Re: [PATCH 00/13] Refactor pci_is_brdige() to simplify code

[Yijing Wang <wangyijing@huawei.com>]

On 2014/4/25 17:42, David Laight wrote:
> From: Yijing Wang
>> This patchset rename the current pci_is_bridge() to pci_has_subordinate(),
>> and introduce a new pci_is_bridge() which determine pci bridge by check
>> dev->hdr_type. The new one is more accurate. PCIe Spec define the pci
>> device is a bridge by the dev->hdr_type = 0x01 || 0x02.
> 
> That is a dangerous rename and is likely to cause difficult to

Hi David,
   I renamed pci_is_bridge() to pci_has_subordinate() because

static inline bool pci_is_bridge(struct pci_dev *pci_dev)
{
	return !!(pci_dev->subordinate);
}

which always check dev->subordinate.

> identify bugs in any code you've missed.

What are you referring to ?


Thanks!
Yijing.






> 
> 	David
> 
> \x04�{.n�+�������+%��lzwm��b�맲��r��zX��\x19\x1e�w��{ay�\x1dʇڙ�,j\a��f���h���z�\x1e�w���\f���j:+v���w�j�m����\a����zZ+�����ݢj"��!�iO��z��v�^\x14\x04\x1a�^[m����\vnƊ��Y&�
> 


-- 
Thanks!
Yijing