Re: [PATCH 1/3] mm, mem-hotplug: fix section mismatch.setup_per_zone_inactive_ratio() should be __meminit.

[Minchan Kim]

On Tue, Apr 12, 2011 at 7:28 PM, KOSAKI Motohiro
<kosaki.motohiro@xxxxxxxxxxxxxx> wrote:
> Hi
>
>> On Mon, Apr 11, 2011 at 5:00 PM, KOSAKI Motohiro
>> <kosaki.motohiro@xxxxxxxxxxxxxx> wrote:
>> > Commit bce7394a3e (page-allocator: reset wmark_min and inactive ratio of
>> > zone when hotplug happens) introduced invalid section references.
>> > Now, setup_per_zone_inactive_ratio() is marked __init and then it can't
>> > be referenced from memory hotplug code.
>> >
>> > Then, this patch marks it as __meminit and also marks caller as __ref.
>> >
>> > Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
>> Reviewed-by: Minchan Kim <minchan.kim@xxxxxxxxx>
>>
>> Just a nitpick.
>>
>> As below comment of __ref said, It would be better to add _why_ such
>> as memory_hotplug.c.
>>
>> "so optimally document why the __ref is needed and why it's OK"
>
> Hmm...
> All of memory_hotplug.c function can call __meminit function. It's
> definition of __meminit.
>
> We can put the same comment to every function in memory_hotplug.c.
> like hotadd_newpgdat().
>
> /* we are OK calling __meminit stuff here - we have CONFIG_MEMORY_HOTPLUG */
> static pg_data_t __ref *hotadd_new_pgdat(int nid, u64 start)
> {
> (snip)
> }
>
> But it has zero information. ;)

It does make sense. Never mind. :)



-- 
Kind regards,
Minchan Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/