Subject: Re: Panic when bringing up network bridge
To: None <port-cobalt@NetBSD.org>
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
List: port-cobalt
Date: 04/27/2006 02:01:21
In article <444F5916.2050009@geekrants.org>
netbsd@geekrants.org wrote:

> 3) vmstat -c 3 -i 1
> http://geekrants.org/francis/netbsd/cobalt/vmstat0.png
> 
> 4) enabled the bridge
> 
> 5) vmstat -c 3 -i 1
> http://geekrants.org/francis/netbsd/cobalt/vmstat1.png

Did you take 3) and 5) on the different powercycle?
(Numbers of interrupts shouldn't be decreased)
If so, can you check them in the single line as I wrote:
# vmstat -c 3 -i 1 ; brconfig bridge0 up ; vmstat -c 3 -i 1
(or with more counts)

> It took quite a while to be able to run the second vmstat.. a couple 
> seconds after enabling the bridge, both green/orange lights for both 
> ethernet ports stayed light up constantly for a good 30 seconds or so, 
> freezing even the serial console. The results were taken are right after 
> it would let me type in the console again.
> 
> Looks like quite an interrupt storm..

Yes, but it doesn't happen on my RaQ2, so it's Qube 2 specific
or there is something which triggers it on your network environment
or settings.

According to your netstat output, there is not RX packet on tlp1,
but does tlp1 work fine if you don't configure bridge?
How much traffic on tlp1?

Here is some stat on my RaQ2 (with 20060425 GENERIC from NetBSD-daily):
---
# uname -v
NetBSD 3.99.18 (GENERIC) #0: Tue Apr 25 10:44:05 UTC 2006  \
builds@b3.netbsd.org:/home/builds/ab/HEAD/cobalt/200604250000Z-obj/\
home/builds/ab/HEAD/src/sys/arch/cobalt/compile/GENERIC
# ifconfig bridge0
bridge0: flags=41<UP,RUNNING> mtu 1500
# brconfig -a
bridge0: flags=41<UP,RUNNING>
        Configuration:
                priority 32768 hellotime 2 fwddelay 15 maxage 20
                ipfilter disabled flags 0x0
        Interfaces:
                tlp1 flags=3<LEARNING,DISCOVER>
                        port 2 priority 128
                tlp0 flags=3<LEARNING,DISCOVER>
                        port 1 priority 128
        Address cache (max cache: 100, timeout: 1200):
                00:08:74:b3:2c:7f tlp0 224 flags=0<>
                00:a0:b0:14:aa:24 tlp0 39 flags=0<>
                00:01:03:ce:74:48 tlp0 17 flags=0<>
                00:e0:a9:10:00:ab tlp0 4294967047 flags=0<>
# vmstat -i
interrupt                                     total     rate
soft serial                                    1974        0
soft net                                       2577        0
soft clock                                     4262        0
cpu hardclock                               1498171      100
cpu level 3                                    1977        0
cpu level 1                                    4619        0
icu irq 4                                         7        0
icu irq 14                                     3998        0
cpu level 2                                     463        0
Total                                       1518048      101

# 
---
Izumi Tsutsui