[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: creating kernel udebs from kernel-source



On Tue, Nov 30, 2004 at 09:17:45AM +0100, Fabio Massimo Di Nitto wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi Sven,
> 
> Sven Luther wrote:
> 
> |
> | Imagine all the .udebs for all the modules plus the kernel,
> 
> Let say for example that i386 builds 14 debs (386, 686, 686-smp,
> k7 and k7-smp flavours) + 1 _all.deb (the docs).
> linux-kernel-di generates approx 40 udebs.
> 
> That means 55 binary packages (and i would say in the worst case
> since i386 is one of the arch that generates more modules than
> any other iirc)

Oh, no, you are oh so wrong. you forgot all the modules .udeb, and they are
loads of them. Like waldi said, it was around 200 packages 6 month ago, and
probably over 400 today, 

> | multiplied by the number of architectures plus the number of
> | architectures having 2.6 kernels.
> 
> ??? this makes no sence to me at all. and we are only talking about one
> kernel.
> 
> You only multiply once by the number of arch.

Nope. you have the 2.4 kernels and the 2.6 kernel, so you add the total number
of kernels which is between the number of arches and two time the number of
arches. Actually, you have to take every available flavours as you mentioned
above. On powerpc this is 5 for 2.4 kernels and 3 for 2.6 ones for a total of
8. (2.4 : powerpc, powerpc-small, power3, power4, apus, 2.6: powerpc, power3,
power4, soon to come ppc64 iserie-legacy, generic and generic-power4
additionally, and i have had at least one case where the UP kernel failed and
a SMP kernel is needed, so maybe we should double some of these).

This makes 8 powerpc flavours, And a quick look at : 

  http://packages.qa.debian.org/l/linux-kernel-di-powerpc-2.6.html

Will show you that powerpc has 114 .udebs, and the 2.4 kernels have 127. Mmm,
easier to get an idea on : 

  http://qa.debian.org/developer.php?login=debian-boot@lists.debian.org

So, for powerpc, this is already 241 .udeb packages, and it is only one arch
out of 11.

Do think become clearer now ? Or maybe i am missing something ?

> |
> | That was by far the most .udebs any package could hold, and it did
> | break.
> 
> eh? how? what is difference of having 40 udebs from package foo or
> 40 udebs + 15 debs from package bar?

The difference is between having (in the powerpc case) 114 .udebs in one
package and 127 in another, over having 241 in the conjunct package, which
breaks the packaging tool if i am not wrong.

> | Ask Colin, he knows about that.
> 
> No please Sven, don't start bouncing people around. Give me a reference
> or dig one for me. It is working here and nothing is broken.

You are welcome to look over the debian-boot archive though, but i belive
Colin, Joeyh and Waldi implemented the thing, so they would be the best to ask
about this, and since Colin works on ubuntu with you, he is the natural person
to ask, also see above about the problems in your calculation. Given the
powerpc packages, i believe even Waldi's 400 .udebs is grossly understated.

Friendly,

Sven Luther



Reply to: