1 |
On Mon, Sep 27, 2021 at 11:47:38PM +0200, Michał Górny wrote: |
2 |
> > Can we consider moving the checks for set A somewhere else, such that we |
3 |
> > don't check the kernel config during package compile & install time, but |
4 |
> > only check it later? This also meaningfully resolves that cases where |
5 |
> > the system that has package building isn't where the packages are being |
6 |
> > used. |
7 |
> I'm not sure if I understand you correctly but if you mean not doing |
8 |
> checks before compiling/installing, then I have to disagree. There is |
9 |
> value in knowing about this kind of problems early (hey, that's why we |
10 |
> have pkg_pretend in the first place!) |
11 |
Ebuilds should be able to call the tool (but it could be made optional |
12 |
easily), which does the checks MORE efficiently than the present eclass |
13 |
code. The ebuilds would be responsible for suitable warnings or failures |
14 |
based on the tool's output. |
15 |
|
16 |
E.g. maybe you're in a rescue environment and you know the tooling will |
17 |
work fine on your final environment. |
18 |
|
19 |
> There's certainly value in knowing 'I need to rebuild my kernel' early |
20 |
> vs learning only after you've spent significant time waiting for some |
21 |
> package to build. |
22 |
|
23 |
One thing to this is if you're doing pkg_pretend for multiple packages |
24 |
in a single emerge call, the tool could greatly amortize the cost of the |
25 |
checks, as well as having them available after merge. |
26 |
|
27 |
Great thought I had would be this tool could ALSO run on boot and warn |
28 |
if some packages are unlikely to work |
29 |
|
30 |
-- |
31 |
Robin Hugh Johnson |
32 |
Gentoo Linux: Dev, Infra Lead, Foundation Treasurer |
33 |
E-Mail : robbat2@g.o |
34 |
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85 |
35 |
GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136 |