Given that compiler optimizations are likely to use new additions to the ISA, I propose that our strategy for deciding whether or not to build gcc should be:
1) use OS provided gcc if it works fine (of course) 2) build our own gcc on OSX (very popular and hopelessly broken, but at least we know the binutils version) 3) build with reduced optimization if the OS gcc is broken but works with less optimization (say, -O2 -march=generic) 4) build our own gcc if the OS gcc is hopelessly broken Automatically reducing optimization levels and switching between different gcc installs could be done by the compilerwrapper. PS: We should ship sage-5.0 first! -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
