GNU logs - #19235, boring messages

Message sent to bug-guile@HIDDEN:

Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Chris Vine <chris@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Sun, 30 Nov 2014 23:30:05 +0000
Resent-Message-ID: <handler.19235.B.14173901488121 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
To: 19235 <at> debbugs.gnu.org
Date: Sun, 30 Nov 2014 23:28:34 +0000
From: Chris Vine <chris@HIDDEN>
Message-ID: <20141130232834.32cbf5b2@HIDDEN>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: list
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>

The make-fresh-user-module procedure leaks memory in guile-2.0.11 as
demonstrated by the attached test case.  This test case should be
invoked either with the "shared" or "fresh" option.  If invoked with
the "fresh" option, it will call make-fresh-user-module on each
iteration through the loop.  On my 32-bit machine it will steadily
accumulate a memory leak before running out of memory on consuming
approximately 2.2G memory, after about 180,000 iterations.  If called
with the "shared" option, it will accumulate no additional memory while
executing, and will execute normally to the end of its iterations.

The question which might be asked is "Would any sane person ever want
to invoke the make-fresh-user-module procedure more than a few times
in a practical program?".  The answer to this question is "Yes", if
guile is being used as an extension framework for a C or C++ program,
and it executes guile extensions as individual tasks, and it is
necessary that top levels should not to be shared.  The execution of
tasks concurrently is one such case, but there can be many cases where
isolated top levels are desirable for tasks executed serially also.

Test case:

----------------------------- snip -----------------------------

/* compile with 'gcc -O2 -Wall `pkg-config --cflags --libs guile-2.0` -o test-guile' */

#include <unistd.h>
#include <libguile.h>
#include <stdio.h>
#include <string.h>

int fresh;

void* func (void* data)
{

  switch (fresh)
    {
      case 0:
	scm_c_eval_string("");
	break;
      default:
	scm_c_eval_string("(set-current-module (make-fresh-user-module))");
    }

  return NULL;
}

int main (int argc, char *argv[])
{

  int count;

  if (argc != 2
      || (strcmp (argv[1], "shared") &&
	  strcmp (argv[1], "fresh")))
    {
      puts ("Usage: test-guile shared | fresh");
      exit (1);
    }

  if (!strcmp (argv[1], "fresh"))
    {
      puts("Invoking make-fresh-user-module");
      fresh = 1;
    }
  else
    puts("Using shared top level");

  for (count = 0; count < 256000; ++count)
    {
      scm_with_guile(func, NULL);

      if (!(count % 100)) {
	printf("%d ", count);
	fflush(stdout);
      }

      usleep(1);
    }

  puts("");
  return 0;
}

Message sent:

Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Mailer: MIME-tools 5.503 (Entity 5.503)
Content-Type: text/plain; charset=utf-8
X-Loop: help-debbugs@HIDDEN
From: help-debbugs@HIDDEN (GNU bug Tracking System)
To: Chris Vine <chris@HIDDEN>
Subject: bug#19235: Acknowledgement (make-fresh-user-module procedure
 leaks memory)
Message-ID: <handler.19235.B.14173901488121.ack <at> debbugs.gnu.org>
References: <20141130232834.32cbf5b2@HIDDEN>
X-Gnu-PR-Message: ack 19235
X-Gnu-PR-Package: guile
Reply-To: 19235 <at> debbugs.gnu.org
Date: Sun, 30 Nov 2014 23:30:05 +0000

Thank you for filing a new bug report with debbugs.gnu.org.

This is an automatically generated reply to let you know your message
has been received.

Your message is being forwarded to the package maintainers and other
interested parties for their attention; they will reply in due course.

Your message has been sent to the package maintainer(s):
 bug-guile@HIDDEN

If you wish to submit further information on this problem, please
send it to 19235 <at> debbugs.gnu.org.

Please do not send mail to help-debbugs@HIDDEN unless you wish
to report a problem with the Bug-tracking system.

--=20
19235: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=3D19235
GNU Bug Tracking System
Contact help-debbugs@HIDDEN with problems

Message sent to bug-guile@HIDDEN:

Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Mark H Weaver <mhw@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Sun, 07 Dec 2014 08:10:02 +0000
Resent-Message-ID: <handler.19235.B19235.141793977414867 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
To: Chris Vine <chris@HIDDEN>
Cc: 19235 <at> debbugs.gnu.org
From: Mark H Weaver <mhw@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
Date: Sun, 07 Dec 2014 03:07:43 -0500
In-Reply-To: <20141130232834.32cbf5b2@HIDDEN> (Chris Vine's message of
 "Sun, 30 Nov 2014 23:28:34 +0000")
Message-ID: <87iohn51dc.fsf@HIDDEN>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.4 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Precedence: list
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>

