Re: [Fwd: What do people think about not installing a stripped,/kernel ?]

From: Alexey Dokuchaev (danfe_at_regency.nsu.ru)
Date: 10/21/04

  • Next message: Robert Watson: "Re: [BETA7-panic] sodealloc(): so_count 1"
    Date: Thu, 21 Oct 2004 16:38:31 +0700
    To: Volker <volker@vwsoft.com>
    
    

    On Thu, Oct 21, 2004 at 07:28:18AM +0200, Volker wrote:
    >
    > <2ct>
    > well, that's the same as the M$ people thought ten years back. "Size &
    > price doesn't matter, so let's waste every Meg we could find."
    >
    > Doing it that way in every corner, you'll have a system which requires
    > plenty of Gigs to install in a few years. A debug kernel by default
    > would just be the beginning of a systematic waste.

    Actually, your expectations are not really what is this all about.
    Debug kernel really helps a lot, especially since there're still lots of
    bugs and regressions in something as bleeding edged as -CURRENT. There
    are users who simply ignore follow-ups that read "please rebuild your
    kernel so we get more information on the problem you're experiencing".
    Oh please, they just don't have time for it, nor desire. They install
    FreeBSD, see the panic, and want to report the problem ASAP, and move on
    doing their actual work, whatever they do for living. We cannot blame
    people for not having time to rebuild their kernels. Still, we're
    really interested in many good-quality bug-reports.

    >
    > Why do you love BSD? Because it's different? Well, I love my beasty
    > because it installs and runs in small to large size systems. And I love
    > it because it's fast. If you blow up everything, your beasty will get
    > slow, fat and ugly.

    I appreciate your magniloquence here, I do feel the same way often, but
    actually it's going in the opposite direction: with the release of 5.3,
    I hope that quite a few developers start devoting time to optimize and
    polish things up, not just bringing new features and functionality in.

    >
    > Personally I would not care about a debugging kernel on my disk but the
    > way poeple think (size doesn't matter, price doesn't matter) it's the
    > very first step into the direction of blowing up everything - because
    > size doesn't matter.

    You're not exactly right here. Particularly, take a look at what Dillon
    has to say about the whole thing:

      http://leaf.dragonflybsd.org/mailarchive/kernel/2004-10/msg00152.html

    ./danfe
    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"


  • Next message: Robert Watson: "Re: [BETA7-panic] sodealloc(): so_count 1"

    Relevant Pages

    • Re: linux USB question
      ... The only rules that really matter are these: What a man CAN do, and what a man CAN'T do! ... I know it>>>>> picks up USB and mem sticks. ... >>>>> has been resolved with the new kernel. ... I spent hours of reading trying to get my flash drive working>>>> and then out of frustration I formatted it vfat in linux and it's ...
      (alt.linux)
    • Re: [Fwd: What do people think about not installing a stripped,/kernel ?]
      ... Disk space is so damn cheap today.... ... > would just be the beginning of a systematic waste. ... > way poeple think (size doesn't matter, price doesn't matter) it's the ...
      (freebsd-current)
    • Re: RFC: [2.6 patch] disallow modular IPv6
      ... It would be difficult to have an opinion on the matter without agreeing ... > way to make a particular device driver appear. ... and then letting that force a change in the base kernel ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [SLE] Repository for kernel 2.6.17?
      ... not run on a non-SMP machine. ... Not a matter of different performance. ... always try the non-smp version of the kernel if there is any problem. ... think this is becoming less and less a troubleshooting step. ...
      (SuSE)
    • Re: [ck] Re: Linus 2.6.23-rc1 -- It does not matter whos code gets merged!
      ... It does not matter who's code gets merged. ... What matters is that the problem gets solved and that the Linux kernel ... And, from a standpoint of ONGOING, long-term innovation: ... There has to be a balance. ...
      (Linux-Kernel)