The patch titled
Subject: mm/hugetlb: fix memfd_pin_folios resv_huge_pages leak
has been added to the -mm mm-hotfixes-unstable branch. Its filename is
mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak.patch
This patch will later appear in the mm-hotfixes-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Steve Sistare <steven.sistare@xxxxxxxxxx>
Subject: mm/hugetlb: fix memfd_pin_folios resv_huge_pages leak
Date: Tue, 3 Sep 2024 07:25:19 -0700
memfd_pin_folios followed by unpin_folios leaves resv_huge_pages elevated
if the pages were not already faulted in. During a normal page fault,
resv_huge_pages is consumed here:
hugetlb_fault()
alloc_hugetlb_folio()
dequeue_hugetlb_folio_vma()
dequeue_hugetlb_folio_nodemask()
dequeue_hugetlb_folio_node_exact()
free_huge_pages--
resv_huge_pages--
During memfd_pin_folios, the page is created by calling
alloc_hugetlb_folio_nodemask instead of alloc_hugetlb_folio, and
resv_huge_pages is not modified:
memfd_alloc_folio()
alloc_hugetlb_folio_nodemask()
dequeue_hugetlb_folio_nodemask()
dequeue_hugetlb_folio_node_exact()
free_huge_pages--
alloc_hugetlb_folio_nodemask has other callers that must not modify
resv_huge_pages. Therefore, to fix, define an alternate version of
alloc_hugetlb_folio_nodemask for this call site that adjusts
resv_huge_pages.
Link: https://lkml.kernel.org/r/1725373521-451395-4-git-send-email-steven.sistare@xxxxxxxxxx
Fixes: 89c1905d9c14 ("mm/gup: introduce memfd_pin_folios() for pinning memfd folios")
Signed-off-by: Steve Sistare <steven.sistare@xxxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Muchun Song <muchun.song@xxxxxxxxx>
Cc: Peter Xu <peterx@xxxxxxxxxx>
Cc: Vivek Kasireddy <vivek.kasireddy@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/hugetlb.h | 10 ++++++++++
mm/hugetlb.c | 17 +++++++++++++++++
mm/memfd.c | 9 ++++-----
3 files changed, 31 insertions(+), 5 deletions(-)
--- a/include/linux/hugetlb.h~mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak
+++ a/include/linux/hugetlb.h
@@ -695,6 +695,9 @@ struct folio *alloc_hugetlb_folio(struct
struct folio *alloc_hugetlb_folio_nodemask(struct hstate *h, int preferred_nid,
nodemask_t *nmask, gfp_t gfp_mask,
bool allow_alloc_fallback);
+struct folio *alloc_hugetlb_folio_reserve(struct hstate *h, int preferred_nid,
+ nodemask_t *nmask, gfp_t gfp_mask);
+
int hugetlb_add_to_page_cache(struct folio *folio, struct address_space *mapping,
pgoff_t idx);
void restore_reserve_on_error(struct hstate *h, struct vm_area_struct *vma,
@@ -1060,6 +1063,13 @@ static inline struct folio *alloc_hugetl
{
return NULL;
}
+
+static inline struct folio *
+alloc_hugetlb_folio_reserve(struct hstate *h, int preferred_nid,
+ nodemask_t *nmask, gfp_t gfp_mask)
+{
+ return NULL;
+}
static inline struct folio *
alloc_hugetlb_folio_nodemask(struct hstate *h, int preferred_nid,
--- a/mm/hugetlb.c~mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak
+++ a/mm/hugetlb.c
@@ -2564,6 +2564,23 @@ struct folio *alloc_buddy_hugetlb_folio_
return folio;
}
+struct folio *alloc_hugetlb_folio_reserve(struct hstate *h, int preferred_nid,
+ nodemask_t *nmask, gfp_t gfp_mask)
+{
+ struct folio *folio;
+
+ spin_lock_irq(&hugetlb_lock);
+ folio = dequeue_hugetlb_folio_nodemask(h, gfp_mask, preferred_nid,
+ nmask);
+ if (folio) {
+ VM_BUG_ON(!h->resv_huge_pages);
+ h->resv_huge_pages--;
+ }
+
+ spin_unlock_irq(&hugetlb_lock);
+ return folio;
+}
+
/* folio migration callback function */
struct folio *alloc_hugetlb_folio_nodemask(struct hstate *h, int preferred_nid,
nodemask_t *nmask, gfp_t gfp_mask, bool allow_alloc_fallback)
--- a/mm/memfd.c~mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak
+++ a/mm/memfd.c
@@ -82,11 +82,10 @@ struct folio *memfd_alloc_folio(struct f
gfp_mask = htlb_alloc_mask(hstate_file(memfd));
gfp_mask &= ~(__GFP_HIGHMEM | __GFP_MOVABLE);
- folio = alloc_hugetlb_folio_nodemask(hstate_file(memfd),
- numa_node_id(),
- NULL,
- gfp_mask,
- false);
+ folio = alloc_hugetlb_folio_reserve(hstate_file(memfd),
+ numa_node_id(),
+ NULL,
+ gfp_mask);
if (folio && folio_try_get(folio)) {
err = hugetlb_add_to_page_cache(folio,
memfd->f_mapping,
_
Patches currently in -mm which might be from steven.sistare@xxxxxxxxxx are
mm-filemap-fix-filemap_get_folios_contig-thp-panic.patch
mm-hugetlb-fix-memfd_pin_folios-free_huge_pages-leak.patch
mm-hugetlb-fix-memfd_pin_folios-resv_huge_pages-leak.patch
mm-gup-fix-memfd_pin_folios-hugetlb-page-allocation.patch
mm-gup-fix-memfd_pin_folios-alloc-race-panic.patch
