[PATCH 4.19 11/25] mm/migrate_device.c: flush TLB while holding PTL

From: Greg Kroah-Hartman
Date: Mon Oct 03 2022 - 04:44:36 EST


From: Alistair Popple <apopple@xxxxxxxxxx>

commit 60bae73708963de4a17231077285bd9ff2f41c44 upstream.

When clearing a PTE the TLB should be flushed whilst still holding the PTL
to avoid a potential race with madvise/munmap/etc. For example consider
the following sequence:

CPU0 CPU1
---- ----

migrate_vma_collect_pmd()
pte_unmap_unlock()
madvise(MADV_DONTNEED)
-> zap_pte_range()
pte_offset_map_lock()
[ PTE not present, TLB not flushed ]
pte_unmap_unlock()
[ page is still accessible via stale TLB ]
flush_tlb_range()

In this case the page may still be accessed via the stale TLB entry after
madvise returns. Fix this by flushing the TLB while holding the PTL.

Fixes: 8c3328f1f36a ("mm/migrate: migrate_vma() unmap page from vma while collecting pages")
Link: https://lkml.kernel.org/r/9f801e9d8d830408f2ca27821f606e09aa856899.1662078528.git-series.apopple@xxxxxxxxxx
Signed-off-by: Alistair Popple <apopple@xxxxxxxxxx>
Reported-by: Nadav Amit <nadav.amit@xxxxxxxxx>
Reviewed-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
Acked-by: David Hildenbrand <david@xxxxxxxxxx>
Acked-by: Peter Xu <peterx@xxxxxxxxxx>
Cc: Alex Sierra <alex.sierra@xxxxxxx>
Cc: Ben Skeggs <bskeggs@xxxxxxxxxx>
Cc: Felix Kuehling <Felix.Kuehling@xxxxxxx>
Cc: huang ying <huang.ying.caritas@xxxxxxxxx>
Cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
Cc: John Hubbard <jhubbard@xxxxxxxxxx>
Cc: Karol Herbst <kherbst@xxxxxxxxxx>
Cc: Logan Gunthorpe <logang@xxxxxxxxxxxx>
Cc: Lyude Paul <lyude@xxxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Paul Mackerras <paulus@xxxxxxxxxx>
Cc: Ralph Campbell <rcampbell@xxxxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
mm/migrate.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2359,13 +2359,14 @@ next:
migrate->dst[migrate->npages] = 0;
migrate->src[migrate->npages++] = mpfn;
}
- arch_leave_lazy_mmu_mode();
- pte_unmap_unlock(ptep - 1, ptl);

/* Only flush the TLB if we actually modified any entries */
if (unmapped)
flush_tlb_range(walk->vma, start, end);

+ arch_leave_lazy_mmu_mode();
+ pte_unmap_unlock(ptep - 1, ptl);
+
return 0;
}