Chris Vine <chris@HIDDEN> writes:

> The make-fresh-user-module procedure leaks memory in guile-2.0.11 as
> demonstrated by the attached test case.  [...]
>
> The question which might be asked is "Would any sane person ever want
> to invoke the make-fresh-user-module procedure more than a few times
> in a practical program?".  The answer to this question is "Yes", if
> guile is being used as an extension framework for a C or C++ program,
> and it executes guile extensions as individual tasks, and it is
> necessary that top levels should not to be shared.  The execution of
> tasks concurrently is one such case, but there can be many cases where
> isolated top levels are desirable for tasks executed serially also.

Unfortunately, Guile modules cannot be garbage collected.  The problem
is that modules are usually referenced by name, not by direct pointers.
Every module must have a name, due to the way our macro expander works.
Modules created by 'make-module' or 'make-fresh-user-module' are named
using gensyms.

We maintain a global map from module names to module objects, and we can
never safely delete from this map, because we cannot prove that the
module name won't be looked up in the future.

I agree that your use case is reasonable.  I'll think about how we might
allow unnamed modules to be collected, but I'm afraid it might be quite
difficult.

In the meantime, I wrote a procedure that uses undocumented interfaces
to forcefully delete a module from the name->module map.  However, I
must emphasize that this procedure is likely to break in a future
version of Guile.  However, it should work in the 2.0.x series.

--8<---------------cut here---------------start------------->8---
;; WARNING: Uses undocumented interfaces; NOT FUTURE PROOF!!
;; This needs (srfi srfi-1)
(define (delete-module! module)
  (let* ((name (module-name module))
         (last-name (last name))
         (parent-name (drop-right name 1))
         (parent (resolve-module parent-name #f))
         (var (module-variable parent last-name)))
    (when (and var
               (variable-bound? var)
               (eqv? (variable-ref var) module))
      (hashq-remove! (module-obarray parent) last-name))
    (hashq-remove! (module-submodules parent) last-name)
    #f))
--8<---------------cut here---------------end--------------->8---

I should mention that creating new modules with 'make-fresh-user-module'
is not thread safe, nor is the procedure above.  Both of them mutate the
same name->module map.  For now, I recommend protecting calls to both of
them with a mutex.

      Mark

Message sent to bug-guile@HIDDEN:

Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Chris Vine <chris@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Sun, 07 Dec 2014 14:20:02 +0000
Resent-Message-ID: <handler.19235.B19235.141796194927448 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
To: Mark H Weaver <mhw@HIDDEN>
Cc: 19235 <at> debbugs.gnu.org
Date: Sun, 7 Dec 2014 14:19:03 +0000
From: Chris Vine <chris@HIDDEN>
Message-ID: <20141207141903.1347c764@HIDDEN>
In-Reply-To: <87iohn51dc.fsf@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: list
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>

On Sun, 07 Dec 2014 03:07:43 -0500
Mark H Weaver <mhw@HIDDEN> wrote:
> Chris Vine <chris@HIDDEN> writes:
> 
> > The make-fresh-user-module procedure leaks memory in guile-2.0.11 as
> > demonstrated by the attached test case.  [...]
> >
> > The question which might be asked is "Would any sane person ever
> > want to invoke the make-fresh-user-module procedure more than a few
> > times in a practical program?".  The answer to this question is
> > "Yes", if guile is being used as an extension framework for a C or
> > C++ program, and it executes guile extensions as individual tasks,
> > and it is necessary that top levels should not to be shared.  The
> > execution of tasks concurrently is one such case, but there can be
> > many cases where isolated top levels are desirable for tasks
> > executed serially also.
> 
> Unfortunately, Guile modules cannot be garbage collected.  The problem
> is that modules are usually referenced by name, not by direct
> pointers. Every module must have a name, due to the way our macro
> expander works. Modules created by 'make-module' or
> 'make-fresh-user-module' are named using gensyms.
> 
> We maintain a global map from module names to module objects, and we
> can never safely delete from this map, because we cannot prove that
> the module name won't be looked up in the future.
> 
> I agree that your use case is reasonable.  I'll think about how we
> might allow unnamed modules to be collected, but I'm afraid it might
> be quite difficult.
> 
> In the meantime, I wrote a procedure that uses undocumented interfaces
> to forcefully delete a module from the name->module map.  However, I
> must emphasize that this procedure is likely to break in a future
> version of Guile.  However, it should work in the 2.0.x series.

