Re: [PATCH 1/9] exofs: osd Swiss army knife

[Boaz Harrosh <bharrosh@panasas.com>]

Andrew Morton wrote:
> On Wed, 31 Dec 2008 17:33:41 +0200 Boaz Harrosh <bharrosh@panasas.com> wrote:
> 
>> Andrew Morton wrote:
>>>> +int prepare_get_attr_list_add_entry(struct osd_request *req,
>>>> +				    uint32_t page_num,
>>>> +				    uint32_t attr_num,
>>>> +				    uint32_t attr_len)
>>>> +{
>>>> +	struct osd_attr attr = {
>>>> +		.page = page_num,
>>> Kernel developers expect a field called "page" to have type `struct
>>> page *'.  osd_attr.page is thus designed to confuse.
>>>
>>>> ...
>>>>
>> Rant below (can be ignored):
>>   This single fix will cause a massive change to the open-osd
>>   initiator patchset, (18 patches), and resubmission .I made the mistake
>>   because this name originates from a file that all naming conventions
>>   are taken from the OSD standard text. However this is no excuse
>>   for using a well known Kernel construct name. I will fix it. And
>>   will be more careful in the future.
> 
> The world wouldn't end if you left the code as-is.  We've done worse things :)

To late I've changed it. I had an Internet outage yesterday so I've only just
pushed the new trees.

I'm glad. Because I found in exofs code, inside the same file, an "u32 page"
next to a "struct page *page" which is really bad. Now attr_page everywhere
is much clearer.

[ As usual:
    git-clone git://git.open-osd.org/linux-open-osd.git linux-next-exofs
  or on the web at:
    http://git.open-osd.org/gitweb.cgi?p=linux-open-osd.git;a=shortlog;h=refs/heads/linux-next-exofs
]

I will submit another round of patches once I address all the other comments.

Thanks
Boaz