mycpp Code Generation

Measure the speedup from mycpp, and the resource usage.

Source code: oil/mycpp/examples

User Time (milliseconds)

Lower ratios are better.

example name gen C++ Python C++ : Python
fib_iter gen 8 900 0.008
modules gen 2 179 0.010
fib_recursive gen 11 896 0.012
loops gen 4 290 0.013
asdl_generated gen 7 364 0.020
parse gen 25 771 0.033
scoped_resource gen 39 1,045 0.038
files gen 4 68 0.057
containers gen 8 119 0.065
tuple_return_value gen 16 188 0.086
length gen 41 207 0.197
cartesian gen 82 329 0.250
classes gen 3 12 0.283
escape gen 100 352 0.283
cgi gen 247 516 0.477
varargs gen 12 12 1.037
control_flow gen 207 114 1.826
gc_stack_roots gen 2 0 inf

Max Resident Set Size (MB)

Lower ratios are better. We use MB (powers of 10), not MiB (powers of 2).

example name gen C++ Python C++ : Python
classes gen 4.5 10.7 0.41
fib_recursive gen 3.4 6.9 0.49
gc_stack_roots gen 3.4 6.9 0.49
modules gen 3.5 7.1 0.50
tuple_return_value gen 3.5 6.9 0.51
scoped_resource gen 3.7 7.1 0.52
parse gen 3.9 7.5 0.53
asdl_generated gen 3.7 6.9 0.53
cartesian gen 3.7 6.9 0.53
cgi gen 3.7 6.9 0.53
length gen 3.7 6.9 0.53
loops gen 3.7 6.9 0.53
control_flow gen 3.8 7.1 0.54
escape gen 3.7 6.8 0.54
fib_iter gen 3.8 6.9 0.55
files gen 3.8 6.9 0.55
containers gen 28.5 47.8 0.59
varargs gen 5.5 6.9 0.79

System Time (milliseconds)

Lower ratios are better.

example name gen C++ Python C++ : Python
classes gen 0 19 0.000
fib_iter gen 0 4 0.000
fib_recursive gen 0 4 0.000
gc_stack_roots gen 0 12 0.000
loops gen 0 4 0.000
modules gen 0 4 0.000
asdl_generated gen 4 16 0.228
parse gen 4 12 0.304
files gen 4 8 0.483
containers gen 8 16 0.490
varargs gen 54 60 0.898
cartesian gen 8 8 0.990
escape gen 8 8 0.998
tuple_return_value gen 4 4 1.013
cgi gen 8 0 inf
length gen 4 0 inf
scoped_resource gen 8 0 inf
control_flow gen 0 0 NA

raw benchmark files