Truncating the sum of squares, and then use shift-and-add algorithm to compute its square root.

Required MPFR testing infra is updated in https://reviews.llvm.org/D87514

# Details

Details

- Reviewers
sivachandra - Commits
- rGf55963d501e4: [libc] Add implementation for hypotf

# Diff Detail

Diff Detail

- Repository
- rG LLVM Github Monorepo

### Event Timeline

libc/test/src/math/hypotf_test.cpp | ||
---|---|---|

22 | The template class FPBits<T> does not support constexpr yet. |

libc/src/math/hypotf.cpp | ||
---|---|---|

2 | With other implementations in | |

100 | I understand why the use of a 64-bit integers is required here. But, I would ideally prefer to restrict a function which deals with 32-bit numbers not have to deal with wider numbers. It surely is not a problem on most modern architectures. So, we can revisit if it ever becomes a problem. FWIW, the AOR implementations also switch to |