-
Notifications
You must be signed in to change notification settings - Fork 6.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
usb: audio: high CPU load when running headphones_microphone sample #26857
Comments
I have done some testing on the headset sample with a nRF5340. I have tested two set ups, one with stereo hp + stereo mic (vanilla sample) and another where I remove one channel from both hp and mic from the overlay file, thus creating a mono device. The starting CPU load before any audio has been played is 10%, then I start the audio and I see that the CPU load goes up to 40% for stereo hp and 37% for mono hp. When I at the same time sink the mic line by recording with Audacity I get 63% for stereo and 57% for mono. When stopping audio and stopping sink I get 33% for both usecases after a while. EDIT: CPU load was measured using |
Ran one more test: Headphones only |
@koffes I see this has been opened for a long time. Is there an update on this? |
@ddavidebor We are waiting for this issue to be resolved as part of #26935 |
Testing again with v3.0.0-rc1 and nRF5340 This was run on an nRF5340 |
Testing on nRF52840 I get: |
#64174 makes rather significant amount of cpu time available to application because it eliminates busy loop inside interrupt while waiting for DMA to finish. |
Is your enhancement proposal related to a problem? Please describe.
Running the mentioned sample on the nRF52840. Vanilla setup, no changes to config.
The CPU loads are:
Idle: 4 %
Playback 33 %
Paused playback: 33 %. Goes down to 26 % after a while
Play and record: 52 %
Describe the solution you'd like
Please try to enable DMA. This will likely reduce CPU load significantly.
@emob-nordic @alexsven
The text was updated successfully, but these errors were encountered: