Slashdot Mirror


Linux 2.3.40 released

Roy Sigurd Karlsbakk writes "Linux kernel 2.3.40 has just been released. You can find a description on what's new in it on kernelnotes.org. It's available on one of the ftp.kernel.org and on most mirrors." Remember, this is a development kernel.

36 of 146 comments (clear)

  1. Buggy networking... by .pentai. · · Score: 2

    I'm not sure about previous relesaes...but I got 2.3.40 earlier this morning, and it seems to have a slightly major bug. Every 15 or so minutes it decides to hose my networking, and requires a reboot to get it going again...for another 15 minutes. Has anyone else noticed this?

    1. Re:Buggy networking... by Da+Penguin · · Score: 2
      Sometimes this has happened to me but I was able to just restart networking by:
      /etc/rc.d/init.d/network restart

      The Great AIP (Artificial Intelligence Project) has started!
      It is open source under the GPL and for Linux

    2. Re:Buggy networking... by ActionListener · · Score: 5

      Hmm... I think you probably clicked "y" for "emulate NT network reliability" in the Networking Options. The new kernel can also emulate AOL 5.0 networking behavior by choosing "n" for "TCP/IP networking."

    3. Re:Buggy networking... by bero-rh · · Score: 2

      Isn't happening here, and I've even used 2.3.40 in a high-load, high-preformance networking benchmark (comparison with Windows 2000 - we won for both static and dynamic web content) and it never crashed.
      I'd call 2.3.40 stable on x86 - it's not very reliable on alpha though.
      Maybe this is a problem with the driver for your network card. Which one are you using?

      --
      This message is provided under the terms outlined at http://www.bero.org/terms.html
    4. Re:Buggy networking... by dennisp · · Score: 2

      What dynamic content would that be? Vbscript ASP vs. mod_perl or jserv?

    5. Re:Buggy networking... by bero-rh · · Score: 2

      perl CGI vs. perl CGI (no mod_perl allowed).

      --
      This message is provided under the terms outlined at http://www.bero.org/terms.html
  2. Two things: by Greyfox · · Score: 2
    1) This is old news -- I've been running 2.3.40 for 3 or 4 days now.

    2) Some kernel.org sites don't seem to have 2.3.40 or even 2.3.39 (I ran across one the other day that only had kernels up to 2.3.28.) How do we know what the latest kernel REALLY is in these circumstances?

    --

    I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    1. Re:Two things: by FauxPasIII · · Score: 2

      Where do you get your kernels ?
      I've been checking twice a day, and just found 2.3.40 today. I check ftp.uk.kernel.org, ftp.us.kernel.org, ftp.sa.kernel.org, and ftp.cc.gatech.edu.

      --
      25% Funny, 25% Insightful, 25% Informative, 25% Troll
    2. Re:Two things: by terjegj · · Score: 4

      >How do we know what the latest kernel REALLY is in these circumstances?

      Try finger:

      $ finger @ftp.kernel.org
      [zeus.kernel.org]
      The latest stable version of the Linux kernel is: 2.2.14
      The latest beta version of the Linux kernel is: 2.3.40
      The latest prepatch (alpha) version *appears* to be: 2.3.41-1

  3. ES1371 working yet? by heroine · · Score: 2

    The soundblaster 128 was revised on Jan 4, 2000 such that either the PCM recording or playback won't work in any of the 3 sound drivers. Also the gain on the PCM output itself is set too high, resulting in clipping at all levels. No control on the mixer has any effect on this.

  4. Re:When is 2.4.0 comming? by bugg · · Score: 2
    Don't base progress on the frequency of releases. If that was your base, than NT4 is being developed at a fairly nice speed, because all of the service packs coming out (sigh)

    Did you read the changelog? There were mainly bug fixes and other "updates" with only a couple new features being added.

    This was just a minor patch, mainly making up for mistakes in the past.

    --
    -bugg
  5. Bet on the 2.4 release date here by CraigMcPherson · · Score: 2

    http://www.linuxnewbie.org/ ubb/Forum22/HTML/000159.html I had initially suggested that entrants ante their /etc/passwd, winner take all, but Sensei had a better idea: the person who guesses the correct date wins a Linuxnewbie t-shirt.

  6. Re:Open Source by QuMa · · Score: 2

    Can you say (or scream) redundant? I mean, yes, linux is nice. I run it too. But is that a reason for posting the devel versions on slashdot. Hell no!

  7. Actually, visit this thread instead by CraigMcPherson · · Score: 2

    http://www.linuxnewbie.org/ ubb/Forum22/HTML/000166.html

    That thread shows what dates are still available. Most of February and March are booked.

  8. Re:huh? by crow · · Score: 4

    When there was a story for every kernel release, it was, indeed, too much. Now /. only reports every once in a while on minor releases. In this case, I think it gives us a good forum to discuss how stable the 2.3 branch is getting and what we might expect to run into if we switch over.