[snip]

> I should mention that creating new modules with
> 'make-fresh-user-module' is not thread safe, nor is the procedure
> above.  Both of them mutate the same name->module map.  For now, I
> recommend protecting calls to both of them with a mutex.

Thanks for that.  Having make-fresh-user-module (and possibly
set-current-module ??) not thread safe should be relatively easy to
deal with in my use case, subject to the next paragraph.  I would need
to call these procedures in C/C++ code before calling scm_eval_string()
to execute a scheme task, but presumably I can obtain a C variable
reference for make-fresh-user-module by calling scm_c_lookup() -
set-current-module already has a C interface provided by libguile.  I
would also need to use a POSIX mutex, but that is fine as guile uses
native threads.

However, that would not be completely effective if guile might also
call make-fresh-user-module internally, since that would be
unprotected.  Are there any circumstances in which guile might do this?

I know from what you said some time ago that guile module loading is not
thread safe.  Without asking you to exercise powers of clairvoyance,
can you think of any other thread safety problems I should be on the
look out for?  The documentation says that "multiple threads can call
scm_with_guile concurrently" and whereas that might literally be true
there seem a number of other things that one might reasonably expect to
do with scm_with_guile that are not.

On the memory leak, the usage case involves a library so I do not think
I can include code which might be liable to breakage at some
indeterminate time.  It would also be quite difficult to implement by
reference to the same mutex as used to protect make-fresh-user-module
because that would mean reimplementing your suggested code on the C/C++
side.  So I think it best just to document the leakage in the
documentation for the library (my one, not guile), and hope that at
some time it will be possible to deal with it in guile.

Chris

Message sent to bug-guile@HIDDEN:

X-Loop: help-debbugs@HIDDEN
Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Chris Vine <chris@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Fri, 26 Dec 2014 18:27:02 +0000
Resent-Message-ID: <handler.19235.B19235.14196183747451 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
X-GNU-PR-Message: followup 19235
X-GNU-PR-Package: guile
X-GNU-PR-Keywords: 
To: Mark H Weaver <mhw@HIDDEN>
Cc: 19235 <at> debbugs.gnu.org
Received: via spool by 19235-submit <at> debbugs.gnu.org id=B19235.14196183747451
          (code B ref 19235); Fri, 26 Dec 2014 18:27:02 +0000
Received: (at 19235) by debbugs.gnu.org; 26 Dec 2014 18:26:14 +0000
Received: from localhost ([127.0.0.1]:58382 helo=debbugs.gnu.org)
	by debbugs.gnu.org with esmtp (Exim 4.80)
	(envelope-from <debbugs-submit-bounces <at> debbugs.gnu.org>)
	id 1Y4ZaY-0001w6-B7
	for submit <at> debbugs.gnu.org; Fri, 26 Dec 2014 13:26:14 -0500
Received: from smtpout5.wanadoo.co.uk ([80.12.242.80]:24047
 helo=smtpout.wanadoo.co.uk) by debbugs.gnu.org with esmtp (Exim 4.80)
 (envelope-from <chris@HIDDEN>) id 1Y4ZaU-0001vs-JH
 for 19235 <at> debbugs.gnu.org; Fri, 26 Dec 2014 13:26:13 -0500
Received: from laptop.homenet ([95.146.111.203]) by mwinf5d66 with ME
 id YJS81p00E4PMKgF03JS8Ln; Fri, 26 Dec 2014 19:26:08 +0100
X-ME-Helo: laptop.homenet
X-ME-Date: Fri, 26 Dec 2014 19:26:08 +0100
X-ME-IP: 95.146.111.203
Received: from bother.homenet (localhost [127.0.0.1])
 by laptop.homenet (Postfix) with ESMTP id 7513E8C1F0;
 Fri, 26 Dec 2014 18:26:08 +0000 (GMT)
Date: Fri, 26 Dec 2014 18:26:08 +0000
From: Chris Vine <chris@HIDDEN>
Message-ID: <20141226182608.24c7dabf@HIDDEN>
In-Reply-To: <20141207141903.1347c764@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN>
 <20141207141903.1347c764@HIDDEN>
X-Mailer: Claws Mail 3.11.1 (GTK+ 2.24.25; i686-pc-linux-gnu)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
X-Spam-Score: 0.0 (/)
X-BeenThere: debbugs-submit <at> debbugs.gnu.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: <debbugs-submit.debbugs.gnu.org>
List-Unsubscribe: <https://debbugs.gnu.org/cgi-bin/mailman/options/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=unsubscribe>
List-Archive: <https://debbugs.gnu.org/cgi-bin/mailman/private/debbugs-submit/>
List-Post: <mailto:debbugs-submit <at> debbugs.gnu.org>
List-Help: <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=help>
List-Subscribe: <https://debbugs.gnu.org/cgi-bin/mailman/listinfo/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=subscribe>
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
X-Spam-Score: 0.0 (/)

As far as I can tell the make-fresh-user-module procedure is not called
by guile itself, and providing a global mutex for it with a binding
enabling it to be called from scheme code seems to work fine.

