http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54677
--- Comment #3 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2012-09-23 10:49:53 UTC --- > I tweaked the dg-error lines and didn't see any problem when I run the > testsuite, weird. Let me double check. Not sure from your PR which are exactly > the excess errors. I don't either. I have done the following: [macbook] f90/bug% g++48 -std=c++0x /opt/gcc/work/gcc/testsuite/g++.dg/cpp0x/decltype32.C >& tmp [macbook] f90/bug% g++48 -m32 -std=c++0x /opt/gcc/work/gcc/testsuite/g++.dg/cpp0x/decltype32.C > & tmp1 [macbook] f90/bug% diff tmp tmp1 [macbook] f90/bug% So the errors seem to be the same with -m32 and -m64, but there is no excess errors with -m32. I don't understand how the 16211 lines of error can pass the tests. BTW see http://gcc.gnu.org/ml/gcc-testresults/2012-09/msg02274.html for a non darwin failure.
