| 1 |
>>>>> On Wed, 05 Jan 2022, Florian Schmaus wrote: |
| 2 |
|
| 3 |
>> That applies to all parallel builds though, not only to ebuilds |
| 4 |
>> inheriting check-reqs.eclass. By tweaking MAKEOPTS, we're basically |
| 5 |
>> telling the user that the --jobs setting in their make.conf is wrong, |
| 6 |
>> in the first place. |
| 7 |
|
| 8 |
> Yes, exactly. And it is a bandaid solution. But I believe that it will |
| 9 |
> do more good than evil. And it is probably only used until portage is |
| 10 |
> able to report to the user that the emerge failed due to OOM (which I |
| 11 |
> believe to be non-trivial to implement, but I am happy to be proven |
| 12 |
> otherwise). |
| 13 |
|
| 14 |
Obviously I disagree. Tweaking the value for a subset of packages isn't |
| 15 |
a solution to the problem. |
| 16 |
|
| 17 |
MAKEOPTS applies to all parallel builds, and users should set it to a |
| 18 |
value suitable for their system (i.e. number of CPUs, available memory, |
| 19 |
etc.). Maybe our documentation needs to be improved? I see that |
| 20 |
make.conf.example says "The suggested number for parallel makes is |
| 21 |
CPUs+1" which may not be the best possible advice. |
| 22 |
|
| 23 |
Ulrich |
Archives