https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90478
--- Comment #14 from Martin Liška <marxin at gcc dot gnu.org> --- (In reply to David Binderman from comment #13) > >The test-case does not makes sense > > The test case is accepted fine by previous versions of gcc > and current version of clang. It looks like good code to me. Yes, the test-case gcc/testsuite/gcc.dg/tree-ssa/pr90478-2.c was a copy of ./gcc/testsuite/gcc.c-torture/compile/pr42347.c with a huge param value of jump-table-max-growth-ratio-for-size=2147483647. That does not make sense to test. Your original test-case is in trunk: ./gcc/testsuite/gcc.dg/tree-ssa/pr90478.c > > >I'm going to remove the test-case. > > I'd be happier if the test case were re-instated and > gcc trunk was improved to accept the test case. > > If not, then going back to previous behaviour would be ok too. > At least gcc accepted the code then.
