[Bug tree-optimization/69196] [5/6 Regression] code size regression with jump threading at -O2

law at redhat dot com gcc-bugzilla@gcc.gnu.org
Sat Jan 16 07:06:00 GMT 2016


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69196

--- Comment #7 from Jeffrey A. Law <law at redhat dot com> ---
It's definitely the FSM threader.  The FSM threader has 3 knobs for tuning.  If
I crank them all down to minimal values, then the trunk & gcc-5 actually
generate slightly smaller code than gcc-4.9.

I don't think Sebastian did a lot of tuning on those params and with the
increased reliance on the FSM threader on the trunk those params likely need
some rethinking.


More information about the Gcc-bugs mailing list