IIRC, some sanitising of MAKE was implemented, as with recursive make such a setting really might lead to over-parallelisation
On Tue, 7 Feb 2023, 20:41 Nils Bruin, <nbr...@sfu.ca> wrote: > On Tuesday, 7 February 2023 at 09:56:06 UTC-8 dim...@gmail.com wrote: > I don't think setting MAKE is a good idea. > > Did you try simply > > make -j8 build > > Thanks! I've tried "touch src/sage/structure/*" -- that triggers quite a > large part of sagemathlib to be rebuilt. It used only 4 processes during > cythonization (when it was running just python processes), but it happily > used in the range of 8 processes during compilation (and that's the much > more expensive phase). > > It reported less than 10 minutes real time for about 60 minutes user. With > the previous single thread, it took about 30 minutes (real and user). I > don't know if that's because this is on a 4-core machine with > hyperthreading or because of other bottlenecks. > > Documentation I was able to find is still recommending MAKE="...". So we > should probably change that now that we have a more make-centric build > system. > > -- > You received this message because you are subscribed to the Google Groups > "sage-support" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to sage-support+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sage-support/7653bddd-bcc0-4724-a685-64f9b5e8ec12n%40googlegroups.com > <https://groups.google.com/d/msgid/sage-support/7653bddd-bcc0-4724-a685-64f9b5e8ec12n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-support/CAAWYfq0rV4_a1tuPBDg3bvV8kVOjr7F4H0Jeki-0hBDABQGhQA%40mail.gmail.com.
