| 1 |
On Sat, 26 Jul 2008 16:56:20 -0500 |
| 2 |
Andrew Gaffney <agaffney@g.o> wrote: |
| 3 |
|
| 4 |
> Duncan wrote: |
| 5 |
> > "--jobs=10 --keep-going --load-average=15" |
| 6 |
> |
| 7 |
> For a dual-dual-core setup, a load average of 4.0 is "fully loaded". |
| 8 |
|
| 9 |
Only in ideal cases, when you have long-running processes |
| 10 |
hammering the cpu and little or no context switches. This isn't the case |
| 11 |
with builds, the actual compile processes that need cpu time usually |
| 12 |
terminate very quickly, which increases the load average. |
| 13 |
I did some benchmarking a while ago with different combinations of -j |
| 14 |
and -l in MAKEOPTS, using the kernel as testcase, and IIRC the fastest |
| 15 |
builds were around -l6.0 (on a dual-core system) and high (or |
| 16 |
unlimited) values for -j (sidenote here: some ebuilds like openoffice |
| 17 |
parse MAKEOPTS to get parameters for their own build systems, but only |
| 18 |
recognize/support -j and ignore -l, so one should still be careful with |
| 19 |
the value for -j). |
| 20 |
|
| 21 |
Marius |
Archives