Let's Talk about PyPy

1.

Let's Talk About David Beazley @dabeaz

2.

PyPy Project • Perhaps you've heard about PyPy • Python
implemented in Python • It is apparently quite a bit faster • How is that possible? Magic???

3.

It's Not Slow Draw the Mandelbrot set Credit: Jeff Preshing
CPython 2.7: 502s _ = ( 255, lambda V ,B,c :c and Y(V*V+B,B, c -1)if(abs(V)<6)else ( 2+c-4*abs(V)**-0.4)/i ) ;v, x=1500,1000;C=range(v*x );import struct;P=struct.pack;M,\ j ='<QIIHHHH',open('M.bmp','wb').write for X in j('BM'+P(M,v*x*3+26,26,12,v,x,1,24))or C: i ,Y=_;j(P('BBB',*(lambda T:(T*80+T**9 *i-950*T **99,T*70-880*T**18+701* T **9 ,T*i**(1-T**45*2)))(sum( [ Y(0,(A%3/3.+X%v+(X/v+ A/3/3.-x/2)/1j)*2.5 /x -2.7,i)**2 for \ A in C [:9]]) /9) ) )

4.

It's Not Slow Draw the Mandelbrot set Credit: Jeff Preshing
CPython 2.7: 502s _ = ( 255, lambda V ,B,c :c and Y(V*V+B,B, c -1)if(abs(V)<6)else ( 2+c-4*abs(V)**-0.4)/i ) ;v, x=1500,1000;C=range(v*x );import struct;P=struct.pack;M,\ j ='<QIIHHHH',open('M.bmp','wb').write for X in j('BM'+P(M,v*x*3+26,26,12,v,x,1,24))or C: i ,Y=_;j(P('BBB',*(lambda T:(T*80+T**9 *i-950*T **99,T*70-880*T**18+701* T **9 ,T*i**(1-T**45*2)))(sum( [ Y(0,(A%3/3.+X%v+(X/v+ A/3/3.-x/2)/1j)*2.5 /x -2.7,i)**2 for \ A in C [:9]]) /9) ) ) PyPy-1.8: 203s Speedup of ~2.5x

5.

It's Not Slow Draw the Mandelbrot set (non-obfuscated) Python 2.7.2
: 14.5s Python 2.7.2+ctypes : 0.95s PyPy-1.8 : 0.42s Yow! That's 34x faster!

6.

I LIKE IT! "Laziness is the ﬁrst great virtue of
a programmer" -- Larry Wall

7.

CPython PyPy One is clearly faster than the other...

8.

CPython PyPy Just in time compilation Translation to C Optimization
One is clearly faster than the other...

9.

CPython PyPy Just in time compilation Translation to C Optimization
... but performance is not what this talk is about. One is clearly faster than the other...

10.

CPython PyPy • Which one can you adjust with a
pocketknife?

11.

pocketknife? ... in the dark

12.

pocketknife? ... in the dark ... under a pressing deadline

13.

pocketknife? ... in the dark ... under a pressing deadline I speak from some experience...

14.

None

15.

Thinking about Tinkering (with PyPy)

16.

Tinkering Matters! CPython

17.

Tinkering Matters! CPython Patches

18.

Tinkering Matters! CPython Patches PEPs

19.

Tinkering Matters! CPython Patches PEPs Extensions python-ideas

20.

"Oh, Interesting..."

21.

Exploring New Ideas ported to An "afternoon hack," with a
big impact parallel Python

22.

Exploring New Ideas ported to An "afternoon hack," with a
big impact parallel Python ... we didn't choose Python for performance.

23.

Tinkering Creates Cool Stuff

24.

CPython PyPy An honest question • Is PyPy something that
YOU can tinker with? • As in YOU... sitting in this room. • Or is it for "evil genuises only?"

25.

Armin Maciej Alex You?

26.

A Confession • PyPy scares me • It's fast. I
get that. • A lot of moving parts • A lot of advanced computer science inside

27.

Tinker Away! I Worry About Complexity ... Abandon all hope

28.

