| 1 |
On Tue, 18 Dec 2012 09:08:53 -0500 |
| 2 |
Michael Mol <mikemol@×××××.com> wrote: |
| 3 |
|
| 4 |
|
| 5 |
This sentence summarizes my understanding of your post nicely: |
| 6 |
|
| 7 |
> Now, why is /usr special? It's because it contains executable code the |
| 8 |
> system might require while launching. |
| 9 |
|
| 10 |
Now there are only two approaches that could solve that problem: |
| 11 |
|
| 12 |
1. Avoid it entirely |
| 13 |
2. Deal with it using any of a variety of bootstrap techniques |
| 14 |
|
| 15 |
#1 is handled by policy, whereby any code the system might require |
| 16 |
while launching is not in /usr. |
| 17 |
|
| 18 |
#2 already has a solution, it's called an init*. Other solutions exist |
| 19 |
but none are as elegant as a throwaway temporary filesystem in RAM. |
| 20 |
|
| 21 |
I should be clear that I do not necessarily support Lennart's |
| 22 |
solutions, but I do support his perception of the problem (at least |
| 23 |
partially). We cannot support situations where *launch* code is |
| 24 |
haphazardly scattered in location X and this must always work for all |
| 25 |
values of X. We already have a remarkably parallel situation in /boot - |
| 26 |
in order to boot at all, the code running at that point in time needs |
| 27 |
to be able to find stuff, and it finds it (by policy) in what we will |
| 28 |
later call /boot. I see this /usr debate as the same thing on a larger |
| 29 |
scale. |
| 30 |
|
| 31 |
-- |
| 32 |
Alan McKinnon |
| 33 |
alan.mckinnon@×××××.com |
Archives