| 1 |
On Fri, Nov 4, 2011 at 3:46 PM, Mike Gilbert <floppym@g.o> wrote: |
| 2 |
> |
| 3 |
> It is good that we warn users about this when they install the package, |
| 4 |
> but I don't think the ebuild should die. |
| 5 |
|
| 6 |
I've always found ebuilds dying at kernel config checks really annoying. |
| 7 |
Checking kernel features at build time (if we die) is broken and |
| 8 |
should be banned IMO: |
| 9 |
|
| 10 |
1) in case of packages not needing kernel headers for building (but |
| 11 |
just checking for a feature) |
| 12 |
a) this breaks the compilation inside chroots on build servers where |
| 13 |
the goal is to just build the package (hint: binpkgs) |
| 14 |
b) it is anyway unable to reliably check if the feature is actually |
| 15 |
available or not (/proc/config.gz might be unavailable, as well as |
| 16 |
kernel sources or headers) |
| 17 |
|
| 18 |
2) in case of packages needing kernel headers (or sources) for building |
| 19 |
a) doing the check in pkg_setup() breaks binpkgs support (the check |
| 20 |
becomes part of a build time task, and not install time) |
| 21 |
|
| 22 |
I'm sure there are other reasons as well... |
| 23 |
|
| 24 |
-- |
| 25 |
Fabio Erculiani |
| 26 |
http://lxnay.com |
Archives