with Condition.wait.  Apparently, Batteries has/had that.
> 
> - Xavier Leroy