Status of Python threading support (GIL removal)?

Lie Ryan lie.1296 at gmail.com
Sat Jun 20 19:52:55 CEST 2009


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jean-Paul Calderone wrote:
> On Sat, 20 Jun 2009 00:07:27 GMT, Lie Ryan <lie.1296 at gmail.com> wrote:
>> [snip]
>>
>> Perhaps we should have more built-in/stdlib operations that can release
>> GIL safely to release GIL by default? And perhaps some builtin/stdlib
>> should receive an optional argument that instruct them to release GIL
>> and by passing this argument, you're making a contract that you wouldn't
>> do certain things that would disturb the builtin/stdlib's operations;
>> the specifics of what operations are prohibited would be noted on their
>> docs.
> 
> There would be a lot less useless discussion if people would do a minimal
> amount of checking to see if the ideas they're suggesting are at all
> feasible.
> 
> Why don't you take a look at the CPython source and see if this is actually
> possible?  If you're not comfortable diving into a C program, then maybe
> you
> could try to become so first, or refrain from making suggestions that rely
> on technical details you know nothing about?
> 
> Jean-Paul

There are 4 types of people's attitude towards discussion on GIL:

Two useless ones:
1. seasoned developers who have run out of ideas and moved to other more
important issues, leaving GIL problems to be solved by the future
generation;
2. newbies who spurts out random ideas, often half-baked, impossible,
unfeasible, or just plain stupid;

and two helpful ones:
3. people who are too tired discussing GIL and just shut their mouth up;
4. people who preached resistance against GIL is futile and converted
people of type 2 into type 3.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAko9IfcACgkQqC3FTmXeMUa11wCdFMmvLM6Y3fx8DPcty27XhuVS
eJkAnAup/G/cQkDML0k49a+SlM1ymvCS
=1cRN
-----END PGP SIGNATURE-----



More information about the Python-list mailing list