Posted by
timothy
on from the jamie-has-the-munchies dept.
punkmac writes "It's that time again to do the thing we all love to do, compile your shiny new kernel. This time its 2.4.20. The changelog can be found at the usual places, and please use a mirror for all your downloads"
Be kind to kernel.org
by
Anonymous Coward
·
· Score: 5, Informative
Don't just use a mirror, get the patch to update your 2.4.19 tree.
Re:how long will it be...
by
silvaran
·
· Score: 5, Informative
It's not a matter of "Linux" having service packs. I'm assuming you use a specific distribution. You have to trust that distribution to release updates for your system as necessary. You can't binary patch the Linux kernel, because no binaries are released, only source. Red Hat, for example, has binary releases of the kernel (included in their distribution), but doesn't bother releasing service packs, instead they release updates as they see fit (pretty much the same thing).
I don't quite understand the obsession with using the latest and greatest. I'm not going to grab this and compile it for my machine; a) 2.4.19 works fine and b) I skimmed through the changelog and didn't really see anything that affected my situation, and c) My roommate will scream at me if I take down the server just to update the kernel by a double-point revision.
You mentioned when you update the kernel/compiler it breaks a lot of applications. You'll have to be more specific. I can see problems when migrating from a 2.2 kernel to a 2.4 kernel, or from gcc 2.x to gcc 3.x, but if you use one of the major distributions, this is trivial. The thought of using "service packs" on an open source operating system indicates to me that you don't have a clear handle on the way things work in Linux.
small VM updates...
by
kinko
·
· Score: 5, Informative
From the changelog....
<hch@lst.de>:
o dump_stack()
o backport yield() and conditional reschedule changes from
o small VM updates from -aa (1/5)
o small VM updates from -aa (2/5)
o small VM updates from -aa (4/5)
o small VM updates from -aa (5/5)
Is this using a Linus definition of small, or a normal definition of small?
Re:Important changes?
by
updog
·
· Score: 5, Informative
There was at least one very important change for me personally - support for KT-400 motherboards (by adding support for VIA8235 southbridge). I had gotten 2.4.20-rc3 for this support when I got my new motherboard, but now it's official.
Incedentally, 2.4.19 will sorta work on a KT-400 board, but disk access is really slow because it can't turn on DMA.
Re:Did they fix the new ptrace vulnerability?
by
Juergen+Kreileder
·
· Score: 4, Informative
I haven't tried it myself yet, but I found no reference to this ptrace vulnerability [google.com] in the changelog. I suspect this is still a problem (it was in 2.4.19).
It was fixed in 2.4.20-rc2, see the "[PATCH] Fix lcall DoS" entry in the ChangeLog or this bk comment (and the corresponding patch).
Re:I'll wait for 2.4.20-ac1 or -ac2. :)
by
darketernal
·
· Score: 5, Informative
You do realize 2.4.20-rc4 was released as 2.4.20 without any changes. Hence, 2.4.20-rc4-ac1 is the -ac1 that you are looking for:)
Re:Apparently
by
mysticalreaper
·
· Score: 5, Informative
Uhm... are you SURE? You sure it's not the new 2.5 development kernel you're talking about? Cause i think if you looked into it, you'd realize that all these massive changes you mentioned, improving threading, I/O, VM and the rest are changes happening in the 2.5 kernel. They are major changes, and very significant. But they're happening in 2.5.
This release is just an incremental release. Bugfixes here, added support there... no major changes, just evolutionary changes.
Good to see the mods are looking for informative comments... however, it's sad when they mod up comments that don't reflect the facts.
Re:Important changes?
by
thesadmac
·
· Score: 4, Informative
I beleive the reiserfs write_file function was added to the reiserfs filesystem. This speeds up writing large files on a reiserfs formatted volume. Which on my boxes is every partition except/boot and swap.
Tips on testing your new kernel
by
goingware
·
· Score: 5, Informative
You should thoroughly test your new kernel before putting it into production. Even if the kernel works well for everyone else, you could personally discover a new bug. You could save yourself a lot of pain by testing first, as well as help the kernel developers.
A while back I wrote a couple articles on Linux kernel testing:
The Open Source Development Lab kindly provides Japanese translations
here.
(The articles are under the GNU Free Documentation License. I would be quite stoked if you copied or translated them. There are articles on other quality topics
here.)
Slashdot Mirror
Don't just use a mirror, get the patch to update your 2.4.19 tree.
It's not a matter of "Linux" having service packs. I'm assuming you use a specific distribution. You have to trust that distribution to release updates for your system as necessary. You can't binary patch the Linux kernel, because no binaries are released, only source. Red Hat, for example, has binary releases of the kernel (included in their distribution), but doesn't bother releasing service packs, instead they release updates as they see fit (pretty much the same thing).
I don't quite understand the obsession with using the latest and greatest. I'm not going to grab this and compile it for my machine; a) 2.4.19 works fine and b) I skimmed through the changelog and didn't really see anything that affected my situation, and c) My roommate will scream at me if I take down the server just to update the kernel by a double-point revision.
You mentioned when you update the kernel/compiler it breaks a lot of applications. You'll have to be more specific. I can see problems when migrating from a 2.2 kernel to a 2.4 kernel, or from gcc 2.x to gcc 3.x, but if you use one of the major distributions, this is trivial. The thought of using "service packs" on an open source operating system indicates to me that you don't have a clear handle on the way things work in Linux.
From the changelog....
<hch@lst.de>:
o dump_stack()
o backport yield() and conditional reschedule changes from
o small VM updates from -aa (1/5)
o small VM updates from -aa (2/5)
o small VM updates from -aa (4/5)
o small VM updates from -aa (5/5)
Is this using a Linus definition of small, or a normal definition of small?
There was at least one very important change for me personally - support for KT-400 motherboards (by adding support for VIA8235 southbridge). I had gotten 2.4.20-rc3 for this support when I got my new motherboard, but now it's official.
Incedentally, 2.4.19 will sorta work on a KT-400 board, but disk access is really slow because it can't turn on DMA.
You do realize 2.4.20-rc4 was released as 2.4.20 without any changes. Hence, 2.4.20-rc4-ac1 is the -ac1 that you are looking for :)
Uhm... are you SURE? You sure it's not the new 2.5 development kernel you're talking about? Cause i think if you looked into it, you'd realize that all these massive changes you mentioned, improving threading, I/O, VM and the rest are changes happening in the 2.5 kernel. They are major changes, and very significant. But they're happening in 2.5.
This release is just an incremental release. Bugfixes here, added support there... no major changes, just evolutionary changes.
Good to see the mods are looking for informative comments... however, it's sad when they mod up comments that don't reflect the facts.
I beleive the reiserfs write_file function was added to the reiserfs filesystem. This speeds up writing large files on a reiserfs formatted volume. Which on my boxes is every partition except /boot and swap.
A while back I wrote a couple articles on Linux kernel testing:
-
Why We Should All Test the New Linux Kernel
-
Using Test Suites to Validate the Linux Kernel
The Open Source Development Lab kindly provides Japanese translations here.(The articles are under the GNU Free Documentation License. I would be quite stoked if you copied or translated them. There are articles on other quality topics here.)
-- Could you use my software consulting serv