[BUG] HiFi3 sat_int32 Saturation Functions Incorrectly Handle INT64_MAX and INT64_MIN Edge Cases

[ShriramShastry]

Describe the bug
A clear and concise description of what the bug is.
What have you tried to diagnose or workaround this issue?

(1) The HiFi3 sat_int32(int64_t x) saturation functions for 32-bit integers are not correctly handling saturation for INT64_MAX and INT64_MIN values. The expected behavior is for these edge cases to return INT32_MAX and INT32_MIN, respectively, matching the original Branching implementation. However, the HiFi3 optimized functions are returning incorrect results for these cases

(2) Another issue is that the number of cycles for sat_int16() exceeds that of the reference branching int16 saturation function..

To Reproduce
Steps to reproduce the behavior: (e.g. list commands or actions used to reproduce the bug)

The HiFi3 saturation functions is here.

static inline int32_t sat_int32(int64_t x)
{
	return (ae_int32)AE_ROUND32F48SSYM(AE_SLAI64(x, 16));
}

The functions fail to handle the edge cases of INT64_MAX and INT64_MIN correctly.

Reproduction Rate
How often does the issue happen ? i.e. 1/10 (once in ten attempts), 1/1000 or all the time.
Always reproducible.

Does the reproduction rate vary with any other configuration or user action, if so please describe and show the new reproduction rate.

Expected behavior
A clear and concise description of what you expected to happen.

With input values of -9223372036854775808 and 9223372036854775807 . It is expected to return -2147483648 and 2147483647 but we are observing 0 and -1

Impact
What impact does this issue have on your progress (e.g., annoyance, showstopper)

Incorrect return values from saturation functions can compromise the correctness, reliability, and performance of a system. It is crucial to ensure that saturation functions behave correctly for all input values, including edge cases, to maintain system integrity and prevent unexpected behavior.

Example
(1) Signal Processing: Incorrect saturation can cause clipping or distortion in the audio output, leading to a poor user experience
(2) Data Integrity: With the saturation logic fails for edge cases like INT64_MAX and INT64_MIN, it could result in corrupted data being processed or stored
(3) Numerical Algorithms: Incorrect saturation can lead to erroneous results, which might propagate through the system and cause larger errors or system failures.
(4) Performance: Since sat_int32() is a part performance critical loop, this will reduce overall performance of the system

Environment

1. Branch name and commit hash of the 2 repositories: sof (firmware/topology) and linux (kernel driver).
   • Kernel: {SHA}
   • SOF: {SHA}
2. Name of the topology file
   • Topology: {FILE}
3. Name of the platform(s) on which the bug is observed.
   • Platform: {PLATFORM}

Screenshots or console output
If applicable, add a screenshot (drag-and-drop an image), or console logs
(cut-and-paste text and put a code fence (```) before and after, to help
explain the issue.

Please also include the relevant sections from the firmware log and kernel log in the report (and attach the full logs for complete reference). Kernel log is taken from dmesg and firmware log from sof-logger. See https://thesofproject.github.io/latest/developer_guides/debugability/logger/index.html

[kv2019i]

@ShriramShastry Does #9344 address this completely or still work to do?

[ShriramShastry]

@ShriramShastry Does #9344 address this completely or still work to do?

Yes, it is address by #9344