On 26 Feb 2024, at 15:55, Zi Yan wrote: > From: Zi Yan <z...@nvidia.com> > > It is used to test split_huge_page_to_list_to_order for pagecache THPs. > Also add test cases for split_huge_page_to_list_to_order via both > debugfs. > > Signed-off-by: Zi Yan <z...@nvidia.com> > --- > mm/huge_memory.c | 34 ++++-- > .../selftests/mm/split_huge_page_test.c | 115 +++++++++++++++++- > 2 files changed, 131 insertions(+), 18 deletions(-)
Hi Andrew, This is the fixup for patch 8. It is based on the discussion with Dan Carpenter at https://lore.kernel.org/linux-mm/7dda9283-b437-4cf8-ab0d-83c330deb9c0@moroto.mountain/. It checks new_order input from debugfs and skips folios early if new_order is greater than the folio order. Thanks. diff --git a/mm/huge_memory.c b/mm/huge_memory.c index a81a09236c16..42d4f62d7760 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3484,6 +3484,9 @@ static int split_huge_pages_pid(int pid, unsigned long vaddr_start, goto next; total++; + + if (new_order >= folio_order(folio)) + goto next; /* * For folios with private, split_huge_page_to_list_to_order() * will try to drop it before split and then check if the folio @@ -3550,6 +3553,9 @@ static int split_huge_pages_in_file(const char *file_path, pgoff_t off_start, total++; nr_pages = folio_nr_pages(folio); + if (new_order >= folio_order(folio)) + goto next; + if (!folio_trylock(folio)) goto next; -- Best Regards, Yan, Zi
signature.asc
Description: OpenPGP digital signature
