1 |
I'm looking to solicit opinions on when it is appropriate for an |
2 |
ebuild to check for kernel config options using linux-info.eclass. I |
3 |
don't think we have any guidelines documented, instead leaving it up |
4 |
to the "common sense" of package maintainers. |
5 |
|
6 |
Adding linux-info calls to pkg_pretend or pkg_setup causes slowdowns |
7 |
when running emerge, so we should do so only when there is a |
8 |
compensating benefit. It doesn't make sense to check for kernel |
9 |
options that are very commonly enabled. But what is "very common"? |
10 |
|
11 |
An obvious example would be CONFIG_INET, which controls IPv4 support |
12 |
in the kernel. It does not make sense to check for that in every |
13 |
package that uses AF_INET sockets. |
14 |
|
15 |
A less obvious example: a user filed a bug against net-misc/dhcpcd |
16 |
today asking that we check for CONFIG_PACKET [1]. My first thought was |
17 |
"why would you ever disable that?". The option description even says |
18 |
"if unsure, say Y". However, I suppose it is technically possible to |
19 |
run a Linux system with it disabled. |
20 |
|
21 |
I think a reasonable rule of thumb would be to assume we can rely on |
22 |
options that are enabled by "make defconfig". If the user chooses to |
23 |
disable them, they are responsible for anything that breaks. |
24 |
|
25 |
Thoughts? |
26 |
|
27 |
[1] https://bugs.gentoo.org/815064 |