lkml.org 
[lkml]   [2004]   [Oct]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Subject2.6.9-rc4-mm1-VP-T7 - horrid death in vortex_init at boot
From
Date
2.6.9-rc4-mm1-VP-T7 plus Ingo's patch to profile.c and 3c59x.c to use
raw_rwlock_t and raw_spinlock_t rather than the non-raw variant. It croaked
when it found the onboard ethernet controller on a Dell Latitude C840 laptop:

lspci says it's a:
02:00.0 Ethernet controller: 3Com Corporation 3c905C-TX/TX-M [Tornado] (rev 78)

Got the following (admittedly truncated - was handwritten and CTS is a pain,
literally) at very early boot:

3c59x: Donald Becker and others. ...
0000:02:00.0: 3Com PCI 3c905C Tornado at 0xec80. Vers LK1.1.19
kernel BUG at net/core/net-sysfs.c:384
process swapper
... registers skipped
call trace:
show_stack
show_registers
die
do_invalid_op
error_code
class_hotplug
kobject_hotplug
kobject_add
class_device_add
netdev_register_sysfs
netdev_run_todo
register_netdev
vortex_probe1
vortex_init_one
pci_device_probe_static
__pci_device_probe
pci_device_probe
bus_match
driver_attach
bus_add_driver
driver_register
pci_register_driver
vortex_init
do_initcalls

The offending code:

static void netdev_release(struct class_device *cd)
{
struct net_device *dev
= container_of(cd, struct net_device, class_dev);

BUG_ON(dev->reg_state != NETREG_RELEASED);

kfree((char *)dev - dev->padded);
}

I'm guessing something broken in the bk-driver-core patch in -rc4-mm1, as
that completely overhauled this stuff.

This ring any bells? If need be, I'll scare up a serial cable and get a
more complete trace - somehow, I don't think netconsole will help here.. ;)

(As an aside, kobject_hotplug calls call_usermodehelper() - which seems
like a Bad Idea if we haven't gotten userspace up and running yet?
[unhandled content-type:application/pgp-signature]
\
 
 \ /
  Last update: 2005-03-22 14:06    [W:0.061 / U:1.008 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site