This new function is defined only on ARM and serves to guarantee a barrier in the WC operation. The barrier means that another run of this loop will not combine with the stores this loop created. On x86 this is happening implicitly because of the spin_unlock(). Suggested-by: "Pavel Shamis" <Pavel.Shamis@xxxxxxx> Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxx> --- drivers/infiniband/hw/mlx5/mem.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/mlx5/mem.c b/drivers/infiniband/hw/mlx5/mem.c index 6191aa833ac2b1..6b29e9ca323ea4 100644 --- a/drivers/infiniband/hw/mlx5/mem.c +++ b/drivers/infiniband/hw/mlx5/mem.c @@ -152,6 +152,7 @@ static int post_send_nop(struct mlx5_ib_dev *dev, struct ib_qp *ibqp, u64 wr_id, for (i = 0; i < 8; i++) mlx5_write64(&mmio_wqe[i * 2], bf->bfreg->map + bf->offset + i * 8); + io_stop_wc(); bf->offset ^= bf->buf_size; base-commit: dd390cba54bbd74fb675a4ac0a78dc23a20d49e2 -- 2.37.1
