From bug 642001 comment 5, we need to make this benchmark faster. One way would be to improve the arguments escape analysis we have. I think v8 might be fast on this mostly because they can allocate |arguments| objects very fast, so this might alternatively be done by waiting for generational GC and then doing arguments a bit differently. function sc_Pair() { } function sc_list() { var res = null; var a = arguments; for (var i = a.length-1; i >= 0; i--) res = sc_Pair(a[i], res); return res; } var t0 = new Date; for (var i = 0; i < 1000000; ++i) sc_list(1, 2, 3); print(new Date - t0);

Maybe have a look how much Bug 616744 would help here.

Fixed by TI, SSA arguments optimization.