On Tue, 22 Mar 2005, Peter Losher wrote:
> We are rebuilding a system which has a Netgear GA621 (using the nge
> driver) installed, and which had run 5.2 & 5.3 pre-releases on amd64
> (it's a Dual Opteron box - Tyan Thunder motherboard)
>
> We are now intending to make it a backup server and it is running
> 5.4-PRE/i386 (cvsupped just under 20 hours ago), and when we try to
> configure nge0, after coming up for around 10 seconds it panics.
The below is a NULL pointer dereference in the kernel. Off-hand, it looks
very likely to be a driver bug. The question is -- where? The best way
to answer this is to compile with DDB/KDB, and do a stack trace from DDB,
or get a dump and do similar things with gdb.
Thanks
Robert N M Watson
>
> -=-
> nge0: <National Semiconductor Gigabit Ethernet> port 0x9800-0x98ff
mem
> 0xfb8ff000-0xfb8fffff irq 29 at device 1.0 on pci1
> nge0: Using TBI
> nge0: 1000baseSX, 1000baseSX-FDX, auto
> -=-
>
> -=-
> % ifconfig fxp0 down
> % ifconfig nge0 204.152.187.8 netmask 255.255.255.0
> %
> %
> %
> %
>
> Fatal trap 12: page fault while in kernel mode
> fault virtual address = 0x8
> fault code = supervisor read, page not present
> instruction pointer = 0x8:0xc055bdda
> stack pointer = 0x10:0xe97c9c9c
> frame pointer = 0x10:0xe97c9cac
> code segment = base 0x0, limit 0xfffff, type 0x1b
> = DPL 0, pres 1, def32 1, gran 1
> processor eflags = interrupt enabled, resume, IOPL = 0
> current process = 39 (irq29: nge0)
> trap number = 12
> panic: page fault
> Uptime: 21m3s
> Cannot dump. No dump device defined.
> Automatic reboot in 15 seconds - press a key on the console to abort
> Rebooting...
> -=-
>
> Any ideas? I have already tried turning of HTT (can't), ACPI (same
> effect). We can't run amd64 since we know the 3ware driver isn't
64-bit
> clean and locks up the system under heavy I/O load. We could just live
> w/ 100baseTX, but we would rather be running GigE...
>
> Best Wishes - Peter
> --
> Peter_Losher@isc.org | ISC | OpenPGP 0xE8048D08 | "The bits must
flow"
>