This also makes it straightforward to incorporate in a thread-safe
way the code you suggested to free stale user modules.  However, as I
mentioned, I am a bit reluctant to incorporate code which might break
in the future.  Is there any possibility that a "delete-module!"
procedure could be included within the public guile API for the next
release of guile?  It seems like something that could be useful to
anyone using non-default user modules in their code.

Chris

Message sent to bug-guile@HIDDEN:

X-Loop: help-debbugs@HIDDEN
Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Andy Wingo <wingo@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Wed, 22 Jun 2016 17:53:02 +0000
Resent-Message-ID: <handler.19235.B19235.14666179352307 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
X-GNU-PR-Message: followup 19235
X-GNU-PR-Package: guile
X-GNU-PR-Keywords: 
To: Chris Vine <chris@HIDDEN>
Cc: Mark H Weaver <mhw@HIDDEN>, ludo@HIDDEN, 15602 <at> debbugs.gnu.org, 19235 <at> debbugs.gnu.org
Received: via spool by 19235-submit <at> debbugs.gnu.org id=B19235.14666179352307
          (code B ref 19235); Wed, 22 Jun 2016 17:53:02 +0000
Received: (at 19235) by debbugs.gnu.org; 22 Jun 2016 17:52:15 +0000
Received: from localhost ([127.0.0.1]:51415 helo=debbugs.gnu.org)
	by debbugs.gnu.org with esmtp (Exim 4.84_2)
	(envelope-from <debbugs-submit-bounces <at> debbugs.gnu.org>)
	id 1bFmJz-0000b4-Fn
	for submit <at> debbugs.gnu.org; Wed, 22 Jun 2016 13:52:15 -0400
Received: from pb-sasl1.pobox.com ([64.147.108.66]:58329
 helo=sasl.smtp.pobox.com) by debbugs.gnu.org with esmtp (Exim 4.84_2)
 (envelope-from <wingo@HIDDEN>)
 id 1bFmJv-0000ao-DG; Wed, 22 Jun 2016 13:52:13 -0400
Received: from sasl.smtp.pobox.com (unknown [127.0.0.1])
 by pb-sasl1.pobox.com (Postfix) with ESMTP id B73731FFC6;
 Wed, 22 Jun 2016 13:52:09 -0400 (EDT)
DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=pobox.com; h=from:to:cc
 :subject:references:date:in-reply-to:message-id:mime-version
 :content-type; s=sasl; bh=uHz8mIV11CKPJO+Gvjt3WnWbTK0=; b=uDFhaM
 YpKzW04+r+tQpmZHEnji5PdvR3LCIYoTc1X3ozbQSRYQWXiSuQcOYoOQ/TnNjTP2
 fAUbvrY+WqSqt3hqVDmB2V74VTQDXlt0XdZON1N6eKt9HMDHtbIVpRq2ivLrbM8o
 E8AWig0NCN5XnW2zSXETulJds2wuu6CjsTcXc=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=pobox.com; h=from:to:cc
 :subject:references:date:in-reply-to:message-id:mime-version
 :content-type; q=dns; s=sasl; b=rIDWMjJzFBxMwmSAj560Gdt2NWWqaQ9L
 BYnBccZ6mTZqxTpKc2nGNKTe4hHCCA1kHoTQbW7x9Qb0OQ/9Fgm8RpD766yBJLTz
 Ro2ULFoZt3sBK2l08aYbL2gZd/1s8Zd/UrKz9tQlLItpkinBMRZPXHhyvnDzK4mc
 HoZlIsW5BNo=
Received: from pb-sasl1.nyi.icgroup.com (unknown [127.0.0.1])
 by pb-sasl1.pobox.com (Postfix) with ESMTP id AFEC91FFC5;
 Wed, 22 Jun 2016 13:52:09 -0400 (EDT)
Received: from clucks (unknown [88.160.190.192])
 (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits))
 (No client certificate requested)
 by pb-sasl1.pobox.com (Postfix) with ESMTPSA id CBAC11FFC4;
 Wed, 22 Jun 2016 13:52:08 -0400 (EDT)
From: Andy Wingo <wingo@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN> <20141207141903.1347c764@HIDDEN>
 <20141226182608.24c7dabf@HIDDEN>
Date: Wed, 22 Jun 2016 19:52:01 +0200
In-Reply-To: <20141226182608.24c7dabf@HIDDEN> (Chris Vine's message of
 "Fri, 26 Dec 2014 18:26:08 +0000")
Message-ID: <87por9ru0u.fsf@HIDDEN>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
X-Pobox-Relay-ID: 0A5E99C2-38A2-11E6-BF3F-C1836462E9F6-02397024!pb-sasl1.pobox.com
X-Spam-Score: -1.4 (-)
X-BeenThere: debbugs-submit <at> debbugs.gnu.org
X-Mailman-Version: 2.1.18
Precedence: list
List-Id: <debbugs-submit.debbugs.gnu.org>
List-Unsubscribe: <https://debbugs.gnu.org/cgi-bin/mailman/options/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=unsubscribe>
List-Archive: <https://debbugs.gnu.org/cgi-bin/mailman/private/debbugs-submit/>
List-Post: <mailto:debbugs-submit <at> debbugs.gnu.org>
List-Help: <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=help>
List-Subscribe: <https://debbugs.gnu.org/cgi-bin/mailman/listinfo/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=subscribe>
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
X-Spam-Score: -1.4 (-)

