1 |
On 05/01/2022 09.28, Ulrich Mueller wrote: |
2 |
>>>>>> On Tue, 04 Jan 2022, Sam James wrote: |
3 |
>> Crank down MAKEOPTS jobs if MAKEOPTS="-jN" is too high for the |
4 |
>> amount of RAM available (uses amount declared as needed |
5 |
>> in the ebuild). Typically should be ~2GB per job. |
6 |
> |
7 |
> Where does this number 2 GB come from? The amount of RAM strongly |
8 |
> depends on the programming language and other factors, so I don't |
9 |
> believe that there's one number that can be used for everything. |
10 |
|
11 |
Surely not, but 2 GiB seems like a good start. I guess it could become |
12 |
an eclass variable that ebuilds could modify later (if the need emerges). |
13 |
|
14 |
|
15 |
>> +# @ECLASS-VARIABLE: CHECKREQS_MEMORY_MANGLE_JOBS |
16 |
>> +# @USER_VARIABLE |
17 |
>> +# @DESCRIPTION: |
18 |
>> +# Allow packages to reduce the number of multiprocessing (e.g. make, ninja) jobs |
19 |
>> +# to lower memory usage. |
20 |
>> +: ${CHECKREQS_MEMORY_MANGLE_JOBS=yes} |
21 |
> |
22 |
> If anything, the feature should be opt-in rather than opt-out. |
23 |
|
24 |
It appears to me that the motivation for this change is to prevent users |
25 |
from running into OOM situations when emerging a package. And I believe |
26 |
that many users, especially novices, are not directly able to determine |
27 |
that portage failed due to OOM, simply because there is no direct hint |
28 |
in the build log. Hence opt-out appears to be sensible to me here. |
29 |
|
30 |
- Flow |