<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta name="Generator" content="Microsoft Exchange Server">  <style></style> </head> <body> <font face="Calibri" size="2"><span style="font-size:11pt;"> <div>Hi ,</div> <div> </div> <div>Thanks for the reply.</div> <div> </div> <div>The platform is based on ARM Cortex A9 core. </div> <div> </div> <div>Regarding the Issue 1 reported :</div> <div> </div> <div>On analyzing the pa_simple_read , the following observations are seen.</div> <div> </div> <div>The delay is casued by pa_stream_peek not retuned read length , not clear on why this could fail since pa write has started.</div> <div> </div> <div>In pa_stream_peek , read length is returned as null , hence blocked at "pa_threaded_mainloop_wait". </div> <div>It comes out of " pa_threaded_mainloop_wait" , and again fails at pa_stream_peek.</div> <div>This looping happens between 5- 8 times then it returns successfully from pa_simple_read.</div> <div> </div> <div>We ran another test code to eliminate the influence of CPU or other application , The test script is as follows :</div> <div> </div> <div>1. paplay -droute_test /home/test441.wav /* route_test is a null sink , play audio to this null sink */</div> <div>2. sleep 1 /* sleep for 1 sec before read */</div> <div>3. ./test_code.sh /* test code to read data from the null sink , added time stamp to measure the duration inside pa_simpe_read */</div> <div> </div> <div>With the above script , the measured time inside <i>first</i> pa_simpe_read is 1.8 sec</div> <div> </div> <div>The issue is not seen if paplay is immediately followed by PA read , i.e</div> <div> </div> <div>1. paplay -droute_test /home/test441.wav /* route_test is a null sink */</div> <div>2. ./test_code.sh /* test code to read data from the null sink , added time stamp to measure the duration inside pa_simpe_read */</div> <div> </div> <div>With this test , the measured time inside pa_simpe_read is few micro seconds and there is no block inside pa_simpe_read.</div> <div> </div> <div>when the test script was executed , no other major CPU consumption in the core.</div> <div> </div> <div>We are trying to understand the impact of starting the pa write earlier than pa read and the cause of this delay.</div> <div> </div> <div>The pa write instance is in an external application and operates independently and the instance cannot be synchronised with pa read start.</div> <div> </div> <div>2. On Issue 2 ,</div> <div> </div> <div> The input and output both are at 44.1 Khz and no resampling of audio involved. CPU is seen when there is no write/read hapenning.</div> <div> </div> <div>Is there any specific PA compile time options to be set for optimized CPU ? please let know .</div> <div> </div> <div>It is ARM cortex A9 at 800 MHz</div> <div> </div> <div> </div> <div>Thanks & Regards</div> <div> </div> <div>Manjula R</div> <div> </div> <div> </div> <div> </div> <div> </div> <div> </div> <div>Message: 4</div> <div>Date: Mon, 9 Feb 2015 18:19:38 +0530</div> <div>From: Arun Raghavan <<a href="mailto:arun@accosted.net">arun@accosted.net</a>></div> <div>To: General PulseAudio Discussion</div> <div> <<a href="mailto:pulseaudio-discuss@lists.freedesktop.org">pulseaudio-discuss@lists.freedesktop.org</a>></div> <div>Cc: "Viswambharan, Vibin \(V.\)" <<a href="mailto:vvibin@visteon.com">vvibin@visteon.com</a>>, "Sampath,</div> <div> NishathKumaran \(N.\)" <<a href="mailto:nsampat2@visteon.com">nsampat2@visteon.com</a>></div> <div>Subject: Re: [pulseaudio-discuss] PA- Performance issue with PA read</div> <div> and CPU consumption in idle state</div> <div>Message-ID:</div> <div> <<a href="mailto:CACuU-+i2YKLcWL81QS-zBL3riU0X_+7RBfkBg-cFcrP8VaejWw@mail.gmail.com">CACuU-+i2YKLcWL81QS-zBL3riU0X_+7RBfkBg-cFcrP8VaejWw@mail.gmail.com</a>></div> <div>Content-Type: text/plain; charset=UTF-8</div> <div> </div> <div>On 4 February 2015 at 16:13, Radhakrishnan, Manjula (M.)</div> <div><<a href="mailto:rmanjula@visteon.com">rmanjula@visteon.com</a>> wrote:</div> <div>></div> <div>> Hi ,</div> <div>></div> <div>></div> <div>></div> <div>> We are observing performance issue with Pulse Audio in our embedded infotainment platform.