Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 21 Oct 2011 17:34:24 +0200 (CEST)
From:      Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl>
To:        Ivan Voras <ivoras@freebsd.org>
Cc:        freebsd-hackers@freebsd.org
Subject:   Re: Measuring memory footprint in C/C++ code on FreeBSD
Message-ID:  <alpine.BSF.2.00.1110211732170.13284@wojtek.tensor.gdynia.pl>
In-Reply-To: <j7rnbt$nfa$1@dough.gmane.org>
References:  <4EA0610B.90206@gmail.com> <20111021084413.GA46039@server.vk2pj.dyndns.org> <4EA1471E.9050501@gmail.com> <j7rhc9$bvb$1@dough.gmane.org> <4EA15004.50308@gmail.com> <j7rnbt$nfa$1@dough.gmane.org>

next in thread | previous in thread | raw e-mail | index | archive | help
>> footprint?
>
> Almost certainly yes. Measuring virtual memory is significantly less
> important for real-world loads. Some of this is very nicely described
> here: https://www.varnish-cache.org/trac/wiki/ArchitectNotes .

definitely.
just run top and compare RES and SIZE fields.

extreme example:

#include <unistd.h>
int blah[1000000000];
main(){sleep(1000);}


run it and see this in top
13317 wojtek        1  45    0  3817M   684K nanslp  0   0:00  0.00% 1





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1110211732170.13284>