On 02/18/2014 01:47 PM, Armin K. wrote: > On 02/18/2014 10:18 AM, Pierre Labastie wrote: >> Le 18/02/2014 05:10, Armin K. a écrit : >>> In BLFS, GCC-4.8.2 page has the following: >>> >>> >>> >>> The instructions below let the build machinery perform a “bootstrap” >>> intentionally. This is necessary if you install the Ada compiler anyway. >>> Even if you don't, a bootstrap is recommended for robustness. >>> >>> >>> >>> Which is, ihmo, a correct thing to do - a bootstrap. But yet, >>> instructions down below contain: >>> >>> --disable-bootstrap \ >>> >>> Is this intentional or what? This disables bootstrapping, but the >>> explanation just before the configure block says that bootstrap is being >>> preformed. >>> >>> Also, I'll check this package and post a patch if necessary. >>> >> Well, I had suppressed the disable-bootstrap when I worked on this, >> but later Bruce put it back (while removing ADA instructions at >> the same time). I consider bootstrapping is the correct thing to >> do, unless maybe you are rebuilding the same version of GCC as in >> LFS. >> >> Regards >> Pierre >> > > You are building GCC-4.8.2 with GNAT Core compiler, which is 4.7.4 > xxxxxx (snapshot), so the bootstrap is highly recommended. I've done a > build and I'll need to make few changes before submitting them, and one > of them is removing disable-bootstrap and mention to add it back if ada > compiler isn't being built and the same gcc version as in LFS is being > used. Does that sound fair enough? >
Okay the build was finished long ago. The full buildlog is available here, along with test results. http://www.linuxfromscratch.org/~krejzi/7.5-systemd-rc1/logs/postlfs/007-gcc.log Just search for "cat <<" phrase when file finishes loading (it's 12 MB). There are few test suite failures. namely FAIL: libmudflap.c++/pass41-frag.cxx ( -O) execution test FAIL: libmudflap.c++/pass41-frag.cxx (-O2) execution test FAIL: libmudflap.c++/pass41-frag.cxx (-O3) execution test These are addressed in LFS, and the instructions need to be brought forward in BLFS, too. FAIL: sourcelocation output - source compiled test FAIL: sourcelocation -findirect-dispatch output - source compiled test FAIL: sourcelocation -O3 output - source compiled test These are Java tests. I am not sure if they were present before but I'm not going to spend my time investigating the reason of the failure. There's also autogen error which has also been fixed in LFS. Other than that, looks good to me. Built without --disable-bootstrap using gnat-gpl-2013. -- Note: My last name is not Krejzi. -- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page