As far as I can see, running gub's make bootstrap
target is completely useless. If you call make lilypond (which is equivalent to `make -f lilypond.make', BTW), all necessary packages are rebuilt – even the stuff from `tools', which doesn't make any sense. Looking into the `*.checksum' files (which gub uses to decide whether it has to rebuild a package), the only differences I see are as follows. --- <make bootstrap>/autoconf.checksum 2019-01-07 22:22:52.950289854 +0100 +++ <make lilypond>/autoconf.checksum 2019-01-08 07:03:16.824272174 +0100 @@ -191,7 +191,7 @@ build_cpu=x86_64 build_dependencies_string=m4;perl;system::gcc;tools::file;tools::librestrict;tools::libtool;tools::make;tools::tar;tools::zlib build_hardware_bits=64 -build_number=0 +build_number=1 build_os=linux build_platform=linux-64 builddir=/home/wl/git/gub/target/tools/build/autoconf-2.63 In other words, `make lilypond' enforces a higher build number in all packages. Alas, my knowledge of gub is too limited to explain that, or even to fix it. So it seems that you can save some hours of compilation time if you don't build the `bootstrap' target. I now wonder whether this is true on my openSuSE GNU/Linux box only ... Werner _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel