Slashdot Mirror


Improve Your GNU/Linux Experience With -mm Patches

An anonymous reader writes "Anyone interested in squeezing maximum performance out of their GNU/Linux operating system and willing to compile a new kernel will be interested in this KernelTrap article about Andrew Morton's -mm patchset. The patchset currently offers better stability and performance than the mainline 2.6.0-test kernel, as well as containing numerous functionality enhancements. Much of the additional functionality is described, as well as providing simple step by step installation instructions. Reading this article, one can certainly understand why Linux creator Linus Torvalds has chosen Andrew to soon become the 2.6 maintainer."

12 of 40 comments (clear)

  1. Re:Oh yes! by Kaladis+Nefarian · · Score: 2, Informative

    Sorry, the Con Kolivas link should be kernel.kolivas.org, but you can find the patches on the other one ;-)

    --
    * Several monkeys are here, playing banjos and wearing small hats.
  2. Re:synaptics touchpad support by Dave2+Wickham · · Score: 3, Informative
    Erm... Because it is?
    [dave@tc4 ~/linux-2.5.74]$ grep Synaptics drivers/input/mouse/Kconfig
    compatible. Support for Synaptics TouchPads is also included.
    For Synaptics TouchPad support in XFree86 you'll need this XFree86
    (standard 2.6.0-test5)
  3. Warning to XFS users by Kaladis+Nefarian · · Score: 5, Informative
    2.6.0-test5-mm3 apparently has a bad snapshot of the XFS code in it. Here's the post from linux-kernel:

    From: Steve Lord <lord at sgi dot com>
    To: Walt H <waltabbyh at comcast dot net>
    Cc: Linux Kernel <linux-kernel at vger dot kernel dot org>,
    Linux XFS Mailing List <linux-xfs at oss dot sgi dot com>

    On Sun, 2003-09-21 at 13:08, Walt H wrote:
    >> Just a follow-up to my earlier post:
    >>
    >> I've put in the xfs code from mm2 into the mm3 tree and all files get
    >> copied and I can manually copy the fstab.backup file afterward. I
    >> realized that the "rebuilding directory inode 256" was the lost+found
    >> directory, which contained 4 old zero length files. That was the key.
    >> XFS under -mm2 doesn't care about old lost+found directories, while -mm3
    >> does. If I removed the source lost+found/ and retried rsync's with -mm3,
    >> it finishes fine and I can copy fstab files. Adding a bogus lost+found
    >> dir with any file in it at the source, and retrying the rsync will lead
    >> to a state where I can't overwrite the existing /etc/fstab file at the
    >> end. So it doesn't look like there's actually any filesystem corruption,
    >> just a strange bug. Hope that helps,
    >>
    >> -Walt
    >>

    If I am correct, test5-mm3 contains a bad version of the xfs code, there
    was a bug where the i_flags field was setup from an uninitialized stack
    variable. mm3 came out during the two days this was in Linus's tree.
    I had some very odd behavior with this code base, rm -r -f would try and
    cd into files and other bizzare things, files could appear to be
    immutable or append only or things they were not. This sounds like
    similar behavior you that you saw. It is fixed in the latest code Linus
    has.

    Steve
    So XFS users should probably go with -mm2 for now.
    --
    * Several monkeys are here, playing banjos and wearing small hats.
    1. Re:Warning to XFS users by rf0 · · Score: 2, Informative

      mm4 has a fix as I've just noticed from the Changelog

      Rus

  4. Re:Tiresome by Anonymous Coward · · Score: 1, Informative

    Development. Series.

    If you want stability you run 2.4, which is the stable kernel.

  5. Re:synaptics touchpad support by Dave2+Wickham · · Score: 2, Informative

    Oh yeah, and it's bad to reply to myself, but the touchpad essentially acts as a PS/2 mouse to anything which doesn't have special drivers.

