[BUG] [v1.8] glitch in test 02_19_TestPipelineCreateDelete16000Hz32b32b8ch

[kkarask]

Describe the bug
Glitch was found near the start of the capture in test 02_19_TestPipelineCreateDelete16000Hz32b32b8ch

Topology

pipe_plb
                        +--------------------+
        +----------+    | +------+   +-----+ |   +----+
        | Host plb |----->|Buffer|-->| Dai |---->|SSP0|
        +----------+    | +------+   +-----+ |   +----+
                        +--------------------+    |  |
                                                  |  |
                        pipe_cap                  |  |
                        +--------------------+    |  |
        +----------+    | +------+   +-----+ |   +----+
        | Host cap |<-----|Buffer|<--| Dai |---->|SSP0|
        +----------+    | +------+   +-----+ |   +----+
                        +--------------------+

To Reproduce
Run test 02_19_TestPipelineCreateDelete16000Hz32b32b8ch with Diagnostic driver
with parameters:
--playback_iterations=75

Reproduction Rate
1/75

Environment

1. Firmware branch name and commit

   • Branch: main, Hash: 9e7a8f9, v1.8

2. Name of the platform(s) on which the bug is observed.

   • Platform: ICL RVP

Screenshots or console output

02_19_TestPipelineCreateDelete16000Hz32b32b8ch.zip

[lgirdwood]

@kkarask when we do see glitches with slim driver, are they spread evenly through the file or always near the start of the test. i.e. above I can see the glitches near the start of the capture.

[keyonjie]

@lgirdwood push out to 1.9?

[wszypelt]

unsupported platform

[antonirx]

on TGL issue does not occur
Tested FW hash: 4943aaa