On Tue Jun 25, 2024 at 7:44 AM EDT, Pankaj Raghav (Samsung) wrote:
> From: Luis Chamberlain <mcgrof@xxxxxxxxxx>
>
> split_folio() and split_folio_to_list() assume order 0, to support
> minorder for non-anonymous folios, we must expand these to check the
> folio mapping order and use that.
>
> Set new_order to be at least minimum folio order if it is set in
> split_huge_page_to_list() so that we can maintain minimum folio order
> requirement in the page cache.
>
> Update the debugfs write files used for testing to ensure the order
> is respected as well. We simply enforce the min order when a file
> mapping is used.
>
> Signed-off-by: Luis Chamberlain <mcgrof@xxxxxxxxxx>
> Signed-off-by: Pankaj Raghav <p.raghav@xxxxxxxxxxx>
> Reviewed-by: Hannes Reinecke <hare@xxxxxxx>
> ---
> There was a discussion about whether we need to consider truncation of
> folio to be considered a split failure or not [1]. The new code has
> retained the existing behaviour of returning a failure if the folio was
> truncated. I think we need to have a separate discussion whethere or not
> to consider it as a failure.
<snip>
>
> +int split_folio_to_list(struct folio *folio, struct list_head *list)
> +{
> + unsigned int min_order = 0;
> +
> + if (!folio_test_anon(folio)) {
> + if (!folio->mapping) {
> + count_vm_event(THP_SPLIT_PAGE_FAILED);
Regardless this folio split is from a truncation or not, you should not
count every folio split as a THP_SPLIT_PAGE_FAILED. Since not every
folio is a THP. You need to do:
if (folio_test_pmd_mappable(folio))
count_vm_event(THP_SPLIT_PAGE_FAILED);
See commit 835c3a25aa37 ("mm: huge_memory: add the missing
folio_test_pmd_mappable() for THP split statistics")
> + return -EBUSY;
> + }
> + min_order = mapping_min_folio_order(folio->mapping);
> + }
> +
> + return split_huge_page_to_list_to_order(&folio->page, list, min_order);
> +}
> +
--
Best Regards,
Yan, Zi
Attachment:
signature.asc
Description: PGP signature
