Diving into byte code optimization in python

Diving into Byte-code
optimization in Python

SciPy India,
IIT Bombay Dec 05th 2011
Chetan Giridhar and Vishal Kanaujia

Fundamentals of Bytecode
• Python source code compiled into Python
byte code by the CPython interpreter
• “.pyc”?
– Automatic compilation : importing a module
– Explicit compilation :
py_compile.compile(“module.py”) – generates
‘module.pyc’
• The module ‘compileall’{}

Attribution-NonCommercial CC BY-NC

Fundamentals | more
• A program doesn't run any faster when it is
read from a ‘.pyc’ file. But, why?
• “.pyc” for a script executed on the command
line
• “.pyc” files – good enough to distribute your
code, but with a caveat!


Compilation phases
• Uses a ‘lexer’ to prepare tokens of Python
code
• A parser arranges token according to the
language grammar and prepares concrete
syntax tree
• Concrete syntax tree is transformed in to AST
• AST is compiled to produce Byte-codes


Python Abstract
Parse Tree
source Syntax
generated
code Tree
Pgen.c
ast.c

compile.c

Executed
Optimized Bytecode
by Python
bytecode generated
VM
ceval.c peephole.c

Python “ast” module
$ cat myast.py $python myast.py
<generator object walk at 0xb784c8ec>
import ast
Module(body=[Expr(value=BinOp
nod = ast.parse('a +2') (left=Name(id='a', ctx=Load()),
print ast.walk(nod) op=Add(),
right=Num(n=2)))])
print ast.dump(nod)

•Convenient for analysis, code transformations and generation
•ASTs are compiled to code objects


A peek into Bytecodes
$ cat scipy.py $ python scipy.py
1 import dis 4 0 LOAD_CONST 1 (10)
3 STORE_FAST 0 (i)
2
3 def foo(): 5 6 LOAD_FAST 0 (i)
9 PRINT_ITEM
4 i = 10 10 PRINT_NEWLINE
5 print i 11 LOAD_CONST 0 (None)
14 RETURN_VALUE
6
7 print dis.dis(foo)

• Bytecode stream: An opcode mix
• Defined in “Python-2.7.2/Include/opcode.h”

Python VM
• Engine to execute Bytecodes
• CPython VM is a C implementation
• Stack based process VM
– PUSH/ POP operations
• Implementation of Python VM?


Python VM: Implementation
Python/ceval.c
--> PyEval_EvalFrameEx()
for(; ;) {
/* Extract opcode and argument */
opcode = NEXTOP();
if (HAS_ARG(opcode))
oparg = NEXTARG();
switch(opcode) {
case LOAD_CONST:
……
}
}

Optimizations

Tools
Getting in to the Problem Space!!


Pyrex
• Python like language to create C module for
Python
• Create your “pyx” files and compile them in “.c”
files
• Import them as modules in your applications
• Pyrex used as:
– speed up the execution of Python code
– Python interface to existing C modules/libraries
• Lot of work for developer
– .py to .pyx?
– thinking in C


Psyco
• An intelligent option – JIT compilation
• Profiles dynamically an application for hot-
spots
• “in-memory” prepares C extension and hook
them appropriately
• Solves “duck” typing
• Memory footprint?
• Support till CPython 2.5


Psyco| Intelligent use
Iterations Without With
Pysco(ms) Pysco(ms)

1000 125 151

100000 12900 12570

• from psyco.classes import *
• pysco.bind(func)


Optimizations

Bytecode level


Why optimize Bytecode?
def foo():
• Python optimization i=0
i =i+1
are ineffective, print i

sometime
4 0 LOAD_CONST 1 (0)
• Duck typing 3 STORE_FAST 0 (i)

5 6 LOAD_FAST 0 (i)
– Run-time types 9 LOAD_CONST 2 (1)
12 BINARY_ADD
– Optimizer misses 13 STORE_FAST 0 (i)

many opportunities 6 16 LOAD_FAST 0 (i)
19 PRINT_ITEM
– TCO 20 PRINT_NEWLINE
21 LOAD_CONST 0 (None)


