On Tue, May 28, 2013 at 07:20:14PM +0200, Peter Bex wrote: > Well, possibly the tests are wrong (flonum equality is Hard). > I'll have a look and see if this can be improved. > If it's indeed an optimization that causes a small difference in > flonums, it shouldn't cause any problems in practice so you should > be able to compile with -O3 and have no trouble.
I just tested with -O3 on NetBSD/amd64 (gcc 4.5.3) and also on Linux/i686 (Debian gcc 4.3.2-1.1) and all tests pass, with and without DEBUGBUILD=1. For now I'd say this is a problem with your particular gcc version. If you or someone else can reproduce this and come up with a patch that would be great. You don't have to edit Makefiles BTW, you can easily force -O3 by running gmake C_COMPILER_OPTIMIZATION_OPTIONS=-O3 ... You can also pass OPTIMIZE_FOR_SPEED=1, which is a catchall setting. If you want to let this stick around, config.make is the place to add these settings. Cheers, Peter -- http://www.more-magic.net _______________________________________________ Chicken-users mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/chicken-users