    To be more specific, I'm considering playing with 2.3, and reading the responses to this story tells me that there are some networking problems that I would prefer to avoid and there might be some sound problems. Consequently, I'll hold off for a while longer.

  9. Closing in on 2.4? by _GNU_ · · Score: 2

    Shouldnt we get a freeze on this soon and move on?
    .40+ is starting to sound stupid :)

    Just installed 2.2.14 on my pIII and I have no need for anything more, but still......... have to try ;)

  10. Why the fsck do you post a devel kernel??? by Le+douanier · · Score: 4


    I thought there was a consensus that it was far better to let other sites post news about new kernel, especially for development kernels.

    Does this kernel have some exceptional new thing integrated beside his number (Oh yes, we have reached the 40th release of 2.3 WHAT A NEWS) to justify to be posted??? Is it the first kernel of a new serie (like 2.2.0 or 2.3.0?)??? Does it fix a big bug??? No, then don't post it.

    Those that are interested by cutting edge kernels know where to find them, those that are not able to find it by themselves probably aren't able to cope with the possible problems that may arise from a development kernel.

    (this is not meant as a flamebait but /. isn't the place for this kind of news, otherwise they may want to post the news that my cat is dead yesterday at 4PM due to an EVIL DVD player that felt upon him).

    --
    "The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
    1. Re:Why the fsck do you post a devel kernel??? by Trepidity · · Score: 2

      That's what I was wondering when I saw it. I don't remember the release of 2.3.32, or 2.3.38, or 2.3.39, or any other development kernel being posted as news here, so I figured there must be something special about 2.3.40. Apparently not.

  11. ALSA is broken too by heroine · · Score: 2

    Yes. ALSA produces the same clipping quite nicely.

  12. Re:SB Live in there yet? by lubricated · · Score: 2

    They fixed it. the emu10k1 (sblive) driver works with the newer experimental kernels again.

    --
    It has been statistically shown that helmets increase the risk of head injury.
  13. PCMCIA is now built-in in the 2.3 series by SurfsUp · · Score: 4

    For anybody running laptops, the big deal with the more recent 2.3 kernels is the built-in PCMCIA support. In the 2.2 series if you want to configure your kernel at all you have to get not only the kernel source but the (big) PCMCIA patch and compile it separately, and there's a somewhat intimidating series of questions you have to answer in the config. It's anything but seamless, and you have to mess around with it every time you upgrade the kernel. In 2.3 it's just one more checkbox item in make xconfig. :-) Progress marches on.

    --
    Life's a bitch but somebody's gotta do it.
  14. The Changelog by dew · · Score: 2
    Here is the Changelog. (Why didn't they link to it directly?)

    It doesn't look like anything really major; just a driver update here, a driver update there. Progress as usual...


    David E. Weekly (dew, Think)

    --

    David E. Weekly
    Code / Think / Teach / Learn
    h4x0r for

  15. Does 2.3.40 fix the bootable MD problem? by Jeffrey+Baker · · Score: 2
    2.3.39 won't boot a software RAID. It just says "Got md request. Not good." and stops there. Does anyone know if .40 fixes this? I see some changes to drivers/block/md.c but they don't look relevant.

    -jwb

    1. Re:Does 2.3.40 fix the bootable MD problem? by James+Manning · · Score: 2

      boot a software RAID meaning boot a kernel on a s/w raid partition? If so, it'll only work with the 0.90 RAID (not the 0.4x that ships in kernels, at least AFAIK) and it will only work with the lilo with RH 6.1 (unless you add a disk= section for md0 geometry).

      kernel.org/pub/linux/daemons/r aid/alpha/ has the 0.90 raidtools and patches for 2.2.11 (works on .11, .12, .13) and www.redhat.com/~mingo/ has patches for 2.2.14 (raid-2.2.14-B1) and 2.3.40 (ibc-ext2-raid-2.3.40-N1)

    2. Re:Does 2.3.40 fix the bootable MD problem? by Jeffrey+Baker · · Score: 2

      What i mean is using a RAID 0 for the root partition. This works fine with raidtool 0.40, kernel 2.2.14, and the stock lilo. However, it doesn't work with kernel 2.3.39. From what I gather on Linux kernel mailing list, this is due to md.c needing update to the new block driver interface. -jwb

  16. Hard drives by / · · Score: 2

    A friend is running a bunch of firewire hard drives on his mac -- he does a lot of video capture and is constantly filling up his drives. He just yanks one out and puts another in, and they're more than fast enough for his purposes. I'd mention the company names if I could remember them.

    --
    "If one is really a superior person, the fact is likely to leak out without too much assistance" -- John Andrew Holmes
  17. Re:Yet another kernel by Cid+Highwind · · Score: 4

    Thats's precisely why I love windows so much! None of that pesky kernel updating. It's nice to know that if a bug is here today, it will be here tomorrow, and next week, and next year. It's good to know that FreeBSD has taken that step in the right direction. Maybe it will be as good as windows someday.

