Re: 2.6.17-mm6

[Andrew Morton <akpm@osdl.org>]

On Mon, 03 Jul 2006 23:00:34 +1200
Reuben Farrelly <reuben-lkml@reub.net> wrote:

> 
> 
> On 3/07/2006 10:03 p.m., Andrew Morton wrote:
> > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17/2.6.17-mm6/
> > 
> > 
> > - A major update to the e1000 driver.
> > 
> > - 1394 updates
> 
> 
> audit: initializing netlink socket (disabled)
> audit(1151924044.008:1): initialized
> SELinux:  Registering netfilter hooks
> Initializing Cryptographic API
> io scheduler noop registered
> io scheduler anticipatory registered
> io scheduler deadline registered (default)
> PCI: Setting latency timer of device 0000:00:1c.0 to 64
> assign_interrupt_mode Found MSI capability
> Allocate Port Service[0000:00:1c.0:pcie0]
> Allocate Port Service[0000:00:1c.0:pcie0]

Looks like the enumeration of the PCIE devices failed to increment something.

> kobject_add failed for 0000:00:1c.0:pcie0 with -EEXIST, don't try to register 
> things with the same name in the same directory.
> 
> Call Trace:
>   [<ffffffff80358298>] kobject_put+0x19/0x21
>   [<ffffffff80358741>] kobject_add+0x181/0x1ac
>   [<ffffffff803aa378>] device_add+0x88/0x4b9
>   [<ffffffff803aa7c2>] device_register+0x19/0x27
>   [<ffffffff80363e90>] pcie_port_device_register+0x3a0/0x3de
>   [<ffffffff8042fa10>] pcibios_set_master+0x7f/0x86
>   [<ffffffff80364004>] pcie_portdrv_probe+0x64/0x80
>   [<ffffffff803612ac>] pci_device_probe+0x4d/0x78
>   [<ffffffff803ac08f>] driver_probe_device+0x5c/0xb4
>   [<ffffffff803ac1c9>] __driver_attach+0x67/0xb9
>   [<ffffffff803ac162>] __driver_attach+0x0/0xb9
>   [<ffffffff803aba4f>] bus_for_each_dev+0x4f/0x79
>   [<ffffffff803abfbc>] driver_attach+0x1c/0x1e
>   [<ffffffff803ab61a>] bus_add_driver+0x7a/0x143
>   [<ffffffff803ac463>] driver_register+0x9f/0xa6
>   [<ffffffff80361497>] __pci_register_driver+0x59/0x7e
>   [<ffffffff806b7a57>] pcie_portdrv_init+0x1c/0x30
>   [<ffffffff80267f3e>] init+0x14e/0x2d0
>   [<ffffffff80260a12>] child_rip+0x8/0x12
>   [<ffffffff80267df0>] init+0x0/0x2d0
>   [<ffffffff80260a0a>] child_rip+0x0/0x12
> 
> The trace shows up 5 times in quick succession, all traces looking the same.......
> 
> Box otherwise boots up OK.

I'd be supecting the changes in drivers/base/core.c.  Does 2.6.17-git19 do
the same thing?