Skip to content

Commit f91d547

Browse files
Barry Songgregkh
Barry Song
authored and
gregkh
committed
mm: fix kernel BUG when userfaultfd_move encounters swapcache
commit c50f8e6 upstream. userfaultfd_move() checks whether the PTE entry is present or a swap entry. - If the PTE entry is present, move_present_pte() handles folio migration by setting: src_folio->index = linear_page_index(dst_vma, dst_addr); - If the PTE entry is a swap entry, move_swap_pte() simply copies the PTE to the new dst_addr. This approach is incorrect because, even if the PTE is a swap entry, it can still reference a folio that remains in the swap cache. This creates a race window between steps 2 and 4. 1. add_to_swap: The folio is added to the swapcache. 2. try_to_unmap: PTEs are converted to swap entries. 3. pageout: The folio is written back. 4. Swapcache is cleared. If userfaultfd_move() occurs in the window between steps 2 and 4, after the swap PTE has been moved to the destination, accessing the destination triggers do_swap_page(), which may locate the folio in the swapcache. However, since the folio's index has not been updated to match the destination VMA, do_swap_page() will detect a mismatch. This can result in two critical issues depending on the system configuration. If KSM is disabled, both small and large folios can trigger a BUG during the add_rmap operation due to: page_pgoff(folio, page) != linear_page_index(vma, address) [ 13.336953] page: refcount:6 mapcount:1 mapping:00000000f43db19c index:0xffffaf150 pfn:0x4667c [ 13.337520] head: order:2 mapcount:1 entire_mapcount:0 nr_pages_mapped:1 pincount:0 [ 13.337716] memcg:ffff00000405f000 [ 13.337849] anon flags: 0x3fffc0000020459(locked|uptodate|dirty|owner_priv_1|head|swapbacked|node=0|zone=0|lastcpupid=0xffff) [ 13.338630] raw: 03fffc0000020459 ffff80008507b538 ffff80008507b538 ffff000006260361 [ 13.338831] raw: 0000000ffffaf150 0000000000004000 0000000600000000 ffff00000405f000 [ 13.339031] head: 03fffc0000020459 ffff80008507b538 ffff80008507b538 ffff000006260361 [ 13.339204] head: 0000000ffffaf150 0000000000004000 0000000600000000 ffff00000405f000 [ 13.339375] head: 03fffc0000000202 fffffdffc0199f01 ffffffff00000000 0000000000000001 [ 13.339546] head: 0000000000000004 0000000000000000 00000000ffffffff 0000000000000000 [ 13.339736] page dumped because: VM_BUG_ON_PAGE(page_pgoff(folio, page) != linear_page_index(vma, address)) [ 13.340190] ------------[ cut here ]------------ [ 13.340316] kernel BUG at mm/rmap.c:1380! [ 13.340683] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP [ 13.340969] Modules linked in: [ 13.341257] CPU: 1 UID: 0 PID: 107 Comm: a.out Not tainted 6.14.0-rc3-gcf42737e247a-dirty torvalds#299 [ 13.341470] Hardware name: linux,dummy-virt (DT) [ 13.341671] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 13.341815] pc : __page_check_anon_rmap+0xa0/0xb0 [ 13.341920] lr : __page_check_anon_rmap+0xa0/0xb0 [ 13.342018] sp : ffff80008752bb20 [ 13.342093] x29: ffff80008752bb20 x28: fffffdffc0199f00 x27: 0000000000000001 [ 13.342404] x26: 0000000000000000 x25: 0000000000000001 x24: 0000000000000001 [ 13.342575] x23: 0000ffffaf0d0000 x22: 0000ffffaf0d0000 x21: fffffdffc0199f00 [ 13.342731] x20: fffffdffc0199f00 x19: ffff000006210700 x18: 00000000ffffffff [ 13.342881] x17: 6c203d2120296567 x16: 6170202c6f696c6f x15: 662866666f67705f [ 13.343033] x14: 6567617028454741 x13: 2929737365726464 x12: ffff800083728ab0 [ 13.343183] x11: ffff800082996bf8 x10: 0000000000000fd7 x9 : ffff80008011bc40 [ 13.343351] x8 : 0000000000017fe8 x7 : 00000000fffff000 x6 : ffff8000829eebf8 [ 13.343498] x5 : c0000000fffff000 x4 : 0000000000000000 x3 : 0000000000000000 [ 13.343645] x2 : 0000000000000000 x1 : ffff0000062db980 x0 : 000000000000005f [ 13.343876] Call trace: [ 13.344045] __page_check_anon_rmap+0xa0/0xb0 (P) [ 13.344234] folio_add_anon_rmap_ptes+0x22c/0x320 [ 13.344333] do_swap_page+0x1060/0x1400 [ 13.344417] __handle_mm_fault+0x61c/0xbc8 [ 13.344504] handle_mm_fault+0xd8/0x2e8 [ 13.344586] do_page_fault+0x20c/0x770 [ 13.344673] do_translation_fault+0xb4/0xf0 [ 13.344759] do_mem_abort+0x48/0xa0 [ 13.344842] el0_da+0x58/0x130 [ 13.344914] el0t_64_sync_handler+0xc4/0x138 [ 13.345002] el0t_64_sync+0x1ac/0x1b0 [ 13.345208] Code: aa1503e0 f000f801 910f6021 97ff5779 (d4210000) [ 13.345504] ---[ end trace 0000000000000000 ]--- [ 13.345715] note: a.out[107] exited with irqs disabled [ 13.345954] note: a.out[107] exited with preempt_count 2 If KSM is enabled, Peter Xu also discovered that do_swap_page() may trigger an unexpected CoW operation for small folios because ksm_might_need_to_copy() allocates a new folio when the folio index does not match linear_page_index(vma, addr). This patch also checks the swapcache when handling swap entries. If a match is found in the swapcache, it processes it similarly to a present PTE. However, there are some differences. For example, the folio is no longer exclusive because folio_try_share_anon_rmap_pte() is performed during unmapping. Furthermore, in the case of swapcache, the folio has already been unmapped, eliminating the risk of concurrent rmap walks and removing the need to acquire src_folio's anon_vma or lock. Note that for large folios, in the swapcache handling path, we directly return -EBUSY since split_folio() will return -EBUSY regardless if the folio is under writeback or unmapped. This is not an urgent issue, so a follow-up patch may address it separately. [v-songbaohua@oppo.com: minor cleanup according to Peter Xu] Link: https://lkml.kernel.org/r/20250226024411.47092-1-21cnbao@gmail.com Link: https://lkml.kernel.org/r/20250226001400.9129-1-21cnbao@gmail.com Fixes: adef440 ("userfaultfd: UFFDIO_MOVE uABI") Signed-off-by: Barry Song <v-songbaohua@oppo.com> Acked-by: Peter Xu <peterx@redhat.com> Reviewed-by: Suren Baghdasaryan <surenb@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Axel Rasmussen <axelrasmussen@google.com> Cc: Brian Geffon <bgeffon@google.com> Cc: Christian Brauner <brauner@kernel.org> Cc: David Hildenbrand <david@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jann Horn <jannh@google.com> Cc: Kalesh Singh <kaleshsingh@google.com> Cc: Liam R. Howlett <Liam.Howlett@oracle.com> Cc: Lokesh Gidra <lokeshgidra@google.com> Cc: Matthew Wilcox (Oracle) <willy@infradead.org> Cc: Michal Hocko <mhocko@suse.com> Cc: Mike Rapoport (IBM) <rppt@kernel.org> Cc: Nicolas Geoffray <ngeoffray@google.com> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Shuah Khan <shuah@kernel.org> Cc: ZhangPeng <zhangpeng362@huawei.com> Cc: Tangquan Zheng <zhengtangquan@oppo.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> [ surenb: resolved merged conflict caused by the difference in move_swap_pte() arguments] Signed-off-by: Suren Baghdasaryan <surenb@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
1 parent 648e04a commit f91d547

