1 |
Alan McKinnon wrote: |
2 |
> On Sunday 08 November 2009 23:20:31 Stroller wrote: |
3 |
> |
4 |
>>> You really need to learn to make your own kernel. ... |
5 |
>>> |
6 |
>> Whilst I agree in principle that a good (slim?) kernel is better and |
7 |
>> your comments on that, I am sceptical whether the majority of people |
8 |
>> have the knowledge to make any significant performance or security |
9 |
>> improvements. |
10 |
>> |
11 |
>> AIUI the kernels shipped by distros like Red Hat, for instance, are |
12 |
>> configured by the very people that work on and maintain the mainline |
13 |
>> kernel tree. How can any of us simple end-users compete with that? |
14 |
>> |
15 |
>> I imagine it to be very easy for any of us normal people to enable or |
16 |
>> disable options that make significant performance impact - but we |
17 |
>> would never know it, because we're not benchtesting it or even |
18 |
>> qualified to assess proper benchtests. |
19 |
>> |
20 |
>> I cannot believe that in a day you could study this subject |
21 |
>> sufficiently to have any reasonable competence on the matter. And thus |
22 |
>> if you do spend only a day, that's wasted time. I would add that the |
23 |
>> kernel is evolving constantly, and in a year's time your knowledge - |
24 |
>> and your .config - is likely to be at least somewhat outdated. |
25 |
>> |
26 |
>> I chose to copy the .config from Knoppix because it's easy to get hold |
27 |
>> of that, but also because it's selected by someone who knows more than |
28 |
>> me, and it is likely to work with any hardware I install into my |
29 |
>> machine or connect by USB. I take Volker's point that a LiveCD .config |
30 |
>> could be the worst possible choice so I'm open to alternatives, but |
31 |
>> I hope those who say I should "learn to make your own kernel" |
32 |
>> appreciate my points over how effectual that will be - sure, I can |
33 |
>> delete my .config and start again with `make menuconfig` and I can go |
34 |
>> through every option and read the help, and I'm sure I'll get just as |
35 |
>> good results as 80% of the people on this list, but I just don't know |
36 |
>> that that's much of an answer. |
37 |
>> |
38 |
> |
39 |
> You are reading way more into the subject than is actually there. |
40 |
> |
41 |
> Red Hat employees do work on mainline and do write kernel code. But finding a |
42 |
> bug, writing new code and fixing security exploits are very different |
43 |
> activities to simply configuring the code that is there. And that is what RH |
44 |
> do - they take the code that is already there, apply whatever backport and |
45 |
> experimental patches suits their distro, then go through menuconfig switching |
46 |
> some things on and some things off. Their needs are different to yours - they |
47 |
> need their kernel to run on just about any hardware on the planet, so they |
48 |
> build a horrendously complex initrd with support for every known boot device, |
49 |
> then build every module that even half-way works. And also enable every known |
50 |
> kernel sub-system (because someone somewhere is going to use it). |
51 |
> |
52 |
> By your analogy, you might consider Red Hat more qualified than you to decide |
53 |
> if you should build an MTA with or without LDAP support. Which is of course |
54 |
> patently ridiculous - if you know you need LDAP then you need it. Otherwise |
55 |
> you don't (and this is not a security issue, it's a features issue) |
56 |
> |
57 |
> If you configure your own kernel, you only need build the bits you use. The |
58 |
> sole benefit for a Gentoo users to using a custom distro kernel is support for |
59 |
> things not in mainline (like some entire FibreChannel product ranges out |
60 |
> there). But please note that even if you copy an RH .config, you do not have |
61 |
> those patches to hand so you will not get those extra features. Unless you |
62 |
> patched the ebuild yourself, in which case you are already au-fait with |
63 |
> building a kernel and we would not be having this discussion. |
64 |
> |
65 |
> In summary, I hear your reasoning and understand your concerns. But it is |
66 |
> flawed and you are worried about something that is not actually there. |
67 |
> |
68 |
> |
69 |
|
70 |
What he said plus this little tidbit of info. When I built my first |
71 |
kernel, I had no howto except for the basic instructions in the Gentoo |
72 |
install guide. This was about 6 years or so ago and there was not a lot |
73 |
on configuring a kernel except for the options Gentoo needed. It took |
74 |
me three tries to get one to boot and work pretty well and all of a hour |
75 |
at most. A lot of that hour was compiling the kernel. |
76 |
|
77 |
You seem to think it takes a rocket scientist to build a kernel, it |
78 |
doesn't. You just have to know what hardware you have and then enable |
79 |
the features you need. Once you get a good one built, using make |
80 |
oldconfig works really well. You can config a kernel in less than five |
81 |
minutes most likely then compile and you are done. If you update fairly |
82 |
regular, make oldconfig will work fine and not cause you the trouble you |
83 |
had in the beginning. |
84 |
|
85 |
Dale |
86 |
|
87 |
:-) :-) |