audio: mixin_mixout_generic: normal_mix: Update src/dst after each copy

[ujfalusi]

The copying part of the functions need to update the src/dst buffer pointers after each iteration to correctly account for the possible wrap of the buffer.

This was discovered while doing pause - 1sec_wait - resume - 3sec_wait cycles and around the 10th iteration audio became 'metallic' sounding.

Fixes: #7478

Thanks @ranj063 to test with XCC which is not affected and pointing me to the correct path.

[lyakh]

Interesting! If I understand the affected algorithm here correctly, those normal_mix_*() functions do mixing and possibly additionally copying. Only if after mixing there are samples for copying (this always happens at least with the first source being mixed in) and when those samples wrap - then we get this bug. So apparently this only happens when pausing

[ujfalusi]

Interesting! If I understand the affected algorithm here correctly, those normal_mix_*() functions do mixing and possibly additionally copying. Only if after mixing there are samples for copying (this always happens at least with the first source being mixed in) and when those samples wrap - then we get this bug. So apparently this only happens when pausing

In a simple case when you have a single path and a single stream then there is only copying, the mixing is skipped as there is nothing to mix.

[kv2019i]

The testbench test fails to apt fetch issue, Zephyr mtl builds fail due to #7516 (now fixed), so that leaves us with https://sof-ci.01.org/sofpr/PR7517/build6472/devicetest/index.html (usual fail) and https://sof-ci.01.org/sofpr/PR7517/build6473/devicetest/index.html (one suspend-resume fail). No fails pointing to this PR, so proceeding with merge.