| 1 |
On Monday 30 September 2013 11:24:58 Neil Bothwick wrote: |
| 2 |
> On Mon, 30 Sep 2013 12:16 +0200, Joost Roeleveld wrote: |
| 3 |
> > > Installing a new kernel does not magically make the old one break. If |
| 4 |
> > > that kernel worked yesterday, it will work today. |
| 5 |
> > |
| 6 |
> > Actually, that is not guaranteed. |
| 7 |
> > I remember a situation in the past where boot-critical software |
| 8 |
> > required a certain minimal kernel-version with specific config-settings. |
| 9 |
> > Without those I could not boot. |
| 10 |
> |
| 11 |
> I don't see how that is an issue with correctly written ebuilds. |
| 12 |
> |
| 13 |
> If you update the kernel, you are increasing the version number and your |
| 14 |
> old one will still work. |
| 15 |
> |
| 16 |
> If you update the software, the ebuild should detect an unsuitable kernel |
| 17 |
> and either warn you or abort. |
| 18 |
|
| 19 |
That is the problem though, the ebuild can't detect that there is an |
| 20 |
unsuitable kernel still available. |
| 21 |
|
| 22 |
> Either way, it is irrelevant whether you are using an initramfs or not. |
| 23 |
|
| 24 |
I agree, my comment was made to point out that a kernel that worked yesterday, |
| 25 |
may no longer work tomorrow. |
| 26 |
|
| 27 |
-- |
| 28 |
Joost |
Archives