http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57698
--- Comment #3 from Jan Hubicka <hubicka at ucw dot cz> --- Hmm, the problem here is that we output errors after early inlining always now, while previously we did only when some other inlining happened in the function (adding extra early inlinable function to the testcase should reproduce the error message on early GCC releases). We can fix by outputting always inline errors only after real inliner was run (when optimizing) but then such testcases won't compile at -O0 that is uncol too. Other alternatives include declaring such testcases invalid, always not outputting diagnostic on previously indirect calls, teaching always inliner to do indirect inlining (that will need to do analysis at -O0), adding disregard_inline_limits attribute... Each of these has pros and cons...
