[libc] Floating point exception error for cospif16/sinpif16 on aarch64.

[michaelrj-google]

Example failure from the buildbot:

[==========] Running 2 tests from 1 test suite.
[ RUN      ] LlvmLibcSinpif16Test.SpecialNumbers
/home/libc-buildbot/libc-aarch64-ubuntu/libc-aarch64-ubuntu/llvm-project/libc/test/src/math/smoke/sinpif16_test.cpp:20: FAILURE
      Expected: __llvm_libc_20_0_0_git::fputil::test_except( static_cast<int>((1 | 2 | 4 | 8 | 16))) & ((1) ? (1) : static_cast<int>((1 | 2 | 4 | 8 | 16)))
      Which is: 0
To be equal to: (1)
      Which is: 1
[  FAILED  ] LlvmLibcSinpif16Test.SpecialNumbers
[ RUN      ] LlvmLibcSinpif16Test.Integers
[       OK ] LlvmLibcSinpif16Test.Integers (9 us)
Ran 2 tests.  PASS: 1  FAIL: 1

Seems like there's a rounding issue.

[llvmbot]

@llvm/issue-subscribers-libc

Author: Michael Jones (michaelrj-google)

Example failure from the buildbot: ``` [==========] Running 2 tests from 1 test suite. [ RUN ] LlvmLibcSinpif16Test.SpecialNumbers /home/libc-buildbot/libc-aarch64-ubuntu/libc-aarch64-ubuntu/llvm-project/libc/test/src/math/smoke/sinpif16_test.cpp:20: FAILURE Expected: __llvm_libc_20_0_0_git::fputil::test_except( static_cast<int>((1 | 2 | 4 | 8 | 16))) & ((1) ? (1) : static_cast<int>((1 | 2 | 4 | 8 | 16))) Which is: 0 To be equal to: (1) Which is: 1 [ FAILED ] LlvmLibcSinpif16Test.SpecialNumbers [ RUN ] LlvmLibcSinpif16Test.Integers [ OK ] LlvmLibcSinpif16Test.Integers (9 us) Ran 2 tests. PASS: 1 FAIL: 1 ``` Seems like there's a rounding issue.