In many ways I think Ludovic was right in #15602 -- we should allow
excursions to isolate changes to the module tree.  Sometimes you want an
excursion to never add a module to the tree.  Sometimes you do, but
maybe all in one go and with a mutex, to avoid races -- like, you could
load a file or evaluate some code in a private fork of the module tree,
but then commit it to the main tree afterwards.  Is that a sensible
thing?

Andy

On Fri 26 Dec 2014 19:26, Chris Vine <chris@HIDDEN> writes:

> As far as I can tell the make-fresh-user-module procedure is not called
> by guile itself, and providing a global mutex for it with a binding
> enabling it to be called from scheme code seems to work fine.
>
> This also makes it straightforward to incorporate in a thread-safe
> way the code you suggested to free stale user modules.  However, as I
> mentioned, I am a bit reluctant to incorporate code which might break
> in the future.  Is there any possibility that a "delete-module!"
> procedure could be included within the public guile API for the next
> release of guile?  It seems like something that could be useful to
> anyone using non-default user modules in their code.
>
> Chris

Message sent to bug-guile@HIDDEN:

X-Loop: help-debbugs@HIDDEN
Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Mark H Weaver <mhw@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Thu, 23 Jun 2016 14:19:02 +0000
Resent-Message-ID: <handler.19235.B19235.146669148911964 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
X-GNU-PR-Message: followup 19235
X-GNU-PR-Package: guile
X-GNU-PR-Keywords: 
To: Andy Wingo <wingo@HIDDEN>
Cc: 15602 <at> debbugs.gnu.org, ludo@HIDDEN, Chris Vine <chris@HIDDEN>, 19235 <at> debbugs.gnu.org
Received: via spool by 19235-submit <at> debbugs.gnu.org id=B19235.146669148911964
          (code B ref 19235); Thu, 23 Jun 2016 14:19:02 +0000
Received: (at 19235) by debbugs.gnu.org; 23 Jun 2016 14:18:09 +0000
Received: from localhost ([127.0.0.1]:52761 helo=debbugs.gnu.org)
	by debbugs.gnu.org with esmtp (Exim 4.84_2)
	(envelope-from <debbugs-submit-bounces <at> debbugs.gnu.org>)
	id 1bG5SK-00036p-Qs
	for submit <at> debbugs.gnu.org; Thu, 23 Jun 2016 10:18:09 -0400
Received: from world.peace.net ([50.252.239.5]:32984)
 by debbugs.gnu.org with esmtp (Exim 4.84_2)
 (envelope-from <mhw@HIDDEN>)
 id 1bG5SJ-000363-0T; Thu, 23 Jun 2016 10:18:07 -0400
Received: from c-73-253-48-168.hsd1.ma.comcast.net ([73.253.48.168] helo=jojen)
 by world.peace.net with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128)
 (Exim 4.84_2) (envelope-from <mhw@HIDDEN>)
 id 1bG5SC-0007iY-6c; Thu, 23 Jun 2016 10:18:00 -0400
From: Mark H Weaver <mhw@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN> <20141207141903.1347c764@HIDDEN>
 <20141226182608.24c7dabf@HIDDEN> <87por9ru0u.fsf@HIDDEN>
Date: Thu, 23 Jun 2016 10:17:48 -0400
In-Reply-To: <87por9ru0u.fsf@HIDDEN> (Andy Wingo's message of "Wed, 22 Jun
 2016 19:52:01 +0200")
Message-ID: <878txwrnub.fsf@HIDDEN>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.95 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
X-Spam-Score: 0.0 (/)
X-BeenThere: debbugs-submit <at> debbugs.gnu.org
X-Mailman-Version: 2.1.18
Precedence: list
List-Id: <debbugs-submit.debbugs.gnu.org>
List-Unsubscribe: <https://debbugs.gnu.org/cgi-bin/mailman/options/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=unsubscribe>
List-Archive: <https://debbugs.gnu.org/cgi-bin/mailman/private/debbugs-submit/>
List-Post: <mailto:debbugs-submit <at> debbugs.gnu.org>
List-Help: <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=help>
List-Subscribe: <https://debbugs.gnu.org/cgi-bin/mailman/listinfo/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=subscribe>
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
X-Spam-Score: 0.0 (/)

