In recent years, there have been significant efforts on the research and development of Massive MIMO (M-MIMO) technologies at the physical layer. So far, however, the understanding of how M-MIMO could affect the performance of network control and optimization algorithms remains rather limited. In this paper, we focus on analyzing the performance of the queue-length-based joint congestion control and scheduling framework (QCS) over M-MIMO cellular networks with limited channel state information (CSI). Our contributions in this paper are two-fold: i) We characterize the scaling performance of the queue-lengths and show that there exists a phase transitioning phenomenon in the steady-state queue-length deviation respect to the CSI quality (reflected in the number of bits $B$ that represent CSI); and ii) We characterize the congestion control rate scaling performance and show that there also exists a phase transitioning phenomenon in steady-state congestion control rate deviation respect to the CSI quality. Collectively, the findings in this paper advance our understanding of the trade-offs between delay, throughput, and the accuracy/complexity of CSI acquisition in M-MIMO cellular network systems.