On 3/9/2020 9:21 PM, Greg KH wrote:
> On Mon, Mar 09, 2020 at 09:11:25PM +0100, Vlastimil Babka wrote:
>>> --- a/mm/memory_hotplug.c
>>> +++ b/mm/memory_hotplug.c
>>> @@ -598,7 +598,13 @@ EXPORT_SYMBOL_GPL(__online_page_free);
>>>
>>> static void generic_online_page(struct page *page, unsigned int order)
>>> {
>>> - kernel_map_pages(page, 1 << order, 1);
>>> + /*
>>> + * Freeing the page with debug_pagealloc enabled will try to unmap it,
>>> + * so we should map it first. This is better than introducing a special
>>> + * case in page freeing fast path.
>>> + */
>>> + if (debug_pagealloc_enabled_static())
>>
>> Won't build on 5.4, see in changelog "Backports for kernel before 5.5 should use
>> debug_pagealloc_enabled() instead."
>
> Builds just fine for me here, are you _sure_ 5.4.y doesn't work?
>
> 5.4.14 got debug_pagealloc_enabled_static() through the backport of
> 8e57f8acbbd1 ("mm, debug_pagealloc: don't rely on static keys too
> early") which you wrote :)
Ah ok, didn't realize that, I probably only checked the mainline :)
> Now if it still needs to be changed, let me know and I will do so.
Nothing needs to change then, sorry for the noise.
> thanks,
>
> greg k-h
>
