d6e3b528b2ae2a55d55749e9ef50b1e12ff34bc3 to 8ad4344ca79f2f248bc5ed70eec72e4b9c4d5e88 · libeigen / eigen · GitLab

Commits on Source 11

Vectorize erf(x) for double. · 5133c836
Rasmus Munk Larsen authored Nov 16, 2024

5133c836
Fix C++20 constexpr test compilation failures · 2fc63808
Tyler Veness authored Nov 18, 2024 and Charles Schlosser committed Nov 18, 2024

2fc63808
Get rid of redundant computation for large arguments to erf(x). · d5eec781
Rasmus Munk Larsen authored Nov 18, 2024

d5eec781
Remove unnecessary check for HasBlend trait. · 0af6ab4b
Rasmus Munk Larsen authored Nov 18, 2024

0af6ab4b
disable fill_n optimization for msvc · 93ec5450
Charles Schlosser authored Nov 19, 2024

93ec5450
Speed up exp(x). · 8ee6f847
Rasmus Munk Larsen authored Nov 19, 2024

8ee6f847
Revert libeigen/eigen@04018007 · 00af4710
Rasmus Munk Larsen authored Nov 19, 2024

00af4710
Prevent premature overflow to infinity in exp(x). The changes also provide a 3-4% speedup. · e7c799b7
Rasmus Munk Larsen authored Nov 19, 2024

e7c799b7
Enable vectorized erf<double>(x) for SSE and AVX, which was accidentally... · 6c6ce9d0
Rasmus Munk Larsen authored Nov 19, 2024
```
Enable vectorized erf<double>(x) for SSE and AVX, which was accidentally removed in merge request 1750.
```
6c6ce9d0
Simplify and speed up pow() by 5-6% · 5610a13b
Rasmus Munk Larsen authored Nov 20, 2024

5610a13b
optimize setConstant, setZero · 8ad4344c
Charles Schlosser authored Nov 22, 2024

8ad4344c