Silence overactive assert in ipa-inline

[Jan Hubicka]

Hi,
with frequencies not being capped by 100 it is easy to run into roundoff errors
that are more than 1.  Maybe I will need to give up on this assert (which would
be pity as it is useful) but for now I just made it bit more tolerant.

Bootstrapped/regtested x86_64-linux.

Honza

	* ipa-fnsummary.c (estimate_node_size_and_time): Be more tolerant for
	roundoff errors.
Index: ipa-fnsummary.c
===================================================================
--- ipa-fnsummary.c	(revision 254884)
+++ ipa-fnsummary.c	(working copy)
@@ -2744,7 +2744,7 @@ estimate_node_size_and_time (struct cgra
   gcc_checking_assert (time >= 0);
   /* nonspecialized_time should be always bigger than specialized time.
      Roundoff issues however may get into the way.  */
-  gcc_checking_assert ((nonspecialized_time - time) >= -1);
+  gcc_checking_assert ((nonspecialized_time - time * 0.99) >= -1);
 
   /* Roundoff issues may make specialized time bigger than nonspecialized
      time.  We do not really want that to happen because some heurstics