On Fri, Dec 06, 2013 at 02:03:24PM -0600, Dan McGhee wrote: > I'm not suggesting any changes to the book. I was asking a question. And > you're right about the time savings. I like FF and TB, but if I could > get it done in one 50+ SBU build, that would be great. And I sure do > appreciate "work-benefit" analyses. > > And I will present my results for use in any way. I need to finish my > UEFI write up for LFS first, but needed to get both e-mail and browser > working in the same partition for convenience sake. > I don't build thunderbird, but for firefox with system nss, nspr using make -jN saves a lot of time. Where N is the number of CPUs that the kernel sees, whether they are real or hyperthreaded. Using more than that always slows things down for me (even mere kernel builds).
Unless you have a specific need for xulrunner, I wouldn't bother building it. Using firefox without xul doesn't build all of xulrunner, so there is a small compile-time saving in that part. Also, the SBUs for firefox/xulrunner are easily affected by small changes in the value of the SBU (e.g. when starting from an old system). But using -j4 I see around 30-33SBU for my own builds / deps / switches, and around 40+ SBU with -j2. Trying to use a single xulrunner build for thunderbird as well as firefox sounds like an interesting idea, but I suggest you go with the easy approach until you are ready to test and document that idea. ĸen -- das eine Mal als Tragödie, dieses Mal als Farce -- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
