Re: compiling 2.6

From: P.T. Breuer (ptb_at_oboe.it.uc3m.es)
Date: 03/18/04


Date: Thu, 18 Mar 2004 06:28:02 +0100

William D. Tallman <wtallman@olypen.com> wrote:
> P.T. Breuer wrote:
> >[WDT]
> >> There are functions said not to be compiled directly into the kernel, and
> >> so not available as modules.
> >
> > This sentence makes no sense. If they are not compiled in the kernel
> > then they are in modules. There is nowhere else!
>
> ROFL!! Hello, Peter! Yep, makes no sense at all: the first 'not' was a
> typo.. <sigh>

Ahhhhhh. Maybe you mean "there is functionality which only becomes
available as parts of a monolithic kernel"? Don't know of any myself,
but of course, there are a lot of kernel options! Yes, there are still
compile choices which don't have "M" as an option, but fewer than in the
past (you try compiling SMP support as a module!).

> >> Obviously, I haven't read all the literature,
> >
> > There's nothing to read. What are you talking about?
>
> I'm talking about all the literature that is applicable to the new kernel.

I don't know of any, except Configure.help.

> Now, you do know that such literature exists, don't you... <grin>

Nope :(. I f you would point me to it, I have a few questions I would
like resolved, like "why has removable media support gone fubar"?

> > In what? There's no conceivable difference, and nobody knows what you
> > are talking about!
>
> No differences? Why a new minor number, then?

Because the new kernel is binary incompatible with the old. Old modules
don't work on new. New don't work on old.

> >> > It doesn't seem to me any less "straightforward".
> >> > The GUI interface to "make xconfig" is a lot nicer.
> >>
> >> Ummm... well, I don't really like GUIs all that much, so I can't
> >> comment.
> >
> > So use make config or make menuconfig or edit .config yourself.

> Well, actually I've gotten to the point with this kernel that I'm
> editing .config, just to make sure that I'm seeing all the choices. I've

That's fine (it's what I do), but it doesn't show you all choices.

> found that choices appear or disappear depending on other choices, and that
> bothers me.

But it's what you should expect! It's always been the case.

> In any case, some functionality formerly available as modules is now said to
> be compiled directly into the kernel.

I don't know of any. Rather, the other way around!

> I gather that this means there is no
> choice in the matter. Also, some stuff is deprecated in favor or new stuff
> (like asla instead of oss; acpi instead of apm; etc).

This is not so - it's simply that much alsa stuff has moved into the
kernel, mutating as it goes, displacing some oss stuff. The result is
somewhat gnarly.

Acpi has NOT replaced apm, though is much improved.

> As I'm very slow of
> wit, it'll take me some time to discover all this myself, so I thought I'd
> ask here first.

I think you are asking "what's new"? That question is answered in some
standard place on the web, but I forget where.

Peter



Relevant Pages

  • Re: Distributions
    ... | general there are a lot of packages for people to use. ... kernel, have to run on a multitude of different systems, they tend to be ... and slower than if you compile those packages, ... can have that stability with virtually any distro. ...
    (Debian-User)
  • Re: Distributions
    ... Ubuntu is based on Debian, ... | general there are a lot of packages for people to use. ... kernel, have to run on a multitude of different systems, they tend to be larger ... and slower than if you compile those packages, ...
    (Debian-User)
  • Problems with custom kernel fbsd 6.2
    ... My kernel for freebsd 6.2 is not compiling but only with my custom config. ... I was able to compile it with the GENERIC kernel included, but my modified one will not compile. ... # Power management support ...
    (freebsd-questions)
  • Re: [PATCH] 2.6 workaround for Athlon/Opteron prefetch errata
    ... >optimisations independently of each other', is fairly simple, (in ... >concept), and elegant, (as it lets you compile the most finely tuned ... >Up to now, selecting a CPU to compile for basically means, "Use ... it's silly to include them all in a kernel for a 386. ...
    (Linux-Kernel)
  • Re: C++ pushback
    ... There are all sorts of macros that use member initialization of that form. ... This does not break the code at run time, this breaks the code at compile time, and should be less painful. ... The kernel relies really _really_ heavily on such structure initializers, and breaking them would effectively break the world as far as the kernel is concerned. ... If they were not, one could simply make a base class having members outlined, and which class does not enforce type safety and is for inheritance only. ...
    (Linux-Kernel)