This is the mail archive of the
libc-help@sourceware.org
mailing list for the glibc project.
Huge chunks of anonymous memory
- From: "Vinu Rajashekhar" <vinutheraj at gmail dot com>
- To: libc-help at sourceware dot org
- Date: Wed, 9 Jul 2008 23:19:32 +0200
- Subject: Huge chunks of anonymous memory
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to :subject:mime-version:content-type:content-transfer-encoding :content-disposition; bh=xD+tJaUYMNyr0rSduwNR/HppF7K0vx6J5I6fFwvB1+I=; b=AtNKOgvvitw430ZjbBxaLmWukTJJ0TqVhgOYIs4uBv6h52JN/GDHyBJnGCVGO3bIWh OfmG6zYkDgYRuSJFH7ZZuVwz5dvIPCfxDV58Ds51/lZrjJz32B74wPI2XurUvIBa31Wa io+3RB3CwJnDBg11V1hJRd3J9zV89s9AfiocE=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:mime-version:content-type :content-transfer-encoding:content-disposition; b=p+JXLY7y+I2yv0sqtNnXk0fZbgQCQpfIKwbXP55klbJQs1p5ivHVJ1QmP1mUd8gxiJ vfHYlBltUyt83i4LJ6Nc//L2+y/AMWbmfn6tFkGTmVFO+yUXltSTHMSJ58wdalnIt0Y9 O9PVUxFKrKgo4yRdSr9xX7Y2AJ6vec29jlcJw=
Hi,
Recently did some instrumentation in the glibc-2.7
nptl code, basically related to take statistics of locking and
unlocking. So the code was in c++ and I modified the make
files and added my source files so that they can be compiled
with glibc nptl.
Since I use the code to get statistics, I have some vectors and stuff
in the c++ code. The problem is the new library uses a lot of memory.
I have done the same thing in the libthr library in FreeBSD, and it didnt
have this problem.
Basically, the problem is that a lot of huge chunks of anonymous memory
is being allocated, and it's not done through "mmap", because I did an "strace"
and I found out. I want to know, if there are some special compiler
options I need
to give, to prevent this anomalous behaviour, or is some problem with paging or
something.
I am on Ubuntu 8.04 x86_64
glibc-2.7
gcc-4.2.3
Can somebody please help me, or forward me to some link ... please.
Cheers,
Vinu