File tree

1 file changed

+66
-9
lines changed

1 file changed

+66
-9
lines changed

‎mm/userfaultfd.c

+66-9
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,7 @@
1818
#include <asm/tlbflush.h>
1919
#include <asm/tlb.h>
2020
#include "internal.h"
21+
#include "swap.h"
2122

2223
static __always_inline
2324
bool validate_dst_vma(struct vm_area_struct *dst_vma, unsigned long dst_end)
@@ -1067,15 +1068,13 @@ static int move_present_pte(struct mm_struct *mm,
10671068
return err;
10681069
}
10691070

1070-
static int move_swap_pte(struct mm_struct *mm,
1071+
static int move_swap_pte(struct mm_struct *mm, struct vm_area_struct *dst_vma,
10711072
unsigned long dst_addr, unsigned long src_addr,
10721073
pte_t *dst_pte, pte_t *src_pte,
10731074
pte_t orig_dst_pte, pte_t orig_src_pte,
1074-
spinlock_t *dst_ptl, spinlock_t *src_ptl)
1075+
spinlock_t *dst_ptl, spinlock_t *src_ptl,
1076+
struct folio *src_folio)
10751077
{
1076-
if (!pte_swp_exclusive(orig_src_pte))
1077-
return -EBUSY;
1078-
10791078
double_pt_lock(dst_ptl, src_ptl);
10801079

