1 |
On Sat, 21 Jan 2012 15:34:39 -0600 |
2 |
Dale <rdalek1967@×××××.com> wrote: |
3 |
|
4 |
> Michał Górny wrote: |
5 |
> > On Sat, 21 Jan 2012 06:28:54 -0600 |
6 |
> > Dale <rdalek1967@×××××.com> wrote: |
7 |
> > |
8 |
> >> Michał Górny wrote: |
9 |
> >>> On Wed, 18 Jan 2012 01:20:03 -0600 |
10 |
> >>> Dale <rdalek1967@×××××.com> wrote: |
11 |
> >>> |
12 |
> >>>> Michał Górny wrote: |
13 |
> >>>>> On Tue, 17 Jan 2012 21:38:26 -0600 |
14 |
> >>>>> Dale<rdalek1967@×××××.com> wrote: |
15 |
> >>>>> |
16 |
> >>>>>> Michał Górny wrote: |
17 |
> >>>>>>> On Tue, 10 Jan 2012 19:14:52 +0100 |
18 |
> >>>>>>> Enrico Weigelt<weigelt@×××××.de> wrote: |
19 |
> >>>>>>> |
20 |
> >>>>>>>> * Micha?? Górny<mgorny@g.o> schrieb: |
21 |
> >>>>>>>> |
22 |
> >>>>>>>>> Does working hard involve compiling even more packages |
23 |
> >>>>>>>>> statically? |
24 |
> >>>>>>>> I guess, he means keeping udev in / ? |
25 |
> >>>>>>> Because adding 80 KiB of initramfs hurts so much? We should |
26 |
> >>>>>>> then put more work just to ensure that admin doesn't have to |
27 |
> >>>>>>> waste 15 minutes to recompile the kernel (if necessary), |
28 |
> >>>>>>> create an initramfs and add it to bootloader config? |
29 |
> >>>>>>> |
30 |
> >>>>>> 80Kbs? You sure about that? I somehow failed to mention this |
31 |
> >>>>>> before. I noticed it when I saw another reply to this post. |
32 |
> >>>>>> Reality check: |
33 |
> >>>>> 80 KiB is enough for mounting plain /usr and booting with it. |
34 |
> >>>>> See tiny-initramfs (but I haven't tested it thoroughly). |
35 |
> >>>>> |
36 |
> >>>> |
37 |
> >>>> My plan is to have /usr on lvm. I think it will end up larger |
38 |
> >>>> and it still adds one more thing to break. |
39 |
> >>>> |
40 |
> >>>> I really wish someone would get a better plan. I think I see a |
41 |
> >>>> garbage dump ahead with lots of Linux distros headed that way. |
42 |
> >>> |
43 |
> >>> Better plan how? LVM requires udev for some reason. Letting rootfs |
44 |
> >>> grow with data unnecessary for a number of users is no good plan |
45 |
> >>> either. Just install that initramfs, be done with it and let us |
46 |
> >>> focus on actual work rather than fixing random breakages. |
47 |
> >>> |
48 |
> >>> We already usually have separate /boot to satisfy the needs of |
49 |
> >>> bootloader. Then you want us to chain yet another filesystem to |
50 |
> >>> satisfy the needs of another layer. Initramfs reuses /boot for |
51 |
> >>> that. |
52 |
> >>> |
53 |
> >> |
54 |
> >> |
55 |
> >> The point is, I don't like initramfs. I don't want to use one. |
56 |
> > |
57 |
> > And I don't like binaries on rootfs. I don't want to have ones. |
58 |
> > |
59 |
> > So we're talking about taste... |
60 |
> |
61 |
> |
62 |
> Actually, we're talking about how things has worked so well for a VERY |
63 |
> long time and there is no need to reinvent the wheel. |
64 |
|
65 |
And required a considerable amount of work which increases due to |
66 |
software getting more complex and users wanting more features. |
67 |
|
68 |
And I don't get 'the wheel' here? What wheel? I'd say we rather want to |
69 |
get rid of the useless fifth wheel. |
70 |
|
71 |
> >> It's funny how I never needed one before either but now things are |
72 |
> >> being broken. It's not LVM that is breaking it either. I wouldn't |
73 |
> >> need the initramfs even if It was on a regular partition until the |
74 |
> >> recent so called "improvements." |
75 |
> > |
76 |
> > ...and your main argument is 'long, long ago someone decided that it |
77 |
> > should match the same taste as mine, so it should be like it |
78 |
> > forever'. Of course, those times there were no such thing as an |
79 |
> > initramfs... |
80 |
> > |
81 |
> |
82 |
> |
83 |
> Then don't break that. Just because someone came up with a initramfs |
84 |
> doesn't mean everyone should be forced to use one. |
85 |
|
86 |
And noone is forced to update the system either. |
87 |
|
88 |
-- |
89 |
Best regards, |
90 |
Michał Górny |