> I know that this question is not new, see > https://mail.python.org/pipermail/python-list/2009-March/530623.html for a > similar problem. But it seems like the problem really exists.
As suggested there, can you isolate the leak using valgrind? Does it happen regardless of the timestepping method that you use? If it is a non-GSL wrapper problem, I would expect the problem to be independent of the timestepper method chosen. If it appears only within a particular timestepper (and hence looks like a bug internal to GSL), can you reproduce it using only the C-based GSL api? - Rhys
