GNU bug report logs - #45705
[feature/native-comp] Excessive memory consumption on windows 10

Previous Next

Full log

View this message in rfc822 format

From: Andrea Corallo <akrl <at> sdf.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: edouard.debry <at> gmail.com, 45705 <at> debbugs.gnu.org
Subject: bug#45705: [feature/native-comp] Excessive memory consumption on windows 10
Date: Fri, 08 Jan 2021 15:50:28 +0000

Eli Zaretskii <eliz <at> gnu.org> writes:

>> Date: Wed, 06 Jan 2021 20:55:35 +0000
>> Cc: 45705 <at> debbugs.gnu.org
>> From: Andrea Corallo via "Bug reports for GNU Emacs,
>>  the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
>> 
>> > On windows 10, I noticed that emacs could use a lot of memory, up to
>> > 6Go and
>> > sometimes more. The memory consumption goes up and down from time to
>> > time while
>> > I am not running any specific program in emacs, apart `lsp-java`
>> >
>> > As my personal computer has 16Go of RAM, I can afford
>> > it. Unfortunately, my work
>> > computer has much less and the whole compute completely freezes at one
>> > point
>> > when using emacs.
>> >
>> > I did not notice that behavior on linux. I do not know if the master
>> > branch has
>> > the same problem. What could be the problem ?
>> 
>> Hi Édouard,
>> 
>> AFAIK was never proved recently Emacs garbage collector is failing to
>> recall memory, so I guess this is just some Lisp program that is
>> allocating a lot of memory keeping then those objects referenced.
>
> IME, 6 GiB is too much for any Lisp program to explain away.
>
> Also, the memory allocator we use on Windows is known to return memory
> to the OS more than glibc on GNU/Linux.  I have never seen my Emacs
> session get anywhere near 1 GiB, let alone more, and my sessions run
> for many weeks without restarting.
>
> Can you tell what kind of memory footprints you see on your system
> with the native-comp branch, after running the session for several
> days or more?

Hi Eli,

consumptions of few GiB is something I've seen more then once for long
standing sessions.  You might be right in this being a memory leak,
indeed I've no prove of that (I think we have none for the other
direction either).

Assuming it's a bug I don't see a priori why this should be a different
one respect the one reported for master.

Regards

  Andrea

Previous Next

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