Re: [Xen-devel] [PATCH 3 of 7] IOMMU: Sanitise pointer work

[Keir Fraser <keir@xxxxxxx>]

On 14/06/2011 10:53, "Andrew Cooper" <andrew.cooper3@xxxxxxxxxx> wrote:

> On 13/06/11 19:13, Keir Fraser wrote:
>> On 13/06/2011 18:02, "Andrew Cooper" <andrew.cooper3@xxxxxxxxxx> wrote:
>> 
>>> Check for null pointers before calling function pointers.
>> But they're never NULL? This one's a bit pointless.
>> 
>>  -- Keir
>> 
> This patch was a direct result of comments regarding my previous attempt
> to make a crash_shutdown iommu_op.  Are we happy to assume that noone in
> the future is going to partially implement an iommu_ops structure?

We'll cross that bridge if we come to it. I can't see anyone implementing
iommu_ops that don't need a call back on suspend and resume.

 -- Keir

> ~Andrew
>>> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>>> 
>>> diff -r 1c3d2e4d06fe -r 68efd418b6f1 xen/drivers/passthrough/iommu.c
>>> --- a/xen/drivers/passthrough/iommu.c Mon Jun 13 17:45:43 2011 +0100
>>> +++ b/xen/drivers/passthrough/iommu.c Mon Jun 13 17:45:43 2011 +0100
>>> @@ -407,17 +407,17 @@ unsigned int iommu_read_apic_from_ire(un
>>>      return ops->read_apic_from_ire(apic, reg);
>>>  }
>>>  
>>> -void iommu_resume()
>>> +void iommu_resume(void)
>>>  {
>>>      const struct iommu_ops *ops = iommu_get_ops();
>>> -    if ( iommu_enabled )
>>> +    if ( iommu_enabled && ops && ops->resume )
>>>          ops->resume();
>>>  }
>>>  
>>> -void iommu_suspend()
>>> +void iommu_suspend(void)
>>>  {
>>>      const struct iommu_ops *ops = iommu_get_ops();
>>> -    if ( iommu_enabled )
>>> +    if ( iommu_enabled && ops && ops->suspend )
>>>          ops->suspend();
>>>  }
>>>  
>>> 
>>> _______________________________________________
>>> Xen-devel mailing list
>>> Xen-devel@xxxxxxxxxxxxxxxxxxx
>>> http://lists.xensource.com/xen-devel
>> 



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel