https://devzone.nordicsemi.com/f/nordic-q-a/28272/uarte-dma-switch-momentnRF52810 1, Conditions: Two dma buffers are used, SHORT = TASK_STARTRX, 1M baud rate (10us / B) Phenomenon: At END_RX interrupt, I called GPIO0 = low. With the logic analyzer, when RXD pin receives num = &#160;MAXCOUNT, GPIO0 will be pulled low afteren-USTelligent Community 13Fri, 15 Dec 2017 11:01:36 GMThttps://devzone.nordicsemi.com/thread/111667?ContentTypeID=1Fri, 15 Dec 2017 11:01:36 GMT137ad170-7792-4731-bb38-c0d22fbe4515:7bf6f709-846e-4a07-ac91-da266b5e9e6crun_ar<p>And you cannot use HW flow control?</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/111669?ContentTypeID=1Fri, 15 Dec 2017 00:13:26 GMT137ad170-7792-4731-bb38-c0d22fbe4515:18ff0d17-c5a7-4b42-90a0-646a3eb4c97dStayhungry<p>Because if the number of bytes accepted did not reach MAXCOUNT, can not generate END_RX, that is, can not know how many bytes received, I set a receive timeout timer, automatically call TASK_STOPRX timeout, resulting in an END_RX, according to my test , Although SHORT = TASK_STARTRX, but after calling TASK_STOPRX, the other side will send the data immediately, and will still lose the data, so I now want to use counter and EVENT RXD_RDY to get the received data, not END_RX, hoping to work</p><div style="clear:both;"></div>https://devzone.nordicsemi.com/thread/111668?ContentTypeID=1Thu, 14 Dec 2017 20:44:21 GMT137ad170-7792-4731-bb38-c0d22fbe4515:d19d6bc9-52a1-4093-9935-2037d65883b6run_ar<p>I would say, yes, and that your guess is correct. But it&#39;s hard to say without seeing your configuration. What do you mean when you say &quot;initiative to call TASK_STOPRX&quot;?</p><div style="clear:both;"></div>