Re: avoiding pci_disable_device()...

[Jeff Garzik <jgarzik@pobox.com>]

Takashi Iwai wrote:
> At Mon, 14 Feb 2005 11:34:13 -0800,
> Greg KH wrote:
> 
>>On Mon, Feb 14, 2005 at 08:24:29PM +0100, Takashi Iwai wrote:
>>
>>>At Mon, 14 Feb 2005 11:06:19 -0800,
>>>Greg KH wrote:
>>>
>>>>>As a result, I have committed the attached patch to libata-2.6.  In many 
>>>>>cases, it is a "semantic fix", addressing the case
>>>>>
>>>>>	* pci_request_regions() indicates hardware is in use
>>>>>	* we rudely disable the in-use hardware
>>>>>
>>>>>that would not occur in practice.
>>>>>
>>>>>But better safe than sorry.  Code cuts cut-n-pasted all over the place.
>>>>>
>>>>>I'm hoping one or two things will happen now:
>>>>>* janitors fix up the other PCI drivers along these lines
>>>>>* improve the PCI API so that pci_request_regions() is axiomatic
>>>>
>>>>Do you have any suggestions for how to do this?
>>>
>>>How about to add an exclusiveness check in pci_enable_device()?
>>>Most drivers suppose that the given pci resources are exclusively
>>>available.
>>
>>You mean only allow pci_enable_device() to work for the first caller of
>>it?  I don't see how that would help this issue out.
> 
> 
> Well, for example, add a new pointer to indicate the driver accessing
> exclusively.  And pci_enable_device() (maybe a new variant would be
> better for compatibility) checks whether this is free.
> 
> The second caller wouldn't reach even to pci_request_regions() because
> of this check.  So, no side-effect of pci_disable_device() in the
> error path.

This doesn't work with a driver that is properly using 
request_resource(), but not using the PCI API.

	Jeff