Re: [PATCH] e100: free IRQ to remove warning when rebooting

["Kok, Auke" <auke-jan.h.kok@intel.com>]

Jeff Garzik wrote:
> Ian Wienand wrote:
>> Hi,
>>
>> When rebooting today I got
>>
>> Will now restart.
>> ACPI: PCI interrupt for device 0000:00:03.0 disabled
>> GSI 20 (level, low) -> CPU 1 (0x0100) vector 53 unregistered
>> Destroying IRQ53 without calling free_irq
>> WARNING: at
>> /home/insecure/ianw/programs/git-kernel/linux-2.6/kernel/irq/chip.c:76
>> dynamic_irq_cleanup()
>>
>> Call Trace:
>>  [<a000000100014340>] show_stack+0x40/0xa0
>>                                 sp=e00000407c927b40 bsp=e00000407c920eb8
>>  [<a0000001000143d0>] dump_stack+0x30/0x60
>>                                 sp=e00000407c927d10 bsp=e00000407c920ea0
>>  [<a0000001000e58e0>] dynamic_irq_cleanup+0x160/0x1e0
>>                                 sp=e00000407c927d10 bsp=e00000407c920e70
>>  [<a0000001000106b0>] destroy_and_reserve_irq+0x30/0xc0
>>                                 sp=e00000407c927d10 bsp=e00000407c920e40
>>  [<a0000001000508f0>] iosapic_unregister_intr+0x5b0/0x5e0
>>                                 sp=e00000407c927d10 bsp=e00000407c920dd8
>>  [<a00000010000aa70>] acpi_unregister_gsi+0x30/0x60
>>                                 sp=e00000407c927d10 bsp=e00000407c920db8
>>  [<a00000010042e300>] acpi_pci_irq_disable+0x140/0x160
>>                                 sp=e00000407c927d10 bsp=e00000407c920d88
>>  [<a000000100774200>] pcibios_disable_device+0xa0/0xc0
>>                                 sp=e00000407c927d20 bsp=e00000407c920d68
>>  [<a0000001003778d0>] pci_disable_device+0x130/0x160
>>                                 sp=e00000407c927d20 bsp=e00000407c920d38
>>  [<a000000100525d20>] e100_shutdown+0x1c0/0x220
>>                                 sp=e00000407c927d30 bsp=e00000407c920d08
>>  [<a00000010037d0e0>] pci_device_shutdown+0x80/0xc0
>>                                 sp=e00000407c927d30 bsp=e00000407c920ce8
>>  [<a0000001004ecb70>] device_shutdown+0xf0/0x180
>>                                 sp=e00000407c927d30 bsp=e00000407c920cc8
>>  [<a0000001000ac4e0>] kernel_restart+0x60/0x120
>>                                 sp=e00000407c927d30 bsp=e00000407c920ca8
>>  [<a0000001000ac990>] sys_reboot+0x3b0/0x480
>>                                 sp=e00000407c927d30 bsp=e00000407c920c30
>>  [<a00000010000b4e0>] ia64_ret_from_syscall+0x0/0x20
>>                                 sp=e00000407c927e30 bsp=e00000407c920c30
>>  [<a000000000010620>] ia64_ivt+0xffffffff00010620/0x400
>>                                 sp=e00000407c928000 bsp=e00000407c920c30
>> Restarting system.
>>
>> I think the solution might be to free the IRQ before the
>> pci_device_shutdown
>>
>> Signed-off-by: Ian Wienand <ianw@gelato.unsw.edu.au>
>>
>> ---
>>
>>  e100.c |    1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/net/e100.c b/drivers/net/e100.c
>> index 3dbaec6..8ae5ac3 100644
>> --- a/drivers/net/e100.c
>> +++ b/drivers/net/e100.c
>> @@ -2782,6 +2782,7 @@ static void e100_shutdown(struct pci_dev *pdev)
>>          pci_enable_wake(pdev, PCI_D3cold, 0);
>>      }
>>  
>> +    free_irq(pdev->irq, netdev);
>>      pci_disable_device(pdev);
>>      pci_set_power_state(pdev, PCI_D3hot);
> 
> agreed, though I think free_irq() should come after pci_disable_device()
> like it does in e100_suspend().
> 
> auke?

I actually think that's unintentional (the free_irq being after
pci_disable_device()) and possibly wrong. Looking at all our other drivers and
various others I see a consistent pattern of enable_device->enable_irq ->
free_irq->disable_device which makes much more sense to me since it's symmetric
and you can't enable irq's on a disabled device anyway.

So, I'd like to ACK this patch and I'll owe you a patch to fix the order up for
the e100_suspend code to change the order - but obviously I'll give that one a
test first.

Acked-by: Auke Kok <auke-jan.h.kok@intel.com>


Thanks,

Auke