<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hi Victor,<br>
<br>
great to hear. I think everybody here appreciates your efforts.<br>
<br>
Do you think there will be any change of merging this back into
CPython?<br>
<br>
Best,<br>
Sven<br>
<br>
<div class="moz-cite-prefix">On 04.11.2015 09:50, Victor Stinner
wrote:<br>
</div>
<blockquote
cite="mid:CAMpsgwagj1Jz=Zue_hNYLtsUg7oJEKvo0hB73A3w-XxH1iJi8Q@mail.gmail.com"
type="cite">
<div dir="ltr">
<div>Hi,<br>
<br>
I'm writing a new "FAT Python" project to try to implement
optimizations in CPython (inlining, constant folding, move
invariants out of loops, etc.) using a "static" optimizer (not
a JIT). For the background, see the thread on python-ideas:<br>
<a moz-do-not-send="true"
href="https://mail.python.org/pipermail/python-ideas/2015-October/036908.html"
target="_blank">https://mail.python.org/pipermail/python-ideas/2015-October/036908.html</a><br>
<br>
</div>
See also the documentation:<br>
<a moz-do-not-send="true"
href="https://hg.python.org/sandbox/fatpython/file/tip/FATPYTHON.rst"
target="_blank">https://hg.python.org/sandbox/fatpython/file/tip/FATPYTHON.rst</a><br>
<a moz-do-not-send="true"
href="https://hg.python.org/sandbox/fatpython/file/tip/ASTOPTIMIZER.rst">https://hg.python.org/sandbox/fatpython/file/tip/ASTOPTIMIZER.rst</a><br>
<div>
<br>
I implemented the most basic optimization to test my code:
replace calls to builtin functions (with constant arguments)
with the result. For example, len("abc") is replaced with 3. I
reached the second milestone: it's now possible to run the
full Python test suite with these optimizations enabled. It
confirms that the optimizations don't break the Python
semantic.<br>
<br>
</div>
<div>Example:<br>
---<br>
>>> def func():<br>
... return len("abc")<br>
... <br>
>>> import dis<br>
>>> dis.dis(func)<br>
2 0 LOAD_GLOBAL 0 (len)<br>
3 LOAD_CONST 1 ('abc')<br>
6 CALL_FUNCTION 1 (1 positional, 0
keyword pair)<br>
9 RETURN_VALUE<br>
<br>
>>> len(func.get_specialized())<br>
1<br>
>>> specialized=func.get_specialized()[0]<br>
>>> dis.dis(specialized['code'])<br>
2 0 LOAD_CONST 1 (3)<br>
3 RETURN_VALUE<br>
>>> len(specialized['guards'])<br>
2<br>
<br>
>>> func()<br>
3<br>
<br>
>>> len=lambda obj: "mock"<br>
>>> func()<br>
'mock'<br>
>>> func.get_specialized()<br>
[]<br>
---<br>
<br>
</div>
<div>The function func() has specialized bytecode which returns
directly 3 instead of calling len("abc"). The specialized
bytecode has two guards dictionary keys:
builtins.__dict__['len'] and globals()['len']. If one of these
keys is modified, the specialized bytecode is simply removed
(when the function is called) and the original bytecode is
executed.<br>
</div>
<div><br>
<br>
You cannot expect any speedup at this milestone, it's just to
validate the implementation. You can only get speedup if you
implement *manually* optimizations. See for example
posixpath.isabs() which inlines manually the call to the
_get_sep() function. More optimizations will be implemented in
the third milestone. I don't know yet if I will be able to
implement constant folding, function inlining and/or moving
invariants out of loops.<br>
<br>
<br>
Download, compile and test FAT Python with:<br>
<br>
hg clone <a moz-do-not-send="true"
href="http://hg.python.org/sandbox/fatpython">http://hg.python.org/sandbox/fatpython</a><br>
./configure && make && ./python -m test
test_astoptimizer test_fat<br>
<br>
<br>
</div>
<div>Currently, only 24 functions are specialized in the
standard library. Calling a builtin function with constant
arguments in not common (it was expected, it's only the first
step for my optimizer). But 161 functions are specialized in
tests.<br>
<br>
</div>
<div>
<br>
</div>
<div>To be honest, I had to modify some tests to make them pass
in FAT mode. But most changes are related to the .pyc
filename, or to the exact size in bytes of dictionary objects.<br>
<br>
</div>
<div>FAT Python is still experimental. Currently, the main bug
is that the AST optimizer can optimize a call to a function
which is not the expected builtin function. I already started
to implement code to understand namespaces (detect global and
local variables), but it's not enough yet to detect when a
builtin is overriden. See TODO.rst for known bugs and
limitations.<br>
</div>
<div><br>
Victor<br>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Python-Dev mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Python-Dev@python.org">Python-Dev@python.org</a>
<a class="moz-txt-link-freetext" href="https://mail.python.org/mailman/listinfo/python-dev">https://mail.python.org/mailman/listinfo/python-dev</a>
Unsubscribe: <a class="moz-txt-link-freetext" href="https://mail.python.org/mailman/options/python-dev/srkunze%40mail.de">https://mail.python.org/mailman/options/python-dev/srkunze%40mail.de</a>
</pre>
</blockquote>
<br>
</body>
</html>