</div> <div> </div> <div>I'm assuming embedded => ARM below.</div> <div> </div> <div>></div> <div>> To brief on the usage of PA in this platform , Pulse Audio is used for Audio routing from an Audio source provider to another Audio Source sink using Pulse Audio using the Null sink and Null sink.monitor.</div> <div>></div> <div>></div> <div>></div> <div>> The audio path is as below.</div> <div>></div> <div>> Audio source provider -> Pulse Audio Nullsink (eg. Route_test) -> Pulse Audio Null source (eg. Route_test.monitor) -> Audio router to sink</div> <div>></div> <div>></div> <div>></div> <div>> There are two behavioral issues observed in the Pulse Audio , below is the description of the issues.</div> <div>></div> <div>></div> <div>></div> <div>> 1. The Audio source provider creates a Pulse stream and writes to the Pulse Audio Nullsink before the Audio router thread has opened the Pulse Stream for reading from nullsink.monitor . In this condition , from the application Audio router , PA reader thread opens the PA stream and calls the PA read API , it is blocked for more than 1 sec inside PA read API before the first read is successful.</div> <div>></div> <div>></div> <div>></div> <div>> What could cause this delay in the first PA read ? Does the writer starting to write to PA first cause a delay/block on the reader when it eventually becomes active ?</div> <div> </div> <div>The delay sounds like a bug. It would be interesting to debug/profile</div> <div>and see where the time is spent.</div> <div> </div> <div>></div> <div>> 2. For the same issue above, on further debugging , On the PA reader side , the PA stream open is called and the stream is kept open always by the reader. In this condition , the delay is no longer observed in the first PA read. But this causes an impact on the CPU.</div> <div>></div> <div>> When the PA stream is always open and no reads are done ,significant CPU usage is observed , PA process consumes 21 - 23 % and the threaded-m1 consumes 5 – 7 % CPU even when there is no read/write happening in PA server.</div> <div>></div> <div>></div> <div>></div> <div>> Why is this high CPU observed and how to optimize the CPU , is there any configuration where the PA stream can be opened and still have low CPU consumption ?</div> <div>></div> <div>> Please provide your insights on the two queries.</div> <div> </div> <div>It would be interesting to know what CPU this is, what frequency it is</div> <div>running at, and what kind of scaling might be happening (ideally,</div> <div>there should be no scaling while you're measuring CPU consumption, and</div> <div>you peg the frequency at some sensible number).</div> <div> </div> <div>Where the CPU utilisation is going would best be determined by</div> <div>profiling (perf top is your friend). Things that could be going wrong</div> <div>vary from bad drivers that wake up the CPU too often, to not having</div> <div>your resampler (speex?) or PulseAudio compiled with the correct CPU</div> <div>optimisations (speex needs to be compiled with correct flags depending</div> <div>on whether you want it to work on float or int samples, and PA has</div> <div>optimisations for ARMv6 and NEON as well).</div> <div> </div> <div>Regards,</div> <div>Arun</div> <div> </div> <div> </div> <div>------------------------------</div> <div> </div> <div>Subject: Digest Footer</div> <div> </div> <div>_______________________________________________</div> <div>pulseaudio-discuss mailing list</div> <div><a href="mailto:pulseaudio-discuss@lists.freedesktop.org">pulseaudio-discuss@lists.freedesktop.org</a></div> <div><a href="http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss">http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss</a></div> <div> </div> <div> </div> <div>------------------------------</div> <div> </div> <div>End of pulseaudio-discuss Digest, Vol 46, Issue 19</div> <div>**************************************************</div> <div> </div> </span></font> </body> </html>