The code for the profiler in the project I mentioned seems to have been custom written for the project. You can see it here:
https://github.com/JuliaLang/julia/blob/master/src/profile.c and here: https://github.com/JuliaLang/julia/blob/master/base/profile.jl There looks to be fairly good separation between the C part, which gathers the actual data, and the high level part, which drives the whole thing and processes the data. I didn't check carefully, but perhaps the C part is of some use to someone trying to do this for at least the C libraries used by Sage. I appreciate that there are technical difficulties at the Cython level and Python stack tracing is different again. The approach can work. I'm not necessarily suggesting it is the approach that should be taken. Bill. On Saturday, 6 September 2014 20:04:25 UTC+2, Volker Braun wrote: > > On Saturday, September 6, 2014 5:44:57 PM UTC+1, Bill Hart wrote: >> >> It takes samples at regular points during the computation >> > > Thats the aforementioned http://trac.sagemath.org/ticket/16777. See that > ticket for technical obstacles of that approach. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
