Re: forcedeth oops

[Jeff Garzik <jeff@garzik.org>]

Andrew Brooks wrote:
> Hello
> 
> I'm getting an oops in forcedeth whenever I shutdown, details below.
> 
> I've tried kernel 2.6.16.59 and the latest forcedeth.c from nvidia.com
> which is package-1.23 version-0.62 date-2007/04/27.
> 
> How can I download the latest forcedeth.c (including 2008-01-13 patches) ?
> It's not in the latest snapshot linux-2.6.24-rc8.
> 
> Also, why is the version on nvidia.com not just older than the one in
> the kernel, but it appears to have forked back in May 2006.  Has there
> been independent development on each version?  They should be the same!

We don't run nvidia.com here :)


> Here's the diff:
> <  *    0.56: 22 Mar 2006: Additional ethtool and moduleparam support.
> <  *    0.57: 14 May 2006: Moved mac address writes to nv_probe and nv_remove.
> <  *    0.58: 20 May 2006: Optimized rx and tx data paths.
> <  *    0.59: 31 May 2006: Added support for sideband management unit.
> <  *    0.60: 31 May 2006: Added support for recoverable error.
> <  *    0.61: 18 Jul 2006: Added support for suspend/resume.
> <  *    0.62: 16 Jan 2007: Fixed statistics, mgmt communication, and low phy speed on S5.
> ---
>>  *    0.56: 22 Mar 2006: Additional ethtool config and moduleparam support.
>>  *    0.57: 14 May 2006: Mac address set in probe/remove and order corrections.
>>  *    0.58: 30 Oct 2006: Added support for sideband management unit.
>>  *    0.59: 30 Oct 2006: Added support for recoverable error.
>>  *    0.60: 20 Jan 2007: Code optimizations for rings, rx & tx data paths, and stats.
> 
> 
> Here's the details of the oops:
> md: md0 switched to read-only mode.
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> printing eip:
> f8ccdd55
> *pde = 36c6a001
> Oops: 0000 [#1]
> SMP
> Modules linked in: nvidia ... forcedeth ... sata_nv
> CPU: 1
> EIP:
> EFLAGS: 00010286 (2.6.16.59 #1)
> EIP is at nv_suspend+0x85/0x350 [forcedeth]
> eax:
> esi:
> ds:
> Process reboot
> Stack:
> Call Trace:
> show_stack_log
> show_registers
> die
> do_page_fault
> error_code
> nv_reboot_handler
> notifier_call_chain
> kernel_restart_prepare
> kernel_restart
> sys_reboot
> sysenter_past_esp
> Code: 8b 8c 3a 98 01 00 00 01 c8 8b ...
> INIT: no more processes left in this runlevel

Please reproduce this problem on a modern kernel (2.6.24-rc) without any 
closed source modules or drivers loaded.  Thanks.

	Jeff