Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Dec 2003 08:30:44 -0500
From:      "Robin P. Blanchard" <robin.blanchard@gactr.uga.edu>
To:        "Doug White" <dwhite@gumbysoft.com>
Cc:        current@freebsd.org
Subject:   RE: Device bge watchdog timeouts
Message-ID:  <EE3D3FBAFFCAED448C21C398FDAD91AC01338E89@EBE1.gc.nat>

next in thread | raw e-mail | index | archive | help
> -----Original Message-----
> From: Doug White [mailto:dwhite@gumbysoft.com]=20
> Sent: Thursday, December 11, 2003 6:45 PM
> To: Robin P. Blanchard
> Cc: current@freebsd.org; Bill Paul
> Subject: Re: Device bge watchdog timeouts
>=20
>=20
> On Thu, 11 Dec 2003, Robin P. Blanchard wrote:
>=20
> > We just repurposed a Dell 2550 from a 4.9-STABLE backup=20
> server (very=20
> > intensive network load) to a 5.2-CURRENT=20
> shoutcast/realserver (minimal=20
> > network load) and are seeing watchdog timeout errors constantly. We=20
> > experience no such trouble in it's -STABLE incarnation (using the=20
> > -STABLE bge driver). We have temporarily given up on the=20
> bge driver in=20
> > -CURRENT and are using the other 100base (fxp) onboard nic.=20
> Has anyone=20
> > got any suggestions on how to get this machine running=20
> smoothly using=20
> > its gig interface (with
> > -CURRENT) ? Thanks in advance.
>=20
> This might be a wider problem with interrupt routing on your=20
> system.  Post the output of 'boot -v' and try it with and=20
> without acpi for kicks.

Good idea. Although I can't take this box down right now as it's happily
streaming our local NPR affiliate. Nevertheless, I like your suspicion:

# vmstat -i
interrupt                          total       rate
irq1: atkbd0                           2          0
irq4: sio0                            22          0
irq6: fdc0                             4          0
irq8: rtc                       18682170        127
irq13: npx0                            1          0
irq14: ata0                           92          0
irq16: fxp0                     12889411         88
irq20: pcm0                     25138308        172
irq30: ahc0                           15          0
irq31: aac0                        17985          0
irq0: clk                      145977654       1000
Total                          202705665       1388

# dmesg |fgrep bge
Preloaded elf module "/boot/modules/if_bge.ko" at 0xc07c01f4.
bge0: <Broadcom BCM5700 Gigabit Ethernet, ASIC rev. 0x7102> mem
0xfeb00000-0xfeb0ffff irq 17 at device 8.0 on pci1
bge0: Ethernet address: 00:06:5b:05:33:99
miibus0: <MII bus> on bge0

Should not device bge / irq 17 be in the vmstat output above ?=20



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?EE3D3FBAFFCAED448C21C398FDAD91AC01338E89>