    Since whan is lack of active development a feature?

    --
    0 1 - just my two bits
  18. Re:Yet another kernel by HoserHead · · Score: 3
    Perhaps some background on the situation, in case people aren't entirely sure what's going on:

    FreeBSD is a kernel /and/ a distribution of software. It has two general branches - Stable and Current (I think). Current is where the bleeding-edge changes go; people who know what they're doing and/or need features only in current track it. Otherwise, you use Stable.

    Linux is a kernel, only. There are many distributions based around it. It is developed in two branches; a stable and development branch. You can tell which branch a kernel is in by the minor version number (Linux kernels are numbered major.minor.patchlevel - Major is iterated very infrequently, after major changes; minor is iterated not so often, when going from development to stable; and patchlevel is the development on the minor version). If the minor version is odd - 1, 3, 5, etc - it is a development kernel, and if it is even - 2, 4, 6 - it is a stable kernel. Development kernels should not be used on production boxes, unless you're willing to support it yourself and deal with problems that may occur. Things can and will break in development kernels. Stable kernels should be used pretty much everywhere, particularly if you are new to Linux.

    Remember: You don't ever need to update a kernel unless it fixes something that's broken, adds support for something you have, or adds a feature you require. Otherwise, stick with what you've got - it's not a prerequisite to update when new versions are released.

  19. Re:Trying to convince investors /. is a Linux site by Le+douanier · · Score: 2


    "Linux hype is what attracts investors' billions these days, so it's important for everyone to pretend their business is "all about Linux"."

    This would be false to pretend that /. is "all about Linux" but /. is definitely biased in favor of posting a lot of Linux news and is frequented by a lot of Linux/Free Software fans, both making /. a Linux related site.

    --
    "The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
  20. It must be completely stable first by CentrX · · Score: 2

    Now this just doesn't make any sense to me. The development kernel must be as stable as it can possibly be made before it becomes a stable kernel. I take pride in that when a Linux kernel is released as stable, it really is stable. If the kernel were to be released before it were ready just so we could "move on" Linux would end up being just like any number of commercial products: released before it's fully ready.

    Chris Hagar

    --

    "The price of freedom is eternal vigilance." - Thomas Jefferson
  21. What's so special about 2.3.40? by RPoet · · Score: 2

    Minor development versions are released almost weekly... Is there anything special about 2.3.40, or will /. from now on announce every minor development release? Let's hope so, at least for the sake of consistancy and integrity (which would be a welcome new feature of slashdots ;).

    --
    "Oppression and harassment is a small price to pay to live in the land of the free." -- Montgomery Burns.
  22. I like occasional updates about the kernel! by EthanW · · Score: 3

    Not everybody follows every latest release of the kernel, but some people like to follow the general progress. When Slashdot posts a story about a new kernel version once in a while it helps people (like me!) keep up with the general progress of the project without being obsessed by it.

    As to complaints of "Go To Freshmeat!" my point is that some people don't want to rabidly follow every release of software, they just want a general feeling for the situation.

    After all, does your life end when Slashdot posts a story about something you already know or is not relvent to you personally?

    Maybe Slashdot needs a filter for kernel release posts, similar to the ones for filtering various authors.

  23. Visor USB syncs on Linux 2.3.40 by doomy · · Score: 5

    Hello Dudes,

    The hackers at Linux USB Visor has managed to get Visor to sync over USB in Linux 2.3.40. They managed to do this with a few extra patches (Which you can get from that site) and help from a freebsd dude. Freebsd dudes hop over there to find out how they did it as well..


    Enjoy syncing the Visor on USB!

    Long live the kernel!
    --

    --
    ...free your source and the rest would follow...
  24. Re:"debian's method" of keeping current by dennisp · · Score: 2

    FreeBSD also allows you to upgrade daily from current.freebsd.org builds without doing make buildworld and make installword after cvsup.

  25. Re:replacing ipchains??? for real? by mick2275 · · Score: 3

    Why:

    there is now no method of sending packets from kernel to user space. Transparent proxying isn't exactly "right". Masquerading is bolted on top of packet filtering, which is what makes building a firewall so complicated.

    The rest of the story:

    The new protocol (iptools (?)) is 'podabe backwards compatible with upchains and ipfwadm.

    --
    Can I bum a .sig off ya?
  26. Re:not just bugfixes by Andreas+Bombe · · Score: 2
    The only things I know of with a firewire interface is digital cameras and some storage solutions.. I guess there are no Linux drivers for the utilities, even if the kernel supports firewire..

    The 1394 subsystem includes a raw1394 driver and there is the libraw1394. This is enough to control digital cameras and to receive the data with a user space program without additional kernel drivers. This has already been done successfully (but there are no finished apps yet).

    Storage solutions usually use the SBP-2 protocol. This has to be done in the kernel, mainly because it's a service to the kernel in form of block devices. That is already worked on, but there is no working code yet.