Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Linux Software

Linux 2.2.4 66

Posted by Hemos from the download-compile dept.
Phaid writes "As if you weren't already being flooded about this, linux 2.2.4 is out. Here's the official Announcement from The Man Himself. There's a small problem with BSD Process Accounting that can be easily fixed with the patch. " You know the routine-go get'em.
This discussion has been archived. No new comments can be posted.

Linux 2.2.4 More

Linux 2.2.4

Comments Filter:
  • Honestly, how long until 2.2.x is done? I wasn't really around for the 2.0.x series... So I am not sure how this is done.
  • 2.2.x will never be 'done' until 2.4.x or 3.0.x comes out - development is a continuous process. Other than the (somewhat embarassing) BSD process accounting compile bug, 2.2.x is extremely stable, much more so than early 2.0.x kernels were.

    Even if all bugs were removed from the kernel, there would still be new releases to add features and drivers. Frequent updates are what keep Linux on the cutting edge.
  • I'm pondering 2.2.4..... 2+2=4, and 2*2=4. Methinks this kernel will have much karma.

    BTW, when are they starting the 2.3.x tree?

  • The 2.0.x kernel is still being fixed. i think 2.0.38 is being worked towards, with alan cox (the maintainer) looking at one more release (2.0.39) before putting it to bed (except for security issues).

    i'm sure some people still use the 1.2.x series, and i bet they patch it on their own. software development is never done. at best it pauses.
  • Since I switched to kernel 2.2.x, I've had two problems:

    - I can't nfs mount my home and mail directories onto my other machines unless I use "nolock"

    - I can't send email using netscape on those other machines (which I suspect might have something to do with the NFS problem). The machine locks up pretty tight, and sometimes has to be rebooted when this happens.

    Does anybody know if 2.2.4 fixes those problems?
    (Hey, give me a break - my server machine is dog-slow and I don't want to spend 5 hours compiling a kernel if it doesn't fix anything).
  • Hi,
    I had some problem mounting the NFS.
    If everything (as in my case) is relying on NFS, it couses all the accounts to freeze.
    My problem was that everything was exported as read-only
    The solution I found is
    on the /etc/exports add the parameters:
    (rw,no_all_squash)

    I hope this helps.

    BTW, where can I find some info on the NFS development status ("minor changes" is not enought)?
    Massimo
  • s/38/37/; s/39/38/;

    Otherwise, yeah... I've still got two boxen that run 1.2.13, because they don't have the resources to really cope with 2.0... they're both low-end 386es with 4M of RAM and hard drive sizes in the double digits.

    Old software never dies...

  • Sparc merge (Score:1)

    by Anonymous Coward
    Thanks to the work of davem, alan and linus, this is the first stock kernel since the 2.1 series to compile on my Sparc. Thanks guys!
  • Aren't you people tired of recompiling a new kernel every week?
  • "2, 2, 4. And noodles" Decker
  • New kernel, whoohoo,
    Let's dance, sing, and parade.

    Why should Slashdot announce kernel updates?
    If anything this is an item for linuxtoday, or maybe the quicky batches.

    Anyone except for me thinks this is redundant?



    ---
  • My dim memory doesn't release connection,
    Where does Decker say that?
  • not everyone who might be interested in kernel updates reads linux today.

    It's a lot more interesting to me than hearing about the latest start wars or whatever.

  • those guys are quite quick: get the newest kernel-debs on netgod.net!
  • I've had some confusion getting the printer to print and the zip drive to zip, although I thought I followed the directions pretty carefully (rmmod lp, insmod ppa, insmod lp). The zip drive works but I can't print.

    If anyone else chains their printer onto their zip drive's pass-through port, I'd greatly appreciate any bonehead tips ("one at a time, bozo!") or other advice. I'd like to be able to print and use the zip drive without constantly shuffling cables, and the documentation indicates that I should be able to do so... but I can't so far.

    Is it just a matter of editing conf.modules?
    (I wasn't sure whether the IRQ and io arguments in parport.txt were generic or "[your_IRQ]...")