10811080
if (!pte_same(ptep_get(src_pte), orig_src_pte) ||
@@ -1084,6 +1083,16 @@ static int move_swap_pte(struct mm_struct *mm,
10841083
return -EAGAIN;
10851084
}
10861085

1086+
/*
1087+
* The src_folio resides in the swapcache, requiring an update to its
1088+
* index and mapping to align with the dst_vma, where a swap-in may
1089+
* occur and hit the swapcache after moving the PTE.
1090+
*/
1091+
if (src_folio) {
1092+
folio_move_anon_rmap(src_folio, dst_vma);
1093+
src_folio->index = linear_page_index(dst_vma, dst_addr);
1094+
}
1095+
10871096
orig_src_pte = ptep_get_and_clear(mm, src_addr, src_pte);
10881097
set_pte_at(mm, dst_addr, dst_pte, orig_src_pte);
10891098
double_pt_unlock(dst_ptl, src_ptl);
@@ -1130,6 +1139,7 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
11301139
__u64 mode)
11311140
{
11321141
swp_entry_t entry;
1142+
struct swap_info_struct *si = NULL;
11331143
pte_t orig_src_pte, orig_dst_pte;
11341144
pte_t src_folio_pte;
11351145
spinlock_t *src_ptl, *dst_ptl;
@@ -1321,6 +1331,8 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
13211331
orig_dst_pte, orig_src_pte,
13221332
dst_ptl, src_ptl, src_folio);
13231333
} else {
1334+
struct folio *folio = NULL;
1335+
13241336
entry = pte_to_swp_entry(orig_src_pte);
13251337
if (non_swap_entry(entry)) {
13261338
if (is_migration_entry(entry)) {
@@ -1334,10 +1346,53 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
13341346
goto out;
13351347
}
13361348

1337-
err = move_swap_pte(mm, dst_addr, src_addr,
1338-
dst_pte, src_pte,
1339-
orig_dst_pte, orig_src_pte,
1340-
dst_ptl, src_ptl);
1349+
if (!pte_swp_exclusive(orig_src_pte)) {
1350+
err = -EBUSY;
1351+
goto out;
1352+
}
1353+
1354+
si = get_swap_device(entry);
1355+
if (unlikely(!si)) {
1356+
err = -EAGAIN;
1357+
goto out;
1358+
}
1359+
/*
1360+
* Verify the existence of the swapcache. If present, the folio's
1361+
* index and mapping must be updated even when the PTE is a swap
1362+
* entry. The anon_vma lock is not taken during this process since
1363+
* the folio has already been unmapped, and the swap entry is
1364+
* exclusive, preventing rmap walks.
1365+
*
1366+
* For large folios, return -EBUSY immediately, as split_folio()
1367+
* also returns -EBUSY when attempting to split unmapped large
1368+
* folios in the swapcache. This issue needs to be resolved
1369+
* separately to allow proper handling.
1370+
*/
1371+
if (!src_folio)
1372+
folio = filemap_get_folio(swap_address_space(entry),
1373+
swap_cache_index(entry));
1374+
if (!IS_ERR_OR_NULL(folio)) {
1375+
if (folio_test_large(folio)) {
1376+
err = -EBUSY;
1377+
folio_put(folio);
1378+
goto out;
1379+
}
1380+
src_folio = folio;
1381+
src_folio_pte = orig_src_pte;
1382+
if (!folio_trylock(src_folio)) {
1383+
pte_unmap(&orig_src_pte);
1384+
pte_unmap(&orig_dst_pte);
1385+
src_pte = dst_pte = NULL;
1386+
put_swap_device(si);
1387+
si = NULL;
1388+
/* now we can block and wait */
1389+
folio_lock(src_folio);
1390+
goto retry;
1391+
}
1392+
}
1393+
err = move_swap_pte(mm, dst_vma, dst_addr, src_addr, dst_pte, src_pte,
1394+
orig_dst_pte, orig_src_pte,
1395+
dst_ptl, src_ptl, src_folio);
13411396
}
13421397

13431398
out:
@@ -1354,6 +1409,8 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
13541409
if (src_pte)
13551410
pte_unmap(src_pte);
13561411
mmu_notifier_invalidate_range_end(&range);
1412+
if (si)
1413+
put_swap_device(si);
13571414

13581415
return err;
13591416
}

0 commit comments

Comments
 (0)