From: Eli Zaretskii <eliz@gnu.org>
To: "Daniel Pettersson" <daniel@dpettersson.net>,
"João Távora" <joaotavora@gmail.com>
Cc: 69241@debbugs.gnu.org
Subject: bug#69241: Fixed patch issues
Date: Sat, 09 Mar 2024 10:56:42 +0200 [thread overview]
Message-ID: <86msr7ye9h.fsf@gnu.org> (raw)
In-Reply-To: <m2bk80dbhh.fsf@dpettersson.net> (message from Daniel Pettersson on Wed, 28 Feb 2024 13:22:50 +0100)
> From: Daniel Pettersson <daniel@dpettersson.net>
> Date: Wed, 28 Feb 2024 13:22:50 +0100
>
> The previous patch included an large regression, where "batch" of
> messages where handled as an LIFO. This patch should fix the issue and
> also removes the dependency on timer.el internal apis. Which makes it
> instead dependent on undocumented behavior that timers of equal time are
> executed in LIFO order.
João, any comments or suggestions? Should I install this?
> >From c4d5ddb9ce5cdb8c283928daf6b166e4ce5a430d Mon Sep 17 00:00:00 2001
> From: Daniel Pettersson <daniel@dpettersson.net>
> Date: Wed, 28 Feb 2024 13:03:56 +0100
> Subject: [PATCH] Jsonrpc: improve performance of process filter function
>
> `run-at-time' keeps `timer-list' list sorted by inserting each timer
> based on the timer value. This means that `timer--time-less-p' needs is
> executed ~ N*N/2 times for each N pending messages. This means that
> jsonrpc becomes unusable for connections that generate a lot messages at
> the same time.
>
> * lisp/jsonrpc.el (Version): Bump to 1.0.25
> (jsonrpc--process-filter): Improve performance
> ---
> lisp/jsonrpc.el | 28 +++++++++++++++++++---------
> 1 file changed, 19 insertions(+), 9 deletions(-)
>
> diff --git a/lisp/jsonrpc.el b/lisp/jsonrpc.el
> index 14fe0447008..5037d8c5b2b 100644
> --- a/lisp/jsonrpc.el
> +++ b/lisp/jsonrpc.el
> @@ -4,7 +4,7 @@
>
> ;; Author: João Távora <joaotavora@gmail.com>
> ;; Keywords: processes, languages, extensions
> -;; Version: 1.0.24
> +;; Version: 1.0.25
> ;; Package-Requires: ((emacs "25.2"))
>
> ;; This is a GNU ELPA :core package. Avoid functionality that is not
> @@ -760,10 +760,11 @@ jsonrpc--process-filter
> (setq message
> (plist-put message :jsonrpc-json
> (buffer-string)))
> - (process-put proc 'jsonrpc-mqueue
> - (nconc (process-get proc
> - 'jsonrpc-mqueue)
> - (list message)))))
> + ;; Put new messages at the front of the queue,
> + ;; this is correct as the order is reversed
> + ;; before putting the timers on `timer-list'.
> + (push message
> + (process-get proc 'jsonrpc-mqueue))))
> (goto-char message-end)
> (let ((inhibit-read-only t))
> (delete-region (point-min) (point)))
> @@ -782,11 +783,20 @@ jsonrpc--process-filter
> ;; non-locally (typically the reply to a request), so do
> ;; this all this processing in top-level loops timer.
> (cl-loop
> + ;; `timer-activate' orders timers by time, which is an
> + ;; very expensive operation when jsonrpc-mqueue is large,
> + ;; therefore the time object is reused for each timer
> + ;; created.
> + with time = (current-time)
> for msg = (pop (process-get proc 'jsonrpc-mqueue)) while msg
> - do (run-at-time 0 nil
> - (lambda (m) (with-temp-buffer
> - (jsonrpc-connection-receive conn m)))
> - msg)))))))
> + do (let ((timer (timer-create)))
> + (timer-set-time timer time)
> + (timer-set-function timer
> + (lambda (conn msg)
> + (with-temp-buffer
> + (jsonrpc-connection-receive conn msg)))
> + (list conn msg))
> + (timer-activate timer))))))))
>
> (defun jsonrpc--remove (conn id &optional deferred-spec)
> "Cancel CONN's continuations for ID, including its timer, if it exists.
> --
> 2.39.3 (Apple Git-145)
>
next prev parent reply other threads:[~2024-03-09 8:56 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-18 1:53 bug#69241: [PATCH] Jsonrpc: improve performance of process filter function Daniel Pettersson
2024-02-28 12:22 ` bug#69241: Fixed patch issues Daniel Pettersson
2024-03-09 8:55 ` Eli Zaretskii
2024-03-09 8:56 ` Eli Zaretskii [this message]
2024-03-09 10:56 ` João Távora
2024-03-10 14:28 ` Daniel Pettersson
2024-03-10 14:41 ` João Távora
2024-03-11 9:44 ` Daniel Pettersson
2024-03-11 13:11 ` Eli Zaretskii
2024-03-11 14:48 ` Daniel Pettersson
2024-03-11 16:27 ` Eli Zaretskii
2024-03-11 19:03 ` Daniel Pettersson
2024-03-11 19:18 ` Eli Zaretskii
2024-03-11 19:39 ` Daniel Pettersson
2024-03-12 3:27 ` Eli Zaretskii
2024-03-12 13:31 ` Eli Zaretskii
2024-03-12 14:34 ` Daniel Pettersson
2024-03-12 15:00 ` Eli Zaretskii
2024-03-12 13:28 ` Eli Zaretskii
2024-03-12 14:33 ` Daniel Pettersson
2024-03-10 16:46 ` Eli Zaretskii
2024-03-10 20:05 ` Daniel Pettersson
2024-03-10 23:30 ` Daniel Pettersson
2024-03-11 12:24 ` Eli Zaretskii
2024-03-11 12:38 ` Daniel Pettersson
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=86msr7ye9h.fsf@gnu.org \
--to=eliz@gnu.org \
--cc=69241@debbugs.gnu.org \
--cc=daniel@dpettersson.net \
--cc=joaotavora@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 external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.