On Tue, 8 Feb 2011, Nicolas Cellier wrote:
Just a detail, Python range(a,b) are semi open [a,b) think like C for(i=0;i<n; i++)
In the Smalltalk proc8:withwith:with:with:: the range was interpreted as [a,b]. This Smalltalk (and Cog) unfairly perform too much work.
And there's another one. In #proc0:block: the loops are (1 to: loops) do: instead of 1 to: loops do:.
Levente
Nicolas
2011/2/8 stephane ducasse stephane.ducasse@gmail.com:
This is really interesting to see that type inferencing can really help. This is definitively a topic I would like to explore in the coming years.
Stef
Hi folks!
Since we are on the verge of 4.2, and we have a brand new Cog VM to play with I felt like dusting off my old Pystone port to Squeak - Sqystone, which I wrote back in 2004.
At that time Squeak was around 5 times faster than CPython. How do we stack up today? Yeah, I know - hardly a good benchmark, they all lie etc etc. :)
I am using Ubuntu 10.10 on a corei7, so this is running on a 64 bit CPU.
Cog!
Squeak 4.2-10966 (soon to be released) + latest Cog r2361 (binary download): Pystone(1.1) time for 50000 passes = 0.06 This machine benchmarks at 833333.3 pystones/second
NOTE: AFAICT running with more passes does not improve it. Also, not sure if I could get more out of this if I built from source on my box.
Regular Squeak
Squeak 4.2-10966 (soon to be released) + regular Squeak VM 4.4.7-2357 (built from src): Pystone(1.1) time for 50000 passes = 0.503 This machine benchmarks at 99403.6 pystones/second
Regular CPython
CPython 3.1.2 (newest in Ubuntu Meerkat, minimal): gokr@quigon:/usr/lib/python3.1/test$ python3.1 pystone.py Pystone(1.1) time for 50000 passes = 0.57 This machine benchmarks at 87719.3 pystones/second
NOTE: 3.2 is reportedly a teeny bit faster. Also not built from source.
Pypy 1.4
wget http://pypy.org/download/pypy-1.4.1-linux64.tar.bz2 gokr@quigon:~/python/pypy-1.4.1-linux64$ ./bin/pypy ./lib-python/2.5.2/test/pystone.py Pystone(1.1) time for 50000 passes = 0.15 This machine benchmarks at 333333 pystones/second gokr@quigon:~/python/pypy-1.4.1-linux64$ ./bin/pypy ./lib-python/2.5.2/test/pystone.py 5000000 Pystone(1.1) time for 5000000 passes = 4.8 This machine benchmarks at 1.04167e+06 pystones/second
NOTE: Also not built from source. Here we run pystone a second time with 100x more loops and get a substantially better number.
Shedskin 0.7
sudo apt-get install g++ libpcre3-dev libgc-dev python-dev sudo dpkg -i shedskin_0.7_all.deb wget http://shedskin.googlecode.com/files/shedskin-examples-0.7.tgz shedskin pystone.py gokr@quigon:~/python/shedskin-examples-0.7$ make g++ -O2 -march=native -fomit-frame-pointer -Wno-deprecated -I. -I/usr/share/shedskin/lib /usr/share/shedskin/lib/builtin.cpp pystone.cpp /usr/share/shedskin/lib/time.cpp /usr/share/shedskin/lib/re.cpp -lgc -lpcre -o pystone gokr@quigon:~/python/gokr@quigon:~/python/shedskin-examples-0.7$ ls -la pystone* -rwxr-xr-x 1 gokr gokr 297329 2011-02-07 23:01 pystone -rw-r--r-- 1 gokr gokr 9193 2011-02-07 23:00 pystone.cpp -rw-r--r-- 1 gokr gokr 1893 2011-02-07 23:00 pystone.hpp -rw-r--r-- 1 gokr gokr 5774 2010-12-11 11:40 pystone.py gokr@quigon:~/python/shedskin-examples-0.7$ ./pystone This machine benchmarks at 2500000.000000 pystones/second
NOTE: I am wondering a bit about this. It tells the same whatever loops I give it... But ok, perhaps it is all fine.
Summary
The regular Squeak VM has not been overrun by CPython in these 6 years time. When I wrote Sqystone Squeak was 5x faster IIRC (can't find the post anymore). Now they are equal more or less, Squeak still a teeny bit faster.
Cog is brutally fast on this one. Compared to CPython and regular Squeak almost 10x faster.
Pypy is about 20% faster than Cog if given enough time to actually start jitting. Cool for the Pypy project! And cool that they aren't that much faster than Cog. :)
Shedskin is the "state of the art" of statically compiling Python via C++ using type inferencing etc etc - so I hear. It is said to be faster than Cython and Psyco. It ends up beating Cog, but "only" by a factor of 3x. I say "only" because that seems pretty good to me given that Cog is a JIT and still pretty young and that Shedskin can only run a subset of Python.
regards, Göran