CPython PyPy An honest question • Is PyPy something that
YOU can tinker with? Honest answer: I don't know

29.

• See if I could teach myself to tinker with
PyPy • From scratch (I'm not a PyPy developer) • Use nothing but the source, online docs, etc. An Experiment:

30.

Constraints A part-time project

31.

Tinkering with PyPy != Using PyPy • If you want
to use it, just run it • It's Python. • Not so interesting (not as much as tinkering) bash % pypy gofaster.py

32.

Tinkering with PyPy != Creating PyPy • submit a useful
bug report (or patch) • Make extensions • Study parts of the implementation (GIL, etc.) • Post messages on pypy-ideas@pypy.org

33.

Where To Start? • Tinkerers use source • You build
it yourself! • You read instructions http://pypy.org Go Download it. Now!

34.

Running py.py • PyPy is written in "Python"... you can
run it bash % python pypy-1.8/pypy/bin/py.py [platform:execute] gcc-4.0 -c -arch x86_64 -O frame-pointer - \ ... PyPy 1.8.0 in StdObjSpace on top of Python 2. (startuptime: 23.23 secs) >>>> • Performance is terrible! • You wouldn't do it except for debugging

35.

Translating PyPy • To get the "real" version, you translate
it • Huh? No makeﬁle? No setup.py? • Already, I'm getting nervous. bash % cd pypy/translator/goal bash % python translate.py -Ojit

36.

Demo

37.

Building PyPy Some Facts: • Movie is @ 64x speed
• Takes a few hours Contrast: Conﬁgure and build CPython-3.2.2 • ./configure; make -j8 • Takes about 90 seconds Immediate Problem: • Finding enough RAM • It takes >4GB 4 cores, 8 GB RAM EC2, m2.xlarge (17GB) What's Actually Happening • Translation of PyPy to C • Creates ~800 C ﬁles • ~10.4 million lines! • 350 Mbytes It might kill the C compiler (or your machine) • Example: gcc-4.2 This is Amazing! • Dare I say "diabolical" • If not intimidating One of the most daunting parts of PyPy • Must redo the process if you make any tweak • An obvious barrier to casual tinkering

38.

RPython • PyPy is actually implemented in "RPython" • RPython
is not an "interpreter", but a restricted subset of the Python language Python rpython • It can run as valid Python code, but that's about the only similarity

39.

RPython • Formal speciﬁcation (in their own words): "RPython is
everything that our translation toolchain can accept"

40.

everything that our translation toolchain can accept" • An analogy "Python is everything that runs without generating a traceback."

41.

everything that our translation toolchain can accept" • An analogy "Python is everything that runs without generating a traceback." • Okay, let's go reading...

42.

Documentation

43.

High-level Docs

44.

Detailed Tech Reports

45.

Detailed Tech Reports To be fair, it was a funded
academic project in PL. (They had to write like this)

46.

Source Code • 454 directories • 5534 ﬁles (4513 .py
source ﬁles) • ~1.25 million non-blank source lines (.py) By The Numbers: It's not so easy to just jump in and make sense of it

47.

Reading Blogs • Recommend start: Andrew Brown • Laurence Tratt
"Fast Enough VMs in Fast Enough Time" "Tutorial: Writing an Interpreter with PyPy" http://bit.ly/fmV2wx http://bit.ly/y8GLqf

48.

Just Do It (Live RPython Coding Demo)

49.

RPython in a Nutshell • RPython is a completely different
language • Python syntax, yes. • Must be compiled (like C, C++, etc.) • Static typing via type inference • Limited set of libraries • If you love Python, you will hate RPython

50.

Type Inference Illustrated def fib(n): if n < 2: return
n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0

51.

n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0 int

52.

n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0 int int

53.

n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0 int int int

54.

n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0 int int int int

55.

n else: return fib(n-1) + fib(n-2) def main(argv): print fib(int(argv[1])) return 0 int int int int int Key Point: Think static typing (like C)

56.

