[libc-commits] [libc] [llvm] [libc][math][c23] Add atan2f16 function (PR #183531)

[Xinlong Chen via libc-commits]

Xinlong-Chen wrote:

> With atan2f16, we can test exhaustively all inputs (~2^31). Can you add an exhaustive test for it in `libc/test/src/math/exhaustive` folder. Notice that it will need to be run manually after built with ninja.

Thanks for the review. I've made the requested changes - please let me know if anything else looks off.

I have implement an exhaustive test for `atan2f16` (in libc/test/src/math/exhaustive/atan2f16_test.cpp) by comparing its output to MPFR over the float16 input range.

The results are as follows (manually run, taking nearly 3h):

```
[==========] Running 1 test from 1 test suite.
[ RUN      ] LlvmLibcAtan2f16ExhaustiveTest.AllInputs
-- Testing for FE_TONEAREST in x range [0x0, 0xfc00), y range [0x0, 0xfc00) --
100% is in process     
Test PASSED
-- Testing for FE_UPWARD in x range [0x0, 0xfc00), y range [0x0, 0xfc00) --
100% is in process     
Test PASSED
-- Testing for FE_DOWNWARD in x range [0x0, 0xfc00), y range [0x0, 0xfc00) --
100% is in process     
Test PASSED
-- Testing for FE_TOWARDZERO in x range [0x0, 0xfc00), y range [0x0, 0xfc00) --
100% is in process     
Test PASSED
[       OK ] LlvmLibcAtan2f16ExhaustiveTest.AllInputs (136240512 ms)
Ran 1 tests.  PASS: 1  FAIL: 0
```

https://github.com/llvm/llvm-project/pull/183531