Peter Jensen smie twierdzic jakoby:

> Well, I've observed "randomness" with -j2+ builds on more than one
> occasion, but it went away when -j1 was set in /etc/make.conf. I think
> it may have something to do with what else the computer is doing at the
> time. If it's only a small difference in timing that will make a
> parallel build fail, it's more likely that it won't have been reported
> yet. The ebuilds that set -j1 are more likely to be obvious cases where
> you're almost guaranteed a failure in parallel makes. That's what's so
> tricky, you need to be *very* careful to get every dependency figured
> out in order to safely build in parallel. With large and complex
> projects that's not always possible (or even considered).
>
one more question: have you (or anyone here) tested diffrence
between -j2 and -j1? i mean - how much faster is -j2 (in average use).
and (if you've tested it), please write if you're using ccache

pozdr. aye

--
"Life's a bitch and so am I, the world owes me, so fuck you." - Green Day
PoLiSh YoUr EnGlIsH:
WYJSC NA LUDZI - GO OUT ON PEOPLE