numexpr for Debian ------------------ It is possible to find more documentation at https://github.com/pydata/numexpr A _detailed_ changelog is available at: https://github.com/pydata/numexpr/blob/master/RELEASE_NOTES.rst Some issue related to threading have been reported on sparc machines. The problem is hard to reproduce and do not affects all installations. For this reason the default number of threads for computations is forced to one on sparc. You can verify if your sparc installation is affected by the problem by running the numexpr test suite. If you are not affected you can safely change the number of threads to be used for computations using the "numexpr.set_num_threads" function. -- Wen Heping Wed, 14 Jul 2010 21:50:27 +0800