Scott Stoddard posted <[EMAIL PROTECTED]>, excerpted below, on Tue, 18 Oct 2005 11:58:05 -0400:
> Duncan wrote: >> changed USE flags vs what you currently have merged.) After each level >> below, I run etc-update, so the number of pending updates doesn't get out >> of hand. Of course, if the current version of any package later in the > > Of course, during those big emerge jobs, you can opt to run etc-update > in another terminal while the emerge is still progressing. I tend to > use this option to save time sorting through the lists while my machine > _isn't_ doing something more productive. /Very/ good point! Me too! In fact, with a dual Opteron, one of the things I do with emerge -p is use the --tree switch as well, to see which emerges depend on each other, then run two or three independent emerge sessions at once, to make full use of the CPUs, while running etc-updates and emerge --pretend --changelog and other such stuff in a forth session. Yes, portage has the jobs thing, but many emerges don't do parallel jobs all that efficiently, or turn them off because it screws up the compile dependency order for certain packages. Checking for dependencies then doing independent emerge sessions where no dependencies exist overcomes that issue. Of course, that only works once one gets past the critical toolchain steps. If you want everything else compiled with the newest gcc, you can't run parallel emerges until gcc itself is merged. However, one can still run etc-update and the like in parallel, even where parallel emerges aren't possible. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman in http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html -- [email protected] mailing list
