On Tue, Jan 8, 2019 at 1:51 AM Nils Bruin <nbr...@sfu.ca> wrote: > > On Monday, January 7, 2019 at 12:48:30 PM UTC-8, Simon King wrote:I did the > implementation a couple of years ago, so, it is possible that >> >> I misremember. However, if I recall correctly, having a reference >> directly from the wrapped method to the cache, rather than going from the >> wrapped method to the instance to which the method is bound and finally >> to an item of the __dict__ of the instance, had a noticeable effect. >> > Yes, by the time you're operating on cython level, saving a few > attribute/dictionary lookups can be quite measurable.
Indeed--I was tired when I wrote my last message and did not mean "either way" the way it reads. I specifically meant that two separate caches seemed unnecessary. But each element having its own reference to the cache can definitely be a speedup. In that case I don't see what the problem is. -- 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 https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
