1 |
On Thursday 27 January 2011 15:05:25 YoYo Siska wrote: |
2 |
> On Thu, Jan 27, 2011 at 03:33:21PM +0200, Nikos Chantziaras wrote: |
3 |
> > On 01/27/2011 03:11 PM, Dale wrote: |
4 |
> > >[...] |
5 |
> > >I am using the -j option for the first time now. I'm updating KDE. It |
6 |
> > >seems to work fine. It doesn't scroll all the stuff like with a regular |
7 |
> > >emerges but this new rig is so fast, I can't read it anyway. I did have |
8 |
> > >a package to fail and it spit out the error for me to read. |
9 |
> > |
10 |
> > You don't need that if you have MAKEOPTS set in your make.conf, |
11 |
> > which is preferred. The -j option of emerge emerges multiple |
12 |
> > packages, while with MAKEOPTS set to "-j4" or whatever does a |
13 |
> > parallel build in the same package (meaning compiling multiple |
14 |
> > source files at the same time). |
15 |
> > |
16 |
> > It's preferred because with "emerge -jN" the last package will only |
17 |
> > use one CPU, while with "-jN" in MAKEOPTS even the last package will |
18 |
> > use N CPUs. Furthermore, emerge can't always build N packages at |
19 |
> > the same time because one can depend on the other, so it will have |
20 |
> > to wait until the dependency is built. |
21 |
> |
22 |
> On the other hand, unpacking, configure and install stages are not |
23 |
> parallel and emerge can do those in parallel for different packages... |
24 |
> The best would be somewhere in the middle ;) |
25 |
> |
26 |
> |
27 |
> There are also the load-average options to -j, i.e.: |
28 |
> MAKEOPTS="-j -l5" emerge -j --load-average=5 .... |
29 |
> |
30 |
> which makes make spawn parallel processes while load average is below 5 |
31 |
> and the same for emerge spawning parallel ebuilds (when make isn't |
32 |
> parallel enough) |
33 |
> |
34 |
> yoyo |
35 |
|
36 |
Hmmm... didn't know about that one yet. |
37 |
Does that mean that by doing it like that, the emerge-process (and compile- |
38 |
processes) will try to keep the load average at 5 and if that is lower, it |
39 |
will keep adding more processes? |
40 |
|
41 |
-- |
42 |
Joost |