GNU bug report logs - #38629
25.2; garbage-collect doesn't reclaim large *compilation*

Previous Next

Package: emacs;

Reported by: Peter Ludemann <peter.ludemann <at> gmail.com>

Date: Sun, 15 Dec 2019 20:40:02 UTC

Severity: important

Merged with 26952, 27498, 31092

Found in versions 25.1, 25.2

Fixed in version 26.1

Done: Paul Eggert <eggert <at> cs.ucla.edu>

Bug is archived. No further changes may be made.

Full log


Message #8 received at 38629 <at> debbugs.gnu.org (full text, mbox):

From: Peter Ludemann <peter.ludemann <at> gmail.com>
To: 38629 <at> debbugs.gnu.org
Subject: Loading/killing the compilation output doesn't lose memory
Date: Sun, 15 Dec 2019 20:16:50 -0800
[Message part 1 (text/plain, inline)]
If I save the *compilation* buffer and restart emacs, memory usage goes up
modestly when I load the file (52MB; it is automatically fontified by emacs
when I load it):

USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
peter    19037  0.6  0.7 441776 115476 ?       Ssl  11:59   2:49 emacs
--daemon
peter    19037  0.6  1.0 495832 169776 ?       Ssl  11:59   2:50 emacs
--daemon

and memory usage returns to about the previous value when I kill the buffer
(I don't need to run 'garbage-collect).

So, the problems seems to be somewhere in the "compile" command:
(a) it uses a *lot* more memory than needed to display the result
(b) it doesn't free that memory when compilation ends or when the
compilation buffer is killed

My "compilation" is actually compile plus 8200 tests, speeding them up by
using GNU-parallel.
[Message part 2 (text/html, inline)]

This bug report was last modified 5 years and 185 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.