unofficial mirror of emacs-devel@gnu.org 
 help / color / mirror / code / Atom feed
From: "John Wiegley" <johnw@gnu.org>
To: Paul Pogonyshev <pogonyshev@gmail.com>
Cc: Emacs developers <emacs-devel@gnu.org>
Subject: Re: Make computational threads leave user interface usable
Date: Wed, 01 Nov 2017 11:12:02 -0700	[thread overview]
Message-ID: <m2h8uduavh.fsf@newartisans.com> (raw)
In-Reply-To: <CAG7Bpar+=Zc9yn6ss2vKnuMCPvJ4bEZ8bJYKb_r1YrKUwgN0Vg@mail.gmail.com> (Paul Pogonyshev's message of "Wed, 1 Nov 2017 16:06:23 +0100")

>>>>> "PP" == Paul Pogonyshev <pogonyshev@gmail.com> writes:

PP> Since several versions Emacs has Lisp threads, but they are not used much,
PP> because 1) only one thread executes at any given time; 2) threads yield
PP> control to each other only with explicit (thread-yield) call or IO blocks.
PP> Which means that it is pointless to start a new thread for heavy
PP> computation: it will lock UI until finished anyway.

I wouldn't say it's pointless, just that it calls for a programming regimen
where you add (thread-yield) calls at appropriate places.

PP> Attached patch tries to solve point 2 only by making threads
PP> automatically yield control to each other from time to time. The patch
PP> is mainly for discussion.

This would introduce the sort of indeterminacy that raised so much objection
to threading support last year. I'm fairly opposed to pre-emptive threading
until we've heard from the field about the success of green threading.

    (make-thread (lambda ()
                   (dotimes (n 10000000)
                     (when (= (% n 1000000) 0)
                       (message "%s" n)))
                   (message "done")))

I think the problem here is not with Emacs' threading model, but how you've
written the code. It is currently a _feature_ that control is not taken away
while engaged in such a computation. For example, this allows delicate state
manipulation without interference from other threads.

If periodic suspension were desired, you could have written:

    (make-thread (lambda ()
                   (dotimes (n 10000000)
                     (thread-yield)
                     (when (= (% n 1000000) 0)
                       (message "%s" n)))
                   (message "done")))

This gives the UI access to Emacs again, but otherwise keeps executing if no
other threads need control.

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2



  parent reply	other threads:[~2017-11-01 18:12 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-01 15:06 Make computational threads leave user interface usable Paul Pogonyshev
2017-11-01 15:10 ` Paul Pogonyshev
2017-11-01 18:12 ` John Wiegley [this message]
2017-11-01 20:03   ` Eli Zaretskii
2017-11-01 21:19   ` Paul Pogonyshev
2017-11-01 21:47     ` John Wiegley
2017-11-03 15:50     ` Stefan Monnier
2017-11-01 19:10 ` Noam Postavsky
2017-11-01 20:16   ` Eli Zaretskii
2017-11-01 20:26     ` Noam Postavsky

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=m2h8uduavh.fsf@newartisans.com \
    --to=johnw@gnu.org \
    --cc=emacs-devel@gnu.org \
    --cc=pogonyshev@gmail.com \
    /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).