Andy Wingo <wingo@HIDDEN> writes:

> In many ways I think Ludovic was right in #15602 -- we should allow
> excursions to isolate changes to the module tree.  Sometimes you want an
> excursion to never add a module to the tree.  Sometimes you do, but
> maybe all in one go and with a mutex, to avoid races -- like, you could
> load a file or evaluate some code in a private fork of the module tree,
> but then commit it to the main tree afterwards.  Is that a sensible
> thing?

Yes, I agree.  In fact, I'd been thinking of something along those lines
to enable thread-safe module loading.  More specifically, I was thinking
that there should be a fluid variable that contains some additional
modules that are not yet committed to the global module tree.

Briefly, when a module is auto-loaded by a thread, the new module would
initially be visible only to that thread, and also to any threads
spawned by that thread during the auto-load.  Any attempts to access the
module from other threads would block until the module is either fully
loaded.

One potential issue that has been troubling me is that in Guile's model,
there's no guarantee that a module will _ever_ finish loading.  The main
program itself could simply run from the auto-load.  That's why I think
it's important to propagate permission to threads created during the
auto-load, but maybe there will still be problems.

    Thoughts?
      Mark

Message sent to bug-guile@HIDDEN:

X-Loop: help-debbugs@HIDDEN
Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: ludo@HIDDEN (Ludovic =?UTF-8?Q?Court=C3=A8s?=)
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Fri, 24 Jun 2016 08:05:02 +0000
Resent-Message-ID: <handler.19235.B19235.14667554854906 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
X-GNU-PR-Message: followup 19235
X-GNU-PR-Package: guile
X-GNU-PR-Keywords: 
To: Mark H Weaver <mhw@HIDDEN>
Cc: Andy Wingo <wingo@HIDDEN>, 15602 <at> debbugs.gnu.org, Chris Vine <chris@HIDDEN>, 19235 <at> debbugs.gnu.org
Received: via spool by 19235-submit <at> debbugs.gnu.org id=B19235.14667554854906
          (code B ref 19235); Fri, 24 Jun 2016 08:05:02 +0000
Received: (at 19235) by debbugs.gnu.org; 24 Jun 2016 08:04:45 +0000
Received: from localhost ([127.0.0.1]:53319 helo=debbugs.gnu.org)
	by debbugs.gnu.org with esmtp (Exim 4.84_2)
	(envelope-from <debbugs-submit-bounces <at> debbugs.gnu.org>)
	id 1bGM6X-0001H4-1x
	for submit <at> debbugs.gnu.org; Fri, 24 Jun 2016 04:04:45 -0400
Received: from eggs.gnu.org ([208.118.235.92]:38775)
 by debbugs.gnu.org with esmtp (Exim 4.84_2)
 (envelope-from <ludo@HIDDEN>) id 1bGM6V-0001Gi-IN
 for 19235 <at> debbugs.gnu.org; Fri, 24 Jun 2016 04:04:43 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
 (envelope-from <ludo@HIDDEN>) id 1bGM6P-0006ps-Mj
 for 19235 <at> debbugs.gnu.org; Fri, 24 Jun 2016 04:04:38 -0400
X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on eggs.gnu.org
X-Spam-Level: 
X-Spam-Status: No, score=-0.6 required=5.0 tests=BAYES_50,RP_MATCHES_RCVD
 autolearn=disabled version=3.3.2
Received: from fencepost.gnu.org ([2001:4830:134:3::e]:54148)
 by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from <ludo@HIDDEN>)
 id 1bGM6H-0006oP-0x; Fri, 24 Jun 2016 04:04:29 -0400
Received: from pluto.bordeaux.inria.fr ([193.50.110.57]:47396 helo=pluto)
 by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_128_CBC_SHA1:128)
 (Exim 4.82) (envelope-from <ludo@HIDDEN>)
 id 1bGM6F-0003tR-VS; Fri, 24 Jun 2016 04:04:28 -0400
From: ludo@HIDDEN (Ludovic =?UTF-8?Q?Court=C3=A8s?=)
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN> <20141207141903.1347c764@HIDDEN>
 <20141226182608.24c7dabf@HIDDEN> <87por9ru0u.fsf@HIDDEN>
 <878txwrnub.fsf@HIDDEN>
X-URL: http://www.fdn.fr/~lcourtes/
X-Revolutionary-Date: 7 Messidor an 224 de la =?UTF-8?Q?R=C3=A9volution?=
X-PGP-Key-ID: 0x090B11993D9AEBB5
X-PGP-Key: http://www.fdn.fr/~lcourtes/ludovic.asc
X-PGP-Fingerprint: 3CE4 6455 8A84 FDC6 9DB4  0CFB 090B 1199 3D9A EBB5
X-OS: x86_64-unknown-linux-gnu
Date: Fri, 24 Jun 2016 10:04:24 +0200
In-Reply-To: <878txwrnub.fsf@HIDDEN> (Mark H. Weaver's message of "Thu, 23
 Jun 2016 10:17:48 -0400")
Message-ID: <8760szau7r.fsf@HIDDEN>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]
X-Received-From: 2001:4830:134:3::e
X-Spam-Score: -6.4 (------)
X-BeenThere: debbugs-submit <at> debbugs.gnu.org
X-Mailman-Version: 2.1.18
Precedence: list
List-Id: <debbugs-submit.debbugs.gnu.org>
List-Unsubscribe: <https://debbugs.gnu.org/cgi-bin/mailman/options/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=unsubscribe>
List-Archive: <https://debbugs.gnu.org/cgi-bin/mailman/private/debbugs-submit/>
List-Post: <mailto:debbugs-submit <at> debbugs.gnu.org>
List-Help: <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=help>
List-Subscribe: <https://debbugs.gnu.org/cgi-bin/mailman/listinfo/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=subscribe>
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
X-Spam-Score: -6.4 (------)

