1 |
On Thursday 27 January 2011 22:18:22 J. Roeleveld wrote: |
2 |
> On Thursday 27 January 2011 23:05:22 Paul Hartman wrote: |
3 |
> > On Thu, Jan 27, 2011 at 3:46 PM, J. Roeleveld <joost@××××××××.org> wrote: |
4 |
|
5 |
> > >> Once, when building my kernel, I accidentally forgot to specify the |
6 |
> > >> number of makes and ran "make -j all". That was a really bad idea, the |
7 |
> > >> system became totally unresponsive for quite a long time, much longer |
8 |
> > >> than normal kernel build time, but it did eventually finish! |
9 |
> > > |
10 |
> > > I have found that multi-core systems with sufficient memory can handle |
11 |
> > > "-j" (no value) a lot better then sindle-core systems. I do on occasion |
12 |
> > > do it with the kernel and can still continue using the system. (For |
13 |
> > > comparison, my desktop is a 4-core AMD64 with 8GB memory) |
14 |
> > |
15 |
> > Strange, in my case it was an i7 920 (4 cores, hyperthreaded, appears |
16 |
> > as 8 CPUs to Linux) with 12GB of RAM. Maybe if I prefixed it |
17 |
> > with"nice" it would not have brought my computer to its knees... or |
18 |
> > maybe related to the schedulers and other kernel voodoo that I don't |
19 |
> > understand. I might try it again someday :) |
20 |
> |
21 |
> That is strange, unless your harddrive is really underperforming? |
22 |
> Or do you have all the options in the kernel selected? |
23 |
> |
24 |
> Btw, HyperThreading doesn't work too well when you have a lot of identical |
25 |
> tasks. In that case, you might end up with lesser performance as there are |
26 |
> no "usable unused" parts in your cores, but the CPU-schedules (the |
27 |
> hardware one for HT) is looking for things to fill those last few bits |
28 |
> with. |
29 |
|
30 |
I'm running i7 Q 720 (4 cores, hyperthreaded) and have MAKEOPTS="-j9" without |
31 |
any slowdown. One or two packages (like OpenOffice) will fail and need -j=1 |
32 |
to emerge. Otherwise no noticeable drop in desktop responsiveness. |
33 |
|
34 |
I have not set up portage niceness so it runs with default value. |
35 |
|
36 |
Given the above what shall I set --load-average as? |
37 |
-- |
38 |
Regards, |
39 |
Mick |