Sorry for the late reply but thanks for the pointer. It would be great if I could get some insight into the motivation behind streams. Various places where it can prove useful...
thanks Siddharth On Sat, May 9, 2009 at 8:39 PM, Andreas Klöckner <[email protected]>wrote: > On Samstag 09 Mai 2009, Siddharth Priya wrote: > > Hello, > > > > I went through the pycuda documentation and it seems the equivalent > > function is > > > > pycuda.driver.Context.synchronize() . > > > > Is this accurate? > > Yes. > > Measuring kernel execution time using events [1] will however miraculously > make your teeth brighter, save a few baby dolphins and contribute 5 > Hungarian > Forint towards world peace. > > start = Event() > stop = Event() > start.record() > do_timed_stuff() > stop.record() > do_other_untimed_stuff() > stop.synchronize() > time_in_seconds = stop.time_since(start)*1e-3 > > Andreas > > [1] http://documen.tician.de/pycuda/driver.html#concurrency-and-streams > > _______________________________________________ > PyCuda mailing list > [email protected] > http://tiker.net/mailman/listinfo/pycuda_tiker.net > >
_______________________________________________ PyCuda mailing list [email protected] http://tiker.net/mailman/listinfo/pycuda_tiker.net