  6. Re:synaptics touchpad support by cpeterso · · Score: 2, Informative


    I was able to use my Dell laptop's touchpad with Linux 2.6.0-test4 without using the -mm tree or applying the new Synaptics driver from XFree86 people. Just add the following kernel boot parameter: psmouse_noext=1

  7. Re:What we need by Anonymous Coward · · Score: 1, Informative

    > actually, except for the address, i think it would be a very good thing to have such easy updates available. Redhat does this, but only on a trial basis, if you want the full thing, you have to pay.

    The "trial basis" is in name only; you "pay" for a renewal of a demo subscription (yeah, I know, dumb name) by filling out an innocuous survey every few months - they email you when you need to do one to keep the sub active. It's questions like "what Red Hat documentation do you consider strongly recommended" or "how did you hear about Red Hat?". A demo account doesn't have the same priority at the Red Hat servers as a paid account, so if there's a frenzy of updating clients, demo accounts will often have to wait a few days, or at least until off-peak hours.

  8. Re:Hmm by Anonymous Coward · · Score: 3, Informative

    > Hmm.. I thought Linux was open source and anyone could damn well do what they please with it. What's stopping me from becoming the Linux 2.6 kernel maintainer?

    Nothing at all is stopping you from becoming a 2.6 kernel maintainer, in the same way that Alan Cox, Dave Jones et al maintain well-respected 'unofficial', but important, variations on the 2.4 tree. You set up a 2.6 tree, incorporate patches that strike you as being particularly useful, and who knows? - others may feel the same, and your tree (and you!) get to the same point as a dj or mm tree.

    No, yours won't be the 'official' (ie, the reference) 2.6 tree - just because you set up a 2.6 tree doesn't obligate Linus or his designated maintainer from accepting your patches - but neither can they/do they/would they prevent others from using your tree if they prove useful, and if your particular patch set does prove widely useful, you've got a good shot at getting them included in the main reference tree.

    And, the more variations on a given kernel tree, the better. They're excellent labs for trying out things that the reference tree maintainers may not want to try just yet, and give good alternative kernel sources for those not up to such an ongoing project. So yes, I'd recommend you do exactly as you seem to be suggesting, and set up and maintain a 2.6 tree with whatever patch set you feel worthwhile.

  9. Re:Better than 2.6?? by gears5665 · · Score: 1, Informative

    My guess is you've never worked on a distributed coding project with over 500 active developers who you've never met face to face. :-P

    Its very hard to keep track of all of the things people want you to add to the kernel and make certain that the things you're adding aren't breaking anything, etc. So the linux kernel development process has a number of branches where things get tested first(in addition to the stable branches of the 2.0,2.2,2.4 kernels that need to be maintained). Its been going like this for years. Alan Cox has a branch(ac), Morton's branch is mm.

  10. Re:Oh yes! by enodev · · Score: 2, Informative

    The interactivity tweaks have just gone mainline: http://linus.bkbits.net:8080/linux-2.5/cset@1.1320 .14.15?nav=index.html|ChangeSet@-1d

  11. Re:My experience with kernel patches by Anonymous Coward · · Score: 1, Informative

    The latest gentoo-sources patch for 2.4.20 can be downloaded here. Warning - really huge directory listing if you browse all the files.)
    Addon patches that update this one can be downloaded here.

    1. Decide if you want the aavm or rmap VM. If you want aa, delete the files ending in .rmap (and vice versa)
    2. I followed what the ebuild script does and deleteted the patches starting with 1, 6 and 8.
    3. Apply the addon patches in the order that they're listed in the ebuild script.
    4. Run your normal make menuconfig/xconfig and turn on the desired options like Preemptible Kernel and Low latency scheduling. I have Timer frequency (HZ) set to 200 since I only have a PII 450. Some people set it to 500 or 1000 for faster CPUs. I also use supermount and grsecurity.

    Some people like the gaming-sources for pure speed but fewer features. The pfeifer-sources are the most bleeding edge patches that eventually become gentoo-sources after testing. Read about the different patches here.