Mark H Weaver <mhw@HIDDEN> skribis:

> Andy Wingo <wingo@HIDDEN> writes:
>
>> In many ways I think Ludovic was right in #15602 -- we should allow
>> excursions to isolate changes to the module tree.  Sometimes you want an
>> excursion to never add a module to the tree.  Sometimes you do, but
>> maybe all in one go and with a mutex, to avoid races -- like, you could
>> load a file or evaluate some code in a private fork of the module tree,
>> but then commit it to the main tree afterwards.  Is that a sensible
>> thing?
>
> Yes, I agree.  In fact, I'd been thinking of something along those lines
> to enable thread-safe module loading.  More specifically, I was thinking
> that there should be a fluid variable that contains some additional
> modules that are not yet committed to the global module tree.
>
> Briefly, when a module is auto-loaded by a thread, the new module would
> initially be visible only to that thread, and also to any threads
> spawned by that thread during the auto-load.  Any attempts to access the
> module from other threads would block until the module is either fully
> loaded.

That sounds like a nice idea.

In the current state of things, perhaps this behavior could be emulated
by running =E2=80=98compile-file=E2=80=99 in a module excursion, and passin=
g it a root
module that=E2=80=99s a copy of =E2=80=98the-root-module=E2=80=99, somethin=
g like that (though
that would probably perform badly.)

> One potential issue that has been troubling me is that in Guile's model,
> there's no guarantee that a module will _ever_ finish loading.

I think the fact that we evaluate all the top-level forms is
problematic.  The R6RS phases were a great idea.  :-)

> The main program itself could simply run from the auto-load.  That's
> why I think it's important to propagate permission to threads created
> during the auto-load, but maybe there will still be problems.

I=E2=80=99m not sure what you mean by =E2=80=9Cpropagate permission=E2=80=
=9D?

Ludo=E2=80=99.

Message sent to bug-guile@HIDDEN:

