leif wrote:
Is there already a ticket for issues with GCC 4.9?
I think I'll open a (small) metaticket soon.
(Release of 4.9.0 is scheduled for Tuesday, so we'll presumably face it soon in some distros.) Depending on when Sage 6.2 gets released, we may also release some 6.1.2 with only fixes for the new GCC version.
(See below.)
So far, Lcalc doesn't build without -fpermissive, and the pre-release version of PPL contained in 6.1.1 at least won't build either (some header / feature test macro weirdness I suppose).
Nope, for PPL 1.1 (included in Sage 6.2.beta*), it's just a name lookup issue (GCC 4.9 is stricter); replacing 'ptrdiff_t' by 'std::ptrdiff_t' a few dozen times is sufficient. (Then also its test suite passes, with '-O3', and the previously affected Sage library extension modules also build.)
For Lcalc 1.23, only a tiny (additional) patch is needed (cf. [1]) to let it and the corresponding Cython extension module build (without '-fpermissive').
I've built Sage 6.2.beta8 from scratch with GCC 4.9.0 RC1 (20140411) with '-O3', and ptestlong passes, on Linux x86_64 that is.
Will retry 6.1.1 (with the patches I now have for 6.2.beta8) later. -leif [1] http://trac.sagemath.org/ticket/12437#comment:10 (related Clang ticket) -- () The ASCII Ribbon Campaign /\ Help Cure HTML E-Mail -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
