Re: Correct way to (re)compile a kernel on Debian Sid
On Thu, Apr 08 2010, Stephen Powell wrote:
> On Thu, 8 Apr 2010 16:01:03 -0400 (EDT), Ivan Marin wrote:
>> On Thu, 8 Apr 2010 10:40:46 -0400 (EDT), Stephen Powell wrote:
>>> It sounds to me like you want to get pristine kernel sources directly
>>> from kernel.org and compile them and run them on a Debian system.
>>> I've never done that, but others tell me that they do it. Of course,
>>> this is not supported by Debian.
>>
>> I understand "not supported" as I can't file bugs from the pristine kernel
>> on the Debian kernel package, and if there's a problem with that, I have to
>> go to the kernel.org bug list, exactly as Stephen pointed. I'm also aware
>> about the changes that the Debian Kernel Team does to the pristine kernel.
>> But all this is about the kernel itself, and not the process of _building_ a
>> new kernel. So a good question is: kernel-package supports the compilation
>> of other kernels than the Debian one?
>
> Yes, I believe so. I myself have never done it, but others on
> this list have.
Indeed. I never use the kernel team kernels (I generally use a
git tree, but I support kernel tarballs)
>>> This is all explained in http://www.wowway.com/~zlinuxman/Kernel.htm
>>> In particular, review Step 10.
>>
>> I will check that, and see if there I can get my way around it. But I
>> still miss the old make-kpkg way... ;-)
Everything the old kernel-package did lives in example scripts
in the examples dir. I confess there is the additional step in
populating /etc/kernel.d, but you get the advantage of being able to
expand what can be done with kernel images.
>
> The new way makes things simpler for the maintainer script. But it requires
> that the user be more knowledgeable in customizing the kernel installation
> process.
True. But it also allows the user more power and flexibility,
and allows users to share their scripts and stuff. I am just waiting
for people to start building on the scaffolding provided :-)
manoj
--
Prof: So the American government went to IBM to come up with a data
encryption standard and they came up with ...Student: EBCDIC!"
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
4096R/C5779A1C E37E 5EC5 2A01 DA25 AD20 05B6 CF48 9438 C577 9A1C
Reply to: