From: Kakadu <kakadu.hafanana@gmail.com>
To: Daniel de Rauglaudre <daniel.de_rauglaudre@inria.fr>
Cc: Caml List <caml-list@inria.fr>
Subject: Re: [Caml-list] OCaml+spacetime crashes during camlp5 compilation
Date: Sun, 19 Feb 2017 00:14:54 +0300 [thread overview]
Message-ID: <CAGmVoG2Nf4kunqc8=L8FWmerbTfTamSPJ6rQehcn8EWuFZ0keg@mail.gmail.com> (raw)
In-Reply-To: <CAGmVoG2vdqnD8R-sc0LW-PZuQxdwHL2RgrKDaW-xWkESOeHYBQ@mail.gmail.com>
Daniel,
It seems that program doesn't happen with ocaml_src/meta/q_MLast.ml
because it is compiled with ocamlc.opt and another q_MLast with
ocamlopt.opt.
Happy hacking,
Kakadu
On Sun, Feb 19, 2017 at 12:02 AM, Kakadu <kakadu.hafanana@gmail.com> wrote:
> I also tried to comment some code in the file. OCamlopt tries to build
> it and fails after eating all free RAM (5Gb).
>
> On Sat, Feb 18, 2017 at 11:52 PM, Gerd Stolpmann <info@gerd-stolpmann.de> wrote:
>> Am Samstag, den 18.02.2017, 11:31 +0100 schrieb Daniel de Rauglaudre:
>>> On Sat, Feb 18, 2017 at 08:49:12AM +0000, Leo White wrote:
>>>
>>> >
>>> > As Gabriel said, it is probably just camlp5 being expensive to
>>> > compile
>>> > and spacetime tipping it over the edge. Have you tried increasing
>>> > the
>>> > stack size limit?
>>> Hmmm... perhaps I could test this bug by decreasing my stack size
>>> limit on my computer?
>>>
>>> The example given seems to show a problem with the file
>>> meta/q_MLast.ml which indeed contains a statement EXTEND which
>>> generates a rather long OCaml expression. Strange that this problem
>>> does not occur with ocaml_src/meta/q_MLast.ml which is a pretty
>>> printed version of the same file (for Camlp5 cold stard).
>>>
>>> I could try to split this EXTEND into several parts to make OCaml
>>> not having to use so much memory?
>>>
>>> BTW, how to show one's stack size limit (Linux) and change it?
>>
>> print: ulimit -s (or just ulimit -a). This is in kilobytes.
>>
>> Change: e.g. ulimit -s 16384 to set it to 16MB.
>>
>> Note that there is a soft and a hard limit, and you cannot increase the
>> soft limit (which is the effective one) beyond the hard limit. Print
>> the hard limit with "ulimit -H -s". Root can also increase the hard
>> limit, however there's sometimes a per-architecture kernel limit.
>>
>> Gerd
>> --
>> ------------------------------------------------------------
>> Gerd Stolpmann, Darmstadt, Germany gerd@gerd-stolpmann.de
>> My OCaml site: http://www.camlcity.org
>> Contact details: http://www.camlcity.org/contact.html
>> Company homepage: http://www.gerd-stolpmann.de
>> ------------------------------------------------------------
>>
>>
next prev parent reply other threads:[~2017-02-18 21:15 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-17 22:00 Kakadu
2017-02-18 1:53 ` Daniel de Rauglaudre
2017-02-18 1:58 ` Gabriel Scherer
2017-02-18 7:39 ` Kakadu
2017-02-18 8:49 ` Leo White
2017-02-18 10:31 ` Daniel de Rauglaudre
2017-02-18 11:17 ` Kakadu
2017-02-18 20:52 ` Gerd Stolpmann
2017-02-18 21:02 ` Kakadu
2017-02-18 21:14 ` Kakadu [this message]
2017-02-19 3:00 ` Daniel de Rauglaudre
2017-02-19 7:34 ` Kakadu
2017-02-19 8:43 ` Daniel de Rauglaudre
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAGmVoG2Nf4kunqc8=L8FWmerbTfTamSPJ6rQehcn8EWuFZ0keg@mail.gmail.com' \
--to=kakadu.hafanana@gmail.com \
--cc=caml-list@inria.fr \
--cc=daniel.de_rauglaudre@inria.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox