2000-11-28 10:35:47

by Lorenzo Allegrucci

[permalink] [raw]
Subject: lmbench on linux-2.4.0-test[4-11]


About 2 times on 3 lmbench-2alpha12 stops at "Local networking".
Below is a 'ps l':

F UID PID PPID PRI NI VSZ RSS WCHAN STAT TTY TIME COMMAND
100 0 207 1 9 0 1980 1204 wait4 S tty1 0:00 -bash
100 0 208 1 14 0 1968 1188 wait4 S tty2 0:00 -bash
000 0 209 1 9 0 1004 456 read_c S tty3 0:00
/sbin/getty 38400 tty3
000 0 210 1 9 0 1004 456 read_c S tty4 0:00
/sbin/getty 38400 tty4
000 0 211 1 9 0 1004 456 read_c S tty5 0:00
/sbin/getty 38400 tty5
000 0 212 1 9 0 1004 456 read_c S tty6 0:00
/sbin/getty 38400 tty6
000 0 4096 207 9 0 1168 588 wait4 S tty1 0:00 make
results
000 0 4255 4096 9 0 1748 852 wait4 S tty1 0:00
/bin/sh ../scripts/results
000 0 4289 4255 9 0 1768 872 wait4 S tty1 0:00
/bin/sh ../../scripts/lmbench CONFIG.lenstra
040 0 4740 1 10 0 1000 292 tcp_da S tty1 0:00
lat_connect -s
040 0 4742 1 9 0 1068 380 wait_f S tty1 0:00 bw_tcp -s
000 0 4753 4289 10 0 1036 480 inet_w S tty1 0:00
lat_connect localhost
100 0 4761 208 17 0 2872 1256 - R tty2 0:00 ps l

lat_connect and bw_tcp are not running.
It looks like a deadlock between connect() and accept().

2.4.0-test[1-2] are not affected, don't know test3.
Athlon UP machine. All kernels compiled with the same
configuration and compiler, gcc 2.95.2

Does anyone confirm this problem?

--
Lorenzo


2000-11-28 10:42:57

by David Miller

[permalink] [raw]
Subject: Re: lmbench on linux-2.4.0-test[4-11]

Date: Tue, 28 Nov 2000 11:08:17 +0100
From: Lorenzo Allegrucci <[email protected]>

Does anyone confirm this problem?

Increase the space between the two numbers configured in
/proc/sys/net/ipv4/ip_local_port_range, try a configuration
such as:

echo "32768 61000" >/proc/sys/net/ipv4/ip_local_port_range

I bet the problem goes away then.

Later,
David S. Miller
[email protected]

2000-11-28 17:51:19

by Lorenzo Allegrucci

[permalink] [raw]
Subject: Re: lmbench on linux-2.4.0-test[4-11]

At 01.57 28/11/00 -0800, you wrote:
> Date: Tue, 28 Nov 2000 11:08:17 +0100
> From: Lorenzo Allegrucci <[email protected]>
>
> Does anyone confirm this problem?
>
>Increase the space between the two numbers configured in
>/proc/sys/net/ipv4/ip_local_port_range, try a configuration
>such as:
>
>echo "32768 61000" >/proc/sys/net/ipv4/ip_local_port_range
>
>I bet the problem goes away then.

ip_local_port_range already comes with such configuration.
Anyhow, now even test1 shows the problem :-(
I can run lmbench on 2.2.x kernels only.
BTW, all local services (smtp, telnet etc) work well.
Any ideas?

--
Lorenzo

2000-11-28 18:01:13

by David Miller

[permalink] [raw]
Subject: Re: lmbench on linux-2.4.0-test[4-11]

Date: Tue, 28 Nov 2000 18:23:52 +0100
From: Lorenzo Allegrucci <[email protected]>

BTW, all local services (smtp, telnet etc) work well. Any ideas?

No ideas, and since the indicated programs from lmbench work
perfectly fine here on my machines, you're going to have to do some
detective work so we can figure out whats wrong.

You need to do some strace'ing of the affected programs and try
to figure out exactly how they get stuck and for what reason.

Later,
David S. Miller
[email protected]