Re: [Python-Dev] cpython: Force 32bit Keccak implementation on SPARC. It look like the Solaris CC

2012-10-14 Thread Antoine Pitrou
On Sun, 14 Oct 2012 02:53:51 +0200 (CEST) christian.heimes python-check...@python.org wrote: http://hg.python.org/cpython/rev/b397b0c08f69 changeset: 79717:b397b0c08f69 parent: 79715:3465ef48d0f5 user:Christian Heimes christ...@cheimes.de date:Sun Oct 14 02:52:01 2012

Re: [Python-Dev] cpython: Force 32bit Keccak implementation on SPARC. It look like the Solaris CC

2012-10-14 Thread martin
Zitat von Antoine Pitrou solip...@pitrou.net: I think SPARC (the architecture) disallows unaligned accesses to memory. I don't think it is specific to the Solaris CC compiler. That's correct. Unaligned memory operations need to use memcpy. Regards, Martin

Re: [Python-Dev] cpython: Force 32bit Keccak implementation on SPARC. It look like the Solaris CC

2012-10-14 Thread Christian Heimes
Am 14.10.2012 12:36, schrieb Antoine Pitrou: I think SPARC (the architecture) disallows unaligned accesses to memory. I don't think it is specific to the Solaris CC compiler. I've updated the comment accordingly. Unaligned access is just used by the opt64 implementation of Keccak. opt32 is more

[Python-Dev] Issues Pending Review

2012-10-14 Thread Michele Orrù
Hello, I have some issues pending 'patch review' hanging for more than two weeks, could somebody please check them out? HTTPServer does not correctly handle bad headers / http://bugs.python.org/issue16083 rlcompleter adds builtins when custom dict is used / http://bugs.python.org/issue5256 a -= b

Re: [Python-Dev] Issues Pending Review

2012-10-14 Thread martin
Zitat von Michele Orrù maker...@gmail.com: Hello, I have some issues pending 'patch review' hanging for more than two weeks, could somebody please check them out? In case nobody picks it up, my 5-for-1 offer still stands: if you review five issues, I'll review one of yours. Regards, Martin

Re: [Python-Dev] [Core-mentorship] Issues Pending Review

2012-10-14 Thread Michele Orrù
In case nobody picks it up, my 5-for-1 offer still stands: if you review five issues, I'll review one of yours. :D merwork saves™ HTTPServer does not correctly handle bad headers / http://bugs.python.org/issue16083 It does not seem clear what should be done and which versions to change.