X-Loop: help-debbugs@HIDDEN
Subject: bug#19235: make-fresh-user-module procedure leaks memory
Resent-From: Mark H Weaver <mhw@HIDDEN>
Original-Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
Resent-CC: bug-guile@HIDDEN
Resent-Date: Sun, 26 Jun 2016 00:52:02 +0000
Resent-Message-ID: <handler.19235.B19235.14669022777030 <at> debbugs.gnu.org>
Resent-Sender: help-debbugs@HIDDEN
X-GNU-PR-Message: followup 19235
X-GNU-PR-Package: guile
X-GNU-PR-Keywords: 
To: ludo@HIDDEN (Ludovic =?UTF-8?Q?Court=C3=A8s?=)
Cc: Andy Wingo <wingo@HIDDEN>, 15602 <at> debbugs.gnu.org, Chris Vine <chris@HIDDEN>, 19235 <at> debbugs.gnu.org
Received: via spool by 19235-submit <at> debbugs.gnu.org id=B19235.14669022777030
          (code B ref 19235); Sun, 26 Jun 2016 00:52:02 +0000
Received: (at 19235) by debbugs.gnu.org; 26 Jun 2016 00:51:17 +0000
Received: from localhost ([127.0.0.1]:55844 helo=debbugs.gnu.org)
	by debbugs.gnu.org with esmtp (Exim 4.84_2)
	(envelope-from <debbugs-submit-bounces <at> debbugs.gnu.org>)
	id 1bGyI9-0001pE-DU
	for submit <at> debbugs.gnu.org; Sat, 25 Jun 2016 20:51:17 -0400
Received: from world.peace.net ([50.252.239.5]:58617)
 by debbugs.gnu.org with esmtp (Exim 4.84_2)
 (envelope-from <mhw@HIDDEN>)
 id 1bGyI7-0001ow-Cc; Sat, 25 Jun 2016 20:51:15 -0400
Received: from pool-71-174-35-80.bstnma.east.verizon.net ([71.174.35.80]
 helo=jojen)
 by world.peace.net with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128)
 (Exim 4.84_2) (envelope-from <mhw@HIDDEN>)
 id 1bGyI1-0001EP-6c; Sat, 25 Jun 2016 20:51:09 -0400
From: Mark H Weaver <mhw@HIDDEN>
References: <20141130232834.32cbf5b2@HIDDEN>
 <87iohn51dc.fsf@HIDDEN> <20141207141903.1347c764@HIDDEN>
 <20141226182608.24c7dabf@HIDDEN> <87por9ru0u.fsf@HIDDEN>
 <878txwrnub.fsf@HIDDEN> <8760szau7r.fsf@HIDDEN>
Date: Sat, 25 Jun 2016 20:50:56 -0400
In-Reply-To: <8760szau7r.fsf@HIDDEN> ("Ludovic
 \=\?utf-8\?Q\?Court\=C3\=A8s\=22'\?\=
 \=\?utf-8\?Q\?s\?\= message of "Fri, 24 Jun 2016 10:04:24 +0200")
Message-ID: <878txsrcwf.fsf@HIDDEN>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.95 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
X-Spam-Score: 0.0 (/)
X-BeenThere: debbugs-submit <at> debbugs.gnu.org
X-Mailman-Version: 2.1.18
Precedence: list
List-Id: <debbugs-submit.debbugs.gnu.org>
List-Unsubscribe: <https://debbugs.gnu.org/cgi-bin/mailman/options/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=unsubscribe>
List-Archive: <https://debbugs.gnu.org/cgi-bin/mailman/private/debbugs-submit/>
List-Post: <mailto:debbugs-submit <at> debbugs.gnu.org>
List-Help: <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=help>
List-Subscribe: <https://debbugs.gnu.org/cgi-bin/mailman/listinfo/debbugs-submit>, 
 <mailto:debbugs-submit-request <at> debbugs.gnu.org?subject=subscribe>
Errors-To: debbugs-submit-bounces <at> debbugs.gnu.org
Sender: "Debbugs-submit" <debbugs-submit-bounces <at> debbugs.gnu.org>
X-Spam-Score: 0.0 (/)

ludo@HIDDEN (Ludovic Court=C3=A8s) writes:

> Mark H Weaver <mhw@HIDDEN> skribis:
>
>> One potential issue that has been troubling me is that in Guile's model,
>> there's no guarantee that a module will _ever_ finish loading.
>
> I think the fact that we evaluate all the top-level forms is
> problematic.  The R6RS phases were a great idea.  :-)
>
>> The main program itself could simply run from the auto-load.  That's
>> why I think it's important to propagate permission to threads created
>> during the auto-load, but maybe there will still be problems.
>
> I=E2=80=99m not sure what you mean by =E2=80=9Cpropagate permission=E2=80=
=9D?

I mean: propagate permission to access the not-yet-committed module.
For example, suppose a program loads a module that runs the main event
loop as a top-level form in its body.  This module will never be
committed to the global module table, because it never finishes loading.
Now suppose that it spawns some new threads.  Those threads should have
access to the module.

Similarly, if a module uses 'par-for-each' to initialize some tables,
the spawned threads should have access to the module being loaded.

      Mark

Last modified: Mon, 25 Nov 2019 12:00:02 UTC

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997 nCipher Corporation Ltd, 1994-97 Ian Jackson.