From: Paul Eggert <eggert@cs.ucla.edu>
To: Eli Zaretskii <eliz@gnu.org>
Cc: michael_heerdegen@web.de, 37321@debbugs.gnu.org
Subject: bug#37321: 27.0.50; Excessive gc in a use case (el-search)
Date: Sat, 14 Sep 2019 10:57:35 -0700 [thread overview]
Message-ID: <d98f0762-d3ff-90e7-07ad-2b081b0aed29@cs.ucla.edu> (raw)
In-Reply-To: <838sqryzc2.fsf@gnu.org>
[-- Attachment #1: Type: text/plain, Size: 193 bytes --]
On 9/14/19 2:57 AM, Eli Zaretskii wrote:
> If we cannot give any specifics, I think at least telling users to
> re-tune their GC tricks would help.
I installed the attached to try to do that.
[-- Attachment #2: 0001-Improve-doc-of-GC-thresholds.patch --]
[-- Type: text/x-patch, Size: 2794 bytes --]
From 1acc0cc9aaf25c808a60cf09cf8a4d1c653c3aa9 Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert@cs.ucla.edu>
Date: Sat, 14 Sep 2019 10:53:24 -0700
Subject: [PATCH] Improve doc of GC thresholds
* doc/lispref/internals.texi (Garbage Collection), etc/NEWS:
Warn that control over GC is only approximate.
---
doc/lispref/internals.texi | 13 ++++++++++---
etc/NEWS | 7 +++++++
2 files changed, 17 insertions(+), 3 deletions(-)
diff --git a/doc/lispref/internals.texi b/doc/lispref/internals.texi
index f85c266ede..c52999e1cd 100644
--- a/doc/lispref/internals.texi
+++ b/doc/lispref/internals.texi
@@ -533,9 +533,6 @@ Garbage Collection
trigger another garbage collection. You can use the result returned by
@code{garbage-collect} to get an information about size of the particular
object type; space allocated to the contents of buffers does not count.
-Note that the subsequent garbage collection does not happen immediately
-when the threshold is exhausted, but only the next time the Lisp interpreter
-is called.
The initial threshold value is @code{GC_DEFAULT_THRESHOLD}, defined in
@file{alloc.c}. Since it's defined in @code{word_size} units, the value
@@ -562,6 +559,16 @@ Garbage Collection
proportion.
@end defopt
+ Control over the garbage collector via @code{gc-cons-threshold} and
+@code{gc-cons-percentage} is only approximate. Although Emacs checks
+for threshold exhaustion regularly, for efficiency reasons it does not
+do so immediately after every change to the heap or to
+@code{gc-cons-threshold} or @code{gc-cons-percentage}, so exhausting
+the threshold does not immediately trigger garbage collection. Also,
+for efficency in threshold calculations Emacs approximates the heap
+size, which counts the bytes used by currently-accessible objects in
+the heap.
+
The value returned by @code{garbage-collect} describes the amount of
memory used by Lisp data, broken down by data type. By contrast, the
function @code{memory-limit} provides information on the total amount of
diff --git a/etc/NEWS b/etc/NEWS
index 94c98a7ebe..252c6bf9b9 100644
--- a/etc/NEWS
+++ b/etc/NEWS
@@ -2429,6 +2429,13 @@ remote systems, which support this check.
+++
** 'memory-limit' now returns a better estimate of memory consumption.
++++
+** When interpreting 'gc-cons-percentage', Emacs now estimates the
+heap size more often and (we hope) more accurately. E.g., formerly
+(progn (let ((gc-cons-percentage 0.8)) BODY1) BODY2) continued to use
+the 0.8 value during BODY2 until the next garbage collection, but that
+is no longer true. Applications may need to re-tune their GC tricks.
+
+++
** New macro 'combine-change-calls' arranges to call the change hooks
('before-change-functions' and 'after-change-functions') just once
--
2.17.1
next prev parent reply other threads:[~2019-09-14 17:57 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-06 13:52 bug#37321: 27.0.50; Excessive gc in a use case (el-search) Michael Heerdegen
2019-09-07 14:23 ` Michael Heerdegen
2019-09-07 15:30 ` Michael Heerdegen
2019-09-08 1:11 ` Paul Eggert
2019-09-08 14:52 ` Michael Heerdegen
2019-09-08 15:25 ` Michael Heerdegen
2019-09-14 8:04 ` Paul Eggert
2019-09-14 8:37 ` Eli Zaretskii
2019-09-14 8:52 ` Paul Eggert
2019-09-14 9:57 ` Eli Zaretskii
2019-09-14 17:57 ` Paul Eggert [this message]
2019-09-14 18:16 ` Eli Zaretskii
2019-09-15 4:33 ` Richard Stallman
2019-09-16 23:53 ` Michael Heerdegen
2019-09-17 0:55 ` Paul Eggert
2019-09-21 0:41 ` Michael Heerdegen
2019-09-21 0:46 ` Michael Heerdegen
2019-09-21 6:19 ` Paul Eggert
2019-09-17 12:47 ` Noam Postavsky
2019-09-21 0:44 ` Michael Heerdegen
2019-09-25 9:42 ` Michael Heerdegen
2019-09-25 20:37 ` Paul Eggert
2019-09-26 11:42 ` Michael Heerdegen
2019-09-26 12:14 ` Eli Zaretskii
2019-09-26 13:03 ` Michael Heerdegen
2019-10-08 8:43 ` Michael Heerdegen
2019-10-08 9:09 ` Eli Zaretskii
2019-10-08 9:11 ` Michael Heerdegen
2019-10-08 9:19 ` Eli Zaretskii
2019-10-08 11:12 ` Michael Heerdegen
2019-10-08 12:11 ` Eli Zaretskii
2019-10-08 12:38 ` Michael Heerdegen
2019-10-08 13:03 ` Eli Zaretskii
2019-10-09 14:47 ` Michael Heerdegen
2019-10-09 15:33 ` Eli Zaretskii
2019-10-09 20:53 ` Paul Eggert
2019-10-10 10:58 ` Michael Heerdegen
2019-10-08 9:22 ` Michael Heerdegen
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=d98f0762-d3ff-90e7-07ad-2b081b0aed29@cs.ucla.edu \
--to=eggert@cs.ucla.edu \
--cc=37321@debbugs.gnu.org \
--cc=eliz@gnu.org \
--cc=michael_heerdegen@web.de \
/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).