Bug#680550: linux-image-2.6.32-5-amd64: kmalloc-32 memory leak for kernel 2.6.32-5-amd64
On 2012-07-21 23:24+0100 Ben Hutchings wrote:
Thanks to the detailed report. Memory leaks are sadly common, and many
have been fixed since Linux 2.6.32. I don't think it's practical to
attempt to match up those many fixes against your list of applications.
Thanks for responding.
Could you please attempt to verify the problem (by periodically
checking kmalloc-32 in /proc/slabinfo on machines you have access to
for a few days) to get a feel for how common this issue is? For
example, it might occur for _all_ desktop users who run
kernel-2.6.32-5-amd64 and not have much to do with my own particular
application mix. After all, this is not an ordinary memory leak.
Instead, it is a leak in a special kind of memory that the kernel
controls directly, and presumably this kernel bug is exercised by just
some subset of the normal kernel calls. If it is a common kernel call
that indirectly generates the kmalloc-32 leak, than most users will
see this issue. On the other hand, if you check on a couple of
different machines running kernel-2.6.32-5-amd64 for the course of
several days and see no kmalloc-32 memory leak, then that is useful
information to help pin down this issue as well.
Can you first test whether the current kernel package in
testing/unstable (linux-image-3.2.0-3-amd64, version 3.2.21-3) fixes the
leak? (You will also need to upgrade linux-base and initramfs-tools,
but nothing else, so this should not disrupt your stable installation.)
I planned to dist-upgrade to Debian testing in any case roughly two
weeks from now when I hope to finish up a project I am currently
working on. I don't want to disrupt that project, but once that is
done just before that dist-upgrade, I will try your idea of just
updating the kernel and its dependencies. After all, although I don't
personally intend to use Debian stable too much longer, others will
continue to use Debian stable for quite some time, so I would like to
help out such users by doing some tests if it turns out (see my
question above for you) the kmalloc-32 memory leak is a common issue
on 64-bit hardware.
If it is fixed there, perhaps you could then test some of the
intermediate kernel versions available from
<http://snapshot.debian.org/package/linux-2.6/> (start with 2.6.37 and
then go forward or backward) to narrow down when this was fixed
upstream. If we know which upstream version fixed the bug, it may be
possible to identify the individual fix to be included in a stable
I would be willing to do those tests as well before the dist-upgrade
if it turns out that linux-image-3.2.0-3-amd64 cures the problem.
For what it is worth my Debian stable 32-bit computer running as an
X-terminal/thin client shows nothing big in /proc/slabinfo
after 7 days uptime.
For that computer, "uname -a" gives
Linux meadowlark 2.6.32-5-686 #1 SMP Sun May 6 04:01:19 UTC 2012 i686
Of course, that computer essentially runs nothing but an X server so
that is a very limited application mix and a heavier application mix
might trigger the issue. On the other hand, the problem might be
limited just to 64-bit hardware. So monitoring /proc/slabinfo on both
32-bit and 64-bit machines running 2.6.32 is required to help pin this down.
Alan W. Irwin
Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).
Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).