On 1/9/26 09:54, Francois Dugast wrote:
From: Matthew Brost <[email protected]>
Add folio_split_unref helper which splits an unreferenced folio
split_unref reads like "split and unref".
You probably want to call this something like "folio_split_frozen" ?
The very definition of "frozen" is "refcount = 0 ", so you can simplify
the documentation.
Are the folios you want to pass in there completely unused (-> free) or
might they still be in use (e.g., migration entries point at them during
folio split)
So I am not sure yet if this should be "folio_split_frozen()" or
"folio_split_freed()" or sth like that.
I'm not CCed on the other patches in the series or the cover letter, so
I don't see the context.
You should describe in this patch here in which context the function is
supposed to be used in later commits.
(refcount == 0) into individual pages. Intended to be called on special
pages (e.g., device-private, DAX, etc.) when returning the folio to the
free page pool.
Cc: Balbir Singh <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Lorenzo Stoakes <[email protected]>
Cc: Zi Yan <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: "Liam R. Howlett" <[email protected]>
Cc: Nico Pache <[email protected]>
Cc: Ryan Roberts <[email protected]>
Cc: Dev Jain <[email protected]>
Cc: Barry Song <[email protected]>
Cc: Lance Yang <[email protected]>
Cc: [email protected]
Cc: [email protected]
Suggested-by: Alistair Popple <[email protected]>
Signed-off-by: Matthew Brost <[email protected]>
Signed-off-by: Francois Dugast <[email protected]>
---
include/linux/huge_mm.h | 1 +
mm/huge_memory.c | 39 +++++++++++++++++++++++++++++++++++++++
2 files changed, 40 insertions(+)
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index a4d9f964dfde..18cb9728d8f1 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -369,6 +369,7 @@ enum split_type {
SPLIT_TYPE_NON_UNIFORM,
};
+void folio_split_unref(struct folio *folio);
int __split_huge_page_to_list_to_order(struct page *page, struct list_head
*list,
unsigned int new_order);
int folio_split_unmapped(struct folio *folio, unsigned int new_order);
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 40cf59301c21..0eb9e6ad8639 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -3580,6 +3580,45 @@ static void __split_folio_to_order(struct folio *folio,
int old_order,
ClearPageCompound(&folio->page);
}
+/**
+ * folio_split_unref() - split an unreferenced folio (refcount == 0)
+ * @folio: the to-be-split folio
+ *
+ * Split an unreferenced folio (refcount == 0) into individual pages.
+ * Intended to be called on special pages (e.g., device-private, DAX, etc.)
+ * when returning the folio to the free page pool.
+ */
+void folio_split_unref(struct folio *folio)
+{
+ struct dev_pagemap *pgmap = page_pgmap(&folio->page);
+ int order, i;
+
+ folio->mapping = NULL;
It's unclear why you mess with the mapping. Usually, throughout a folio
split, we populate the folio->mapping to all split folios.
--
Cheers
David
