| 1 |
> On 5 Jan 2022, at 19:18, Kai Krakow <kai@××××××××.de> wrote: |
| 2 |
> |
| 3 |
> Am Mi., 5. Jan. 2022 um 19:22 Uhr schrieb Ulrich Mueller <ulm@g.o <mailto:ulm@g.o>>: |
| 4 |
>> |
| 5 |
>>>>>>> [...] |
| 6 |
>> That applies to all parallel builds though, not only to ebuilds |
| 7 |
>> inheriting check-reqs.eclass. By tweaking MAKEOPTS, we're basically |
| 8 |
>> telling the user that the --jobs setting in their make.conf is wrong, |
| 9 |
>> in the first place. |
| 10 |
> |
| 11 |
> Well, I'm using a safe combination of jobs and load-average, maybe the |
| 12 |
> documentation should be tweaked instead. |
| 13 |
> |
| 14 |
|
| 15 |
I think "safe" is doing some heavy lifting here... |
| 16 |
|
| 17 |
> I'm using |
| 18 |
> [...] |
| 19 |
|
| 20 |
> |
| 21 |
> The "--jobs" parameter is mostly a safe-guard against "make" or |
| 22 |
> "emerge" overshooting the system resources which would happen if |
| 23 |
> running unconstrained without "--load-average". The latter parameter |
| 24 |
> OTOH tunes the parallel building processes automatically to the |
| 25 |
> available resources. If the system starves of memory, thus starts to |
| 26 |
> swap, load will increase, and make will reduce the jobs. It works |
| 27 |
> pretty well. |
| 28 |
> |
| 29 |
> I've chosen the emerge loadavg limit slightly higher so a heavy ebuild |
| 30 |
> won't starve emerge from running configure phases of parallel ebuilds. |
| 31 |
> |
| 32 |
|
| 33 |
... because it's quite hard for this logic to work correctly enough |
| 34 |
of the time without jobserver integration (https://bugs.gentoo.org/692576 <https://bugs.gentoo.org/692576>). |
| 35 |
|
| 36 |
But indeed, I'd say you're not the target audience for this (but I appreciate |
| 37 |
the input). |
| 38 |
|
| 39 |
Best, |
| 40 |
sam |
Archives