On Wednesday 16 April 2008 15:39:41 Jonathan Worthington wrote: > chromatic wrote:
> > It helps the PIR Ackerman benchmark by 4.67%. parrot_pass_args gets more > > expensive, but next_arg_sig and everything else except for > > Parrot_init_arg_indexes_and_sig_pmc gets called much, much less. I > > played with the patch a little bit, but didn't get it much faster. > So is the optimization worth applying, do you think? Seems like a > relatively small gain once we take it away from a benchmark that we'd > expect it to directly help... > > It may be possible to avoid the Parrot_init_arg_indexes_and_sig_pmc > > calls, but I couldn't immediately see how. > I pondered this too, but didn't really look into it properly. If we can do this, the optimization might be worth it. -- c
