Smoke: integer addition

Sum 1..N in a tight loop, the smallest workload in the suite. It exists to surface interpreter dispatch overhead at the smallest possible payload.

Runtime Compile time

Runtime · median per inner-loop window

median of 7 runs

Native Gocompiled

25.1 µs0.05×

Pipitbytecode VM

465 µsbaseline

CPython 3.14bytecode VM

1.54 ms3.31×

PyPy 7.3tracing JIT

509 µs1.09×

Ttengobytecode VM

3.03 ms6.50×

Sscriggobytecode VM

2.61 ms5.61×

Mmvmbytecode VM

1.01 ms2.16×

YyaegiAST walker

1.60 ms3.44×

Full statistics

Runner	N	Compile	Runtime	P95	Stddev	RSS	vs pipit	Status
Native Gocompiled	7	108 ms	25.1 µs	25.7 µs	270 ns	71 MiB	0.05×	OK
Pipitbytecode VM	7	422 µs	465 µs	541 µs	37.0 µs	54 MiB	1.00×	OK
CPython 3.14bytecode VM	7	226 µs	1.54 ms	1.65 ms	59.9 µs	n/a	3.31×	OK
PyPy 7.3tracing JIT	7	172 µs	509 µs	525 µs	6.85 µs	n/a	1.09×	OK
tengobytecode VM	7	110 µs	3.03 ms	3.68 ms	228 µs	59 MiB	6.50×	OK
scriggobytecode VM	7	85.8 µs	2.61 ms	2.66 ms	28.3 µs	59 MiB	5.61×	OK
mvmbytecode VM	7	166 µs	1.01 ms	1.76 ms	291 µs	59 MiB	2.16×	OK
yaegiAST walker	7	167 µs	1.60 ms	2.21 ms	349 µs	59 MiB	3.44×	OK

Workload & symmetry rules

Workload

Sum 1..1,000,000 in a single tight loop, print the result mod 2^32.

Why this benchmark exists

It's the simplest possible interpreter workload. If dispatch dominates anything, it dominates this. End-to-end mode is therefore mostly startup time; inner-loop mode is mostly add+jump throughput.

Symmetry rules

One loop, two locals, no function calls in the hot path.
No sum() / numpy / strconv shortcuts.

Source code

pipit / Go
pipit_source.go
native Go
native_main.go
CPython / PyPy
cpython.py
tengo
script.tengo