Performance regression in timeseries.SortIndex.time_sort_index

[TomAugspurger]

import pandas as pd
import numpy as np

N = 10 ** 5
idx = pd.date_range(start="1/1/2000", periods=N, freq="s")
s = pd.Series(np.random.randn(N), index=idx)
%timeit s.sort_index()

# 1.0.2
108 µs ± 8.27 µs per loop (mean ± std. dev. of 7 runs, 10000 loops each)
# master
225 µs ± 8.36 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

According to https://pandas.pydata.org/speed/pandas/index.html#timeseries.SortIndex.time_sort_index?p-monotonic=True&commits=f683473a156f032a64a1d7edcebde21c42a8702d-085860a49f3a87aa4e24b3115b50b85c4b3c5676, the first slow commit is #33755, which just bumps Cython in numpydev... So probably not actually that commit.