On Thu, Apr 11, 2024 at 05:00:33PM +0800, Miaohe Lin wrote:
> But as code changes, the above page lock shift is gone. And I think below logic can't
> trigger now. As we hold extra page refcnt so page can't be coallesced into a new THP or Slab page.
So I was wrong.
I did not check the slap part, but I did check the code that tries to
coallesce pages.
We have the following check in hpage_collapse_scan_pmd():
/*
*
* Here the check may be racy:
* it may see total_mapcount > refcount in some cases?
* But such case is ephemeral we could always retry collapse
* later. However it may report false positive if the page
* has excessive GUP pins (i.e. 512). Anyway the same check
* will be done again later the risk seems low.
*/
if (!is_refcount_suitable(folio)) {
result = SCAN_PAGE_COUNT;
goto out_unmap;
}
So I guess that since we hold an extra refcount from memory-failure
code, this page cannot be part of a THP anymore.
If we confirm that the same goes for slab, I would replace that with
your warn check.
In this way, if we ever trigger that warn path, as least we will know
that that possibility exists.
So it is a good way to get rid of false assumptions and code burden.
--
Oscar Salvador
SUSE Labs
