Re: kernel is always too big....
From: Dave Uhring (daveuhring_at_yahoo.com)
Date: 07/19/03
- Next message: Andreas Janssen: "Re: soundserver/ logout"
- Previous message: Michael Heiming: "Re: kernel is always too big...."
- In reply to: Peter T. Breuer: "Re: kernel is always too big...."
- Next in thread: ne...: "Re: kernel is always too big...."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sat, 19 Jul 2003 03:40:57 -0500
On Sat, 19 Jul 2003 09:15:38 +0200, Peter T. Breuer wrote:
> I meant I follow cvs. They fix bugs faster than I find any. It's too
> hard to locate a bug in a strange fs code even if you spot the symptom
> and get a repeatable test for it. Well .. almost.
Despite the barbs shot against XFS by that fellow from gentoo, I have
never had a problem with XFS. But I use Linux on my desktop PeeCee and
that probably obviates some of the problems of XFS. My servers run
Solaris and OpenBSD. In fact, the machine from which this article arises
is a Sun Ultra 1 running Solaris 9.
> Well, that was ONE bug. I recall two others, vaguely (link trick, and
> recursive modprobe). I'm not going to break sweat over it.
So your statement "?? Modutils has no interesting bugs. Never has had,
as far as I know." requires correction?
The utilities are probably stable at this point. But the use of -any-
otherwise unrequired utility carries risk with it.
> I'm not worried - if anyone finds a hole in modutils, it will be fixed
> within 10 minutes of publication. It's that basic! Why should I worry
> about it in that case? That would be like worrying in case anybody
> finds out that air is deadly to human life under some circumstances.
Air is indeed deadly to human life for some people - those who dive in
deep waters and return too swiftly to the surface.
> 3com have hysterically always supported linux and communicated with
> linux people. It's just that they change their chips like nobody's
> business and have their own bugs that they don't know about. It's all
> perfectly innocent.
Naming a NIC which uses a different architecture with the same
nomenclature as the previous generation is not innocent. It is mindful of
the behavior of LinkSys with their LNE-100TX series of 4 separate chips
and god_only_knows how many different chips sold by SMC.
Use supported hardware and forget about having to modprobe some
experimental module. You might enjoy fixing such things but most people
simply do not have that capabilty.
Do you think the OP now knows where to find bzImage?
- Next message: Andreas Janssen: "Re: soundserver/ logout"
- Previous message: Michael Heiming: "Re: kernel is always too big...."
- In reply to: Peter T. Breuer: "Re: kernel is always too big...."
- Next in thread: ne...: "Re: kernel is always too big...."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