# file1.py def name2(args): statement statement def name3(args): statement statement
statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Now think about the whole program Type inference + restrictions

57.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement There is a single spark... Entry point

58.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Entry point Exploration Begins

59.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Exploration Begins Entry point

60.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Entry point Exploration Begins Whole program type annotation!

61.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement All reachable control- ﬂow paths are followed Entry point Whole program type annotation!

62.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Entry point Imagine this diagram, but with tens of thousands of functions Whole program type annotation!

63.

statement statement def name1(args): statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement Key Insight: Entry point All of the reachable code is RPython

64.

RPython def name1(args): statement statement statement # file1.py def name2(args):
statement statement def name3(args): statement statement statement statement def name1(args): statement statement statement def name3(args): statement statement statement statement # file2.py def name1(args): statement statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement PyPy Source def name1(args): statement statement # file3.py def name2(args): statement statement def name3(args): statement statement def name4(args): statement statement class B(object): def method1(self,args): statement statement statement def method2(self,args): statement statement def name1(args): statement statement def name2(args): statement statement def name4(args): statement statement Translation C Compile pypy-c Entry point

65.

None

66.

Understanding Translation • The translation process will blow your mind
• Full understanding by mortals is probably futile • Snakes + Souls of Ph.D. students inside? • Let's look at a small taste...

67.

A Function def fib(n): if n < 2: return n
else: return fib(n-1) + fib(n-2) Obvious question: How does it translate to C?

68.

Traditional Compiler def fib(n): if n < 2: return n
else: return fib(n-1) + fib(n-2) Lexer Parser IR C You might think it's like a traditional compiler.

69.

Traditional Compiler def fib(n): if n < 2: return n
else: return fib(n-1) + fib(n-2) Lexer Parser IR C You might think it's like a traditional compiler. (and you would be wrong)

70.

Traditional Compiler IR def fib(n): if n < 2: return
n else: return fib(n-1) + fib(n-2) Lexer Parser C Insight: Python already parsed the code! ... so don't do it again. ?????

71.

RPython Translation IR C Translation occurs directly from Python code
objects >>> fib.__code__.co_code '|\x00\x00d\x01\x00k\x00\x00r\x10\x00 d\x02\x00St\x00\x00|\x00\x00d\x02\x00 \x18\x83\x01\x00t\x00\x00|\x00\x00d \x01\x00\x18\x83\x01\x00\x17Sd\x00\x00S'

72.

Bytecode Interpretation CPython • Python has a bytecode interpreter •
Core of the eval loop (written in C). bytecode interpreter runtime

73.

Bytecode Interpretation CPython • It executes the bytecode bytecode interpreter
runtime >>> fib.__code__.co_code '|\x00\x00d\x01\x00k\x00\x00r\x1 d\x02\x00St\x00\x00|\x00\x00d\x \x18\x83\x01\x00t\x00\x00|\x00\ \x01\x00\x18\x83\x01\x00\x17Sd\

74.

Bytecode Interpretation PyPy • PyPy has a bytecode interpreter too
• Written in pure Python (that's the whole idea) bytecode interpreter runtime >>> fib.__code__.co_code '|\x00\x00d\x01\x00k\x00\x00r\x1 d\x02\x00St\x00\x00|\x00\x00d\x \x18\x83\x01\x00t\x00\x00|\x00\ \x01\x00\x18\x83\x01\x00\x17Sd\

75.

Bytecode Interpretation PyPy runtime bytecode interpreter • Bytecode interpreter is
modular • Also used by the translate.py program translate.py bytecode interpreter abstract runtime

76.

Just to be clear... PyPy translates itself using its own
bytecode interpreter

77.

Abstract Interpretation translate.py bytecode interpreter abstract runtime 2 0 LOAD_FAST
0 (n) 3 LOAD_CONST 1 (2) 6 COMPARE_OP 0 (<) 9 POP_JUMP_IF_FALSE 16 3 12 LOAD_CONST 2 (1) 15 RETURN_VALUE 5 >> 16 LOAD_GLOBAL 0 (fib) 19 LOAD_FAST 0 (n) 22 LOAD_CONST 2 (1) 25 BINARY_SUBTRACT 26 CALL_FUNCTION 1 29 LOAD_GLOBAL 0 (fib) 32 LOAD_FAST 0 (n) 35 LOAD_CONST 1 (2) 38 BINARY_SUBTRACT 39 CALL_FUNCTION 1 42 BINARY_ADD 43 RETURN_VALUE 44 LOAD_CONST 0 (None) 47 RETURN_VALUE Translator runs the code "in the abstract"

78.

