1 |
On 3/14/12 10:58 AM, Matthew Summers wrote: |
2 |
|
3 |
> __Everyone__ is already using an initramfs, therefore there are no |
4 |
> initramfs-less systems anymore (it may just be empty). Every single |
5 |
> person reading this thread that has not already done so needs to |
6 |
> immediately go read the relevant documentation located in |
7 |
> /usr/src/linux/Documentation/filesystems/ramfs-rootfs-initramfs.txt, |
8 |
> then and only then can a real discourse be had. |
9 |
|
10 |
Yawn, I don't and I won't since I don't need it. Why should I? |
11 |
|
12 |
> Why is an in-kernel initramfs so bad anyway? I am baffled. Its quite |
13 |
> nice to have a minimal recovery env in case mounting fails, etc, etc, |
14 |
> etc. |
15 |
|
16 |
Because at least for me is *totally* pointless. |
17 |
|
18 |
My main system is with a single partition so I shouldn't care much, I |
19 |
have a system that has a separate /usr so probably I'll have *some* pain |
20 |
once I'll upgrade it if I don't merge /usr and / partitions before. |
21 |
|
22 |
Still the whole idea brings us back to the freebsd "everything in /usr" |
23 |
while would make more sense go the hurd way "everything in /" if there |
24 |
is a sound reason to merge those. Beside the whole |
25 |
/usr/share/id-data-du-jour-my-udev-rule-might-need and the I-want-glib |
26 |
and I-want-dbus bandwagon I hadn't seen any compelling reason. |
27 |
|
28 |
Having anything as complex as dbus for early boot sounds dangerous or frail. |
29 |
|
30 |
lu |