[Bug ipa/58721] [4.9 Regression] The subroutine perdida is no longer inlined in fatigue.f90
hjl.tools at gmail dot com
gcc-bugzilla@gcc.gnu.org
Thu Nov 28 19:19:00 GMT 2013
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58721
H.J. Lu <hjl.tools at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Last reconfirmed| |2013-11-28
CC| |hjl.tools at gmail dot com
Ever confirmed|0 |1
--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Yuri Rumyantsev from comment #1)
> We found the cause of performance degradation - bug was introduced by
> r202567, namely in callback function "check_callers":
>
> was
>
> if (!has_hot_call && cgraph_maybe_hot_edge_p (e))
>
> must be
>
> if (!(*(bool*)has_hot_call) && cgraph_maybe_hot_edge_p (e))
>
> After apllying this fix routine is inlined.
It looks obvious. Please submit a patch. Thanks.
More information about the Gcc-bugs
mailing list