Abstract Interpretation Translator runs the code "in the abstract" 2
0 LOAD_FAST 0 (n) 3 LOAD_CONST 1 (2) 6 COMPARE_OP 0 (<) 9 POP_JUMP_IF_FALSE 16 3 12 LOAD_CONST 2 (1) 15 RETURN_VALUE 5 >> 16 LOAD_GLOBAL 0 (fib) 19 LOAD_FAST 0 (n) 22 LOAD_CONST 2 (1) 25 BINARY_SUBTRACT 26 CALL_FUNCTION 1 29 LOAD_GLOBAL 0 (fib) 32 LOAD_FAST 0 (n) 35 LOAD_CONST 1 (2) 38 BINARY_SUBTRACT 39 CALL_FUNCTION 1 42 BINARY_ADD 43 RETURN_VALUE 44 LOAD_CONST 0 (None) 47 RETURN_VALUE 2 0 LOAD_FAST 0 (n) 3 LOAD_CONST 1 (2) 6 COMPARE_OP 0 (<) 9 POP_JUMP_IF_FALSE 16 3 12 LOAD_CONST 2 (1) 15 RETURN_VALUE 5 >> 16 LOAD_GLOBAL 0 (fib) 19 LOAD_FAST 0 (n) 22 LOAD_CONST 2 (1) 25 BINARY_SUBTRACT 26 CALL_FUNCTION 1 29 LOAD_GLOBAL 0 (fib) 32 LOAD_FAST 0 (n) 35 LOAD_CONST 1 (2) 38 BINARY_SUBTRACT 39 CALL_FUNCTION 1 42 BINARY_ADD 43 RETURN_VALUE 44 LOAD_CONST 0 (None) 47 RETURN_VALUE Translator runs the code "in the abstract" translate.py bytecode interpreter abstract runtime

79.

0 LOAD_FAST 0 (n) 3 LOAD_CONST 1 (2) 6 COMPARE_OP 0 (<) 9 POP_JUMP_IF_FALSE 16 3 12 LOAD_CONST 2 (1) 15 RETURN_VALUE 5 >> 16 LOAD_GLOBAL 0 (fib) 19 LOAD_FAST 0 (n) 22 LOAD_CONST 2 (1) 25 BINARY_SUBTRACT 26 CALL_FUNCTION 1 29 LOAD_GLOBAL 0 (fib) 32 LOAD_FAST 0 (n) 35 LOAD_CONST 1 (2) 38 BINARY_SUBTRACT 39 CALL_FUNCTION 1 42 BINARY_ADD 43 RETURN_VALUE 44 LOAD_CONST 0 (None) 47 RETURN_VALUE translate.py bytecode interpreter abstract runtime

80.


81.


82.


83.


84.


85.


86.

Abstract Interpretation A Control ﬂow graph is constructed translate.py bytecode
interpreter abstract runtime

87.

Abstract Interpretation C Code The full details are "hairy"

88.

"You are in a maze of twisty little passages, all
alike." (and a huge green ﬁerce snake bars the way)

89.

Understanding the Source • Two different languages co-exist (same syntax)
# file1.py def name2(args): statement statement def name3(args): statement statement statement statement def name1(args): statement statement statement Full Python???? RPython???? Which is it?

90.

Understanding the Source • Two different languages co-exist (same syntax)
# file1.py def name2(args): statement statement def name3(args): statement statement statement statement def name1(args): statement statement statement Full Python???? RPython???? Which is it? (You can't look in isolation)

91.

