mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-04 17:08:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 13

gh-125985: Fix cmodule_function() scaling benchmark (#128460)

Browse source 
Add a separate benchmark that measures the effect of
`_PyObject_LookupSpecial()` on scaling.

In the process of cleaning up the scaling benchmarks for inclusion, I
unintentionally changed the "cmodule_function" benchmark to pass an
`int` to `math.floor()` instead of a `float`, which causes it to use the
`_PyObject_LookupSpecial()` code path. `_PyObject_LookupSpecial()` has
its own scaling issues that we want to measure separately from calling a
function on a C module.
This commit is contained in:
Sam Gross 2025-01-03 16:48:47 -05:00 • committed by GitHub
parent b75ed951d4
commit f1574859d7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 11 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

14
Tools/ftscalingbench/ftscalingbench.py
View file

@ -54,8 +54,16 @@ def object_cfunction():
@register_benchmark
def cmodule_function():
for i in range(1000 * WORK_SCALE):
math.floor(i * i)
N = 1000 * WORK_SCALE
for i in range(N):
math.cos(i / N)
@register_benchmark
def object_lookup_special():
# round() uses `_PyObject_LookupSpecial()` internally.
N = 1000 * WORK_SCALE
for i in range(N):
round(i / N)
@register_benchmark
def mult_constant():
@ -206,7 +214,7 @@ def benchmark(func):
color = "\x1b[33m" # yellow
reset_color = "\x1b[0m"
print(f"{color}{func.__name__:<18} {round(factor, 1):>4}x {direction}{reset_color}")
print(f"{color}{func.__name__:<25} {round(factor, 1):>4}x {direction}{reset_color}")
def determine_num_threads_and_affinity():
if sys.platform != "linux":