[-next] i2c: Fix missing pci_disable_device() on error in ali1535_setup()

Fix the missing pci_disable_device() before return
from ali1535_setup() in the error handling case.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zou Wei <zou_wei@huawei.com>
---
 drivers/i2c/busses/i2c-ali1535.c | 1 +
 1 file changed, 1 insertion(+)

Hi Wei,

On Mon, 07 Jun 2021 11:21:08 +0800, Zou Wei wrote:
> Fix the missing pci_disable_device() before return
> from ali1535_setup() in the error handling case.
> 
> Reported-by: Hulk Robot <hulkci@huawei.com>
> Signed-off-by: Zou Wei <zou_wei@huawei.com>
> ---
>  drivers/i2c/busses/i2c-ali1535.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c
> index fb93152..bdbaf79 100644
> --- a/drivers/i2c/busses/i2c-ali1535.c
> +++ b/drivers/i2c/busses/i2c-ali1535.c
> @@ -206,6 +206,7 @@ static int ali1535_setup(struct pci_dev *dev)
>  exit_free:
>  	release_region(ali1535_smba, ALI1535_SMB_IOSIZE);
>  exit:
> +	pci_disable_device(dev);
>  	return retval;
>  }
>  

We don't actually want to disable the PCI device. Maybe it was already
enabled before the driver was loaded, and maybe the BIOS needs the
device when the system is being shut down. You'll notice that we do not
call pci_disable_device(dev) in ali1535_remove(), so there's no reason
to do it in the error path.

As a matter of fact the i2c-i801 driver, which is used on the same kind
of hardware, has the following note in its remove function:

	/*
	 * do not call pci_disable_device(dev) since it can cause hard hangs on
	 * some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010)
	 */

So this is a nack from me, sorry.

Hi Jean,

Thanks for your review and detailed explanation.

On 2021/6/8 17:46, Jean Delvare wrote:
> Hi Wei,
> 
> On Mon, 07 Jun 2021 11:21:08 +0800, Zou Wei wrote:
>> Fix the missing pci_disable_device() before return
>> from ali1535_setup() in the error handling case.
>>
>> Reported-by: Hulk Robot <hulkci@huawei.com>
>> Signed-off-by: Zou Wei <zou_wei@huawei.com>
>> ---
>>   drivers/i2c/busses/i2c-ali1535.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c
>> index fb93152..bdbaf79 100644
>> --- a/drivers/i2c/busses/i2c-ali1535.c
>> +++ b/drivers/i2c/busses/i2c-ali1535.c
>> @@ -206,6 +206,7 @@ static int ali1535_setup(struct pci_dev *dev)
>>   exit_free:
>>   	release_region(ali1535_smba, ALI1535_SMB_IOSIZE);
>>   exit:
>> +	pci_disable_device(dev);
>>   	return retval;
>>   }
>>   
> 
> We don't actually want to disable the PCI device. Maybe it was already
> enabled before the driver was loaded, and maybe the BIOS needs the
> device when the system is being shut down. You'll notice that we do not
> call pci_disable_device(dev) in ali1535_remove(), so there's no reason
> to do it in the error path.
> 
> As a matter of fact the i2c-i801 driver, which is used on the same kind
> of hardware, has the following note in its remove function:
> 
> 	/*
> 	 * do not call pci_disable_device(dev) since it can cause hard hangs on
> 	 * some systems during power-off (eg. Fujitsu-Siemens Lifebook E8010)
> 	 */
> 
> So this is a nack from me, sorry.
>