  • What other forum can turn the release of some new source code into an intense debate? Someone's usually got an angle on it that's interesting...
  • You don't have to -- as the saying goes...

    if it ain't broke don't fix it.

    Basically, if the new kernel adds little or nothing that you need, then don't bother with it.
    Personally, I'm waiting for reiserfs to make it into the kernel -- its stable, and I have been running it with 2.2.2acSomething for some time now (sine 2.2.2 came out)

    p.s. anybody know how one persuades knfsd to do anything?
  • I don't know about menuconfig, but if you copy your old .config file into the new kernel directory, and do a "make oldconfig", it'll load most options out of the old .config. New options it'll stop and ask you about.
  • copy the old config file to /usr/src/linux/.config (replace /usr/src/linux with your linux source directory if different) and use make oldconfig
  • I'll have to assume that aslot of slashdot reader do read linuxtoday,
    or the quicky batches.
    If they're not reading it, they can make it appear in their little slashbox.

    Regarding the starwars, I just turn them off.
    It would be a whole lot better to have a kernel subject (with a corn kernel picture),
    or sofware updates subject,
    and i would turn it off happily.


    ---
  • Hmm... I know someone with that exact card who used 2.2.2 with no problems. Actually... I'm not sure if he ever used 2.2.2, but I know he used 2.2.1 and 2.2.3 without difficulty.
  • Well, depends on your definition of "done". Alan is currently working on 2.0.37 and plans on having 2.0.38 out the middle of the year, and that will be it for 2.0; expect similar for 2.2 after 2.4 comes out.
    2.4 is due before the beginning of next year; Linus wants this one to be short.
    Finally, I have no clue when the 2.3 series starts...my guess would be pretty soon, since Linus now has 2.2 all sync'd up and hopefully the nasty bugs fixed, it's time to go wild again. When 2.3 comes out, 2.2 will probably be considered done in the sense of stable.

  • Big Deal (Score:1)

    by jnik ( 1733 )
    I like having these announced; they don't come up on Freshmeat that fast. Plus, there's usually a lot of discussion and I like to see what problems people are having before I recompile (especially true when 2.3 starts).
  • > Aren't you people tired of recompiling a new kernel every week?

    I don't. And I don't imagine that everyone else does. I like to stay reasonably up to date, but I'm running 2.2.1 on my home system. Just because a new kernel is released doesn't mean that everyone *MUST* install it. I probably will install 2.2.4 at some point soon, when I'm done tweaking some other stuff on my system. But I'm in no hurry, 2.2.1 is working just great for me.

    Eric
  • Everyone seems to be glossing over the main reason for kernel recompiles: it's just plain fun. Something about messing with the brain of yer OS is cool. Not to mention, you can brag to your friends about have a shiny new kernel.

  • The ole chinese man doesn't quite get English, repeating "two", so Decker shows with fingers.
  • I think it's referring to the scene when Deckard (not "Decker"!) is at the noodle stall and orders four bowls. The old man insists in Japanese that two is enough.

  • Error and fix (Score:4)

    by korpiq ( 8532 ) <-.@korpiq.ik i . fi> on Wednesday March 24, 1999 @07:08AM (#1965649) Homepage
    The next mail in that thread said following, and guess what? At least I had to apply that patch to get it to compile. Assume a 2.2.4-oops soon as The Man returns from vacation.


    Date: Tue, 23 Mar 1999 19:35:04 -0500
    From: Arvind Sankar

    acct.c: In function `sys_acct':
    acct.c:197: too few arguments to function `filp_close'
    acct.c:203: too few arguments to function `filp_close'

    Should the second parameter be NULL? (guessing wildly after looking at binfmt*)

    Yes, here is the fix

    --- linux/kernel/acct.c.~1~ Tue Mar 23 16:14:01 1999
    +++ linux/kernel/acct.c Tue Mar 23 16:34:55 1999
    @@ -194,13 +194,13 @@
    }
    if (old_acct) {
    do_acct_process(0,old_acct);
    - filp_close(old_acct);
    + filp_close(old_acct, NULL);
    }
    out:
    unlock_kernel();
    return error;
    out_err:
    - filp_close(file);
    + filp_close(file, NULL);
    goto out;
    }

  • I have a similar card... there was a problem with one of the kernel releases, but it was fixed. I wish I could tell you which one exactly, but I have been getting all of the patches in the 2.2.x series, so I dont recall ;)

    I am running 2.2.3 and it works...

