mm/rmap: introduce and use hugetlb_try_share_anon_rmap()
authorDavid Hildenbrand <david@redhat.com>
Wed, 20 Dec 2023 22:44:29 +0000 (23:44 +0100)
committerAndrew Morton <akpm@linux-foundation.org>
Fri, 29 Dec 2023 19:58:48 +0000 (11:58 -0800)
commit0c2ec32bf0b2f0d7ccb98c53ee5d255d68e73595
tree0c59e12d15cb796b80755874b694e29a2f735d5e
parentebe2e35ec0f256372c158a18de459fb60070b313
mm/rmap: introduce and use hugetlb_try_share_anon_rmap()

hugetlb rmap handling differs quite a lot from "ordinary" rmap code.  For
example, hugetlb currently only supports entire mappings, and treats any
mapping as mapped using a single "logical PTE".  Let's move it out of the
way so we can overhaul our "ordinary" rmap.  implementation/interface.

So let's introduce and use hugetlb_try_dup_anon_rmap() to make all hugetlb
handling use dedicated hugetlb_* rmap functions.

Add sanity checks that we end up with the right folios in the right
functions.

Note that try_to_unmap_one() does not need care.  Easy to spot because
among all that nasty hugetlb special-casing in that function, we're not
using set_huge_pte_at() on the anon path -- well, and that code assumes
that we would want to swapout.

Link: https://lkml.kernel.org/r/20231220224504.646757-6-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Yin Fengwei <fengwei.yin@intel.com>
Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
include/linux/rmap.h
mm/rmap.c