http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58721
--- Comment #4 from Yuri Rumyantsev <ysrumyan at gmail dot com> --- It turned out that proposed fix does not help trunk compilers since now another huge routine is inlined firstly (read_input) and for perdida we got the following message: not inlinable: iztaccihuatl/17 -> perdida.constprop/122, --param large-function-growth limit reached Note that passing --param large-function-growth=130 will restore performance. I assume that additional heuristics must be designed to inline once called functions which are called inside inner loop.
