unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
From: Juri Linkov <juri@jurta.org>
Cc: ttn@gnu.org, teirllm@dms.auburn.edu, emacs-devel@gnu.org
Subject: Re: GC (was: lists.texi)
Date: Sat, 25 Jun 2005 00:54:05 +0300	[thread overview]
Message-ID: <87y88zv3vm.fsf@jurta.org> (raw)
In-Reply-To: <u3br7h39r.fsf@gnu.org> (Eli Zaretskii's message of "Fri, 24 Jun 2005 23:08:00 +0200")

>> It helps to increase the value of gc-cons-threshold at least tenfolds
>> to run slow GC less often.
>
> Yes, but then Emacs itself slows down, and when GC eventually
> happens, it, too, takes a very long time.

In my tests, when I have the default gc-cons-threshold set to 400000,
GC takes 1 sec.  When I increase it 100 times to 40000000, GC takes
the same 1 sec (and not 100 sec as if there were linear dependence).
And there is no slowdown.

> I used to run with gc-cons-threshold customized to a very large value,
> but stopped doing that, since the advantages were far too minor and
> disadvantages far too annoying.

Could you tell all disadvantages? (except of obvious one of memory use
which users with large memory can tolerate).

>> Anyway, it would be good to mention this problem in the elisp manual.
>
> You mean, user's manual, right?  This has nothing to do with Lisp
> programming, it's a usage issue.

No.  `gc-cons-threshold' and advice for its usage are documented in
the Emacs Lisp reference manual.  I don't object to documenting it
in the user's manual as well.

>> Currently it suggests increasing `gc-cons-threshold' for a program
>> that creates lots of Lisp data.  There are too many such Emacs
>> packages
>
> As long as ``lots of Lisp data'' isn't defined in some quantitative
> terms, one cannot claim that ``too many Emacs packages'' do this.

In quantitative terms ``lots of Lisp data'' could mean the default
value of `gc-cons-threshold'.  Then ``too many Emacs packages'' means
that with garbage-collection-messages equal to t, many Emacs commands
display the ``Garbage collecting...'' message.

>> In addition to this problem, the manual could also mention the
>> problem of GC duration increasing with the size of allocated Lisp
>> objects.
>
> Isn't it obvious that GC, at least its mark phase, is approximately
> linear in the number of live Lisp objects?

Yes, GC is linear in the number of *live* Lisp objects, but not garbage.
IIUC, the number of live objects GC traverses doesn't depend on the
value of `gc-cons-threshold'.

>> The advice for this problem could be the same: to set
>> `gc-cons-threshold' to a larger value.
>
> Based on my experience, I would object to such advice.

Your advice to look at `gc-cons-threshold' helped me enormously.
Now with a large value of `gc-cons-threshold' I have no slowdown
anymore.  It would be good to share this information with other
Emacs users by mentioning it in the documentation.

-- 
Juri Linkov
http://www.jurta.org/emacs/

  reply	other threads:[~2005-06-24 21:54 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-06-18 23:19 lists.texi Luc Teirlinck
2005-06-19  0:01 ` lists.texi Luc Teirlinck
2005-06-19  0:15 ` lists.texi Luc Teirlinck
2005-06-19  0:37   ` lists.texi Luc Teirlinck
2005-06-19  6:37     ` lists.texi David Kastrup
2005-06-19 15:55     ` lists.texi Richard Stallman
2005-06-19 17:47       ` lists.texi Luc Teirlinck
2005-06-20 17:52         ` lists.texi Richard Stallman
2005-06-20 23:12           ` lists.texi Luc Teirlinck
2005-06-21  5:13             ` lists.texi David Kastrup
2005-06-21 15:13             ` lists.texi Richard M. Stallman
2005-06-21 16:35             ` lists.texi Thien-Thi Nguyen
2005-06-21 19:00               ` lists.texi Luc Teirlinck
2005-06-21 21:56                 ` lists.texi Thien-Thi Nguyen
2005-06-21 19:45               ` lists.texi Luc Teirlinck
2005-06-21 20:58               ` lists.texi Luc Teirlinck
2005-06-21 22:09                 ` lists.texi Thien-Thi Nguyen
2005-06-22 16:28                 ` lists.texi Juri Linkov
2005-06-22 19:27                   ` lists.texi Eli Zaretskii
2005-06-22 18:44                     ` lists.texi Luc Teirlinck
2005-06-22 20:25                     ` lists.texi Luc Teirlinck
2005-06-23 16:53                       ` lists.texi Richard M. Stallman
2005-06-24 19:02                       ` GC (was: lists.texi) Juri Linkov
2005-06-24 19:02                     ` Juri Linkov
2005-06-24 21:08                       ` Eli Zaretskii
2005-06-24 21:54                         ` Juri Linkov [this message]
2005-06-24 23:52                           ` Luc Teirlinck
2005-06-25  0:51                             ` Miles Bader
2005-06-25  9:48                           ` Eli Zaretskii
2005-06-25 11:58                             ` GC Adrian Aichner
2005-06-25 12:53                               ` GC Miles Bader
2005-06-25 21:53                                 ` GC Adrian Aichner
2005-06-26  0:02                                   ` GC Miles Bader
2005-06-26  8:20                                     ` GC Adrian Aichner
2005-06-26 18:51                                       ` GC Eli Zaretskii
2005-06-26 23:43                                         ` GC Juri Linkov
2005-06-27  5:38                                         ` GC Richard M. Stallman
2005-06-26 22:42                                       ` GC Richard M. Stallman
2005-06-25 12:15                             ` GC (was: lists.texi) Miles Bader
2005-06-25 13:10                               ` GC Gaëtan LEURENT
2005-06-25 14:48                                 ` GC Eli Zaretskii
2005-06-25 14:45                               ` GC (was: lists.texi) Eli Zaretskii
2005-06-25 16:40                               ` Richard M. Stallman
2005-06-28  4:55                             ` GC Stefan Monnier
2005-06-28 21:29                               ` GC Richard M. Stallman
2005-07-11 17:00                                 ` GC Stefan Monnier

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

  List information: https://www.gnu.org/software/emacs/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87y88zv3vm.fsf@jurta.org \
    --to=juri@jurta.org \
    --cc=emacs-devel@gnu.org \
    --cc=teirllm@dms.auburn.edu \
    --cc=ttn@gnu.org \
    /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.
Code repositories for project(s) associated with this public inbox

	https://git.savannah.gnu.org/cgit/emacs.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).