run the benchmark directly, maybe codspeed will handle benchmarking.

tests/test_performance.py

@@ -2,6 +2,7 @@
 import time
 import timeit
 
+import pytest
 
 from uncertainties import ufloat
 
@@ -38,7 +39,7 @@ def test_complexity():
         assert 0.9 * log10(n / n0) < log10(t / t0) < 1.1 * log10(n / n0)
 
 
-def test_speed(benchmark):
-    num_list = (10, 100, 1000, 10000, 100000)
-    for num in num_list:
-        benchmark(lambda: time_ufloat_sum_benchmark(num))
+@pytest.mark.parametrize("num", (10, 100, 1000, 10000, 100000))
+@pytest.mark.benchmark
+def test_speed(num):
+    ufloat_sum_benchmark(num)