[PATCH 5.0 096/117] IB/mlx5: Reset access mask when looping inside page fault handler

[Greg Kroah-Hartman]

From: Moni Shoua <monis@xxxxxxxxxxxx>

commit 1abe186ed8a6593069bc122da55fc684383fdc1c upstream.

If page-fault handler spans multiple MRs then the access mask needs to
be reset before each MR handling or otherwise write access will be
granted to mapped pages instead of read-only.

Cc: <stable@xxxxxxxxxxxxxxx> # 3.19
Fixes: 7bdf65d411c1 ("IB/mlx5: Handle page faults")
Reported-by: Jerome Glisse <jglisse@xxxxxxxxxx>
Signed-off-by: Moni Shoua <monis@xxxxxxxxxxxx>
Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx>
Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
 drivers/infiniband/hw/mlx5/odp.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/infiniband/hw/mlx5/odp.c
+++ b/drivers/infiniband/hw/mlx5/odp.c
@@ -560,7 +560,7 @@ static int pagefault_mr(struct mlx5_ib_d
 	struct ib_umem_odp *odp_mr = to_ib_umem_odp(mr->umem);
 	bool downgrade = flags & MLX5_PF_FLAGS_DOWNGRADE;
 	bool prefetch = flags & MLX5_PF_FLAGS_PREFETCH;
-	u64 access_mask = ODP_READ_ALLOWED_BIT;
+	u64 access_mask;
 	u64 start_idx, page_mask;
 	struct ib_umem_odp *odp;
 	size_t size;
@@ -582,6 +582,7 @@ next_mr:
 	page_shift = mr->umem->page_shift;
 	page_mask = ~(BIT(page_shift) - 1);
 	start_idx = (io_virt - (mr->mmkey.iova & page_mask)) >> page_shift;
+	access_mask = ODP_READ_ALLOWED_BIT;
 
 	if (prefetch && !downgrade && !mr->umem->writable) {
 		/* prefetch with write-access must