Skip to content

Skip test when hardware sample rates mismatch in data path test#2357

Open
q23175401 wants to merge 1 commit intomainfrom
skip_sample_rate_mismatch
Open

Skip test when hardware sample rates mismatch in data path test#2357
q23175401 wants to merge 1 commit intomainfrom
skip_sample_rate_mismatch

Conversation

@q23175401
Copy link
Copy Markdown
Collaborator

This adds a check to skip tests where the input and output hardware sample rates don't match when using a USB device (direct loopback), to avoid test failures caused by the mismatch.

This adds a check to skip tests where the input and output hardware
sample rates don't match when using a USB device (direct loopback),
to avoid test failures caused by the mismatch.
@q23175401 q23175401 requested a review from robertwu1 March 26, 2026 09:18
@q23175401 q23175401 self-assigned this Mar 26, 2026
@robertwu1
Copy link
Copy Markdown
Collaborator

LGTM! Thanks for doing this!

@flamme
Copy link
Copy Markdown
Collaborator

flamme commented Mar 26, 2026

This doesn't really make sense to me. The hardware sample rate API returns the configuration of the particular input or output stream configuration in the HAL side. This doesn't mean the USB device is opened in different configuration. The DSP can always open the device at a different configuration and do conversion. Why it is needed to skip the test?

@robertwu1
Copy link
Copy Markdown
Collaborator

The issue is that if the DSP has one hardware input sample rate and a different output sample rate, it can be very hard for it to do conversions efficiently. This is just like on the app side where we are opening the stream at the same input sample rate and output sample rate. Obviously this should work in all cases but in reality it's hard for some of these cases for this to work correctly, especially with something like a peripheral attached.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

3 participants