From: Benedikt Meurer <benedikt.meurer@googlemail.com>
To: caml-list@yquem.inria.fr
Subject: Re: [Caml-list] OCaml GC [was Is OCaml fast?]
Date: Sun, 28 Nov 2010 15:29:08 +0100 [thread overview]
Message-ID: <F909DA08-5F0C-4B3D-8FD2-4EB948098176@googlemail.com> (raw)
In-Reply-To: <4CF25878.4060202@univ-savoie.fr>
Speaking of the OCaml GC in general, wouldn't it make sense to replace the current generational collector with a collector framework that requires less copying in the common case. For example, dividing the heap into two parts, "large object space" and "small object space", where LOS is managed by mark-sweep/mark-compact (could use the current major heap) and SOS is managed by a recent mark-copy/mark-region collector (Immix [1] comes to mind here). That way objects would no longer need to be copied between generations, and using Immix may also reduce cache misses and improve locality of small/medium sized objects (with proper evacuation heuristics). This should be doable with a moderate amount of work, and should require no incompatible changes, while opening the door for concurrent/parallel garbage collection (this would require incompatible changes then, replacing caml_young_ptr/caml_young_limit with TLS vars, etc.).
Benedikt
[1] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.144.3640
next prev parent reply other threads:[~2010-11-28 14:29 UTC|newest]
Thread overview: 113+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-22 13:21 Is OCaml fast? Thanassis Tsiodras
2010-11-22 13:35 ` [Caml-list] " Gregory Bellier
2010-11-22 13:39 ` Lukasz Stafiniak
2010-11-22 13:42 ` Thomas Fischbacher
2010-11-22 13:43 ` Sylvain Le Gall
2010-11-22 13:55 ` [Caml-list] " Dario Teixeira
2010-11-23 2:11 ` Isaac Gouy
[not found] ` <999386690.746882.1290478813177.JavaMail.root@zmbs4.inria.fr>
2010-11-23 9:12 ` [Caml-list] " Fabrice Le Fessant
2010-11-23 17:56 ` Isaac Gouy
2010-11-23 19:54 ` [Caml-list] " Mark Diekhans
2010-11-23 20:04 ` Isaac Gouy
2010-11-23 22:56 ` [Caml-list] " oliver
2010-11-23 23:54 ` Jon Harrop
2010-11-24 1:24 ` Erik de Castro Lopo
2010-11-25 11:17 ` Jon Harrop
2010-11-22 14:04 ` Gerd Stolpmann
2010-11-22 14:22 ` [was: Re: Is OCaml fast?] OCaml Shootout task force Sylvain Le Gall
2010-11-22 14:36 ` [Caml-list] Is OCaml fast? bluestorm
2010-11-22 15:01 ` Török Edwin
2010-11-22 15:54 ` Goswin von Brederlow
2010-11-22 15:02 ` Gerd Stolpmann
[not found] ` <582306206.731582.1290438133628.JavaMail.root@zmbs4.inria.fr>
2010-11-22 16:46 ` Fabrice Le Fessant
2010-11-22 18:33 ` Török Edwin
2010-11-27 21:11 ` Pierre Etchemaïté
2010-11-28 13:26 ` OCaml GC [was Is OCaml fast?] Christophe Raffalli
2010-11-28 14:29 ` Benedikt Meurer [this message]
2010-11-28 18:57 ` [Caml-list] " Eray Ozkural
2010-11-28 19:40 ` Jon Harrop
2010-11-28 19:59 ` Benedikt Meurer
2010-11-28 23:34 ` Jon Harrop
2010-11-29 12:11 ` Benedikt Meurer
2010-11-29 22:27 ` Török Edwin
[not found] ` <916556265.243293.1291069665032.JavaMail.root@zmbs1.inria.fr>
2010-12-02 12:57 ` Damien Doligez
2010-12-02 15:07 ` Török Edwin
2010-12-03 21:42 ` [Caml-list] optimizing caml_modify [was OCaml GC [was Is OCaml fast?]] ygrek
2010-11-23 2:03 ` Is OCaml fast? Isaac Gouy
2010-11-23 10:37 ` [Caml-list] " Christophe TROESTLER
2010-11-23 15:50 ` Jon Harrop
2010-11-23 18:06 ` Isaac Gouy
2010-11-23 21:34 ` [Caml-list] " Jon Harrop
2010-11-23 16:08 ` Jon Harrop
2010-11-23 18:03 ` Isaac Gouy
2010-11-23 19:14 ` [Caml-list] " Török Edwin
2010-11-23 20:17 ` Isaac Gouy
2010-11-23 21:14 ` [Caml-list] " Christophe TROESTLER
2010-11-23 21:55 ` Isaac Gouy
2010-11-23 22:25 ` [Caml-list] " Richard Jones
2010-11-24 0:11 ` Isaac Gouy
2010-11-23 17:53 ` Isaac Gouy
2010-11-23 19:24 ` [Caml-list] " Gerd Stolpmann
2010-11-23 20:28 ` Isaac Gouy
2010-11-23 20:55 ` [Caml-list] " Gerd Stolpmann
2010-11-23 21:32 ` Isaac Gouy
2010-11-24 22:28 ` [Caml-list] " Goswin von Brederlow
2010-11-23 23:21 ` evil sloot
2010-11-24 6:54 ` [Caml-list] " David Rajchenbach-Teller
2010-11-22 17:02 ` [Caml-list] " Oliver Bandel
2010-11-22 17:08 ` David Rajchenbach-Teller
2010-11-22 17:23 ` Oliver Bandel
2010-11-22 17:54 ` David Rajchenbach-Teller
2010-11-22 23:55 ` Jeff Schultz
2010-11-22 23:28 ` Eray Ozkural
2010-11-23 2:01 ` Isaac Gouy
2010-11-23 23:27 ` [Caml-list] " oliver
2010-11-24 0:23 ` Isaac Gouy
2010-11-24 1:36 ` [Caml-list] " Eray Ozkural
2010-11-24 2:13 ` Isaac Gouy
2010-11-24 4:39 ` [Caml-list] " Jeff Meister
2010-11-24 6:22 ` Andrew
2010-11-24 7:16 ` Isaac Gouy
2010-11-24 6:50 ` Isaac Gouy
2010-11-24 10:24 ` [Caml-list] " Christophe Troestler
2010-11-24 11:33 ` Eray Ozkural
2010-11-24 17:32 ` Isaac Gouy
2010-11-24 17:57 ` [Caml-list] " Christophe Raffalli
2010-11-24 19:07 ` Ed Keith
2010-11-24 19:13 ` Isaac Gouy
2010-11-24 19:17 ` [Caml-list] " David Rajchenbach-Teller
2010-11-24 22:25 ` Oliver Bandel
2010-11-25 16:59 ` Stefan Monnier
2010-11-25 18:21 ` Isaac Gouy
2010-11-25 22:11 ` [Caml-list] " Jon Harrop
[not found] ` <1534555381.33107.1290723160355.JavaMail.root@zmbs4.inria.fr>
2010-11-25 22:50 ` Fabrice Le Fessant
2010-11-26 20:25 ` Isaac Gouy
2010-11-27 18:55 ` Stefan Monnier
2010-11-28 18:14 ` [Caml-list] " oliver
2010-11-29 14:19 ` Gerd Stolpmann
2010-11-29 16:12 ` Threading and SharedMem (Re: [Caml-list] Re: Is OCaml fast?) Oliver Bandel
2010-11-29 16:24 ` Gerd Stolpmann
2010-11-29 16:33 ` Oliver Bandel
2010-11-27 15:58 ` [Caml-list] Re: Is OCaml fast? Christophe Raffalli
2010-11-28 18:17 ` oliver
2010-11-29 7:33 ` Christophe Raffalli
2010-11-29 11:25 ` Jon Harrop
2010-11-29 11:44 ` oliver
2010-11-29 17:29 ` Christophe Raffalli
2010-11-24 6:55 ` David Rajchenbach-Teller
2010-11-24 7:23 ` Isaac Gouy
2010-11-23 2:06 ` Isaac Gouy
[not found] ` <1110536178.728445.1290434684177.JavaMail.root@zmbs4.inria.fr>
2010-11-22 16:39 ` [Caml-list] " Fabrice Le Fessant
2010-11-22 17:21 ` Philippe Strauss
2010-11-22 18:33 ` Oliver Bandel
2010-11-23 1:58 ` Isaac Gouy
2010-11-24 10:29 ` [Caml-list] " David Allsopp
2010-11-24 18:39 ` Isaac Gouy
2010-11-24 20:59 ` [Caml-list] " David Allsopp
2010-11-25 0:16 ` Isaac Gouy
2010-11-24 14:07 ` [Caml-list] " Cedric Cellier
2010-11-24 14:34 ` Vincent Aravantinos
2010-11-24 15:30 ` Thanassis Tsiodras
2010-11-24 16:26 ` Oliver Bandel
2010-11-24 16:27 ` Vincent Aravantinos
2010-11-24 18:16 ` Isaac Gouy
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=F909DA08-5F0C-4B3D-8FD2-4EB948098176@googlemail.com \
--to=benedikt.meurer@googlemail.com \
--cc=caml-list@yquem.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;
as well as URLs for NNTP newsgroup(s).