  • not necessary. 224-2^(2+4)=2*2^4+2*2^(2+4) cancels this.
  • some tulips want newer tulip.c from Beckers site. file in kernel is rather old.

    exact url is at the top of that tulip.c, grab the "newer experimental driver with more features".
  • 2.2.2 came out on 2/22

    That was like, cool or something.
  • 2.2.2 came out on 22/2/1999. Sounds like another Pentium floating-point bug. ;-)
  • Uhhh...read clue-point-two...

    Linux's NFS client implementation (allegedly) works now. One thing that does work is that the client actually uses an NFS lock server, whereas 2.0 didn't. So if you aren't running a lockd on your NFS server, you need the nolock mount option.

  • Huh? I have it automated. If my test system ever goes down, I get whatever kernel was most recently released when it reboots. Since it generally only crashes due to kernel bugs, this is a pretty fun way to keep up.

    One system crashed on 2.1.128, it rebooted in 2.1.131, and is still running 2.1.131...
  • Some of us are _paid_ to test out these kernels.

    We have customers who don't care about kernel releases as long as they work, so we keep track of which ones do work (with specific test cases for each customer).
  • My 2.2.3 kernel configuration is saved in a file called /root/kernelconfig. If I upgrade the source to 2.2.4, can I load my 2.2.3 config in "make menuconfig", or do I have to go thru and set everything all over again? How are new/updated options handled?

    In general it's a good idea to browse the make xconfig menu and make sure all your options are still the way you want them. Basically, any options in your .config that are no longer supported are ignored, whilst new options that you don't specify use the default value. Still, the only time I've had to modify my old config when moving from 2.2.x to 2.2.(x+1) was to disable an option that broke between releases.

    The short answer for 2.2.3->2.2.4 is yes, you can use the 2.2.3 config file, as long as you first patch kernel/acct.c if you enabled BSD process accounting (if you don't enable BSD process accounting you don't need the patch). I had no problems whatsoever doing that.
  • I am running rpc.mountd and rpc.nfsd from the nfs-server-2.2beta37-1 rpm. Doesn't that do the locking? I definitely compiled in locking support in the kernel on the server.
  • Probably better sent to a newsgroup appropriate to the subject, but what the heck....

    With the 2.2.x and 2.1.x kernel, there is no need to separate the printer and Zip drivers into separate modules. Compile them right in there, as part of the kernel. They work fine together. Of coure, when I did it (before my networked printer. ;) ), I was using /dev/lpX, not /dev/parX. They have the same major/minor, though. You might try using /dev/lpX instead, and see if that works.

    If you're still in the dark ag..I mean, using 2.0.x, then yes, you will need them as separate modules.
  • I'll agree with that... I found there to be quite a few improvements from 2.0.35 to this one... but I also foound it to be quite unstable, even with upgrading to the latest versions of all packages. (X freezes completely, and a STOP-A boot is the only way of getting the machine back). I switched back to 2.0.35 for now... (don't really have the time to deal with it for now).
  • Nope, lockd does the locking.

    spoke:~> /etc/rc.d/init.d/nfs status
    rpc.kmountd (pid 125) is running...
    nfsd (pid 135 134) is running...
    lockd (pid 106) is running...

    No problems here.
    Grab HJLs latest knfsd package (currently 1.2), compile it and install it, and replace your current /etc/init.d/nfs and /etc/init.d/nfsfs with the rc.nfsd and rc.nfsfs in the knfsd-1.2 package.

    Grab the latest knfsd package at:
    ftp://ftp.varesearch.com/pub/support/hjl/knfsd/

    -Dave
  • Nah, I'm running 2.2.4, it's just that I have a local printer passed-through my zip drive on the same port, and the printer isn't working. It's not a big deal I guess, I can just print and backup at different times.

    Thanks!

    ps. 2.2.4 is noticeably snappier for me, and I haven't even re-upgraded my pgcc-ized version of XFree86 since I switched to Debian... I should do that too. Hell, right now I'm on the phone with Oracle to see whether they'll support our server if I put 2.2.4 on there too.

  • I just posted this under the "Technical Question?" thread. Here's the recipe:
    1. Get HJL's knfsd-1.2 package from ftp://ftp.varesearch.com/pub/support/hjl/knfsd/
    1a. Untar it.
    2. Compile it with `make`
    3. Install it with `make install`
    4. Replace /etc/init.d/nfs and /etc/init.d/nfs with the etc/rc.nfsd and etc/rc.nfsfs in the knfsd package.
    5. Make sure portmap is setup to start BEFORE nfs starts.
    6. Start nfs `/etc/init.d/nfs start`
    7. Check to make sure it all worked `/etc/init.d/nfs status`
  • There's two things you can try. First of all, try using the DEC driver, (leave both the DEC and Tulip drivers enabled). It's the CONFIG_DE4X5 option in your .config, also right above the Tulip driver in the menuconfig. If that doesn't detect your card either, grab the latest driver from Donald Beckers site and replace tulip.c with it. Someone else posted the URL to Donald's site.

    -Dave
  • So, does anyone know the full story on the PNP Soundblaster 16? Mine only works for 8-bit sounds, but won't make a noise at all with a 16-bit sound.

    I've read all the FAQs and everything else, but it seems that there is no solution to the problem. I upgraded to the Linux 2.2.3 kernel because I heard a rumor that it knew about my strange sound card, but that was false. 2.2.4 doesn't look like it has anything new for my soundcard.

    Will I ever be able to use my card, or should I just go out and get a new one?
  • Have you compiled sound card support into the kernel? How about trying the linuxpnp stuff? That brought my card to life, although I still need to get some tools running under Enlightenment for it to be fully running. Check out PNP Tools [demon.co.uk]. I added them to my startup scripts, and the card is alive now. Now I just need a volume control... (ouch!) I'm still using 2.0.36, so unless 2.2.x has the PNP init stuff built in this ought to work.
  • Can anyone else reproduce the crash that occurs when you buffer a movie larger than your VM in xanim?

  • I agree with jnik -- Kernel updates should be announced on slashdot.

    It's not the news itself that's important -- it's the comments.

  • If slashdot stopped announcing kernel updates, I (and many others, I'm sure) would have to start reading yet another web page regularly to find out 1) when kernels are released and 2) what problems the new releases have.

    One can always turn off the "Linux" topic in the /. user preferences.
  • I encountered the same error, and was wildly
    bewondered. That's because so many people seem
    to have no problems at all here.

    How the *** can that be?


    BTW: I'm installing NT now in a vmware-box.
    But compiling the 2.2.4 now makes that install a
    little bit slow ;-)

    Cheers,

    Jeroen
  • [root@fuse linux]# cat .config | grep BSD
    # CONFIG_BSD_PROCESS_ACCT is not set
    # CONFIG_BSD_DISKLABEL is not set
    [root@fuse linux]#

    That'll be why 2.2.4 compile on mine - I'm not using BSD process accounting.

    James
  • Yes.

    With the bugfix, i could compile 2.2.4 but not
    do make bzImage ...
    So now I'm shutting of BSD (thanks) and trying again....

    Jeroen

  • BTW: I'm installing NT now in a vmware-box.
    But compiling the 2.2.4 now makes that install a
    little bit slow ;-)

    I want that software! You claim it really executes software in different OS's in parallel? My gosh!

    Any problems? How does it work? How does it hook the OS's to do the swap dance? Can you play mp3's without breaks?
  • Yes it is...

    Really cool. Runs matlab nearly as fast as under nt..

    This morning, i'm gonna install ms-office...

    Jeroen

Slashdot Top Deals

Nothing ever becomes real till it is experienced -- even a proverb is no proverb to you till your life has illustrated it. -- John Keats

Close