| 1 |
On Wed, Jul 24, 2013 at 04:40:38PM -0400, Rich Freeman wrote: |
| 2 |
> Also, not all fixes are equal. The ones that are the biggest concern |
| 3 |
> are security fixes. |
| 4 |
|
| 5 |
How do you _know_ which fixes are security fixes? |
| 6 |
|
| 7 |
> If you tell me that the kernel has a new exploit |
| 8 |
> 2x/week then I'll start to wonder when the kernel team started |
| 9 |
> outsourcing to MS. Most fixes provide no benefit to most users. |
| 10 |
> Upgrading kernels on Gentoo is not automatic either, especially if you |
| 11 |
> have an initramfs, complex configuration, modules in outside packages |
| 12 |
> (nvidia, virtualization, etc), and so on. |
| 13 |
|
| 14 |
I'm releasing, on the average, 3 new kernel versions a week, with 100+ |
| 15 |
patches in them (for different branches.) Sometimes more. Please tell |
| 16 |
me exactly how you are going to evaluate which fixes I make are security |
| 17 |
fixes, and you know which to pick and choose from. |
| 18 |
|
| 19 |
Trust me, it's a hard problem, people have tried it in the past, and |
| 20 |
failed horribly :) |
| 21 |
|
| 22 |
> It just seems like we should be able to get by without a semiweekly |
| 23 |
> kernel upgrade on our "stable" branch. |
| 24 |
|
| 25 |
You want me to slow down and do releases in larger chunks then? Hah, |
| 26 |
not a chance... |
| 27 |
|
| 28 |
good luck, |
| 29 |
|
| 30 |
greg k-h |
Archives