Re: [PATCH] x86/msi: Don't dereference pdev before its NULL check

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 23/07/15 12:45, Jan Beulich wrote:
>>>> On 23.07.15 at 12:54, <andrew.cooper3@citrix.com> wrote:
>> --- a/xen/arch/x86/msi.c
>> +++ b/xen/arch/x86/msi.c
>> @@ -1345,7 +1345,7 @@ int pci_restore_msi_state(struct pci_dev *pdev)
>>      struct msi_desc *entry, *tmp;
>>      struct irq_desc *desc;
>>      struct msi_msg msg;
>> -    u8 slot = PCI_SLOT(pdev->devfn), func = PCI_FUNC(pdev->devfn);
>> +    u8 slot, func;
>>      unsigned int type = 0, pos = 0;
>>      u16 control = 0;
>>  
>> @@ -1357,6 +1357,9 @@ int pci_restore_msi_state(struct pci_dev *pdev)
>>      if ( !pdev )
>>          return -EINVAL;
>>  
>> +    slot = PCI_SLOT(pdev->devfn);
>> +    func = PCI_FUNC(pdev->devfn);
> I think this is the wrong way round - see the alternative patch just
> sent.
>
> Btw, how did you spot this so quickly now but not during review?
> Any tool helping with that?

Ah - I accidentally neglected to say that it was Coverity.  It is hooked
automatically into all XenServer builds, and I am being proactive
keeping my upstream branch up-to-date, given the volume of code going in
at the moment.

~Andrew