Re: Memory leak in network buffers 2.1.125ac3

Kurt Garloff (garloff@kg1.ping.de)
Wed, 11 Nov 1998 08:05:24 +0100

On Wed, Nov 11, 1998 at 02:03:19AM +0100, Kurt Garloff wrote:
> Nov 11 01:46:30 kg1 kernel: Networking buffers in use : 144
> Nov 11 01:46:30 kg1 kernel: Total network buffer allocations : 16120011
> Nov 11 01:46:30 kg1 kernel: Total failed network buffer allocs : 2
> Nov 11 01:46:30 kg1 kernel: IP fragment buffer size : 0
> root@kg1:~ # uptime
> 1:52am up 4 days, 8:28, 2 users, load average: 3.66, 3.85, 3.86
> ...
> Kernel is 2.1.125ac3, SMP (Dual PII-350), w/ HZ = 400, scheduler bigpatch 4
> from Rik and compiled with egcs-1.1a (-O2, -march=pentiumpro -fschedule-insns2).

Recompiled w/ -O2 -mpentiumpro -fschedule-insns2: No change.

When I go single user (runlevel S), the leakage stops. (I can see it with
Shift-Scroll and issueing klogd -o -f -.) As soon as I start the syslogd
(1.3.30), the leakage goes on! So it must be somewhere in the socket or
the surrounding layers.

Still searching ...

-- 
Kurt Garloff <K.Garloff@ping.de>  (Dortmund, FRG)
PGP key on http://student.physik.uni-dortmund.de/homepages/garloff
Unix IS user friendly - it's just selective about who its friends are!


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/