rguillebert

9 ms·

31.
by rguillebert 9y ago
Then just use the version of PyPy with a GIL?
32.
by rguillebert 9y ago
C extensions will still run under the GIL
33.
by rguillebert 9y ago
It's not the PyPy developers' job to make every Python library threadsafe, people writing libraries will have to make their code threadsafe, like in every other language.
34.
by rguillebert 9y ago
I used to work in tech in France (now I'm in the UK) and I had to pee in a cup, so YMMV I guess.
35.
by rguillebert 10y ago
how so?
36.
by rguillebert 11y ago
This is reference counting too
37.
by rguillebert 12y ago
It says in the blog post that it's 1% faster than CPython (which means that it's as fast as CPython really) on their benchmarks, while PyPy is much faster than that on the same benchmarks. Check speed.pypy.org and speed.pyston.org
38.
by rguillebert 12y ago
Having just one implementation is hurting the ecosystem by restricting the amount of things you can do with Python
39.
by rguillebert 12y ago
PyPy optimizes that out, you have a thing that looks exactly like a list, but is implemented differently under the hood
40.
by rguillebert 12y ago
your days must be really short, it takes an hour on my machine
41.
PyPy-STM: first “interesting” release(morepypy.blogspot.com)
242 points by rguillebert 12y ago|19 comments
42.
by rguillebert 12y ago
you need something to allow concurrent access to internal interpreter data structures...
43.
by rguillebert 12y ago
I think you're talking about the c extension api.
44.
by rguillebert 12y ago
Another approach should be used for those, but to be clear I wasn't talking about the entire numpy library, I'm talking about things like numpy.fft
45.
by rguillebert 12y ago
What would you replace it with ?
46.
by rguillebert 12y ago
Jython is usually slower than CPython I believe, it has no GIL though.
47.
by rguillebert 12y ago
Help us make your code faster, report it please :)
48.
by rguillebert 12y ago
Well, I'm not talking about every C extension in the world, I'm talking about thin bindings around C routines which is what numpy has for fft for example, someone wrote a basic equivalent for numpypy in a few hours with no prior n
49.
by rguillebert 12y ago
yes
50.
by rguillebert 12y ago
You can use these C and Fortran routines on PyPy, just not with the CPython C extension API.
51.
by rguillebert 12y ago
it looks like a static site...
52.
PyPy STM blog post : STMGC-C7 with PyPy(morepypy.blogspot.com)
2 points by rguillebert 13y ago|0 comments
53.
by rguillebert 13y ago
And it's the marginal tax rate (beyond 1 million €).
54.
by rguillebert 13y ago
Since when does french-bashing need to be factually correct ?
55.
by rguillebert 13y ago
That's CPython specific though, PyPy doesn't need it.
56.
by rguillebert 13y ago
PyPy does the same thing as using __slots__ on CPython automatically, no need to use __slots__ to take advantage of the JIT.
57.
by rguillebert 13y ago
This tax doesn't even exist, it has been rejected by the constitutional council (the French supreme court)
58.
by rguillebert 13y ago
I think it's in the norm of French salaries (or I don't know good companies)
59.
by rguillebert 13y ago
Because CPython has no JIT :)
60.
PyPy Status Blog: PyPy 2.0 alpha for ARM(morepypy.blogspot.fr)
6 points by rguillebert 13y ago|0 comments

More ›