On Sun Mar 09 22:28:10 2014, bulk88 wrote: > On Tue Feb 11 22:16:17 2014, tonyc wrote: > > - avoid doing the same gimme test twice, which could confuse a reader > > Now there are 2 EXTENDs which are quite fat (for C) to do, scaling, 4 > pushes, comparison, reading interp globals. My code has just 1 EXTEND > call, which I like. The perl implementation strikes the speed vs size balance heavily on the speed side - using macros and inline functions which produce faster but significantly larger code. So I think adding code like: const SSize_t stack_grow = 1 + (-(gimme_v == G_ARRAY) & avlastindex); to save a few bytes from the extra EXTEND() expansion (while slighly slowing it down) doesn't match how perl's implementation has balanced speed vs size. > Since you fixed the leak a while ago, Yes, I wanted to make sure the fix at least made it into 5.20. Tony --- via perlbug: queue: perl5 status: open https://rt.perl.org/Ticket/Display.html?id=120939