total_time in .pymon vs terminal output

[martinsvat]

Describe the bug
Dear, I ran into an inconsistency in results. Actually, the run-time in terminal's output shows different values than total-time in .pymon.

To Reproduce

import time
import pytest


def wrapper_function(foo: int) -> None:
    accumulator = []
    for _ in range(1000):
        accumulator = accumulator + [_]


class TestClass:
    @staticmethod
    def setup_stuff(foo: int) -> int:
        if foo == 1:
            time.sleep(10)
        return foo

    @pytest.mark.parametrize("foo", [1])
    @pytest.mark.parametrize("rounds, iterations", [(1, 1), (1, 10), (10, 1), (10, 10)])
    def test(self, benchmark, foo: int, rounds: int, iterations: int) -> None:
        TestClass.setup_stuff(foo)
        benchmark.pedantic(
            wrapper_function,
            kwargs=dict(foo=foo),
            iterations=iterations,
            rounds=rounds,
            warmup_rounds=0,
        )

executed with pytest mwe.py -v
output

============================= test session starts ==============================
platform linux -- Python 3.12.8, pytest-8.3.3, pluggy-1.5.0 -- /home/.....
cachedir: .pytest_cache
benchmark: 5.1.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: /home/azureuser/projects/bigplan-algorithm
plugins: monitor-1.6.1, benchmark-5.1.0, xdist-3.6.1
collecting ... collected 4 items

mwe.py::TestClass::test[1-1-1] PASSED                                    [ 25%]
mwe.py::TestClass::test[1-10-1] PASSED                                   [ 50%]
mwe.py::TestClass::test[10-1-1] PASSED                                   [ 75%]
mwe.py::TestClass::test[10-10-1] PASSED                                  [100%]


------------------------------------------------------------------------------------------ benchmark: 4 tests -----------------------------------------------------------------------------------------
Name (time in us)            Min                   Max                  Mean              StdDev                Median                IQR            Outliers         OPS            Rounds  Iterations
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
test[10-1-1]            849.1300 (1.0)      1,331.9500 (1.45)       905.0879 (1.05)     150.1471 (inf)        858.7840 (1.0)      14.8370 (inf)           1;1  1,104.8651 (0.96)         10           1
test[10-10-1]           852.9680 (1.00)       937.6166 (1.02)       865.4228 (1.0)       25.4941 (inf)        858.9587 (1.00)      5.1887 (inf)           1;1  1,155.5045 (1.0)          10          10
test[1-10-1]            919.9207 (1.08)       919.9207 (1.0)        919.9207 (1.06)       0.0000 (1.0)        919.9207 (1.07)      0.0000 (1.0)           0;0  1,087.0502 (0.94)          1          10
test[1-1-1]           1,395.1670 (1.64)     1,395.1670 (1.52)     1,395.1670 (1.61)       0.0000 (1.0)      1,395.1670 (1.62)      0.0000 (1.0)           0;0    716.7601 (0.62)          1           1
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Legend:
  Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile.
  OPS: Operations Per Second, computed as 1 / Mean
============================== 4 passed in 40.21s ==============================

Expected behavior
Now, when I open .pymon file and find that test, I will end up with four runs, each one taking little bit over 10
seconds, i.e.

item_variant           total_time
test[1, 1, 1]             10.005931615829468
test[1, 10, 1]           10.01454782485962
test[10, 1, 1]           10.013768434524536
test[10, 10, 1]         10.092158317565918

I wonder why the database stores the whole run-time of the tested method (given parameters input). Pytest measures each single round, iteration, etc., providing expected time of a single benchmark.pednatic(...) call. However, total_time incorporates even setup time, i.e. TestClass::setup_stuff.

If this is a feature, can I somehow get run-time per benchmark.pedantic(...)? Sure, I can divide total_time by rounds * iteration but that would not be correct as setup time would be divided as well.

While this is a trivial MWE with a trivial example, my real use-case is dealing with much bigger data. Let's imagine that I'm passing path to a huge file. In the setup phase of a test method, I load this file. Passing a loaded file directly via parametrization/fixtures is possible in theory; however, then, I would need to load all huge files to memory, keep them as parameters, and then use in the test. (Besides memory / cost requirements, it could force some memory processes on background, e.g. swaps).

Desktop (please complete the following information):

linux 
Python 3.12.8
pytest-8.3.3
monitor-1.6.1
benchmark-5.1.0
xdist-3.6.1

Additional context
If this is a feature with expected behavior, please refer me to the manual.