On Sun, Nov 21, 2010 at 00:54, Leonardo Santagada <[email protected]> wrote: > here is the latest version of the code http://paste.pocoo.org/show/294079/ > > Things apear to be working except that when I run the tests > http://paste.pocoo.org/show/294087/ the kevent one fails. Supposedly > it should give me back the event setting kevent.data to how much you > can read on the file and it is returning 0 to me. > > Maybe I did something dumb while declaring kevent_array or in > kevent_struct, but I'm kind of lost. Python 3.1 show the expected > behavior (of course, using its much more high level interface).
> Any tips? Have you tried using strace on the two programs, to see and compare the actual invocations? Variations of strace -e kqueue,kevent,file,network might help selecting relevant data (-e syscall1,syscall2,...,syscallN shows only those syscalls; file and network are syscall categories). If you are not on Linux but on other *BSDes (and it seems you must be, since Linux has other syscalls), you might still use strace or equivalents: http://www.cyberciti.biz/tips/ktrace-freebsd-macosx-tool-howto.html Best regards -- Paolo Giarrusso - Ph.D. Student http://www.informatik.uni-marburg.de/~pgiarrusso/ _______________________________________________ [email protected] http://codespeak.net/mailman/listinfo/pypy-dev