Understanding the Source def cast_object_to_ptr(PTR, object): """NOT_RPYTHON: hack. The object
may Limited to casting a given object to """ if isinstance(PTR, lltype.Ptr): TO = PTR.TO else: TO = PTR ... • PyPy uses doc strings to help you sort it out • It is enforced by the translator (an assertion)

92.

Understanding the Source • Deeper question: Why would you have
mixed code? # file1.py def name2(args): statement statement def name3(args): statement statement statement statement def name1(args): statement statement statement RPython Python • Head throbbing...

93.

Execution Contexts # file1.py def name2(args): statement statement def name3(args):
statement statement statement statement def name1(args): statement statement statement Translation (Python) Executable (C)

94.

Execution Contexts # file1.py def name2(args): statement statement def name3(args):
statement statement statement statement def name1(args): statement statement statement Translation (Python) Executable (C) • At translation, the code separates Metaprogramming Implementation • decorators • metaclasses • exec()

95.

Example def decorator(func): statements ... def wrapper(*args,**kwargs): statements ... return
func(*args,**kwargs) return wrapper @decorator def func(args): statements ...

96.

Example def decorator(func): statements ... def wrapper(*args,**kwargs): statements ... return
func(*args,**kwargs) return wrapper @decorator def func(args): statements ... Python RPython Python RPython

97.

Rules of Thumb • Code that executes at import time
can make use of all Python features • Code reachable through the entry point (target) is RPython • Keeping it straight is hard (for me anyways)

98.

But Wait, There's More!

99.

Foreign Code • PyPy is written in "Python", but can
access external C code and libraries • os, math, time, threads, etc. • There is a highly developed FFI mechanism • Plus a conﬁguration system (think autoconf)

100.

Example (Accessing Foreign Functions)

101.

A Quandary • How do I end this talk? •
I've only talked about RPython • When do we get to the PyPy?

102.

A Realization I still don't know know how PyPy works!
Score: PyPy: 1 Dave: 0

103.

A Deeper Realization I don't even know how CPython works!

104.

WAT!?! WAT!?!

105.

A Clariﬁcation I do know how to use the tools
that make CPython

106.

A Clariﬁcation I do know how to use the tools
that make CPython • ANSI C • Makeﬁles • Algorithms • Data Structures

107.

The Challenge PyPy has a different set of tools •
RPython • translate.py • Metaprogramming • Foreign Functions

108.

So how to end this talk?

109.

Wait! I used to be an evil professor!

110.

Figuring out how PyPy works is left as an exercise!
(You'll learn a lot)

111.

Postscript

112.

Postscript Let's talk about Ruby! threads.each { |aThread| aThread.join }

113.

Breaking GILs • As you know, I like breaking GILs
• You know, global interpreter locks • As in threads and stuff... • I love it!

114.

A Benchmark • Message-passing with a CPU-bound thread C :
1.11s Python 2.7 : 1.60s Ruby 1.9 : 5839.4s • Don't concern yourself with the details • Ruby 3600x slower than Python? • What's that all about? Let's go tinker!

115.

Tinkering with Ruby • It was pretty straightforward • Finding
the GIL didn't take long • An afternoon of ﬁddling around (Search for my talk at RuPy 2011) • Caused by a more extreme case of the thread priority inversion that's in Python 3.3

116.

Just to be clear... I couldn't write a real Ruby
program to save my life right now.

117.

A PyPy Benchmark • A similar message-passing benchmark Python 2.7
: 15.6s PyPy-1.6 : 6689.2s (428x slower) • Huh? What's that all about? • No idea! Or even how to look. • That is the whole reason for this talk

118.

Parting Words • Can you tinker with PyPy? • Honest
answer: I still don't know • Should you try to go tinker with it anyways? • YES! • You will ﬁnd interesting things inside

119.

"My God, it's full of stars!"

120.

"My God, it's full of stars!" (or VMs?)

121.

Thanks! • Hope you learned at least one new thing
• Special thanks: • Alex Gaynor • Maciej Fijalkowski • Chipy • Twitter: @dabeaz

Let's Talk about PyPy

Let's Talk about PyPy

Want more?

Transcript