[PATCH v2 1/1] userfaultfd: fix move_pages_pte() splitting folio under RCU read lock

[Suren Baghdasaryan <surenb@google.com>]

While testing the split PMD path with lockdep enabled I've got an
"Invalid wait context" error caused by split_huge_page_to_list() trying
to lock anon_vma->rwsem while inside RCU read section. The issues is due
to move_pages_pte() calling split_folio() under RCU read lock. Fix this
by unmapping the PTEs and exiting RCU read section before splitting the
folio and then retrying. The same retry pattern is used when locking the
folio or anon_vma in this function. After splitting the large folio we
unlock and release it because after the split the old folio might not be
the one that contains the src_addr.

Fixes: 94b01c885131 ("userfaultfd: UFFDIO_MOVE uABI")
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
---
Changes from v1 [1]:
1. Reset src_folio and src_folio_pte after folio is split, per Peter Xu

[1] https://lore.kernel.org/all/20231230025607.2476912-1-surenb@google.com/

 mm/userfaultfd.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
index 5e718014e671..216ab4c8621f 100644
--- a/mm/userfaultfd.c
+++ b/mm/userfaultfd.c
@@ -1078,9 +1078,18 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
 
 		/* at this point we have src_folio locked */
 		if (folio_test_large(src_folio)) {
+			/* split_folio() can block */
+			pte_unmap(&orig_src_pte);
+			pte_unmap(&orig_dst_pte);
+			src_pte = dst_pte = NULL;
 			err = split_folio(src_folio);
 			if (err)
 				goto out;
+			/* have to reacquire the folio after it got split */
+			folio_unlock(src_folio);
+			folio_put(src_folio);
+			src_folio = NULL;
+			goto retry;
 		}
 
 		if (!src_anon_vma) {
-- 
2.43.0.472.g3155946c3a-goog