Optimizations: Tail Recursive Calls
$ cat runbytecode3.py $ python runbytecode3.py

import dis 5 1 LOAD_CONST 10
2 STORE_FAST x

6 4 LOAD_FAST x
def foo(): 5 LOAD_CONST 10
x = 10 6 COMPARE_OP ==
7 POP_JUMP_IF_FALSE to 17
if x == 10:
7 9 LOAD_CONST 0
x=0 10 STORE_FAST x
foo()
8 12 LOAD_GLOBAL foo
13 CALL_FUNCTION 0
14 POP_TOP
print dis.dis(foo) 15 JUMP_FORWARD to 17
>> 17 LOAD_CONST None
18 RETURN_VALUE
None


BytePlay: Do it yourself!
• Experiment with Bytecodes
• Generates, recompiles and runs code on-the-
fly
• Very useful to evaluate different code
optimizations
• Example


Playing with Bytecode
$cat runbytecode2.py $ python runbytecode2.py
reassembled ‘byteplay.py' imported.
from byteplay import *
from pprint import pprint 5 1 LOAD_CONST 10 <<<-------------------------
2 STORE_FAST x
def foo():
x = 10 6 4 LOAD_FAST x
5 PRINT_ITEM
print x
6 PRINT_NEWLINE
7 LOAD_CONST None
c = Code.from_code(foo.func_code) 8 RETURN_VALUE
print printcodelist(c.code) None

# Modify the byte code 5 1 LOAD_CONST 1000 <<<---------------------------
c.code[0] = (LOAD_CONST, 1000) 2 STORE_FAST x

foo.func_code = c.to_code() 6 4 LOAD_FAST x
5 PRINT_ITEM
print printcodelist(c.code) 6 PRINT_NEWLINE
7 LOAD_CONST None
8 RETURN_VALUE
# Call the modified function
None
foo()
1000


Going forward

PyPy


Does PyPy help?
• Python interpreter in Python itself
• A target function needs to be written with
subset of Python (Restricted Python)
• PyPy can translate this function to runtime of
your choice ☺
– Options: C/POSIX, CLI/.NET and Java/JVM
• Seamless and transparent


PyPy | more
from pypy.translator.interactive import @compdec
Translation
def fact(n):
class compdec:
if 0 == n:
def __init__(self, func): return 1
self.func = func return n * fact(n -1)
self.argtypes = None fact(10)

def __call__(self, *args):
argtypes = tuple(type(arg) for arg in
args)
if argtypes != self.argtypes:
self.argtypes = argtypes
t = Translation(self.func)
t.annotate(argtypes)
self.cfunc = t.compile_c() What PyPy does?
•The translation of the RPython function to C.
•Invoking the C compiler to create a C extension module.
return self.cfunc(*args)
•Importing the compiled function back into the Python interpreter.


Recommendations
• Always profile your applications
• 90/10 rule of performance gain
• Performance critical portion could be written
as C extensions
– Wrappers like SWIG could be used to bridge
Python/C applications
– Pyrex, Psyco, and PyPy
– Tuning Bytecodes manually
• Evaluate and use ☺


References
• http://docs.python.org/release/2.5.2/lib/module-dis.html
• http://www.cosc.canterbury.ac.nz/greg.ewing/python/Pyre
x/
• http://www.dalkescientific.com/writings/diary/archive/201
0/02/22/instrumenting_the_ast.html
• http://cs.utsa.edu/~danlo/teaching/cs4713/lecture/node7.
html
• http://www.devshed.com/c/a/Python/How-Python-Runs-
Programs/4/
• http://www.enthought.com/~ischnell/paper.html
• http://codespeak.net/pypy/dist/pypy/doc/translation.html
• http://effbot.org/pyref/type-code.htm


Questions

Thank you for your time and attention ☺

• Please share your feedback/ comments/ suggestions to us at:
• cjgiridhar@gmail.com , http://technobeans.com
• vishalkanaujia@gmail.com, http://freethreads.wordpress.com


Backup slides


Bytecodes | more
• Atomic unit of execution (thread safe)
• A Python module is compiled and saved in
form of a “pyc” file
• An optimization; avoid compilation phase for
future uses of a module
• If source file is changed, pyc is recompiled
• “pyc” file is coded in bytecode!


Code Objects
• Bytecode representation
• Immutable objects (No ref to mutable objects)
• Function object (Code objects that reference
global variables)
• “co_code” = Bytecodes


Benefits of using Bytecode
• Architecture neutral code
• Portability
• Simple optimizations


Diving into byte code optimization in python

Recommended

More Related Content

What's hot (20)

Similar to Diving into byte code optimization in python (20)

More from Chetan Giridhar (9)

Recently uploaded (20)

Diving into byte code optimization in python