Re: Compiling a kernel
* Jameson C. Burt (email@example.com) wrote:
> You have gotten a couple DIFFERENT approaches to installing a kernel on
> At least one comment should send up a warning:
> if the approach becomes too intricate, or requires specialization,
> very likely you read minute details that you should never need learn
> (unless you're creating Debian packages).
Yes, I am trying to avoid extremely involved approaches just so I
don't do things I won't know how to undo later. I always try to
understand just how I can backtrack after some action in order to
avoid real lasting irreparable damage. Eventually that was what
became of my foray into Gentoo, or at least partly, and I would hope
to avoid doing that again.
> It's surprising how many new options get added with each new kernel,
> so "make oldconfig" can take 30 minutes, or even 2 hours if you go from a running
> 2.4.* kernel to a 2.6.18 kernel (then, you need to rethink all this).
It has certainly surprised me, again and again. If I go any period at
all with a kernel, by the time I go back to recompile a new one I am
mired in reading trying to understand what all of these new options
do. This time I had a string of errors and then a kernel panic in the
process of hammering out what some of it out.
> I recently installed (largely the Debian way) kernel 184.108.40.206,
> which isn't available in the current Debian testing version (etch)
> and anything beyond 2.6.16 probably never will be in etch
> since etch is being stabilized to become "stable" in the next few
My install of etch has 2.6.17, so they must be moving pretty fast.
But, like you, I compiled the 220.127.116.11 today and it is running fine at this
time. Thanks to all the help I received from those such as yourself
here at this list.
> Each time, I sequentially change my name with a sequence number from
> jameson2 to jameson3, jameson4, jameson5, ...,
> so my modules go into a different directory
> (modules get installed into their own directories like /lib/18.104.22.168-jameson2-oldconfig/)
> This lets lets grub run my old kernels with their modules in their own
> /lib/* directory.
> Each such module directory uses about 50MB.
Many thanks for this tip. Since my new kernel was different from the
previous, I upgraded from 2.6.17 to 2.6.18, I did not need to worry
about that. But, it will be very handy when I need to compile again
and don't want to clobber all of my modules. This is the second very
useful tip in that area I have read on this list.
> On compiling with --initrd, I finally drank the coolade last year.
> Before, I tried to have no modules, compiling needed modules into the
> kernel itself.
> Since almost everyone now compiles oodles of modules, mostly uneeded,
> but unknown uneeded, instructions now usually talk about modules.
> As others in this thread have mentioned,
> when booting, some necessary modules must be accessed right away,
> so you are better off compiling with --initrd.
Or you could use my method mentioned above, and just wait for kernel
panics. ;-) I actually did manage to have a fairly modular kernel with
those absolute necessities builtin. And because the old kernel and
modules are still on board if something fatal does go wrong at least I
can go back and begin again. Or, I hope I can.
> You mentioned NVidia.
> I presume you have an Nvidia card with 3D acceleration.
> Without using NVidia's binary driver, you will find XWindows using the
> open-source "nv" WWindows driver,
> and being slow even though you don't run games.
> To install NVidia's driver, you either most use an older kernel
> corresponding to Debian's packages that help with NVidia's binaries,
> or you must run an NVidia executable.
> I have tried both.
> Since I use kernels newer than Debian has in testing (etch),
> I go to nvidia.com, get NVidia's driver, then run it.
I actually did this with the Debian kernel. It is always what I did
with Slack and so I am used to it. I had installed Ubuntu a couple of
weeks ago, (no offense to the Ubuntu folks but I wasn't crazy about
it), and I worked the better part of two days trying to get my nvidia
card working. I read page after page of howtos and fora or list
messages, and none agreed on how to do it, except that you would hose
your system if you used the drivers from nvidia. Well, I hosed the
system doing it the "ubuntu" way, because just as I try to avoid I
ended up doing things I didn't understand and so could not undo. One
error stacked up on another and pretty soon it was bye-bye Ubuntu.
With this install of Debian I decided to stick to what I know, and
grabbed the binary installer direct from NVidia. I ran that, and in
less than two minutes I was up and running. No complaints from Debian
and no complaints from NVidia. Oh, except for one because I had tried
to compile it before I updated the system fully and so I had a gcc
which didn't match the kernel, but that was fixed in no time.
> While getting an executable from nvidia.com directly worked well,
> I noticed that it must have overwitten a few Debian installed files.
> So when I upgrade Debian, I am prepared (but haven't seen)
> for possible graphics oddities, in which case I would rerun nvidia.com's
> executable to re-ovewrite those files.
Exactly what I have planned to do in the case there is a problem.
But, so far, there hasn't been one. And perhaps that will remain the
> You have heard from many of us, some complementing others information,
> some giving non-Debian approaches, and you don't really know who to
> listen to.
> YOU NEED TO GO TO A DEBIAN DOCUMENT (which has more thought even than
> these email-lists); eg,
> "Compiling a New Kernel" at
> which corresponds to
> www.debian.org --> "installation manual"
Many thanks for the info and the links. I will go and see what they
say, for future reference. I am sure that there are immense loads of
info out there, and I have tried to glean what I can before troubling
the list. Though, I do have to compliment this list in general for
being one of the most generous I have ever had the good fortune to
interact with. The reputation which preceded Debian users was that of
an arrogant lot with keyboard shortcuts to type RTFM with one stroke.
But, this has been the opposite of my experience reading and posting
here. If this list is any indication of the Debian user community in
general, it is certainly one of the finest.
Many thanks again for all the help, and I will be sure to read the
info at the links you included.