Re: [PATCH 09/10] memory-hotplug: enable memory hotplug to handle hugepage

[Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>]

On Tue, Mar 26, 2013 at 05:31:51PM +0530, Aneesh Kumar K.V wrote:
> Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> writes:
> 
> > +/* Returns true for head pages of in-use hugepages, otherwise returns false. */
> > +bool is_hugepage_movable(struct page *hpage)
> > +{
> > +	struct page *page;
> > +	struct hstate *h;
> > +	bool ret = false;
> > +
> > +	VM_BUG_ON(!PageHuge(hpage));
> > +	/*
> > +	 * This function can be called for a tail page because memory hotplug
> > +	 * scans movability of pages by pfn range of a memory block.
> > +	 * Larger hugepages (1GB for x86_64) are larger than memory block, so
> > +	 * the scan can start at the tail page of larger hugepages.
> > +	 * 1GB hugepage is not movable now, so we return with false for now.
> > +	 */
> > +	if (PageTail(hpage))
> > +		return false;
> > +	h = page_hstate(hpage);
> > +	spin_lock(&hugetlb_lock);
> > +	list_for_each_entry(page, &h->hugepage_activelist, lru)
> > +		if (page == hpage) {
> > +			ret = true;
> > +			break;
> > +		}
> > +	spin_unlock(&hugetlb_lock);
> > +	return ret;
> > +}
> > +
> 
> May be is_hugepage_active() ?

Yes, it would be nice.

Thanks,
Naoya