Tomasz Leman

@kv2019i wrote: > We of course have the option to not use this, and expose driver-specific interfaces instead and make the abstractions on SOF side (i.e. make ipc-zephyr.c -> ipc-zephyr-intel-adsp.c...

@lgirdwood topic for discussion: At what point can we start removing tests that have already been ported?

Skipping migration of `rstrcmp` and `rstrlen` tests for now. In Zephyr builds, these functions are just simple wrappers around standard C library functions (`strlen` and `strcmp`), so the tests wouldn't...

## Math Tests Migration Strategy ### Current CMock Structure Analysis The math tests are organized into 8 subcategories with clear separation: ``` sof/test/cmocka/src/math/ ├── arithmetic/ # Codecs & Advanced Math...

To be done after the migration is completed: - Refactor functions in sof/src/math/numbers.c: constants as function arguments + update documentation

@lgirdwood - First PR submitted for legacy test cleanup https://github.com/thesofproject/sof/pull/10367 Removes legacy cmocka tests for `list` and `fast-get` components after coverage verification: - **List**: 25/25 lines (ztest) vs 22/22 lines...

> Is this a regression? The issue might have been hidden due to the fuzzer used in CI having limited code coverage. I am not certain about the coverage achieved...

> any input on correct priority for this bug? Should this block v2.12 release? I am not sure how to prioritize this bug. However, I believe it is not a...

@kv2019i @andyross I have identified the same problem using a fuzzer with UndefinedBehaviorSanitizer: ```console INFO: Running with entropic power schedule (0xFF, 100). INFO: 29455 files found in ./ipc3_corpus INFO: -max_len...

> but yeah, there's a NULL check just before this. @kv2019i The checks you are talking about only check the beginning of the list. The null pointer appears later when...